THE IMPACT OF NATIONAL RULES ON UNILATERAL

CONDUCT THAT DIVERGE FROM ARTICLE 102 OF THE

TREATY ON THE FUNCTIONING OF THE EUROPEAN
UNION (TFEU)

FINAL REPORT

RESEARCH TEAM: PROJECT MANAGEMENT

Andrea Renda (coordinator) Anne-Claire Zirnhelt
Sally van Siclen
Ioannis Kokkoris
Consuelo Pacchioli
Lorna Schrefler
Claudia Desogus
Can Selcuki

Done in Bruges and Brussels, 21 November 2012

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

This study was produced by the College of Europe and the Centre for European
Policy Studies for the European Commission and represents its authors' views on
the subject matter. These views have not been adopted or in any way approved by
the Commission and should not be relied upon as a statement of the
Commission's views. The European Commission does not guarantee the accuracy
of the data included in this report, nor does it accept responsibility for any use
made thereof. This study is intended to describe the legal situation at December
2011.

PAGE 2 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

TABLE OF CONTENTS
MAIN FINDINGS 5
EXECUTIVE SUMMARY 7
INTRODUCTION 17
1.1 Methodology 20
1.2 Preliminary clarifications 21
1.2.1 Which national legal rules? 21
1.2.2 What does “stricter than Article 102 TFEU” mean? 25
1.3 Structure of the Report 25

2. MAPPING NATIONAL LEGAL RULES ON UNILATERAL

CONDUCT: THE RESULTS OF OUR LEGAL ANALYSIS 27
2.1 Countries applying stricter standards for the definition of
dominance and/or abuse of dominance 27
2.1.1 Countries using market share thresholds for the
presumption of dominance 28
2.1.2 Countries having per se rules for conducts subject to a
rule of reason under EU competition law 40
2.1.3 Concluding remarks 41
2.2 Specific types of conduct and legal rules 42
2.2.1 Abuse of economic dependence or superior bargaining
position 42
2.2.2 Rules on sales below cost 69
2.2.3 Tied sales 79
2.2.4 Unfair competition and consumer protection laws 91
2.3 Sectoral regulation 97
2.3.1 The grocery sector 97
2.3.2 Tying and other potentially unfair practices in retail
financial services 101
2.3.3 Network infrastructure 106
2.4 Concluding remarks: a persisting legal fragmentation 107

3. UNILATERAL CONDUCT, BUSINESS STRATEGY AND

CONSUMER WELFARE: LITERATURE AND EVIDENCE 111
3.1 Diverging standards for defining dominance and/or abuse 111
3.2 Assessing the impact of economic dependence/superior
bargaining position 115
3.3 Assessing the impact of sales below cost 121
3.4 Assessing the macroeconomic impact of competition rules 131
3.4.1 Studies that analyze the effect of antitrust on GDP
growth 135

PAGE 3 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

3.4.2 Studies that focus more on the impact of antitrust on

productivity growth 136

4. AN ASSESSMENT OF THE IMPACT OF NATIONAL RULES THAT

ARE STRICTER THAN ART. 102 TFEU 138
4.1 Indication of most affected enterprises 138
4.1.1 The business survey and workshops 139
4.1.2 Affected enterprises 147
4.2 Impact of stricter national rules 148
4.2.1 Stricter definitions of dominance 148
4.2.2 Rules on abuse of economic dependence 152
4.2.1.1. Views of market players 152
4.2.1.2. Costs of compliance of those whom the rules are
designed to constrain 155
4.2.1.3. Quantitative estimates of costs of compliance 156
4.2.1.4. Suppliers´estimates of the costs of convergence 158
4.2.3 Rules on “sales below cost” 160
4.2.4 Rules on tying 162
4.2.5 Impact of the divergence of stricter national rules on
relationships between retailers and suppliers 163
4.2.6 Possible different impacts of certain national rules 164
4.2.7 Impact of national sector specific unilateral conduct
rules 165
4.3 Assessment of the broader economic impact 168
4.3.1 The impact on businesses not involved in grocery
retailing or supply, or network infrastructure 168
4.3.2 The impact on consumers 170
4.3.3 An econometric analysis of the impact of sales below
cost rules 171
4.3.4 The impact on public authorities 176
4.3.5 The impact on broader economic variables 177
4.4 Conclusions 178

5. MAIN FINDINGS 182

PAGE 4 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

MAIN FINDINGS
 There is a remarkable fragmentation of legal rules on unilateral conduct,
which affects as many as 18 Member States. Legal rules on unilateral conduct
are remarkably fragmented in the EU27, with a minority of EU Member States
featuring national legislation broadly aligned with EU competition law, and as
many as 18 countries exhibiting different types of diverging rules or standards
that are stricter than Article 102 TFEU.
 The enterprises that appear to be most affected by the current legal
fragmentation are of three types: enterprises active in certain Member States
in grocery retailing, and their suppliers; suppliers of intermediate products;
and enterprises active in network infrastructure such as electricity, pipeline
gas, and telecommunications.
 Such fragmentation can hamper the internal market, in particular by
increasing legal uncertainty and raising compliance costs for affected
undertakings. Companies wishing to engage in cross-border trade are
hindered by (i) the existence of a wide divergence in the interpretation of the
concepts of dominance and abuse thereof; (ii) the legal uncertainty
surrounding the exact features of national legislation and its enforcement; and
(iii) the need to change contracts and commercial conduct to adhere to
national rules.
 Most national rules that are stricter than Article 102 TFEU do not
predominantly pursue the same objectives of Article 102 TFEU. As a matter of
fact, many of these rules are conceived to protect certain competitors, rather
than the competitive process. In some cases this is due to consolidated
national traditions, but in other cases laws are simply transplanted from other
legal systems.
 Most national rules that are stricter than Article 102 TFEU appear to be
harmful for consumers. This is due to reduced product variety created by the
lack of a Single Market; but also to the fact that many of the stricter rules on
unilateral conduct are meant to protect smaller competitors, rather than the
competitive process, most often leading to price increases for end consumers.
This is particularly the case for national rules on abuse of economic
dependence/superior bargaining power, as well as rules on sales below cost,
which – as we estimated through an econometric analysis – lead to artificial
price increases of up to 3%.
 Extending the convergence rule of Article 3(2) Reg. CE 1/2003 to Article 102
TFEU would not have significant negative consequences for consumers,
businesses, and public authorities. National competition authorities would not
be significantly affected; small retailers that are meant to be protected by some
rules on sales below cost would not be put at a disadvantage; and consumers
would gain in a number of respects, as a result of the removal of existing
fragmentation in national rules.
 At the same time, extending the convergence rule of Article 3(2) Reg. CE
1/2003 to Article 102 TFEU would only partially solve the problem of
fragmentation, since national governments might retain an incentive to
pursue objectives that are inconsistent with EU competition law and the Single

PAGE 5 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Market through legislation and regulation, which is not covered by the

convergence rule.

PAGE 6 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

EXECUTIVE SUMMARY

C
ompetition law is one of the key pillars of the EU internal Market, as
attested by the fact that rules on competition were included already in the
1957 Treaty of Rome, which established the European Community. The
application of consistent and well-designed rules on competition throughout the
EU member states is essential for companies wishing to engage in cross-border
trade or establish themselves in more than one member state, as well as for
European consumers who deserve to have access to product and service markets
under similar conditions and with the maximum possible variety of choice. At the
same time, as recalled by Commissioner Almunia in a recent speech delivered in
May 2010, “competition policy is not something that can be pursued only at EU
level … it is crucial that enforcement takes place both at EU level and within the
Member States, in a coherent and consistent manner”1.
Against this background, EU competition rules on unilateral conduct co-exist
with national rules that often apply different standards and criteria to assess
dominance, or address conducts such as the abuse of economic dependence or
superior bargaining power, sales below cost, tying and other commercial
practices that are considered harmful for competition, or for certain types of
players active along the value chain (e.g. small producers or retailers). This is
possible since Reg. CE 1/2003 – while introducing a “convergence rule” (Article 3
para. 2) that seeks to create a level-playing field by providing for a single
standard of assessment for agreements, concerted practices and decisions by
associations of undertakings (Article 101 TFEU) – does not prevent member
states from enacting or maintaining rules that are stricter than Article 102 TFEU
(formerly Article 82 EC). More specifically, Article 3(2) of Reg. CE 1/2003 allows
Member States to enact or preserve legislation on unilateral conduct that is
stricter than Article 102 TFEU; however, the same paragraph does not allow
member States to apply rules that exactly mirror Article 102 TFEU in a stricter
way than what is done at EU level under the case law of the Court of Justice of
the EU.
In 2008-2009, the European Commission carried out an assessment of the
functioning of Reg. CE 1/2003, which was accompanied by an extensive
consultation period, during which some stakeholders have expressed the view
that the current fragmentation of legal approaches to unilateral conduct in the
EU27 can represent an obstacle to the full achievement of the EU Single Market2.
The existing divergence between national and EU rules on unilateral conduct
takes different forms:
 First, even when the national rules otherwise mirror EU competition rules,
stricter criteria are applied at national level to assess dominance and/or abuse
of dominance. This should not be allowed under Article 3(2) Reg. CE 1/2003.
 Second, some national rules focus on the bargaining strength of one of the
parties in a commercial relationship along the value chain, regardless of
whether that party is dominant in a relevant product market. This is typically

1 See the speech by Commissioner Almunia, “Competition and Consumers: the future of EU
Competition Policy, Brussels, 12 May 2010, SPEECH/2010/233.
2 See http://ec.europa.eu/competition/consultations/2008_regulation_1_2003/index.html

PAGE 7 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the case of national rules on abuse of economic dependence/superior

bargaining power, currently in force in several countries. These rules were
explicitly mentioned as examples of allowed rules by Reg. CE 1/2003, as will
be explained below.
 Third, certain national rules aim at prohibiting certain unilateral conduct
(irrespective of whether the company is dominant in a relevant market) that
can harm consumers by affecting i.a. competition and/or consumer choice.
This is the case for those countries that ban selling below cost or tied sales.
 Fourth, some Member States have enacted legislation in specific sectors –
most notably, in the food sector, in retail financial services and in network
industries – which is stricter than Article 102 TFEU.
A key question, to which this study can merely contribute with background
information without ultimately resolving it, was precisely which of those rules
came within the scope of the convergence exception of Reg. CE 1/2003. This
issue triggered a very lively debate among practitioners, scholars, officers of
competition authorities and industry representatives during the four workshops
we organized during the period in which this study was drafted. Some
commentators even questioned that rules on abuse of economic dependence can
indeed be considered as predominantly pursuing the same goals of Article 102
TFEU, which should be the criterion for including them under the scope of a
future convergence rule. As a result, the boundary between those stricter national
laws on unilateral conduct, which are inside – “competition rules” – and outside
is subject to interpretation: an expansive definition of what it means to “protect
competition on the market” would bring very many laws indeed within the scope
of the convergence exception. As a matter of fact, many laws we have found at
national level cannot be said to predominantly pursue the same objectives of
Article 102 TFEU: many of these laws seem to us to have been enacted with the
more or less explicit goal to protect smaller competitors in large retail markets,
rather than protecting the competitive process to the benefit of end consumers.
In light of the work been carried out elsewhere to reduce barriers to the internal
market, and the benefits of advocacy for reduction of unnecessarily competition-
suppressing rules, information which arguably lies outside the scope of this study
is nevertheless reported.

COMPARATIVE LEGAL ANALYSIS: MAIN FINDINGS

1. A REMARKABLE LEGAL FRAGMENTATION

Our analysis of the existing national legal provisions on unilateral conduct is

based on the legislations in force until November 2012 and shows that, as far as
rules on unilateral conduct are concerned, the EU27 still exhibits a remarkable
fragmentation, which can potentially hamper the Single Market, forcing
businesses wishing to engage in cross-border trade to adapt their strategy and
conduct based on the rules in force in different national territories. Based on our
findings:
 In three countries – Austria, Germany and Lithuania – the definition of
dominance contained in the competition law can be considered as
significantly stricter than the one adopted in Article 102 TFEU. In Hungary a

PAGE 8 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

specific piece of legislation outside the competition law domain introduces a

concept akin to dominance, although it differs a lot from the latter.
 In Austria, Bulgaria, Estonia, Poland and Romania the definition of
dominance and/or abuse is only slightly stricter than Art. 102 TFEU, since
there is a larger role attributed to market shares as an indication of
dominance. Among these countries, Poland seems to be the less diverging
one, as the analysis of market shares is complemented by other factors.
 Abuse of economic dependence is regulated in as many as twelve countries
(Austria, Cyprus, Czech Republic, France, Germany, Greece, Hungary, Italy,
Latvia, Portugal, Romania, Spain), with different modalities and standards
used, and different degrees of involvement of competition authorities.
 Sales below cost are strictly regulated in seven national jurisdictions (Austria,
Bulgaria, France, Germany, Latvia, Portugal, Romania), and have been
subject to specific sectoral rules especially in the grocery sector in the past
years.
 Tying has been subject to a per se prohibition with different scope and criteria
in ten Member States (Bulgaria, Cyprus, France, Germany, Hungary, Ireland,
Poland, Portugal, Romania, Slovakia): seven of these countries will be able to
maintain these provisions in the field of retail financial services (Bulgaria,
France, Germany, Hungary, Poland, Portugal, Romania), where the EU
Unfair Commercial Practices Directive leaves room for exceptions to the
maximum harmonization rule.
 Finally, in at least three jurisdictions (Belgium, Hungary, Romania) we found
national unfair competition laws that impose stricter requirements on
unilateral conduct compared to Article 102 TFEU.

2. FRAGMENTATION AFFECTS EIGHTEEN MEMBER STATES

A number of countries appear to contribute to different degrees to the

fragmentation of legal rules on unilateral conduct. Overall, as many as 18
Member States appear as having enacted national rules that are stricter than
Article 102 TFEU.
 Hungary and Germany are the countries that have the highest number of
rules that are stricter than Article 102 TFEU (7 for Hungary, 6 for Germany).
In Hungary, the divergence is the result of rules which, except for conditional
sales, do not belong to competition law but are found in other bodies of law:
in the Act on Trade that disciplines a particular form of market power on the
buyer side, the abuse of economic dependence and tying, as well as in sectoral
rules applied to the grocery sector, financial services and network industries.
In Germany the divergence is due to rules on dominance, economic
dependence, sales below cost, and tying, plus rules applicable in the grocery
sector and in network industries.
 Two countries, Romania and Slovakia, exhibits five national rules that
diverge from Art. 102 TFEU, though to a slightly lesser extent than the
countries listed above. In Romania, the most diverging rules are on
dominance, economic dependence, sales below cost, tying, and unfair
competition. Slovakia exhibits slightly diverging rules on the abuse of

PAGE 9 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

economic dependence and unfair competition, as well as sectoral rules in the

grocery, financial and network industries.
 In Austria, France and Portugal there are four instances of rules that
diverge from Art. 102 TFEU. In Austria the most diverging rules are on the
definition of dominance, on economic dependence and on sales below cost. In
addition, there are sectoral rules in the network industries which follow
different criteria, although these laws transpose EU Directives into national
law and therefore should be considered in line with the relevant European
legislation. In France, divergence is due to rules on abuse of economic
dependence, sales below cost, tying and sectoral rules on financial services.
Portuguese law diverges in the fields of economic dependence, sales below
cost, and tying.
 Bulgaria, Poland all have three types of national rules that can be
considered stricter than Art. 102 TFEU. Bulgaria diverges mostly due to
slightly stricter criteria for assessing dominance, sales below cost and tying.
Poland features slightly diverging criteria for the finding of dominance, but
also significantly diverging legislation on tying and in the field of financial
services.
 Finally, other countries feature only one or two types of diverging rules,
including Belgium, Cyprus, the Czech Republic, Estonia, Greece,
Ireland, Latvia Lithuania and Spain. The Czech Republic features
relevant national rules on economic dependence disciplined in a very
controversial legislation in the grocery sector. Similarly, in Latvia stricter
rules are provided for by competition law in relation to economic dependence
specifically at the retail level.

The figure below graphically summarises our findings on the degree of

fragmentation generated by diverging legal rules on unilateral conduct in the
EU27. The darker the colour, the greater the number of legal provisions found at
national level, which appear stricter than Article 102 TFEU. In order to build this
graph, we have only considered the number of provisions that are potentially
stricter than Article 102 TFEU, without accounting for any measure of the actual
extent to which each rule diverges from the EU rule.

PAGE 10 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The fragmentation of legal rules on unilateral conduct in the EU27

Europe – Countries

THE IMPACT OF NATIONAL RULES ON UNILATERAL

CONDUCT: MAIN FINDINGS
The legal provisions identified above exhibit varying degrees of enforcement and
are stricter than Article 102 TFUE in different ways. Their impact on industry
players is difficult to assess, and seems unevenly distributed across industry
sectors and stages of the value chain. Our analysis led to the following findings.

3. FRAGMENTATION CREATES LEGAL UNCERTAINTY TO THE DETRIMENT OF

BUSINESS CONDUCT AND THE INTERNAL MARKET

The uncertainty generated by the existing legal fragmentation seems to exert a

significant impact in all sectors, both due to the uncertain treatment of conduct
that would fall under Article 102 TFEU at national level; and due to the difficulty
in interpreting patterns of enforcement and interpretation of the wide variety of
national rules that are stricter than EU rules on abuse of dominance. Although
quantifying this uncertainty in monetary terms would prove prohibitively
difficult, we have gathered consistent evidence from surveyed, interviewed and
consulted stakeholders, which points at significant chilling effects. Companies
have reported that they refrain from adopting behaviour that they consider to be
pro-competitive (and that would be considered as such under EU competition
law), since they fear that national authorities would challenge that behaviour in
one or more Member States. The difficulty to coordinate, manage and monitor
compliance with national rules creates significant administrative costs for

PAGE 11 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

businesses operating in more than one Member State some of them try to solve
the problem by hiring local lawyers and changing contractual documents and
market behaviour in every country; but other companies prefer to standardize
their compliance programmes in a way that leads them to under-exploit their
competitive potential, for example by refraining from engaging in pro-
competitive short-term promotional activities, or by keeping prices artificially
high for fear of being considered would-be predators.

More in detail: Multinational, typically intermediate goods manufacturers

described the qualitative effects above of divergent definitions of dominance, and
how diverging legal standards can increase compliance costs for businesses.
Others, typically multinational retailers described how they modified their
business processes to achieve compliance with divergent rules, including the
modification of contract terms, costs of renegotiation, training of personnel, and
the modification of accounting and invoicing systems. In addition to the one-time
costs of adapting to new or amended legislation, the results of the contract
renegotiations had significant negative impacts on profits. Based on estimates
provided by two large retailers, we have been able to scale up compliance costs to
the EU level: our figures of €80 000 to €160 000 for direct compliance costs, and
of €45-70 million for foregone profits, if one assumes the stricter rules apply to
ten retailers in ten Member States, is significant relative to usual profit margins
in the retailing sector. These figures are based on estimates independently
provided in confidence to the researchers by two large grocery retailers who incur
these costs in their operations. Alternative assumptions, e.g., only 5 large affected
retailers or fewer Member States, would have a proportionate effect on the
“headline figures,” as pointed out in Section 4.2.1.3 of this Report.3

4. DIVERGENCE IN THE DEFINITIONS OF DOMINANCE AND/OR ABUSE CAUSE

SERIOUS CONCERNS TO BUSINESSES

Not only the existence of stricter national rules on unilateral conduct, but also the
divergent interpretation of rules on abuse of dominance at national level is
creating uncertainty for businesses active in EU Member States. The existence of
national rules that heavily rely on market shares as a proxy for dominance; laws
that explicitly set very low thresholds above which dominance is presumed; or
legal provisions which apply a per se rule to conduct that would be appraised
under an effects-based approach at the EU level must be considered as a serious

33As described in detail in the body of the report, these estimates are based on figures provided,
independently, to the researchers from two large grocery groups that have operations in several
EU Member states. The groups did not provide documents, but rather provided their estimates.
As detailed later, assumptions were made to ”scale up” these two undertakings´ estimates to the
universe of “similar” undertakings within the EU. Alternative assumptions are also reasonable
and would give different ”headline figures.” The details of the calculations are provided so one
may easily make alternative scaling up assumptions to test the sensitivity of the ”headline figures”
to the assumptions. However, the interviews conducted by the researchers of representatives of
similar grocery groups (similar in the sense of being active both in EU-15 and post-EU-15
Member states, and expressing similar concerns regarding the qualitative effects on their
operations of the same laws, or laws that have provisions that have the same qualitative effects)
indicate that some sort of scaling up is required because these interviewed grocery groups
indicated the same sort of qualitative effects as did the grocery groups providing the quantitative
estimates, but did not provide quantitative estimates.

PAGE 12 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

problem, which hampers the Internal Market. It is worth recalling, in this

respect, that Article 4(3) TFEU obliges national authorities to avoid interfering
with EU objectives, including competition, and so might in theory prohibit
national authority from adopting decisions that run counter to the objectives of
Article 102 TFEU.

5. ABUSE OF ECONOMIC DEPENDENCE/SUPERIOR BARGAINING POWER: LITTLE

BENEFITS, UNINTENDED CONSEQUENCES

National rules prohibiting abuse of economic dependence/superior bargaining

power are obviously a major concern of grocery retailers and their suppliers.
Grocery retailers find that these rules significantly increase their costs of doing
business, and in particular:
 Problems of legal uncertainty, related to the rule-making process, including
whether stakeholders are consulted about proposed rules, whether the impact
of the proposed rules is assessed, whether the proposed rules fit into the
existing legal framework, and whether the rules are sufficiently
understandable to be complied with by those to whom they apply.
 High cost of the system of rules, especially in certain member states.
Legislation on abuse of economic dependence/superior bargaining position
appears to have had an unintended negative effect on SMEs who supply large
retailers. As a result of these rules, large retailers tend to shift purchases away
from smaller towards larger suppliers so as to reduce the retailers’ exposure to
the risk that they may be accused of abuse of economic dependence/superior
bargaining position. If SMEs provide regional specialities, or variety valued by
consumers, then consumers too are inadvertently harmed.
Against this background, rules on abuse of economic dependence/superior
bargaining position are costly and do not appear to, in practice, aid the small
local suppliers they are designed to protect. Nevertheless, suppliers to grocery
retailers argued for retaining and strengthening rules against abuse of economic
dependence. Convergence with Article 102 TFEU would, if these rules are
deemed to come under the rubric of “competition rules,” reduce these costs and
distortions, but may harm certain suppliers.

6. RULES ON SALES BELOW COST CAN REDUCE COMPETITION IN RETAILING AND

KEEP PRICES ARTIFICIALLY HIGH
Evidence from various studies indicates that prohibitions of sales below cost
reduce competition in retailing. It is worth noting that such prohibitions are in at
least one Member state (France) combined with a separate restrictive law, a
prohibition of discrimination, in a way that creates a retail price floor. This
combination is particularly harmful for price competition, also since it does not
mirror the current (flexible) approach adopted at the EU level towards
discriminatory behaviour under Article 102(c), and as such creates a remarkable
confusion and inconsistency in the application of competition rules in the EU. In
addition, the use of cost (or sometimes, purchase price) benchmarks across
Member States varies so much that compliance with these rules requires
enormous care, and reportedly leads businesses to keep prices artificially high. If
such prohibitions were removed, then retail prices are estimated to fall by up to

PAGE 13 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

three percent. And this price fall would be the result of increased price
competition.
Even more importantly, rules prohibiting sales below cost, which are chiefly
aimed at protecting small retailers, appear not to perform the protective function.
Suppliers negotiate different deals with different distributors. Large firms may
well have found ways to evade the rules. Nevertheless, small retailers persist
through offering a differentiated experience to that of the large retailers.
Importantly, any attempt to protect small retailers through an application of
provisions that ban sales below cost is ineffective if the cost benchmark used is
referred to the larger company: due to economies of scale, even prices above
costs for the larger company could prove unsustainable for smaller retailers:
consequently, protection of smaller retailers cannot be effectively achieved by
rules on unilateral conduct that predominantly pursue the same goal of Article
102 TFEU.
In summary, as shown in this Study, rules prohibiting sales below cost have a
significantly harmful effect on consumer prices, and they appear not to have a
detectable beneficial effect on small retailers.

7. THREE SETS OF ENTERPRISES APPEAR TO BE PARTICULARLY AFFECTED

Our study has identified three sets of enterprises that are particularly affected by
stricter national rules on unilateral conduct than are others.
 Enterprises active in certain Member States in grocery retailing, and their
suppliers, who were affected by stricter rules on abuses of economic
dependence and sales below cost. The evident goal of national rules is to
protect smaller suppliers and retailers, respectively, as was done also, with
rather harmful consequences, in other jurisdictions in the past – e.g. in the
US with the 1936 Robinson-Patman Act. Research carried out for the
purposes of this Study has uncovered a problem related to the distribution of
market power along the grocery value chain, which we believe cannot, and
should not, be fixed through the application of national rules that pursue
competition law objectives.
 Suppliers of intermediate products, who were affected mostly by stricter
definitions of dominance. These players mostly face the dilemma of whether
to standardize their compliance programs across the EU27 by tailoring them
to the country featuring the strictest laws (with a resulting chilling effect in all
other countries); or to face significant adaptation costs by hiring lawyers,
modifying contracts, changing competitive strategy and retraining personnel
whenever they wish to enter a new national market in the EU.
 Enterprises active in network infrastructure such as electricity, pipeline gas,
and telecommunications, who were affected by competition rules aimed at
their specific sectors. In these sectors, problems are peculiar, due to the
overlap between the ex post application of competition rules, and the ex ante
application of regulatory measures that borrow, to varying degrees, concepts,
notions and definitions from EU competition law. For example, the definition
of Significant Market Power (SMP, i.e. dominance) in the telecommunications
follows slightly different criteria depending on the national regulatory
authority; and the same can be said for the choice of remedies that follow

PAGE 14 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

from a finding of SMP, with some countries imposing structural separation of

the network, and others being more lenient in the application of access policy.

8. CONSUMERS ARE HARMED BY SOME OF THE NATIONAL RULES

Consumers could be affected by a possible extension of the convergence rule to
national competition laws concerning unilateral conduct by changes in price, in
variety or speed and direction of innovation. The possible change in price can be
estimated from several studies that examined the effect of removal of sales below
cost prohibitions on retail prices. If convergence implied that such prohibitions
would be removed, then they serve to estimate the effect on price of convergence
in those Member States with an effective sales-below-cost prohibition.
Consumers in those Member States with effective sales-below-cost laws would
likely benefit by a decrease in prices of a few percentage points, if convergence to
Article 102 TFEU had the effect of removing these laws, with the important
exception of instances of predatory pricing. This, too, significantly affects the
Single Market.

9. THE IMPACT ON NATIONAL AUTHORITIES IS LIKELY TO BE SMALL

We have estimated whether an extension of the convergence rule to rules on
unilateral conduct would harm national competition authorities, for example by
imposing on them a duty to analyse the effects of the conducts where they
currently rely on a less burdensome per se approach or on string presumptions
triggered by market share thresholds. When interviewed, these authorities
themselves either estimated that their actual additional costs would be small, or
pointed out that they had a duty to investigate that did not solely rely on legal
presumptions, such as market share thresholds. Thus, we conclude that the
direct economic impact on national competition authorities of a possible
extension of the convergence rule to national competition laws concerning
unilateral conduct would be small.

10. EXTENDING THE CONVERGENCE RULE TO UNILATERAL CONDUCT MIGHT

ONLY PARTLY SOLVE THE PROBLEM OF LEGAL FRAGMENTATION
Our Study concludes that the current legal fragmentation is harmful for
consumers, for most businesses, and overall for the Single Market. At the same
time, it was apparent from opinions collected that simply extending the
convergence rule to rules on unilateral conduct would not ensure that national
authorities do not resort to other means to obtain the purported effect of
legislation that is currently in place. In many circumstances, “equilibrating
tendencies” such as over- or under- enforcement of legislation, direct sectoral
regulation and soft law might lead to the same level of divergence that is
currently observed, even under a convergence rule. Accordingly, we simply
observe that safeguard measures would be needed in order to ensure that the
intended effect is indeed achieved.

PAGE 15 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

THE IMPACT OF NATIONAL RULES ON UNILATERAL

CONDUCT THAT DIVERGE FROM ARTICLE 102 OF THE
TREATY ON THE FUNCTIONING OF THE EUROPEAN
UNION (TFEU)

PAGE 16 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

INTRODUCTION
Competition law is one of the key pillars of the EU internal Market, as attested by
the fact that rules on competition were included already in the 1957 Treaty of
Rome, which established the European Community. The application of consistent
and well-designed rules on competition throughout the EU member states is
essential for companies wishing to engage in cross-border trade or establish
themselves in more than one member state, as well as for European consumers
who deserve to have access to product and service markets under similar
conditions and with the maximum possible variety of choice. At the same time, as
recalled by Commissioner Almunia in a recent speech delivered in May 2010,
“competition policy is not something that can be pursued only at EU level … it is
crucial that enforcement takes place both at EU level and within the Member
States, in a coherent and consistent manner”4.
Against this background, EU competition rules on unilateral conduct co-exist
with national rules that apply different standards and criteria to assess
dominance; and rules that ban certain conduct regardless of whether the related
undertaking is dominant, or regardless of any evidence of effect. This is possible
since Reg. CE 1/2003 – while introducing a “convergence rule” (Article 3 para. 2)
that seeks to create a level-playing field by providing for a single standard of
assessment for agreements, concerted practices and decisions by associations of
undertakings (Article 101 TFEU) – does not prevent member states from
enacting or maintaining rules that are stricter than Article 102 TFEU (formerly
Article 82 EC).
In 2008-2009, the European Commission carried out an assessment of the
functioning of Reg. CE 1/2003, which was accompanied by an extensive
consultation period, during which some stakeholders have expressed the view
that the current fragmentation of legal approaches to unilateral conduct in the
EU27 can represent an obstacle to the full achievement of the EU Single Market5.
The existing divergence between national and EU rules on unilateral conduct
takes different forms:
 First, even when the national rules mirror EU competition rules, firms
wishing to engage in cross-border trade may face significant problems due to
the stricter criteria applied at national level to assess dominance and/or
abuse of dominance. As will be shown in the next sections of this Study, this
occurs in countries such as Austria, Bulgaria, Germany and Lithuania – where
the definition of dominance contained in the competition law can be
considered as significantly stricter than the one adopted in Article 102 TFEU;
as well as in Austria, Estonia, Poland and Romania, where there is either a
reliance on stricter market share thresholds or the use of per se rules where
EU competition law would adopt an effects-based approach.

4 See the speech by Commissioner Joaquim Almunia, “Competition and Consumers: the future
of EU Competition Policy, Brussels, 12 May 2010, SPEECH/2010/233.
5 See the responses to the consultation on “Antitrust: Preparation of the report on the
functioning of Regulation 1/2003”, available online at the following website:
http://ec.europa.eu/competition/consultations/2008_regulation_1_2003/index.html. And see
the Communication from the European Commission to the European Parliament and the Council,
“Report on the functioning of Regulation 1/2003”, COM(2009) 206 final, 29.4.2009.

PAGE 17 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 Second, some national rules focus on the bargaining strength of one of the
parties in a commercial relationship along the value chain, regardless of
whether that party is dominant or not in a relevant product market. This is
typically the case of national rules on abuse of economic dependence/superior
bargaining power currently in force in countries like Austria, Cyprus, Czech
Republic, France, Germany, Greece, Hungary, Latvia, Portugal, Romania,
Spain; and – with somewhat different features (see Section 2 below) – in
Italy. Recital 8 of Reg. CE 1/2003 explicitly mentions national provisions,
which prohibit or impose sanctions on abusive behaviour toward
economically dependent undertakings6. Besides rules concerning specifically
the abuse of economic dependence, some national provisions regulate
behaviour labelled as 'abuse of superior bargaining power' or 'abuse of
significant influence'. The aim of these kinds of rules is essentially to regulate
disparities of bargaining power in distribution relationships, including where
neither the supplier nor the distributor holds a dominant position on a
specific market. It should be noticed that these rules are not always included
in competition provisions but sometimes belong to other bodies of law, as we
are going to indicate in detail for each rule.
 Third, certain national rules aim at prohibiting certain unilateral conduct that
can harm consumers by affecting i.a. competition and/or consumer choice.
This is the case for those countries that ban selling below cost or tied sales,
although in almost all sectors of the economy some of these practices (e.g.
tied sales) are now considered incompatible with the Unfair Commercial
Practices Directive7. As will be shown below, sales below cost are strictly
regulated in seven national jurisdictions, while tied sales have been subject to
per se prohibitions in as many as ten Member States. However, in most of the
cases provisions dealing with tying have different scope and criteria with
respect to competition law. Finally, in at least three countries unfair
competition laws impose stricter requirements on unilateral conduct
compared to Article 102 TFEU.
 Fourth, some Member States have enacted legislation in specific sectors –
most notably, in the food sector, in retail financial services and in network
industries – which is stricter than Article 102 TFEU.
Whether such difference may constitute an obstacle to the EU internal market
depends on a number of different factors, such as:
 The extent to which the undertakings affected formulate their business
strategies at a pan-European or regional level8. The more such strategies

6 The wording of the Recital is: These stricter national laws may include provisions which
prohibit or impose sanctions on abusive behaviour toward economically dependent undertakings.
7 Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005
concerning unfair business-to-consumer commercial practices in the internal market and
amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and 2002/65/EC of the
European Parliament and of the Council and Regulation (EC) No 2006/2004 of the European
Parliament and of the Council (‘Unfair Commercial Practices Directive’), OJ L 149, 11.6.2005, p.
22–39. See Renda et al. (2009), Final Report of the Study on Tying and other potentially unfair
practices in retail financial services, a report for the European Commission DG Internal Market,
available online at http://ec.europa.eu/internal_market/consultations/docs/
2010/tying/report_en.pdf.
8 As recalled by the Commission in the Report on the functioning of Regulation 1/2003
{SEC(2009)574} at points 21 and 22 the last sentence of Article 3(2) of the Regulation contains

PAGE 18 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

develop a standard strategy for more than one national market, the greater
the potential impact on their conduct in the presence of different standards.
Fragmented rules at national level may imply (i) the need to adapt to each
national situation, thus incurring adaptation costs; or (ii) constraints on the
business conduct of those firms that decide not to face adaptation costs and
instead adopt in all countries a single strategy tailored to the country that
features the “strictest” rules.
 The actual differences between national and EU rules, their distribution
across member states and the extent to which they regulate issues that are
considered to be sensitive by the (potentially) affected undertakings. The
greater the size of the markets involved, especially those that feature the most
restrictive rules, the greater the possible impact on business conduct.
 The extent to which national rules on unilateral conduct are applied to
conduct capable of affecting trade between EU member states.
 Whether the rules apply to all sectors or only some specific sectors of the
economy;
 The extent to which the rules are actually enforced in practice, which can
depend on the frequency of litigation on those rules, but also on the frequency
of settlements of controversies regarding those rules9.
 Whether the rules are applied as stand-alone rules or in combination with
other rules that already apply at national level: for example, rules on the
abuse of superior bargaining position may be found also in national contract
law. In such cases, where abuse of superior bargaining position rules are
included in national contract law, national contract law will apply. Other rules
may overlap with consumer protection rules or provisions on unfair trade.
More generally, the actual impact of those rules depends significantly on
whether they are the core subject of litigation, or on the other hand they are
referred to only marginally in existing cases or settlements;
 Whether, absent those rules, further impediments would persist in national
legislation, which would create similar obstacles to those created by the rules
at hand – in other words, whether the “net impact” of those rules is
significantly hampering the internal market.
 Whether and to what extent consumers can shop across borders for a specific
type of (individual or bundled) good or service: if this is the case, then the
impact of national rules may be milder.
The European Commission has retained the College of Europe and the Centre for
European Policy Studies in order to conduct a study on the potential impact of
the national rules on unilateral conduct that are stricter than Article 102 TFEU.

an exception from the level playing field and implies that undertakings doing cross-border
business in the internal market may be subjected to a variety of standards as to their unilateral
behaviour.
9 Even if practical implementation is not frequent, the mere existence of the rule may affect
business strategies, due to the deterrent effect of the rule or simply, as pointed out by the
Commission, because “firms have to make sure that their practices comply with all the legal
standards in those Member States in which they have activities”. See Communication from the
European Commission to the European Parliament and the Council, “Report on the functioning
of Regulation 1/2003”, COM(2009) 206 final, 29.4.2009.

PAGE 19 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

In this Final Report, we illustrate the results of our analysis, by providing: a

mapping of existing rules on unilateral conduct in the EU27; a review of the
existing literature and empirical evidence of market effects generated by the
current variety of rules in force; the results of a business survey launched for the
purpose of assessing the impact of the existing national rules on business
strategy; and an assessment of the impact of those rules on the Internal Market.
Below, we illustrate the main methodological steps we have followed in the
present report.

1.1 Methodology
This report is the result of a collective effort by a team of economists and lawyers,
which made use of a variety of methodologies. In particular:
 The mapping of national legal rules and their enforcement has been carried
out through a survey, made possible thanks to the availability of a network of
national legal experts, who received two sets of questions related to their
national legal system during the summer of 2010 – a preliminary (first-stage)
questionnaire and a more detailed, second-stage questionnaire. The two
questionnaires are attached at Annex I and Annex II of this Final Report. The
results were then analysed by our legal experts in the core team, including
most notably Dr. Ioannis Kokkoris and Dr. Claudia Desogus. A key difficulty
in this phase of the study was distinguishing the legal rules that would fall
within the scope of the present study, as opposed to legal rules that would fall
outside the scope of the study. Section 2 below contains a detailed description
of the criteria we have used to identify the rules that were subject to in-depth
scrutiny in our report.
 The economic analysis of the impact of different types of legal rules on
unilateral conduct was carried out through desk research by our core team of
economists, including Dr. Sally van Siclen, Consuelo Pacchioli and Can
Selcuki. Literature on diverging standards to define dominance and abuse, as
well as economic literature on each of the individual types of conducts subject
to this report was extensively analysed and is reported in Section 3 below.
 An extensive survey of industry stakeholders has been conducted by the team
members, and coupled with bilateral meetings with several stakeholders and
a series of ad hoc workshops on the subject matter of this Report. Workshops
were held during September and October 2010 in Stockholm, Rome, and
Brussels.
 A variety of estimation techniques has been considered before the choice was
made that the difference-in-difference methodology would be the most
appropriate to provide an assessment of the potential impact of certain
national legal rules on price levels and business conduct, subject to data
availability. This choice was validated by members of the Chief economist
team at the European Commission, DG Competition. In this report, we have
been able to analyse one natural experiment, related to sales below cost in the
Irish grocery sector.
 Finally, a mix of qualitative and quantitative assessment is used to provide a
final analysis of the impact of national legal rules on business conduct and on

PAGE 20 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the Single Market. Limited data availability and, to some extent, also limited
interest on the side of business stakeholders in a number of countries and
sectors have represented a key constraint on the road towards a full
quantification of the impact of existing rules.
The specific features of each of the methodological steps followed during our
analysis are illustrated in more detail below, in the relevant sections of this study.

1.2 Preliminary clarifications

1.2.1 Which national legal rules?

Before we start illustrating the main findings of our analysis, it is important to

clarify which national rules were included in the scope of this Report. This
question is less straightforward than it may seem at first blush: as a matter of
fact, dominant undertakings are subject to several types of national rules on
unilateral conduct, including rules belonging to contract law, tort law, etc. Some
of these rules clearly fall outside the scope of this study (e.g. pure contract law
rules, such as rules on penalty clauses or rules on gross disparity); at the same
time, other rules clearly fall within the scope of the study (e.g. rules on economic
dependency and sales below cost or prohibitions of tying). But for several types of
rules that affect unilateral conduct, there was initially a degree of uncertainty on
the need to include them in this Report or not. Additional difficulties were
related to the fact that legal rules are subject to interpretation by national judges,
which in some cases leads to a better refinement of the actual scope of those
rules, and the objectives they pursue.
In line with Article 3(3) of Reg. CE 1/2003, the relevant national laws for the
purposes of our study are those that predominantly pursue an objective not
different from those of Articles 101 and 102 TFEU. This does not mean that such
provisions should be necessarily included in national competition laws; but it
means that state measures that predominantly pursue other legitimate objectives
should not be included in our analysis. Further clarification on the scope of the
study can be found in Recitals 8 and 9 of Reg. CE1/2003:
 Recital 8 of Reg. CE 1/2003 provides an example of a law deemed to
constitute a stricter national competition law with respect to undertakings’
unilateral conduct, that is, “provisions which prohibit or impose sanctions on
abusive behaviour toward economically dependent undertakings.”
 Recital 9 of Reg. CE 1/2003 notes that the Regulation does not preclude
Member States from implementing national legislation which protects
legitimate interests other than competition objectives, provided such
legislation is compatible with general principles and other provisions of EU
law. The recital provides an example of a law deemed to pursue
predominantly an objective different from that of competition law: “national
legislation that prohibits or imposes sanctions on acts of unfair trading
practice, be they unilateral or contractual. Such legislation pursues a
specific objective, irrespective of the actual or presumed effects of such acts
on competition on the market. This is particularly the case of legislation
which prohibits undertakings from imposing on their trading partners,

PAGE 21 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

obtaining or attempting to obtain from them terms and conditions that are
unjustified, disproportionate or without consideration.”
Furthermore, clarifications are also found in the Commission Staff Working
Paper accompanying the Report on the Functioning of Reg. CE 1/2003,
SEC(2009)574. There:
 The Commission Staff Working Paper accompanying the Report on the
Functioning of Reg. CE 1/2003, SEC(2009)574 states that while, “Concepts
such as the abuse of economic dependency exist in the competition laws of
some Member States….it is not clear whether competition law is the
appropriate instrument to address concerns arising from e.g. disparities in
bargaining power (see paragraph 181 below). [footnote omitted] As regards
specifically the laws concerning resale below cost or at loss, certain studies
have suggested that this kind of regulation results in price increases and loss
of consumer welfare. [footnote omitted]” (para. 178)
 The Commission points out the difficulty of distinguishing between stricter
competition rules on unilateral conduct and laws on unfair trading practices.
They indicate that if the objective is to regulate contractual relationships
between undertakings by stipulating the terms and conditions that for
instance suppliers must offer to distributors, rather than regulating their
competitive behaviour on the market, then “the proper classification
appears to be that of laws concerning unfair practices.” This is distinguished
from, “rules combating excessive market power or protecting small
undertakings in a market against their larger competitors.” [emphasis
added] The latter are more likely to be considered competition law provisions.
(para. 181)
 Further, “It has been submitted that the main distinctive feature is whether
the aim of the provision is limited to regulating a contractual relationship
with a view to protecting a weaker party against a stronger party or
whether competition on the market is taken into account either in the
elaboration of the rule or its application.” [emphasis added] (footnote 213)
The Terms of Reference (TOR) of the contract under which this study was
undertaken provide the context. As requested by the European Commission, this
study provides an overview of stricter national provisions that could be regarded
as competition rules on unilateral conduct in the 27 Member States of the
European Union. This overview is based on an assessment of the aim and
contents of the national provisions and, therefore, is independent of whether
these rules are included in national competition provisions. The questions to be
studied are whether stricter national competition rules than those of Article 102
of the TFEU, as permitted under Article 3 of Reg. CE 1/2003, have an economic
impact, and what would be the economic impact of their convergence with Article
102 TFEU. In accordance with the definitions in Article 3, attention is limited to
those national laws that predominantly pursue an objective similar to those of
Articles 101 and 102 TFEU. According to the TOR, and indeed recital 9 of Reg. CE
1/2003, “Articles 101 and 102 of the Treaty [have] as their (main) objective the
protection of competition on the market.” (p. 2).
In case of doubt as to whether a national provision could be qualified as a
competition rule as envisaged by Article 3(2), last sentence, of Reg. CE 1/2003,
we have included it in the study and indicated the reason(s) for our reservations.

PAGE 22 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

In practice, this “demarcation” exercise proved quite challenging, especially since

drawing a line that distinguishes between competition provisions and non-
competition provisions is in some cases a difficult task. For example:
 Rules that refer to imbalances in the bargaining position of parties located on
the value chain – such as some rules on abuse of economic dependence or
abuse of superior bargaining power – do not necessarily pursue the same
goals as the EU competition rules, but are nevertheless imposing constraints
on dominant and non-dominant undertakings to the extent that they cannot
be ignored by a study such as the present one. This was confirmed also by the
fact that several stakeholders highlighted the existence of these rules as a
potential obstacle to the Single Market during the consultation on the
functioning of Reg. CE 1/2003 in 2009.
 Rules on contractual relationships could simultaneously protect a weaker
party against a stronger one and restrict competition. For example, rules
prohibiting discrimination at the upstream market and sales below cost at the
downstream market, both of which can be viewed as protecting weaker
parties, can dampen competition in the downstream market. H0wever, some
scholars doubt whether these rules – especially those on sales below cost –
can be interpreted as stricter than Article 102 TFEU10.
Accordingly, during our empirical analysis, remaining always within the scope of
the study as defined above has proven to be difficult. In particular, enterprises
may not necessarily distinguish between rules on unfair practices and rules on
competition when they demand greater convergence.11 The national laws cited by
businesses in interviews as causing them to modify their business strategies or to
incur extra costs fell into both categories. For example, some laws cited applied
to enterprises based on a nominal turnover threshold rather than any threshold
related to market power such as the share of the relevant market held by a given
undertaking.
Moreover, it must be recalled that there are other current European initiatives
aimed at rules and conduct which may be relevant for this Study. For example,
the relations between manufacturers and distributors, especially those involved
in the food supply chain, are the subject of the reports and consultation on
“Towards more efficient and fairer retail services in the Internal Market for
2020.”12 There, the Commission analyses the functioning of the entire retail
chain from suppliers to consumers and identifies a series of issues which may
prevent the retail sector realising its full potential, including i.a. divergent rules,
the risk of unfair commercial practices between different actors along the supply
chain, a lack of information and transparency on pricing and quality labels and
inadequate enforcement. The pattern of participation of companies in our

10 See i.a. Mateus, A. (2011), Ensuring a More Level Playing Field in Competition
Enforcement Throughout the European Union, (European Competition Law Review, Vol. 31,
Issue 12, 2010; discussing whether a law that prohibits sales below cost should be considered
‘stricter’ in the terms of limiting further the behaviour of a dominant firm; or ‘stricter’ in the sense
that protects small enterprises from competition; and adding that “in the last sense, they further
restrict competition which is not, from our perspective, the intent of the Treaty founders”.
11 The business community is reported in the Staff Report as calling for, “an extension of the
convergence rule to national laws covering unilateral conduct.”(para. 177)
12 COM(2010) 355 final, Report from the Commission to the European Parliament, the Council,

the European Economic and Social Committee of the Regions, Retail market monitoring report
“Towards more efficient and fairer retail services in the internal market for 2020.”

PAGE 23 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

voluntary questionnaire and interviews indicate that the types of companies on

which that work is focused were also the most concerned with national rules on
unilateral conduct that diverged from Article 102 TFEU.
To conclude, based on available references such as the text and Recitals of Reg.
CE 1/2003 as well as the Report on the Functioning of Reg. CE 1/2003 and the
accompanying Staff Working Paper, the study focuses on those national laws that
restrict unilateral conduct and that have as their predominant objective the
protection of competition on the market13. Given that there are cases in which
national provisions have slightly diverging or even multiple objectives, only part
of which overlap with those of Article 102 TFEU, we have chosen to be over-
rather than under-inclusive, and to include those laws in the analysis. In light of
the ongoing work elsewhere to reduce barriers to the internal market, and the
benefits of advocacy for reduction of unnecessarily competition-suppressing
rules, where we collect information about competition-suppressing rules which
arguably lie outside the scope of this study, strictly defined, we report it.

Article 3, Regulation CE no. 1/2003

Relationship between Articles 81 and 82 of the Treaty [now Articles
101 and 102 TFEU] and national competition laws

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.
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.

13 “Protection of competition on the market” is a rather broad phrase, which potentially

encompasses several, more specific objectives. Richard Whish, for example, cites a number of
possible purposes of competition law within the European Union: protecting consumer interests,
dispersal of economic power and redistribution of wealth, protecting small rivals or small
businesses, and integration of a single European market, among others. Such a broad definition
of purpose would imply that several national laws have predominantly the same objective as—at
least one objective of—competition law. See Whish, R. (2003), Competition Law, 5th ed. Reed
Elsevier, pp. 17-21.

PAGE 24 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

1.2.2 What does “stricter than Article 102 TFEU” mean? Establishing
a benchmark

Answering this question is impossible without establishing a benchmark. As a

matter of fact, in competition law the interpretation of the key articles of the
Treaty (such as Art. 102 TFEU) is provided by the decisions of the European
Commission, the General Court and most importantly the Court of Justice of the
European Union. Article 102 TFEU has been in force since the Treaty of Rome in
1957. Since then, the EU approach to the conducts that form the subject matter of
this legal provision (i.e. unilateral conduct by dominant undertakings) has
evolved significantly. Important recent cases such as i.e. Deutsche Telekom,
Telefonica, Wanadoo, Magill, IMS Health, Microsoft, Intel and many others
have given the Commission an important opportunity to clarify its approach to
these conducts.
Since 2005, the Commission has also worked on a review of its approach to
Article 102 TFEU, which – after the publication of a Discussion Paper in 2005
and an extensive stakeholder consultation – culminated in the publication of a
Guidance paper in December 2008, which outlines the Commission’s
enforcement priorities as regards the treatment of exclusionary abuses to what is
now Article 102 TFEU14. The key feature of the Guidance document is the focus
on the notion of “anticompetitive foreclosure”, which implies both evidence of
actual or likely foreclosure of competitors from the relevant market, and likely
consumer harm. Accordingly, one important methodological question in
assessing the impact of national rules that diverge from Article 102 TFEU is what
type of enforcement we consider as the benchmark for each of the conducts, and
in particular whether the established orientations of the Court of Justice, as well
as the emerging orientation of the Commission in the Guidance Paper on its
enforcement priorities should be taken as a benchmark.
In what follows, we will use the former benchmark: the definition of dominance
and/or abuse, as well as the current approach to the scrutiny of unilateral
conduct as developed by the jurisprudence of the Court of Justice will be
considered as the key reference in the appraisal of the “strictness” of the national
rule or standard relative to the EU one.

1.3 Structure of the Report

The remainder of this Final Report is structured as follows.
Section 2 below provides information on the national rules on unilateral conduct
in force, which are stricter than Art. 102 TFEU. We provide also information on
the extent to which those rules are enforced in practice, as well as summary
tables containing basic information on the rules in place, and an indication of the
countries in which such rules exist.

14 After having undergone legal-linguistic revision, the Guidance was adopted in all EU
languages on 9 February 2009. The document, the Commission's “Guidance on its enforcement
priorities in applying Article 82 (EC) to abusive exclusionary conduct by dominant undertakings”
SEC(2009)864, is available online at
http://ec.europa.eu/competition/antitrust/art82/index.html.

PAGE 25 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Section 3 reflects on existing legal and economic literature as well as existing

empirical studies and sectoral investigations by competition authorities or other
enforcers, with the aim to describe the potential impact of certain legal rules on
price levels and other market characteristics.
Section 4 uses the findings of Sections 2 and 3 to provide an assessment of the
impact of national legal rules that are stricter than article 102 TFEU. We provide
this assessment by means of both qualitative considerations and quantitative
analysis.
In Section 5, we summarize the main findings of this report and comment on
possible future policy directions.
The Report is accompanied by eight Annexes, which contain the two
questionnaires sent to national legal experts (Annex I and II); the questionnaire
sent to industry stakeholders (Annex III); a list of selected references (Annex IV);
a description of the methodology we used to carry out our econometric study
(Annex V): country reports, which describe more in detail the legal rules in place
in the member states (Annex VI); summary tables on national legal rules (Annex
VII); and resources of selected national competition authorities (Annex VII).

PAGE 26 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2. MAPPING NATIONAL LEGAL RULES ON UNILATERAL

CONDUCT: THE RESULTS OF OUR LEGAL ANALYSIS

This Section is the result of (i) an in depth analysis of the national competition
rules in force at the time of completion of this study (November 2012) that
diverge from Article 102 TFEU as identified in the introductory section above;
and (ii) an analysis of the national diverging standards for assessing dominance
and/or the impact of specific conducts covered by Article 102. Moreover, the
Section consists of an (iii) in depth analysis of the enforcement of specific
national rules on unilateral conduct, including the case law of the courts, as well
as details on private enforcement where relevant and available, and the decisions
of National Competition Authorities (NCAs) and/or of other competent
authorities15. This is coupled with the (iv) analysis of the compliance and
implementation of the above mentioned-divergent rules with respect to Article
102 TFEU together with the stricter national legal rules. This Section also
assesses whether, absent those rules, other national rules would still affect the
enforcement of competition rules on unilateral conduct in a way that
substantially diverges from Article 102 TFEU (for example, in national contract
law, tort law or unfair trade law).
The next sections present a cross-country comparative analysis of the treatment
of national rules on unilateral conduct that are stricter than Article 102 TFEU.
The analysis starts with a brief account of the enforcement of competition
legislation related to Article 102 TFEU in Member States before it addresses in
detail the enforcement of stricter rules on dominance/abuse of dominance and of
provisions on abuse of economic dependence/abuse of superior bargaining
position, sales below cost, tied sales and sector-specific rules. Annex VI to this
Final Report contains a more detailed analysis of the legal provisions in force in
the EU member states.

2.1 Countries applying stricter standards for the definition

of dominance and/or abuse of dominance
A number of EU member states have rules in place on abuse of dominance, which
can be considered as relying on stricter standards than Article 102 TFEU.
National legal rules differ in two main ways:
(i) Some countries use rather low thresholds, beyond which a relative
presumption of dominance applies, i.e. the allegedly dominant
undertaking has the burden of proving that it has no dominant position.
This is particularly the case in Austria, Bulgaria, Germany.
(ii) Some countries apply per se rules to conducts that are subject to an
effects-based approach in EU competition law.

15In some cases, we have also considered the role of sectoral inquiries and market analyses
performed by national competition authorities or sectoral regulators (e.g. in the financial services
sector, or in the food sector). These initiatives have in some circumstances triggered compliance
with competition rules or have led to subsequent investigations on the unilateral conduct of
specific undertakings.

PAGE 27 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.1.1 Countries using low market share thresholds for the

presumption of dominance

In our analysis, we found five countries (Austria, Bulgaria, Czech Republic,

Germany and Lithuania) in which the definition of dominance contained in the
law can be considered as significantly stricter than the one adopted in Article 102
TFEU16. In other four countries (Estonia, Lithuania, Poland, Romania) the
national rules only slightly differ from Article 102 TFEU, since a relative
presumption of dominance has been established in the law whenever the
undertaking concerned holds a market share of 40% (or 30%, in Lithuania). In
Latvia the 40% threshold was present in the law until 2008. Afterwards the
threshold was abolished and therefore the presumption does not apply anymore.
Where present, such threshold mirrors the one mentioned in the European
Commission’s “Guidance on its enforcement priorities in applying Article 82 (EC)
to abusive exclusionary conduct by dominant undertakings”17.

2.2.4.1. Austria

In Austria, competition is regulated by the new Cartel Act of 2005 (KartellGesetz,

published in the Federal Gazette I 62/2005, in force since the 1st of January
2006). The amendment to the previous law aimed at harmonizing Austrian
substantive competition law with EU competition rules. The objective of the law
is the protection of free competition, although many commentators argue that
the protection of customers is also part of the objectives of the law.
The abuse of dominance is prohibited by section 5 of the Cartel Act and enforcing
authorities are supposed to apply the same standards applied at EU level.
However, Article 4(2) of the Cartel Act provides that the defendant undertaking
bears the burden of proving that it is not dominant on the relevant market if:
1) It has a market share of above 30%, or
2) It holds a market share of more than 5%, provided that it is exposed to
competition of not more than two other competitors; or
3) It holds a market share of more than 5% and belongs to the four biggest
undertakings on the relevant market that have a joint market share of at least
80% (CR4 > 80%).
The definition of the abuse of dominant position contained in Art. 5 of
the Austrian Cartel Act follows the EU provisions. A minor difference can be
found on the fact that at para.1, no.5, of the mentioned Article, it is explicitly
stated that an abuse of a dominant position may consist, inter alia, in the sale of
goods at a price below the purchase/acquisition price.
As reported by experts, abuse of dominance cases have traditionally been
vigorously pursued in Austria18. The investigating authority is the Federal

16 A recent amendment in the Maltese competition law removed a rebuttable presumption of

dominance and a reference to two forms of abusive conduct which are not found in the non-
exhaustive list provided in Article 102 TFEU. See further:
http://ec.europa.eu/competition/ecn/brief/03_2011/mt_act.pdf
17 OJ C 45, 24.2.2009, p. 7–20, published on 24 February 2009.
18 Redlinger and Kuhnert, in http://www.dwpv.com/images/2007_Anitrust_Year_In_Review_-

_ABA_Section_of_International_Law.pdf

PAGE 28 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

competition authority19, and private enforcement is also possible – with the

Austrian Supreme Court acting as the Upper Cartel Court, i.e. the appeals court
against decisions of the Cartel Court (see Annex VI for examples)20.
Private enforcement of competition law rules is possible under several legal
bases. The Cartel Act foresees in § 36 (1) that a third party with a justified interest
can file claims to the Cartel Court. However, the Cartel Act itself contains only
provisions on cease and desist filings concerning anticompetitive behaviour.
Third parties may file a claim for damage compensation based on the Austrian
Unfair Competition Act (UWG) or the Austrian General Civil Code (ABGB –
Allgemeines Bürgerliches Gesetzbuch).
Regarding claims for a damage compensation by third parties, the legal basis for
these actions are §§ 1295 et seq. ABGB. Generally, European competition law
provisions are “safeguard clauses”, which can be subject to compensation claims.
This has the advantage that according to § 1298 ABGB the burden of proof lies
not on the damaged third party (the opposite way remains for claims according
to the UWG). Furthermore, actions for injunctive relief can be based mainly on §
1 UWG in conjunction with § 14 para. 1 or § 15, para. 1 of the UWG. These actions
can be filed by competitors, whose legal sphere was violated by anticompetitive
conduct, and partly by consumers21. Private actions are more frequently brought
before both the specialized Cartel Court and the regular Commercial Courts to
obtain injunctive relief. Before, private actions were mainly aimed at uncovering
and stopping illegal competitive practices, whereas now they are more and more
aimed at recovering the damages suffered as a consequence of the
anticompetitive conduct. We have information about meaningful cases at this
regard, and these are reported in Annex VI, but we do not have precise statistical
information about the number of cases dealt with by the Austrian Cartel
Authority and/or by the Courts, nor about settled cases, since even in the annual
reports of the Austrian Federal Cartel Authority there is no specific information
regarding the total amount of cases referring to certain conducts. However, the
most frequent conducts subject to scrutiny are price discrimination practices,
refusals to supply and tying practices, particularly in the telecommunications
market The least frequent conducts are predatory pricing practices, due to the
difficulties entailed by the standard of proof concerning the alleged infringement.

2.2.4.2. Germany

In Germany, competition is regulated in the German Act against Restraints of

Competition (Gesetz gegen Wettbewerbsbeschränkungen, hereinafter the

19 For an overview of recent cases, see http://www.bwb.gv.at/NR/rdonlyres/39CEDC6D-8ADE-

45EE-9293-C5B5171916BA/34739/Annualreport20072008.pdf;
http://www.bwb.gv.at/NR/rdonlyres/39CEDC6D-8ADE-45EE-9293-
C5B5171916BA/34739/Annualreport20092010.pdf.
20 The Cartel Court acts as public enforcer, its decisions being subject to judicial review as well as

as traditional civil court for claims brought in the framework of private enforcement.
21 Under Austrian law, only the subjects that were directly harmed are entitled to damages. Where

the latter have passed-on damages, the subjects who were indirectly harmed (e.g. indirect
purchasers in cartel cases) can, under certain conditions, sue for damages if they can prove
causation. The “passing-on defence” is allowed. However, to our knowledge there is no final
decision dealing with these issues against the background of a private anti-trust law enforcement
case.

PAGE 29 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

“GWB”) as amended on the 1st July 2005. The question of the objective of the
GWB is very controversial. As stated in Section 1 of the GWB, the main objective
is to assure “freedom of competition” and prevent economic power from
hindering competition. However, some prohibitions (e.g. § 20) appear to be
loosely connected to the objective of freedom of competition, and are more
geared towards the protection of Small and Medium-sized Enterprises. For these
reasons, the standards set forth by some provisions contained in the GWB are
eventually stricter than those indicated in Art 102 TFEU, and will be analysed in
the relevant sections below.
Under German law, and in line with EU competition law, dominance is defined
as a position of economic strength enjoyed by an undertaking, which enables it to
behave independently of its competitors - at least to a certain extent. According
to § 19(2) GWB the test of dominance must be fully and positively assessed with
the help of the given criteria.
According to § 19(2), sentence 1, GWB single dominance exists, where an
undertaking
- has no competitors or is not exposed to any substantial competition (Nr.
1), or
- has a paramount market position in relation to its competitors; for this
purpose, account shall be taken in particular of its market share, its
financial power, its access to supplies or markets, its links with other
undertakings, legal or factual barriers to market entry by other
undertakings, actual or potential competition by undertakings
established within or outside the scope of application of the GWB, its
ability to shift its supply or demand to other goods or commercial
services, as well as the ability of the opposite market side to resort to
other undertakings (Nr. 2).

The criteria set in § 19 (2), sentence 1, Nr. 2, GWB are not cumulative in the sense
that all of them must be prevalent for a finding of dominance.

According to § 19(2), sentence 2, GWB collective dominance exists, where:

two or more undertakings are dominant insofar as no substantial
competition exists between them with respect to certain kinds of goods or
commercial services and they jointly satisfy the conditions of [§ 19 (2)]
sentence 1.

The cases of dominance of (groups of) several undertakings (collective

dominance) are examined by the Bundeskartellamt more often in the context of
merger control (§ 36 GWB), for which § 19(2) and (3) GWB apply.

In the Bundeskartellamt’s work and the case law of the German courts, the
market shares of the firm under investigation and of its competitors - including
their development over the past years - are used as a starting point for the
analysis. German competition law contains - unlike Art. 102 TFEU – restricted
rebuttable market share presumptions under § 19(3) GWB:

1. An undertaking is presumed to be dominant if it has a market share of at

least one third (single dominance).

PAGE 30 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2. A group of undertakings is presumed to be dominant if it:

a. consists of three or fewer undertakings reaching a combined
market share of 50 percent, or
b. consists of five or fewer undertakings reaching a combined
market share of two thirds, unless the undertakings demonstrate
that the conditions of competition may be expected to maintain
substantial competition between them, or that the number of
undertakings has no paramount market position in relation to
the remaining competitors (collective dominance).
The presumption under § 19(3) GWB contains a partly stricter standard of
dominance than those set forth in Art. 102 TFEU in case of “non-liquet” (= “it is
not clear”), i.e. when a fact is not considered proven, nor it can be considered
rebutted, due to insufficient or uncertain evidence. However, it is restricted to
such a situation. To the Bundeskartellamt’s knowledge there has been no case
since the introduction of § 19(3) GWB in 1973, where the presumption has taken
effect by its own force. In view of the corresponding criteria for assessing
dominance under § 19(2) GWB and Art. 102 TFEU and the German case practice
in abuse cases over the years, it cannot be established that the notion of
dominance under § 19 GWB is fulfilled more easily than the one under Art. 102
TFEU. The competent authorities are the Cartel Offices on the federal
(Bundeskartellamt) and the regional level22.
In the last published reporting period 2007-2008 there have been 98 new cases
of abuse of dominant positions investigated the Bundeskartellamt. Only 6 cases
have been concluded with binding decisions of the Bundeskartellamt. The other
cases have been regulated by informal settlements or have been discontinued. In
2009-2010 approximately 70 new abuse cases (approximately 35 concerning
dominance) were investigated by the Bundeskartellamt. Approximately two
cases ended with commitment decisions, one with a prohibition decision, 6 with
the undertakings´ renouncement of the alleged abuse of a dominant position and
8 with a decision stating no grounds for action. Around 20 assumed cases were
not opened or dismissed on other grounds. 29 new cases relating to §19 GWB
were opened and approximately 14 cases were closed with commitment
decisions.
Approximately 8 new boycott cases were initiated with one leading to a fining
decision. Several cases ended with the undertakings’ renouncement of the alleged
conduct. In 2009-2010 approximately 160 new abuse cases – approximately half
of them concerning dominance - have been opened by the Länder competition
authorities (Landeskartellbehörden). Approximately 9 abuse of dominance cases
ended with commitment decisions, 30 with the undertakings´ renouncement of
the alleged abuse of a dominant position and Around 180 pending or new
assumed abuse cases were not opened or dismissed on other grounds. Around

22 Whether the Bundeskartellamt or the relevant Land competition authority

(Landeskartellbehörde) is responsible for enforcing the prohibition of abusive practices depends
on whether the effect of a competition restraint extends beyond the territory of a federal Land (in
this case the Bundeskartellamt is, in principle, the competent authority) or not (in this case the
cartel authorities of the Länder are, in principle, responsible). Indications of possibly abusive
practices pursued by a certain company are brought to the Bundeskartellamt’s attention by
competitors, suppliers and customers.

PAGE 31 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

circa 3 new boycott cases (2 cases dismissed in total) and 9 resale at a loss cases
(1 commitment decision, 1 with the undertakings´ renouncement of the alleged
conduct, 4 cases were dismissed in total) were initiated.23
In 2005, the German legislature amended the GWB, with the specific intention of
facilitating private antitrust enforcement. Although private damages actions
based on cartel infringements are still not widespread, they have become more
common.
§ 33(3) GWB provides that whoever commits an infringement of the provisions of
the GWB or of Articles 101 or 102 TFEU shall be liable for damages arising
therefrom. The party affected by the breach of competition law (the party
concerned) may request a court order that the party that infringed German or
European competition law should refrain from an antitrust violation. This means
that the claimant can either request that the defendant ceases a given activity or
that the defendant performs a certain activity, e.g., to supply the claimant.
Damages claims may either be filed as an action for affirmative relief or as an
action for a declaratory judgment. The action for affirmative relief may be the
appropriate type of action if the plaintiff – at the time the claim is made – is able
to calculate the exact amount of damages suffered as a result of the infringement.
If the plaintiff is unable to quantify the damages suffered due to information only
held by the defendant the plaintiff can bring an action in stages (where the
plaintiff in the first stage brings a non quantified claim for performance in
connection with an action for full disclosure and quantifies the loss in the second
stage). If the plaintiff is not able to determine the exact amount of loss suffered,
the plaintiff may apply for a declaratory judgment that, if successful, will give rise
to the opportunity to sue the defendant for the exact amount of money
corresponding to the loss at a later date. Moreover, a declaratory judgment often
forms the basis for a settlement.
An example of recent judgments of German courts is the decision of the Higher
Regional Court Düsseldorf concerning a damages claim against Deutsche Post for
abusing its dominant position24.
As for the private enforcement, Peyer (2010) in his survey of private antitrust
litigation in Germany between 2005 and 2009 reports the impression “that
litigation often takes place between small or medium sized companies which may
hint at the use of economic dependence rather than dominance in antitrust
proceedings”25. Figure 2 below reports the findings of Peyer (2010), which show
that Sections 19-21 of the GWB – unfortunately not distinguished between abuse
of dominance and abuse of economic dependence – accounted for the lion’s share
of private litigation cases in Germany. Further hints at the possible role of abuse
of economic dependence comes from additional evidence that refusal to deal was
the single most frequent allegation (60 cases) in Peyer’s sample. This, of course,
only suggests that the provisions contained in Sections 19-21 of the GWB are
significantly enforced also through private litigation: however, it does not

23 See: Tasks of the Bundeskartellamt (brochure) -

http://www.bundeskartellamt.de/wEnglisch/Publications/TaskW3DnavidW2647.php - see
statistics on page 37.
24 Higher Regional Court of Düsseldorf, judgment dated 30 September 2009, file number VI – U

(Kart.) 17/08 (V), WuW/E DE/R 2763 – 2769.

25 Peyer (2009) at 42.

PAGE 32 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

constitute sufficient evidence that such enforcement ultimately results in the

application of stricter criteria for unilateral conduct compared to Article 102
TFEU. As matter of fact, German authorities have underlined that the existence
of potentially stricter criteria does not translate, in practice, into a more
aggressive application of competition rules compared to the relevant EU
provisions26.

Figure 1 – Primary allegations in German competition litigation,

regulated v. unregulated sectors

Source: Peyer (2010)

2.2.4.3. Hungary

Another interesting case is Hungary, where competition law provisions are

regulated in the Act LVII of 1996 on the Prohibition of Unfair Restrictive Market
Practices. The Act entered into force on 1 January 1997 and was lastly amended
in 2009 (or, the “Hungarian Competition Act”).”).
In the Hungarian Competition Act, abuse of dominance is approached in a very
similar way to the EU definition and interpretation27. However, another law, Act
CLXIV of 2005 on Trade, entered into force on 1st June 2006, is relevant in this
context. The law covers all trading activities performed in Hungarian territory
and aims at protecting the interests of the traders and the consumers, but also
the total welfare as well.

26 See presentation by Charlotte Zapfe at the III Annual Conference of the Recent Developments
in Competition Enforcement, Autorità Garante della Concorrenza e del Mercato, Rome, 6 May
2011.
27 Section 21 of the Act LVII of 1996 on the Prohibition of Unfair Market Behaviour and the

Restriction of Competition.

PAGE 33 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Subsection 5 of Section 7 of the law introduced a concept akin to abuse of

dominance – the ‘abuse of significant market power’ – which uses a minimum
threshold (yearly net turnover of 100 billion Hungarian Forints) to identify
undertakings holding significant market power. This legal provision is essentially
aimed at addressing cases of abuse of buyer power, by means of different and
stand-alone legislation separate from the competition law. This legislation, in
force since 1 June 2006, explicitly prohibits the abuse of significant market
power against suppliers, and is enforced by the Hungarian Competition
Authority, which uses separate forms for notifications based on the Trade Act.
Companies having significant market power may be required to adopt fair market
practices in dealing with suppliers, to draw up self-regulatory standards or codes
of conduct governing such practices, along with the procedures to be applied in
connection with any violation of these standards and codes.
The Hungarian Competition Authority28 (Gazdasági Versenyhivatal)
(hereinafter referred as the “GVH)”) is the competent enforcing authority.
The Hungarian Competition Authority’s first investigation based on the Act on
Trade dealt with the large retail distribution sector, where the company involved
was Tesco. In this case GVH investigated whether Tesco abused its significant
market power by limiting the number of shelf and display stocking agencies
working for Tesco and by requiring its suppliers to cover many of the costs of
these agencies. Tesco employed third party shelf stocking agencies to put out
products in retail outlets for consumers, and the suppliers had to enter into
contracts with these agencies. Tesco also employed a coordinating agency to
control the work of the shelf stocking agencies. The coordinating agency had to
pay a rent for use of the instruments required for the stocking. The stocking
agencies had to pay hourly rates to the coordinating agency. Finally, the suppliers
had to pay for the services of the stocking agencies. After a while Tesco changed
its policy in relation to stocking agencies and reduced the number of the stocking
agencies from 40-50 to 6 - selected exclusively by the firm. Because of the small
number of the stocking agencies with the obligation to use the coordinating
agency the terms of the contracts between Tesco and the stocking agencies could
result in disadvantageous terms for the suppliers. The GVH established that
Tesco has a significant market power on the relevant market. With its significant
market power, the firm was able to force its suppliers to hire the services of third
party suppliers - namely the stocking agencies, and also to prescribe undue risk
pooling contract conditions resulting in one-sided advantages for Tesco as
against the supplier, meaning in particular the charging of costs, such as storage,
advertising, marketing and other costs to the supplier. As a result of the
investigation, no fine was imposed on Tesco, but the firm has undertaken certain
undertakings for the future. In accordance with the procedure rules in the
Competition Act, in the course of the competition investigation, Tesco agreed
with two undertakings to ensure compliance of its practices with the provisions
of the Act on Trade.
Private enforcement
Before 1 June 2009, legal and natural persons that suffered damage in
connection with violation of the HCA provisions were only able to initiate civil
law proceedings on the basis of the general civil law provisions. As of the

28 In some English translations you may find “Office of Economic Competition”.

PAGE 34 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

enactment of the amendment (1 June 2009), the HCA contains specific

provisions. It now provides the legal bases for private enforcement. Article 92 (1)
of the HCA authorizes the GVH to file civil actions against business entities.
Moreover the Article 93 of the HCA creates the legal background for private
enforcements. This article refers to the general liability rules of the Hungarian
civil law which can be found in Article 339 of Act IV of 1959 on the Civil Code of
the Republic of Hungary.
Therefore, currently two options are available to subjects who want to file a claim
to recover damages suffered as a consequence of competition law infringements:
i) for certain acts of unfair competition indicated in Chapter II of the HCA, such
as the breach of business secrets, calls for boycott, special claims may be brought
exclusively in front of civil courts (the plaintiff may request, inter alia, the
establishment of the infringement, its termination, its discontinuation as well as
compensation for any damages, etc.) on the basis of the provisions contained in
Chapters IV-V of the HCA and/or Article 101-102 of the TFEU..
ii) private enforcement claims may be lodged against an undertaking that
committed an infringement of competition law (including the prohibition of anti-
competitive agreements, the prohibition of the abuse of a dominant position and
unfair commercial practices) on the basis of the provisions of the Hungarian Civil
Code (including those relating to contract law, such as the nullity of contracts,
and tort law, such as compensation for damages provisions).
The HCA expressly stated that a prior decision of the GVH establishing an
infringement of competition law is not a condition precedent for a litigation to
take place for the breach of competition law; so-called “stand-alone” actions are
therefore expressly permitted, while “follow-on” actions (litigation initiated after
a respective decision of the competition authority) are also possible, and are, in
fact, encouraged.
The possibility of private enforcement of competition cases in Hungary is
relatively new. In addition, although theoretically possible, it is unlikely that
private parties will go before a judge without a infringement decision from the
GVH. However, recent statistics of the GVH shows that the number of cases of
abuse of dominant position has been decreasing since 2004 (30 in 2004, 25 in
2005, 33 in 2006, 13 in 2007, 7 in 2008, 1 in 2009, 3 in 2010, but administrative
court review is still pending). In addition to this, it should be considered that the
level of intervention from the Courts in this field is very low in Hungary, since the
jurisprudence in this field is not really developed. These reasons together explain
why there is no case law available at the moment (e.g. in 2009 no new civil court
action on abuse of dominant position was filed).
So far, there has not been any private enforcement of the Act on Trade.

2.2.4.4. Lithuania

Moreover, in Lithuania the first Law on competition was adopted on 15

September 1992 and was later amended in several occasions, including in 2009.
The objective of the legislation is to “protect freedom of fair competition” (Article
1(1)). To what extent, if at all, this objective includes either consumer or total
welfare has never been explicitly discussed in the case law. The enforcement of
prohibition of abuse of a dominant position is entrusted solely to the national

PAGE 35 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

competition authority – the Competition Council of the Republic of Lithuania. All

decisions made by the Competition Council can be appealed to the Vilnius
Regional Administrative Court.
The Law deals with the abuse of dominant position by one or more undertakings
at Art. 9, but provides a definition of ‘dominance’ at Art. 3, which at paragraph 11
reads as follows: “Dominant position” means the position of one or more
undertakings in the relevant market directly facing no competition or enabling
to make a unilateral decisive influence in such relevant market by effectively
restricting competition.’ This paragraph also provides for a legal presumption of
dominance: ‘Unless proven otherwise, an undertaking (except for an
undertaking engaged in retail trade) with the market share of not less than 40%
shall be considered to enjoy a dominant position within the relevant market.
Unless proved otherwise, each of a group of three or a smaller number of
undertakings (except for undertakings engaged in retail trade) with the largest
shares of the relevant market, jointly holding 70% or more of the relevant
market shall be considered to enjoy a dominant position.’
Among the amendments that are important to the assessment of a dominant
position of undertakings, it should be mentioned that the Law passed on 24
September 2009, which came into force on 1 January 2010, modified Article
3(11) in respect to undertakings holding a dominant position in the retail market.
Having considered that undertakings engaged in retail trading may acquire
market power even before they reach the 40% relevant market share established
threshold, a last sentence was added to this paragraph: ‘Unless proved otherwise,
an undertaking engaged in retail trade with the market share of not less than
30% shall be considered to enjoy a dominant position within the relevant
market. Unless proved otherwise, each of a group of three or a smaller number
of undertakings engaged in retail trade with the largest shares of the relevant
market, jointly holding 55% or more of the relevant market shall be considered
to enjoy a dominant position’.
In respect of these undertakings, thus, the threshold on which the presumption
of the dominance is based was reduced to 30% share of the relevant market for
single dominance and to 55% for collective dominance. However, there are no
reported cases where the Competition Council has successfully proven the
existence of dominance of an undertaking with a market share of 30%.
Noteworthy is that the Law on Competition was amended with the 30% market
share presumption specifically for large supermarkets, but this presumption
applies expressis verbis to any undertaking engaging in retail trade29.
Private antitrust actions may be based on Article 50 of the Law on Competition.
The claimant is entitled to the full recovery of damages caused by a defendant’s
unlawful conduct.
A further legal basis for private antitrust actions can be found in general tort law
implemented by the Civil Code. The invalidity of agreements for antitrust reasons
can also be based on provisions of the Civil Code. Claims for compensation of
damages are heard in civil courts in accordance with the Code of Civil Procedure.

29 It must also be recalled that Market shares by themselves do not necessarily entail the
dominance of the undertaking where other circumstances are established that the undertaking is
still facing competition in the market and is in no position to exercise a unilateral decisive
influence on it. See Annex VI for more details.

PAGE 36 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Private antitrust litigation in Lithuania is developing slowly. A relatively large

number of competition law-related cases are tried in courts each year, but private
antitrust cases account for only a small percentage of them. To our knowledge
none of them concerned an abuse of dominant position. Also, private
enforcement of the DPRT provision is for the time being non-existent.
The slow development of private antitrust litigation is partly due to the fact that it
is difficult to prove an infringement of competition rules privately, especially in
complicated, highly technical cases in which proof requires a lot of data. Due to
the complexity of these cases, the courts tend to doubt privately gathered
evidence.

2.2.4.5. Countries with an explicit market share threshold

In addition to these five countries, in other countries thresholds have been set,
which correspond to the same threshold of 40% identified by the European
Commission as a reference minimum value for a finding of dominance30:
 In Bulgaria, a similar threshold was applied in Bulgaria under the previous
Competition Law, in force until 2008 when the new Competition Law
entered into force. The new law, as much as the previous one, covers all
sectors and aims at protecting free competition and market entry. In
particular, Article 17 of the repealed Competition Law set forth a
presumption of dominance when the market share of the undertaking is
above 35%. The new competition law in force does not provide for a market
share threshold, which automatically determines whether an undertaking is
dominant. However, the market share is considered to be one of the main
criteria for the assessment of market power, as stated by the Competition
Commission in the Decision № 624/23.06.2009:“indicator of market power
is the market share of undertaking, and in the process of assessment, its
stability along time and the market shares of the competitors shall be also
taken into account.” However, in case an undertaking has over 70% of the
relevant market, such market share becomes the main criterion to prove the
market power of the undertaking. This has been affirmed by the Commission
for the Protection of Competition in Decision No. 816 as of July 13, 2010
stated that: “the market share of undertaking, especially if over 70%, shall be
the main criterion for assessment of the market power and the possibility for
independent economic conduct”. Accordingly, despite the removal of a
market share threshold for the presumption of dominance, Bulgarian
legislation still allows for stricter standards in the determination of
dominance, in the form of a greater reliance on market shares as indicators of
dominance, which may entail a weaker role of other concepts, such as
potential competition of contestable markets in the identification of

30Such rules may be argued to be stricter than Art. 102 TFEU since the latter does not refer
explicitly to any threshold for establishing dominance, although the Commission Guidance
Document, consistently with the Court of Justice, has declared that a finding of dominance is very
unlikely with market shares below 40%. The National rules analyzed here explicitly consider the
40% threshold to establish a presumption of dominance, which might be more or less easy to
rebut depending on the country.

PAGE 37 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

dominance.31 However, it must be recalled that most of the cases examined

by the Commission for Protection of Competition reportedly refer to
dominance exceeding 80% of the relevant market (e.g. Commission
Decisions No. 81/2006; No. 152/2006; No. 628/2007, etc.).
 In Estonia, Article 13(1) of the Competition Act specifies that a dominant
position is presumed if an undertaking or several undertakings operating in
the same market account for at least 40% of the turnover in the relevant
market32.
 In Poland, Article 4(10) of the Act of 16 February 2007 on Competition and
Consumer Protection – entered into force on 21 April 2007 – does not include
any fixed threshold determining whether an undertaking is dominant.
However, it establishes a presumption of dominance if the undertaking holds
a share of more than 40% of the relevant market. Nevertheless, this aspect
cannot be seen considered as a significantly stricter rule, since the presumed
market share is contestable and further defined conditions must be fulfilled
before dominance is finally established.
Moreover, whilst EU law defines dominance as the ability of an undertaking
to prevent effective competition by affording it the power to behave to an
“appreciable extent” independently, Polish legislation refers to a “significant”
independence. Although the list of possible abuses of dominant positions is
longer in the Polish Act on Competition and Consumer Protection, both the
list of the Polish and European legislation is not enumerative. Thus, all in all,
despite the presence of a presumption of dominance, Polish competition law
cannot be said to adopt significantly stricter standards for the definition of
dominance.
 The Romanian Competition Council (hereinafter the “RCC”) had issued
guidance on the application of Article 6 of the Romanian Competition Law
(hereinafter the “RCL”),33 including on the definition of dominance. The
Regulation was abrogated since August 2010, by the RCC’s President Order
no. 389. Article 2 (3) of the Regulation for the Application of the Provisions of
Articles 5 and 634 defined dominance. Following the decree of the President of
the RCC no. 87 of April 16, 2004, the above regulation had been a major
source for the interpretation and the application of the definition of
dominance under Article 6 RCL. The RCC’s guidance defined dominance in
Article 2(3) of the Regulation for the Application of the Provisions of Articles
5 and 6 of the RCL no. 21/1996 regarding anti-competitive practices.35
Accordingly, the Romanian definition of dominance referred to “the situation
where an economic agent is able, to a considerable extent, to behave
independently from its competitors and customers in the relevant market”.

31 However, with Decision No. 6454 of May 17, 2010 the Supreme Administrative Court
stipulated that “the Rules referring to the determination of the dominant position are provided
for in Article 20 and they represent a complex of different criteria including, but not limited to,
the market share“.
32 According to article 13 (2) of the Competition Act undertakings with special or exclusive rights

or in control of essential facilities specified in §§ 14 and 15 of the Act are deemed to be

undertakings in a dominant position.
33 Romanian Competition Law no. 21 of 1996, OJ I-742 of August, 16, 2005.
34 Published in the Official Journal no. 430 of May 13, 2004.
35 Published in OJ no. 430 of May 13, 2004.

PAGE 38 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Notwithstanding the abrogation of the Regulation, the RCC’s definition of

dominance remains similar to the EU jurisprudential definition given in Case
C-85/76 Hoffman-LaRoche,36 namely, the “ability of an undertaking to
behave to a significant degree independently from its competitors,
customers and consumers” and it has been referred in practice.37 However,
the concept of competitors also includes potential competition. Also, the
definition does not make any reference to consumers. Another difference with
Article 102 TFEU lies in the fact that the concept of dominance does not
require the existence of a “substantial market power over a period of time”
by the allegedly dominant economic agent or firm. Thus, in practice, the RCC
examines the relative size of the competitors’ market shares and the exercise
of its power “over a specific period of time”.38 The RCL no. 21/1996 does not
define any market share thresholds. However, the recent amendment to this
law, particularly to the Article 6 on the prohibition of the abuse of a dominant
position, introduced a relative negative legal presumption of market
dominance.39 According to Article 6(3) of the newly amended RCL, it is
presumed that one or more undertakings are not dominant where its market
share or their joint market shares do not exceed 40% in the relevant market,
during the given period of analysis, and unless such a presumption is refuted.
Market shares give a relevant indication on the market structure, but this
indication is not the only applicable condition in order to define the
dominance. In case of volatile markets, this indication could be fast changed.
Therefore, dominance must be assessed in the context of combined factors.
Taken separately, as done in the cases exemplified in the study, these factors
might not be necessarily determinative. This analysis may lead to the finding
of dominance even in the presence of very low market share thresholds. We
refer to Annex VI for relevant case law in this respect.

2.2.4.6. Countries where the previous market share threshold was abolished

 In Latvia until 2008 the definition of dominance used to refer to two

cumulative criteria – a threshold of 40% and the ability to significantly
hinder, restrict or distort competition in any relevant market for a sufficient
period of time by acting with full or partial independence from competitors,
clients, suppliers or consumers; however, in its decision No. 71 of 20 June
200740, the Latvian Competition Council pointed out that the 40% threshold
restricted its ability to apply competition law in a effective manner, and from
2008 the reference to this threshold has been repealed by the amending law.
 In Portugal the first Competition Act of 1983 foresaw a presumption of
individual dominant position starting at 30% market share (article 14(3)),
and a presumption of collective dominant position starting at 50% (for
markets with 3 or less undertakings) or 65% (for markets with 4 or 5

36 Case C-85/76 Hoffman-LaRoche 1979 ECR 461, 3 CMLR 211, para 38.
37 RCC decision no. 23 Sasha Distribution v Heineken Romania June 25, 2007, para 7.
38 Romanian Competition Council decision no. 77 Atlas Telecom Network Romania v RDS & RCS

of April 25, 2005, para 10 (3).

39 One or more undertakings are presumed not to be dominant if the market share or the joint

market shares in the relevant market do not exceed 40%, until this is refuted.
40 Text of the decision available in Latvian language http://www.kp.gov.lv/uploaded_files/

2007/DV71_2006.pdf, page 23.

PAGE 39 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

undertakings). These presumptions were maintained in the second

Portuguese Competition Act of 1993 (article 3(3)). The current Portuguese
Competition Act, however, has eliminated any such presumption (Article
6(2)). The Lisbon Commercial Court has recently held that these are not
concepts of dominance, but merely examples. It built its own concept of
dominance on the basis of EU case-law that an undertaking holds a dominant
position when its market power is significant and stable in time, granting it
economic power and independence such that it may act on the market
without having to take into account the possible reactions of competitors
and/or of consumers, being able, inter alia, to modify the price of the product
or service to its own benefit. An appeal is pending which may or may not
address this interpretation.41 The limited practice of the Portuguese
Competition Authority in what concerns abuse of a dominant position, and
the non-publication of such decisions, makes it difficult to assess the extent
to which a presumption has been applied in practice. The issue does not
appear to have been addressed in court proceedings either. A recent
judgment has noted that market share is the strongest structural criterion to
be used when assessing the existence of dominance, and it referred to (and
apparently adhered to) EU case law42. Additional data, also on the private
enforcement of all the above-mentioned rules, can be found in Annex VI.

2.1.2 Countries having per se rules for abusive conduct by dominant

undertakings, which are subject to a rule of reason under EU
competition law

Some European countries have enacted competition legislation that differs from
EU competition law in the treatment of individual types of abuses, in some cases
resorting to per se prohibitions for conduct that is subject to a more effects-based
appraisal in EU competition law. An apparent good example is Austria, where
the explicit inclusion of the case of “sales below the purchase cost price” as a case
of abuse of dominant position in Article 5 of the Cartel Act may render the
definition of abuse stricter then EU competition law. However, so far the Cartel
Court and Supreme Cartel Court have interpreted the relevant provision by
recurring to the effects-based principles developed by the European case law with
regard to predatory pricing43. In addition, a number of countries have enacted
legislation aimed at prohibiting per se unilateral conduct such as tying and
bundling. Normally, these rules are found outside national competition
legislation, e.g. in consumer protection laws or in the general commercial or civil
codes, and these provisions apply irrespective of the degree of market power of
the allegedly infringing company. We will describe these national rules below, at
section 2.2.3.

41 Judgment of the Lisbon Commercial Court of 2 March 2010, PT Comunicações (Ducts) case, p.
142
42 Judgment of the Lisbon Commercial Court of 2 March 2010, PT Comunicações (Ducts) case, p.

146-147.
43 Supreme Cartel Court 18.6.1998, 16 Ok 5/98, Supreme Cartel Court 9.10.2000, 16 Ok 6/00;

Supreme Cartel Court, 16.12.2002, 16 Ok 11/02; Supreme Cartel Court 17.10.2005, 16 Ok 43/05;
Supreme Court of Justice 8.4.2008, 4 Ob 23/08y.

PAGE 40 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.1.3 Concluding remarks

Figure 1 below summarizes the main findings of this section, which looked at the
main objectives and scope of competition-related legislation in Member States. A
more detailed analysis is included in Annex VI.
As clearly emerges from the picture, there is a remarkable degree of
fragmentation between national approaches to competition legislation, in
particular as regards the definition of dominance. Countries in red are those
where the definition is significantly stricter, whereas countries coloured in
orange feature only minor differences with the definition of dominance under
Article 102 TFEU.
While acknowledging that drawing a demarcation line between “significantly”
and “slightly” stricter is certainly difficult, in the graphical representation below
we have tried to differentiate between rules that provide clearly stricter criteria
already in the text of the law; and countries where the main difference with EU
legislation is merely that the 40% market share threshold is made explicit in the
text of the law, and as such may be given more weight in the implementation of
the legal provision compared to what occurs in the enforcement of EU
competition rules.
In the case of Portugal, the fact that market share thresholds have been used in
the past, but have been eliminated both in the current competition law and are
used with direct reference to EU case-law by national courts, led us to exclude
this country from the list of Member States in which stricter criteria are used to
define dominance and/or abuse.
As a consequence, in Figure 2 below five countries are considered as adopting
significantly stricter criteria for the definition of dominance and/or abuse
(Austria, Bulgaria, Czech Republic, Germany and Lithuania); in other three
countries (Estonia, Poland, Romania) the national rules only slightly differ from
Article 102 TFEU.

PAGE 41 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Figure 2 ––Countries
Europe Countrieswith stricter definition of dominance and/or
abuse

Significantly Stricter definition of dominance

Slightly Stricter definition of dominance

2.2 Specific types of conduct and legal rules

Besides the existence of diverging criteria and thresholds for the definition of
dominance and/or abuse in national competition legislation, the landscape of
national rules on unilateral conduct also features numerous national legal
provisions that focus on specific types of conduct, often prohibiting conduct
regardless of the market power held by the undertaking concerned. Below, we
look at national rules on abuse of economic dependence/abuse of superior
bargaining power; national rules on tied sales; national provisions on sales below
cost; and specific rules applicable to given sectors of the economy.

2.2.1 Abuse of economic dependence or superior bargaining position

Several countries have enacted rules on abuse of economic dependence or

superior bargaining power, mostly to capture situations in which one contractual
party holds a structural advantage and superior negotiating strength compared to
its counter-party, as can often occur in industrial sub-contracting or, more
generally, in complex vertical production chains44. These rules are sometimes

44 The terms “abuse of superior bargaining power” and “abuse of economic dependence” will be
grouped together as appropriate in this Report. Some countries use the former expression,
whereas others use the latter: in general, rules that deal with either behaviour focus on the
imbalance of bargaining strength between the parties in a commercial relationship (often, a long-
term contract); the term “economic dependence” is meant to identify situations in which this
superior bargaining strength is due to the fact than one of the two parties is significantly
dependent on the commercial relationship with the counter-party; in this respect, this concept is

PAGE 42 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

qualified as purely contract law rules, or as competition rules that target abuse of
a relative dominant position. As reported in Cseres (2010), “Some jurisdictions,
for example Germany, employ specific provisions in their competition law
prohibiting abuse of superior buying power, others employ them in other specific
contexts such as tort liability under commercial code like France, again in other
jurisdictions a private civil remedy exists (Italy) or separate administrative
regulation of retail chains. A separate administrative act is often the legislative
model opted for by the CEECs, like in Hungary and in the Czech Republic. To the
contrary, in Latvia the provision is part of the competition law. The enforcement
of these rules rest with the respective NCAs except in the Slovak Republic where,
when the related provisions were in force, the Slovak Antimonopoly Office
refused to be the controlling body. Below, we provide details on national
provisions that tackle the issue of abuse of economic dependence or superior
bargaining power”45. A more detailed description is contained in Annex VI.

2.2.4.1. Austria

In Austria, the Cartel Act explicitly incorporates the concept of economic

dependence in its definition of dominant position applying to both sides of the
market (buyers and sellers)46. Under Article 4(3) of the Cartel Act a firm is
regarded as dominant when it has a superior position in the market in relation to
its purchaser or supplier. Such position is considered to be present in particular if
these firms are dependent on the maintenance of business relations in order to
avoid severe economic disadvantages. The Cartel Act specifically provides that
“dominance is also given in a situation where a company has superior position
on the market vis-à-vis their suppliers or customers, in particular when the
affected suppliers or customers are dependent on the maintenance of business
relations with this company in order to avoid very heavy financial losses.” The
Austrian concept of “economic dependency” applies to both sides of the market
(buyers and sellers). Recent cases include:
 Case 16Ok6/08 decided by the Supreme Court (Der Oberste Gerichtshof,
hereinafter the “OGH”) on 16 July 2008, in which the plaintiff – a cinema
owner - claimed against the defendant – a film distributor and owner of
several cinemas in Austria – an abuse of its dominant position resolving from
economic dependency. The defendant was obliged to receive certain
“blockbuster” movies by the plaintiff in order to get the revenues necessary
for maintaining its business. The defendant obtained exclusive rights for the
distribution of a movie “Asterix” in Austria and did not wish to rent its rights
to the plaintiff. While the defendant had a monopoly on these rights, the
defendant stated omission due to abuses of a dominant position, as the

narrower than the concept of superior bargaining position, which can occur also due to a broader
range of circumstances as there is no requirement of dependence (e.g. market power, coercion).
See i.a. Jenny (2008), Abuse of dominance, at
http://www.unctad.org/sections/ditc_ccpb/docs/ditc_ccpb0008_en.pdf.
45 KJ Cseres, The impact of Regulation 1/2003 in the new Member States, (2010) 6(2)

CompLRev., 145-182.
46 The Austrian Cartel Act does not explicitly regulate the abuse of superior bargaining position.

However, the formulation in Article 4(3) allows dealing also with abuse of superior bargaining
position in business-to-business relations. It follows that the abuse of a superior bargaining
position is part of the definition of a dominant position in Article 4 of the Austrian Cartel Act.

PAGE 43 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

defendant could promote the movie in his own cinemas, and claimed to
receive a copy of the abovementioned movie. The Cartel Court confirmed the
plaintiff’s claims in first and second instance, despite the fact that there was
no factual and objective reason against distributing or ordering another copy
of the movie. According to the Court, the defendant abused both his
monopoly regarding the movie and the plaintiff’s dependency.
 Case 16Ok5/09 decided on 3 June 2009, in which the plaintiff – a creamery
and milk producer – claimed against the defendant – the fourth biggest
creamery in Austria – an abuse of its dominant position resolving from
economic dependency. The defendant was obliged to supply the plaintiff with
milk and bottles over a certain period of time. By the end of 2007, the
defendant raised the prices for its goods above the average. The plaintiff had
no alternative due to the lack of other suppliers to his place of business and
ran the risk of losses. When the defendant prolonged the deadline of supply
without prior notice, which has not been accepted by the plaintiff, a part of
the hitherto agreed goods has not been delivered (caps for the bottles). The
Cartel Court negated the plaintiff’s claims in first and second instance due to
wrong management without a decisive influence of the defendant’s actions.
The increase in prices had been publicised before the prolongation of the
contract and the plaintiff may not make his business development dependent
on just the defendant’s future services.
The investigating authority is the Federal cartel authority. The Upper Regional
Court acts as Cartel Court. There are no finalized decisions on the abuse of
superior bargaining position. As a matter of fact, the Austrian Federal
Competition Authority started investigations in the food commerce sector with
particular focus on the issue of buyer power of big supermarket chains in the
Austrian market (see below, Section 2.3.1)47. Against this background, there is
evidence of cases being resolved by the national competition authority through
the announcement of commitments adopted by the undertaking under
investigation48 49. Unfortunately, it was not possible to obtain information about
the number of settled cases. Private enforcement is also possible but we have no
specific information about the number of legal actions initiated.

47 As a matter of fact, the recent ICN report on abuse of a superior bargaining position,
completed in 2008, the Austrian Competition Authority reported no case law on this subject
matter. See http://www.internationalcompetitionnetwork.org/uploads/library/doc386.pdf, page
20.
48 Structural and behavioural commitments are quite common. Structural commitments include,

for example, the abandonment of stores and branches in favour of a competing undertaking.
Another structural commitment would be the reassurance not to close down a production facility.
Behavioural commitments would be, for example, the reassurance of the notifying parties to offer
products of a competitor on a wholesale level on comparable and non-discriminating conditions.
49 Article 27(1) of the Cartel Act allows the Cartel Court to replace the measures according to

Article 26 by announcing binding commitments of the participating undertakings and

consolidation of undertakings as legally binding, if it is to be expected that these commitments
exclude henceforward violations. Recent case law suggests that both structural and behavioural
commitments are quite common. Structural commitments include, for example, the
abandonment of stores and branches in favour of a competing undertaking. Another structural
commitment would be the reassurance not to close down a production facility. Behavioural
commitments would be, for example, the reassurance of the notifying parties to offer products of
a competitor on a wholesale level on comparable and non-discriminating conditions.

PAGE 44 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Private enforcement of the prohibition of abuse of economic dependence can

occur on the basis of the provisions indicated in the previous section. There are
several examples of private enforcement of the provision regulating the abuse of
dominant position under Article 4(3). Two main cases that can be considered
representative of the private enforcement in relation to the abuse of dominant
position: OGH –16Ok5/09 and OGH - 16Ok6/08. Details of the cases are
provided in Annex VI.
In addition, behaviour that can be defined as abuse of superior bargaining
position is tackled by the Austrian Federal Law for the Improvement of Local
Supplies and the Competitive Conditions (“Nahversorgungsgesetz”, NVG),
which prohibits a number of practices, e.g., discriminatory practices or
demanding payments or services without equivalent. This Law - even though
primarily intended to apply to the grocery sector - applies to all other sectors50, if
not stated otherwise (in § 2a NVG, a specific provision for gas and electricity
market). And although it is not primarily geared to protect competition, we
report it only for the sake of completeness. The provisions included in the NVG
are subject to scrutiny by the Austrian Federal cartel authority, although the only
possible sanction is a cease and desist order, with no fines51.
On 16 July, 2008, the Austrian Supreme Court ruled that a supplier having a
certain level of market power must apply similar conditions to all undertakings
that buy and resell goods (i.e. authorized resellers) irrespective of whether the
goods have undergone processing or working prior to reselling them. The
defendant, the Bavarian public law institution “Bayrische Staatsforsten”,
manages Bavaria’s state-owned forests. The Bavarian state forests supplied
the Klauser Group, an undertaking operating five sawmills in Bavaria, with
sawtimber on conditions very favourable to the Klauser Group. The claimant, the
Austrian association of timber industry, filed an application with the Austrian
Cartel Court asking for an injunction that would stop the discriminatory
behaviour, based on § 2 of the NVG. Rejecting arguments that § 2 NVG should be
applied to undertakings having market power only, the Supreme Court holds that
the said provision in the NVG does not contain such a restriction. The Supreme
Court also stressed that such a restriction would leave no independent scope for §
2 NVG beyond general Austrian Cartel Law. That said, the Supreme Court does
not elaborate on whether the scope of § 2 NVG - the wording of the provision
obliges any supplier, no matter how powerful he is - should be interpreted in a
way that the provision presupposes at least a certain amount of market power.
Supreme Court refers to two earlier judgments, in which it held that the aim of
the NVG is to ensure that small customers are supplied on the same terms as big
customers in order to stay competitive52; and that small customers may not be
put at a disadvantage53. The Supreme Court concluded that, though the
enactment of the NVG and the rulings quoted refer to the food trade, § 1 and § 2
NVG pursue broader objectives, as confirmed by the heading of the said
provisions (“fair business conduct”).

50 The last case, where an application has been filed on the basis of the NVG before the Cartel
Court concerned the timber industry. KOG 9.5.2010, 16 Ok 1/10
51 See the Report on abuse of superior bargaining position of the ICN (2008), available online
at http://www.internationalcompetitionnetwork.org/uploads/library/doc386.pdf.
52 Austrian Supreme Court (Oberste Gerichtshof), 21 January, 2003, Case 4 Ob 210/02i.
53 Austrian Supreme Court (Oberste Gerichtshof), 3 April, 2001, Case 4 Ob 34/01f).

PAGE 45 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

In line with this orientation, the Austrian Higher Cartel Court has recently
decided that the NVG is a complementary legislation to the antitrust rules and is
to be applied also to foreign acts if they significantly affect the Austrian market54.
More importantly, the Court allowed the application of Section 2 of the NVG,
which generally prohibits undertakings from offering discriminating sales
conditions to commercial resellers, beyond its teleological core, the food retail
industry, and offered a very broad interpretation of ‘commercial resellers’,
including effectively all undertakings with the exception of end-users. This
decision was widely understood (and criticized) as a general prohibition to
discriminate in business-to-business relationships55.

2.2.4.2. Cyprus

The current Competition Law of Cyprus at Article 6(2) prohibits the abusive
exploitation of a relationship of economic dependence. This provision was
introduced in the law on the protection of competition (Law 207/89) in exactly
the same form as it currently exists nowadays and was first introduced by the
amending law, Law 111(I)/ 199956. As evidenced by the wording of Article 6(2) of
the Law, the following stakeholders are intended to be protected: any
undertaking which has the capacity of a customer, supplier, producer,
representative, distributor or business partner. Also, three elements must be
proved in order for Article 6(2) of the Competition Law to be applicable: i) a
relationship of economic dependence; ii) an abuse of such a relationship by the
dominant party in the relationship; iii) and the absence of an equivalent
alternative solution to which the abused may resort to.
The enforcing authority is the Cyprus competition commission. However, the
competition authority has applied the section on the prohibition of an abusive
exploitation of a relationship of economic dependence under Article 6(2) only in
11 cases over a period of 21 years. There is no case where the competition
authority has examined ex officio an infringement of the relevant section,
contrary to the ex officio investigations which it has initiated over the years
against concerted practices and abuses of a dominant position.
Nevertheless, as regards the complaints brought on the basis of this section
which prohibits an abusive exploitation of a relationship of economic
dependence, in all cases save for the cases where the complainant withdrew its
complaint, the competition authority has made a finding of an infringement.

54 Austrian Supreme Court as Higher Cartel Court, 16 Jul. 2008 – 16 Ok 3/08, available at
www.ris.bka.gv.at
55 Austrian Supreme Court as Higher Cartel Court, 9 Jun. 2010 – 16 Ok 1/10, available at

www.ris.bka.gv.at.
56 The relevant article reads as follows: ‘Any abuse by one or more undertakings, of a relationship

of economic dependence where an undertaking stands compared to that or those undertakings,

which has the position of a customer, supplier, producer, representative, distributor or business
partner, even vis-à-vis one specific type of goods or services and which does not possess an
equivalent alternative solution, is prohibited. This abuse of a relationship of economic
dependence may, in particular, be found in the imposition of unfair trading conditions, in the
application of discretionary treatment, in the interruption of trade relationships by assumption or
transfer of the activities developed within these trade relationships in a way which substantially
affects competition or in the sudden and inexcusable interruption of long-term trade
relationships.’

PAGE 46 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Further, only two cases involving this section (covering both dominant and non-
dominant companies) have been appealed to the Supreme Court, where in one
case the Court did not examine the merits of abusive exploitation (Tengerakis),
whilst in the second one the decision of the competition authority was upheld,
both at first and at second instance (Loizou v Hellenic Petroleum Cyprus
Limited). We do not have information about any settlement case. Private
enforcement is possible, but we could not collect any evidence of cases where
Article 6(2) was at stake.

2.2.4.3. The Czech Republic

In the Czech Republic, the Act No. 395/2009 Coll. on Significant Market Power
in the Sale of Agricultural and Food Products and its Abuse (hereinafter the
“Significant Market Power Act”) enacted on 3 November 2009 and in force since
1 February 2010 addresses abuse of economic dependence. Despite its name, it is
based upon the “economic dependence” concept (also with some elements of the
“superior bargaining position” concept). The Significant Market Power Act
concerns only sales of agricultural and food products. It was enacted to tackle
commercial retail chains allegedly abusing their bargaining position against
smaller suppliers from the agricultural and food sector. The Act was enacted to
protect these. It is not necessary for the supplier to be a “smaller” undertaking
(an SME), as the Act protects all suppliers in the agricultural and food sector
under the following conditions. The Act only applies to retail chains with a
turnover above CZK 5 billion. The Office for the Protection of Competition will be
responsible for supervising compliance with the legislation. Where it declares
that a retailer has abused a significant market power, it may impose a fine of up
to CZK 10 million or up to 10% of its net annual turnover.
Under Article 3, para. 1, of the Act, there are two main conditions that must both
be fulfilled before the legislation applies:
 The supplier must be in a position of economic dependence on the purchaser
as a result of the market situation; and
 The purchaser is able to impose unilaterally beneficial trade conditions on the
supplier (that is, it has superior bargaining power).
Under Article 3(2) of the Significant Market Power Act, “significant market
power shall be deemed to be a relation between a buyer and a supplier [...] and
in which the buyer may impose unilaterally beneficial trade conditions on the
supplier”.
There are six categories of prohibited behaviour, each the subject of detailed
rules in an appendix to the legislation:
 Rules for issuing invoices;
 General business terms and conditions;
 Conditions in contracts between retail chains and suppliers;
 Conditions of sale;
 Prohibited practices between retail chains and suppliers;
 Other customs in business dealings between retail chains and suppliers.

PAGE 47 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

One of the most criticized aspects of the Act is the time limit to pay within 30
days of delivery. In addition, it is not permissible to sell at a cost lower than the
purchase price (see below, Section 2.2.2). A distributor cannot claim fees before
an actual order is made, or automatically profit from more favourable terms than
the supplier may have agreed with other competitors. Furthermore, a distributor
will be obliged to compensate a supplier for any financial losses incurred
following an attempt, whether successful or not, to gain any benefit or payment
that is not derived from any actual service, or is disproportionate to the services
provided57.
The enforcing body is the Czech national competition authority. Cases dealt with
by the authority under the new legislation are estimated to be fewer than 10 by
the consulted expert. According to the information publicly available, 5
administrative proceedings have been initiated against following retail chains:
Kaufland (2 administrative proceedings started in 2010 have been joined to a
single proceeding), Ahold (proceeding started in 2010) and 3 new proceedings
against Tesco, Globus and Lidl (because of due date of invoices longer than 30
days). In the first case mentioned, Kaufland was imposed a fine amounting to
approx. CZK 14 million (approx. EUR 560,000). Kaufland breached the Act
when it negotiated payment terms longer than 30 days with an absolute majority
of its suppliers. In addition to the financial penalties the Office imposed on
Kaufland also corrective measures requiring removing from their contracts with
suppliers all provisions which are inconsistent with the law within 90 days from
the effective date of the decision. The proceedings against Ahold, however, was
discontinued as the company fulfilled all obligations imposed by the Office. These
latter proceedings are still pending so no final decision has been issued.
Under article 6, par. 2, of the Significant Market Power Act, the Office may decide
to terminate proceedings on condition that the parties to the proceedings jointly
propose commitments in favour of restoring effective competition. A decision
under article 6, par. 2, of the Significant Market Power Act may only be reached if
(i) the commitments proposed are sufficient, (ii) the harmful situation will be
eliminated by their fulfilment, and (iii) the abuse of significant market power did
not result in a substantial distortion of competition. There haven’t been
settlement cases so far.
The Minister of Industry has recently proposed to the Government to repeal
completely the Significant Market Power Act at the end of 2011. The Czech
Competition Office has recently begun a public consultation on the proposed
abolition of the Significant Market Power Act and related draft amendments to
the Czech Competition Act and the Act on Prices58. In fact, the ÚOHS, after
finding unfair practices between retailers and their suppliers, feels that there is
the need for better regulation of significant market power. At the same time, the
UOHS feels it is important to write a law that will not restrict businesses or create
an unpredictable environment. For these reasons, the UOHS would like to meld

57 This outcome is consistent with a decision of the Supreme Court of the Czech Republic on 30
April 2009 (23 Cdo 2184/2007). In that matter, a supplier was obliged under an agreement with
a distributor to contribute to the latter’s investments and the cost of opening a new store, without
receiving anything in return. In addition, there was no agreement that stipulated the amount of
goods the distributor was obliged to buy from the supplier or for how long. In ruling in favor of
the supplier, the Supreme Court stated, “there is no justification for such a relationship between
two independent businesses, where one contributes to the operating costs of another.”
58 See Clifford Chance, Antitrust Review, March-April 2011.

PAGE 48 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

these responsibilities into the Act on the Protection of Competition and the Act
on Prices.
The Significant Market Power Act has no provisions concerning its private
enforcement. The private enforcement of the specific duties set forth by
Significant Market Power Act might be theoretically realised based on general
rules on the compensation for damage or unfair competition. However, we are
not aware of any case of private enforcement of Significant Market Power Act
rules so far. Therefore, we cannot consider the rate of success of plaintiffs or the
easiness of meeting the burden of proof. Also, for the above-mentioned reason,
and due to the absence of any administrative decision appealed before the courts,
there are no cases we can consider as representatives of the case law on this
matter.
The Significant Market Power Act has no provisions concerning its private
enforcement. However, the private enforcement of the rules of the Significant
Market Power Act is possible based on general private law statutes described in
the previous section. To our knowledge there has not been any single case of the
private enforcement of Significant Market Power Act in the Czech Republic. The
Act is however relatively recent – it came into force on 1 February 2010 –
therefore, if any action has been filed, it is probably pending before lower courts
whose decisions are not available.
Some commentators emphasized the difficulties in the implementation of the Act
on Significant Market Power due to the fact that the list of prohibited activities is
extensive and difficult to follow. Based on such criticism, it was proposed to
repeal the Act and to replace it with new provisions contained in the Act on
Protection of Competition and in the Act on Prices.
However, all the attempts to repeal this Act ultimately failed. Therefore the Act is
currently in force in the original version that was first enacted, even if there are
still some political discussions regarding possible changes or adjustments.

Other than these provisions related to the “Significant Market Power Act”, there
is no explicit provision under Czech law on the abuse of a superior bargaining
position. However, Article 2, par. 3, of the Pricing Act of 1990, which entered into
force on 1 January, 1991, states that “neither the seller nor the buyer can abuse
its [more advantageous] economic position to gain an unfair profit” (since 1991,
the wording of this provision has been slightly modified several times). This
provision was last amended by Act No. 403/2009 Coll. which entered in force on
18 November, 2009 and which was intended to further clarify the provision of
article 2 par. 3 of the Pricing Act. From this last amendment of the Act on
Pricing, article 2 par. 4 defines the term “more advantageous economic position”;
article 2 par. 5 determines what “unfair profit” is. This amendment added the
words “more advantageous” in the definition. The above provisions are intended
to protect all parties (both businesses and consumers) which have a weaker
economic position as a result of inappropriate (unfair) pricing by undertakings
which have a substantially more advantageous market position.
The decisions of the competent body, the Ministry of Finance, are not publicly
available. Only appeals and subsequent judicial reviews of administrative
decisions by the Supreme Administrative Court are publicly available. However,
it is possible to affirm with a certain degree of certainty that the provisions of

PAGE 49 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

article 2 par. 3 of the Act on Pricing are not enforced very often. According to the
Annual Reports on Price Control prepared by the Ministry of Finance and the
Government, there were eight inspections carried out in the whole Czech
Republic by the Tax Directorates in 2008 (12 in 2007, 13 in 2006, 16 in 2005); in
two cases in 2008 a breach was revealed and a fine was imposed (6 in 2007, 6 in
2006, 6 in 2005). Therefore, in total in 2005, 2006, 2007 and 2008, 20 decisions
were issued by the Tax Directorates declaring that article 2 par. 3 of the Act on
Pricing had been breached and imposing a fine. Seven of these 20 decisions were
overturned on appeal. Some were, however, affirmed later at first instance.
There are no specific private enforcement provisions related to the Article 2 par.
3 of the Act on Pricing. However, the enforcement of these provisions is in
general possible based on following general rules:
- the agreement which is contrary to the aforementioned provision is void
for inappropriate pricing,
- it is possible to claim for the unjust enrichment in case of breach of the
aforementioned provision (or, in certain case, to claim for the
compensation for damage).
According to our information, the private enforcement of these rules is very rare.
There is only one case related to the private enforcement of the aforementioned
rule published in the public web database of the Supreme Court – Judgment No.
29 Cdo 1102/2008. The Supreme Court dealt with the alleged excessive price
charged by a constructor to a client for the construction of a family house. The
Supreme Court cancelled, upon the client’s request, the Appeal Court’s decision
based on the fact it omitted to consider the case (and the contract for work’s legal
validity) under the abuse of economic position rules set forth in the Act on
Pricing (Article 2 par. 3). However, the Supreme Court emphasized that it is
necessary to prove that the contractor really had a specific economic position to
abuse and that “it is not possible to conclude on the existence of the abuse of
economic position based merely on the constructor’s position in the contractual
relationship (that is, that the constructor is in the position of a professional
dealing with a consumer)”. Nevertheless, the Supreme Court gave no specific
guidance on how to consider the existence of such a position. Moreover, the case
was decided under the legal rules existing before the amendment of the Act on
Pricing No. 403/2009 Coll. For this reason, it is difficult to provide an evaluation
of level of enforcement of the provision and of the rate of success of plaintiffs.

2.2.4.4. France

In France, abuse of economic dependence was dealt originally under legislation

on restrictive trade practices governed by civil law within Article L. 442-6 of the
Code of commercial law and Article L. 420-2 of the French Code of Commercial
Law. Article L. 420-2 of the Commercial code was modified by law no. 420 the 15
of May 2001 and Article L. 442-6 of the Commercial code has been modified by
the law no. 874/ 2010 of the 27th of July 2010. The latter no longer contains any
provision on abuse of economic dependence. Article L. 442-6 of the Code of
commercial Law established the responsibility of the undertakings that "submit
or attempt to submit a trading partner requirement creating a significant
imbalance in rights and obligations of Parties" and requires it to repair the
resulting damage. The abuse of economic dependence/superior bargaining

PAGE 50 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

position is now exclusively tackled by L. 420-2 of the Code of commercial Law.

However, some older cases still refer to abuse of superior bargaining position
under L. 442-6-I of the commercial code. The provisions cover all sectors and
aim at protecting trade partners.
The Commission d’examen des pratiques commerciales, in its notice on abuses
in commercial relations in a question related to the “significant imbalance”,
states that the client must not use its superior bargaining position to ask
systematically to its supplier for a price decrease for the only reason that the
supplier sold its product to a competitive distributor59. According to the General
Directorate for Competition Policy, Consumer Affairs and Fraud Control
(DGCCRF), the fact that an undertaking imposes on a trading partner obligations
that create a significant contractual imbalance does not automatically entail that
the undertaking holds a superior bargaining power. At the same time, the
DGCCRF does not consider that the superior bargaining power is a precondition
of application of Article L. 442-6, I, 2 of the Commercial Code. Indeed the
dependence relation no longer seems to be a requirement for the application of
Article L 442-6 but only a condition of appreciation of the imbalance as
confirmed also by the case law (TA Lille 6 January 2010).
Pursuant to Article L. 420-260, “…The abuse by an undertaking or group of
undertakings of the state of economic dependence in which a client or supplier
undertaking finds itself in respect of the above shall also be prohibited when it is
likely to affect the operation or structure of competition. These abuses may in
particular consist of refusals to sell, linked sales or the discriminatory practices
referred to in Article L. 442-6”. The provision is intended to repress the practice
of unbalanced contracts implemented by an undertaking or group of
undertakings, who dominate in one or more markets but hold a dominant
position in the market as a whole. The Article L. 420-2 is enforced by the
Autorité de la concurrence (previously Conseil de la concurrence), which
indicated that the state of economic dependence stems from an aggregation of
cumulative criteria: (i) notoriety of the supplier's brand; (ii) importance of the
supplier's market share; (iii) significance of the supplier's market share in the
sales figures of the company in question, provided this market share is not the
result of a deliberate choice by the corporate customer; (iv) difficulty for the
company to find other suppliers of equivalent products.61
Very few cases of abuse of economic dependence have been rendered (under 10),
since the four criteria required by the case law are rarely met. Also, the burden of
proof obviously falls on the "abused" company (e.g., Cour de cassation, 12
January 1999) and, in practical terms, the assessment of the existence of a state
of economic dependency (and, further, of an abuse of it) is very restrictively made
and require economic evidence and in-depth economic and financial analysis on
a case-by case basis. Even in the case where an economic dependence has been
demonstrated by the Competition authority like in the Trivial Pursuit case 89-D-
38, the Court of appeal overturned the judgment. This explains why the
enforcement of article L. 420-2 II of the Commercial Code has been limited

59 See http://www.pratiques-commerciales.minefi.gouv.fr/questions/abus.htm.
60 Act No 420 of 15 May 2001, Article 66, Official Gazette of 16 May 2001.
61 Nollet L. (2003), “France: Anticompetitive Practices”, ECLR, 24(7), N116-117. The last criterion

is normally the decisive one.

PAGE 51 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

together with a lack of complaints due to dependent undertakings’ fears of

reprisal.
There are no specific settlement provisions for the abuse of economic
dependence. However, settlement is possible for abuse of economic dependence
case based on the general settlement procedure called ‘negotiated settlement
route’ under Article L. 464-2 III of the Commercial Code, where undertakings
can benefit from fine reductions as a result of a party’s not contesting the
existence of the alleged practices, which applies to cartels and abuse of dominant
position. In this case, the maximum fine that may be imposed is reduced by half.
In addition to this reduced fine ceiling, if the undertakings also offer
commitments to modify their behaviour in the future, the Autorité may grant a
reduction to the actual fine. Undertakings may also, pursuant to Ordinance
2004-1173 of 4 November 2004, offer commitments to remedy the situation and
avoid a decision ruling on the existence of an infringement under Article L. 464-2
I of the Commercial Code.
It should be recalled that pursuant the Ordinance of 13 November 2009, that
implements the Law on the Modernization of the Economy (LME) of 4 August
2008 (which re-designed the competences of the Conseil de la concurrence, now
Autorité de la concurrence) provides that the French minister of economy is
entitled to settle and order measures as regards practices that only affect local
markets. The minister is in charge of such cases if they do not come under
articles 101 and 102 TFEU and if the combined turnover in France of all the
undertakings concerned does not exceed €100 million and the turnover of each
of the undertakings in France does not exceed €50 million.
We do not have specific information about settled cases on abuse of economic
dependence62.
In terms of private enforcement, as a first general comment, judges have always
had a very restrictive attitude towards plaintiffs claiming the existence of a
situation of economic dependence under art. L. 420-2 of the Code of Commerce.
For example, the Cour de Cassation, in an appeal brought by a distributor of
audiovisual products, stated that for a distributor, a state of economic
dependence was defined as the situation in which an undertaking had no
opportunity at all of replacing its supplier or suppliers so as to meet its demand
for stock on comparable technical and economic terms. The mere fact that a
distributor obtained a very large proportion, or even all, of its stock from a single
supplier was not enough to establish a state of economic dependence within the
meaning of Art. L. 420-2 of the Commercial Code.63
However, in some cases, courts have accepted plaintiffs’ claims. In a case
involving a franchise agreement for a supermarket operated under a trade name
belonging to the Promodès Group, the Cour de Cassation concluded, first of all,
that the franchisee was in a position of economic dependence as its purchases of
supplies from the franchiser represented two thirds of the value of its turnover.
The Court then held that the franchisor abused this dependence by taking

62 However, we would like to recall here the case France Télécom of 2007 (decision 07-D-33 of
15 October 2007), where the company, accused of having abused its dominant position,
benefitted from the negotiated settlement procedure, as an example of the application of this
procedure in relation to unilateral conducts.
63 Concurrence SA v Sony SA, 2005, E.C.C. 4.

PAGE 52 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

responsibility for the management and accounting services of the franchised

store, in return for a substantial increase in the franchise fees. In addition, the
franchisee was obliged to place orders without prior knowledge of purchase
prices.64 Thus, the franchisor acted in a manner contrary to the franchisee's
interests. The Court also held that, given that the agreement was a permanent
contract, the franchisee was unable to safeguard itself from the influence of the
franchisor, as termination of the agreement would inevitably lead to the
termination of the management leasing agreement thus resulting in the
franchisee being effectively prevented from using any alternative sources of
supply.65
For what concerns the abuse of a superior bargaining position, the approach
adopted by French courts has traditionally been less restrictive than the one
adopted for abused of economic dependence. This is mostly due to the fact that
Article L. 442-6 had been created exactly with the purpose to protect smaller
retailers against large chains. Before the rule was abrogated, there were
reportedly numerous cases based on this provision.
Today there are not many cases concerning the abuse of superior bargaining
position which have been recognized by either the Autorité or one of the
competent civil or commercial courts. One of them is the TC Roubaix-Tourcoing
case. As regards settlement provisions for cases of abuse of superior bargaining
position pursuant to Article L. 464-7, the DGCCRF, which is a department of the
Minister for Economic Affairs, can require companies to put an end to the
practices referred to in Articles L. 420-1, L. 420-2 and L. 420-5 when (i) the
practice affects a “local market”; (ii) the practice does not fall within the purview
of Articles 81 and 82 EC; and (iii) the turnover of each involved firm is below €50
millions and their aggregate turnover is below €100 millions. The DGCCRF may
also, under the same conditions, propose a compromise. Private enforcement is
possible on the basis of tortious or contractual liability disciplined in the French
Civil Code (article 1382 et seq. of the Civil Code for tortious liability, article 1134
et seq. of the Civil Code for contractual liability), but reportedly underdeveloped
in France66. In Annex VI we report a number of private cases in which the
prohibition of abuse of economic dependence was at stake.

2.2.4.5. Germany

Germany was the first country in Europe to introduce provisions on the abuse of
economic dependence back in 1973. The motivation for such an introduction was
essentially the need to prevent big oil corporations from discriminating against
small oil stations during the oil crisis of the early 1970s. Essentially, the
prohibition of abuse of economic dependence applies to dominant undertakings,
but extends to non-dominant ones whenever the latter deal with small and
medium-sized enterprises (SMEs), which are explicitly mentioned as the
beneficiaries of this set of legal provisions. Generally, a SME is considered to be

64 The franchisor also attempted and in fact obtained a power of attorney and a banking
signature from its franchisee.
65 Utzschneider Y. (1998), “France: franchise agreement - position of economic dependence”,
ECLR, 19(5), N82.
66 See, i.a., the Report on France of the Global Legal Group’s International Comparative Legal

Guide to dominance, at http://www.iclg.co.uk/khadmin/Publications/pdf/2977.pdf.

PAGE 53 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

dependent if it cannot reasonably switch to other suppliers or purchasers, or

switching is not sufficiently possible. More than three decades of case law have
led to the establishment of different types of economic dependence: (i)
dependence on a product range or on a particularly strong (must-have) brand;
(ii) business-related dependence, mostly linked to long-term contractual
relationships; (iii) shortage dependence67; and (iv) technical dependence68.
A number of relevant provisions are found in German law.
 Rules applicable only to dominant undertakings. Sections 19(1), 19(4) and
20(1) GWB deal with the abuse of a dominant position, i.e. exclusionary,
exploitative and discriminatory conduct. For the pruposes of enforcement of
these provisions, economic dependence and superior bargaining position can
only constitute an element of dominance, and as such would be relevant for a
finding of anticompetitive conduct. §§ 19(1) and (4) and 20(1) GWB thus
cover only dominant companies69.
 Rules applicable also to non-dominant undertakings. Other provisions
contemplate conducts which can be implemented both by undertakings
holding market power as specified by § 20(2) and 20(3) GWB70; and by non-
dominant undertakings with a “superior market power in relation to small
and medium-sized competitors”, as specified by Art. 20(4) GWB71.
Private enforcement of this rule of this rule is very much developed.
The court practice on this provision starts with the well-known case Rossignol
(WUW/E BGH 1391) of 1975, where the Court of Appeal held that Rossignol, a
distributor of sporty items, had a relatively dominant market position due to the
reputation of the trademark of its products. Being in this sense a leader in the
market, the Court held that Rossignol was obliged to supply the retailer. This

67 E.g., when a business depends, in shortage periods, on companies that have access to raw
materials and who favour customers related to them over unrelated customers.
68 See Kellezi (2008), at 61-62.
69 As stated in Section 2.1.1 above, the definition of dominance adopted in Germany is – at least in

principle – stricter than the corresponding EU definition at Art. 102 TFEU and related case law.
70 Sections 20(2) and 20(3), sentence 2, GWB deal with undertakings holding market power
and hindering or discriminating against dependent suppliers and purchasers. Section 20(3) GWB
deals with the invitation by dominant or powerful buyers to dependent companies to grant
advantages (actual or potential passive discrimination).
71 §20 (1) GWB states that dominant undertakings shall not directly or indirectly hinder in an

unfair manner another undertaking in business activities which are usually open to similar
undertakings, nor directly or indirectly treat it differently from similar undertakings without any
objective justification; according to § 20(2), paragraph 1 shall also apply to undertakings and
associations of undertakings insofar as small or medium-sized enterprises as suppliers or
purchasers of certain kinds of goods or commercial services depend on them in such a way that
sufficient and reasonable possibilities of resorting to other undertakings do not exist. This means
that §20(2) applies also to non-dominant companies. A supplier of a certain type of goods or
commercial services is presumed to depend on a purchaser, if, in addition to the price reductions
or other considerations customary in the trade, the purchaser regularly receives special benefits
not granted to similar purchasers (§ 20(2), sentence 2 GWB). Those enterprises and associations
are also prohibited from using their market position to cause other enterprises in business
activities to accord them preferential terms in the absence of facts justifying such terms,
according to § 20 (3) GWB. Cases such as Rossignol and subsequent cases have led to the
clarification that a situation of economic dependence is compatible with a situation of
“considerable competition” in the relevant market: accordingly, the provision goes beyond the
scope of application of Article 102 TFEU. See KZR 1/75, Rossignol, November 20, 1975, WuW/E
BGH page 1391.

PAGE 54 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

case gave rise to the case law on abuse of economic dependence. The progeny of
this case has been classified as referring to the s. c. abus de dependence de
gamme72. Another sector of application of such provision was the market for TVs.
Many producers have been involved in court cases: Nordmende, Graetz, Saba,
Telefunken, Grundig, Lowewe Opta, Philips and Sony. They were all producing
high quality TVs for quite high prices compared to the prices charged for other
imported products. Individually each producer had a very small market share.
Thus, they were not dominant, non even collectively. At the same, it was clear
that for a specialized retailer it was fundamental to have these renowned
trademarks in order for him to be competitive in the market. Based on this, the
German case law has created also the s. c. dependence from the leader73. IN all
these cases, the situation of dependence is evaluated in light of the reputation of
the brand in the market, of the marketing activity performed by the producer,
fidelity of consumers to the brand, etc. These criteria give an idea of the injury
that the dependent undertaking would suffer, in terms of damage to its image, if
denied access to such products. Dependence is also found in cases dealing with
the termination of long-standing commercial relations. This is particularly the
case for those undertakings that base all, or most of, their economic activity on
one supplier only, as it happens often for car dealers. However, German Courts
take into account the behaviour of the dependent undertaking in determining its
own situation, i.e. they look at whether the lack of alternative suppliers is the
result of the voluntary lack of diversification in the sources of supply. As it is
explained in further details in Annex VI, the case law has also identified the
situation where an undertaking is dependent on another one due to the shortage
of a given product in the market74. In particular, in this case the risk is that the
supplier privileges firms belonging to its group, or to its distribution system, or
companies to which it is vertically integrated, to the detriment of independent
resellers. All in all, it appears clear that the prohibition of unfair hindrance
include in § 20(2) has been often used as an instrument to control the selective
distribution systems and the practices imposed by the producers within them. In
fact, often judges felt that the application of this provision was the only way to
sanction cases where the privileges provided by a system of distribution were
abused by a producer that did not have a dominant position, nor did it entered an
anticompetitive agreement with the distributors75.
Here below we provide two cases that may be considered relevant.
In the recent decision Presse Grosso regarding the press distribution system the
BGH decided that a conduct represents a forbidden discrimination according to §
20 I GWB when it has negative consequences on the competitive position of the
supposedly discriminated party. For decades the German press and media
distribution has been based on exclusive distribution rights for around 70 local
companies (so called Presse Grossisten) that carry out the distribution of
newspapers, magazines and similar products. Each company distributes all the

72 Cf BGH, Grundig, WuW/E, 2419.

73 See WuW/E BGH 1814 (Allkauf-Saba);WuW/E OLG 3508 (Allkauf-Saba II)
74 Cf. Agip II, WuE/E OLG 1499.
75 See Bornkamm, PME et pratiques prohibées en-deça de seuil de la dominance. Point de vue

allemand (1), in Concurrences, 2007, no. 2, p. 74, and Reh, PME et pratiques prohibées en-deça
de seuil de la dominance. Point de vue allemand (2), in Concurrences, 2007, no. 2, p. 75,
specifically referring to the Adalat case, as an example of the incapability of EU competition law
rules to catch such kind anticompetitive practices.

PAGE 55 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

press products in its region. The German editors used this channel to reach any
town and distribution point in Germany. On August of 2004 the association of
German editors made a statement defending this system in order to guarantee a
complete distribution of any newspaper and therefore to preserve the variety of
press in Germany and the freedom of information of the German citizens. The
German editor Bauer Media Group, one of the main players in Germany and
Europe, did not adhere to this declaration. In the year 2009 Bauer decided to
finish its exclusive distribution agreement with the local distribution company
Grade, for the region of Hamburg and to carry out the distribution through its
own subsidiary. Grade filed a claim before the local courts of Hamburg arguing
that the decision of Bauer to finish this contract was not possible due to the
declaration of the year 2004. Furthermore Grade considered that it was being
discriminated by Bauer due to the distribution through the subsidiary and to the
denial of Bauer to supply Grade with newspapers. The regional court stated the
claim while the second instance denied the right of Grade to continue with the
contract. Finally the BGH confirmed the second decision. According to the BGH
the contract was correctly finished by Bauer. In particular, Bauer is not bound to
the declaration of the year 2004 since it did not enter into it. Furthermore there
is no discrimination according to § 20 GWB. Bauer can continue delivering its
newspapers to other companies in other regions on an exclusivity basis since this
does not disturb the competition possibilities of Grade and this company
continues distributing the products of the other editors.
The Regional High Court of Karlsruhe considered in Scout that a selective
distribution system is also allowed in other sectors apart from the luxury
products sector if the requirements to the reseller are objective and conditioned
by the product. Therefore the distribution through an internet auction platform
as Ebay can be prohibited by the producer. This decision refers to the resale of
satchels of the German company Scout. The Court considered that the selection
of authorised distributors can be based on objective reasons. Therefore the
producer can prohibit the resale of its products through normal Ebay auctions.
The court does not refer to the distribution through an official Ebay shop.
In particular, Peyer (2010) has collected information on 449 competition law
actions including both appeal and first instance litigation., and found that 55.9%
of those related to Sections 19-21 GWB. The author himself states that “the data
does not reveal whether the abuse allegations – representing more than 50 per
cent of all anticompetitive behaviour – are predominantly base on dominance
(section 19 GWB) or dependency (section 20 GWB)”76. However, anecdotal
evidence seems to suggest that the level of enforcement of Section 20 is quite
high, and can explain the high percentage of private cases brought under Sections
19-21 GWB.
There are generally no settlement provisions in cases according to the GWB. The
only explicit provisions on settlement agreements are applicable in cases
according to § 15 of the Unfair Competition Act (UCA). Nevertheless, the
Bundeskartellamt has established and further developed criteria and
requirements for so-called commitment decisions.77 According to §32b GWB, the
cartel authority may by way of a decision declare commitments to be binding on

76 See Peyer (2010), at http://www.tilburguniversity.edu/research/institutes-and-research-

groups/tilec/meetingsandevents/03cleen/peyerempirical.pdf
77 Fine proceedings against coffee roasters on account of price fixing-B11-18/08 of 18 Dec. 2009.

PAGE 56 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the undertakings pursuant to § 32 GWB. These commitments are often applied in

cases of Electricity- and Gas Supply contracts where an infringement according
to § 19 (4) No. 2 GWB (so-called exploitative abuse) is apparent. There are no
settlement cases reported regarding abuses of economic dependence/superior
bargaining position of general relevance. Most of known settlement cases relate
to the gas and electricity sectors.
Overall, Sections 20(2) and 20(3) lead to a significant amount of litigation in
Germany, as reported by our national experts. This leads also to sanctioning
conduct that emerges within vertical relationships along the value chain, in
particular for what concerns selective distribution. In a number of circumstances,
this has led to sanctioning behaviour – mostly in the presence of vertical
integration along the value chain – that would not qualify as anti-competitive
neither under Article 101 TFEU, nor under Article 102 TFEU. Annex VI contains
a more detailed explanation of these cases.

2.2.4.6. Greece

In Greece, the abuse of economic dependence is addressed by Act No. 146/1914

on "Unfair Competition" (hereinafter the “AUC”), which intends to protect
individual traders from unfair practices by their competitors contravening
"bonos mores". The AUC is not based on Articles 101 and 102 of TFEU.
Originally, the prohibition of abuse of a relationship of economic dependence
was inserted in this law by Art. 16 of Law no. 2000/1991, which added a new
paragraph to Art. 2 of competition legislation Law no. 703/1977. It was
incorporated as a separate article and substituted by Art. 1(2) of Law no.
2296/1995. It was abolished by Art. 1(1) of Law no. 2837/2000, and came into
force again by the aforementioned Art. 1 of Law no. 3373/2005.
The new Competition Act Law no. 3959/2011 has no provision on abuse of
economic dependence. It was abolished again in 2009 (Law no. 3784/2009) and
now is inserted as article 18a in the AUC pursuant to which any such claims need
to be brought before civil courts and not before the competition authority. The
preparatory committee of Law no. 3784/2009 explains that the reason for
abolishing Art. 2A of the former Competition Act on abuse of economic
dependence and bringing it within the AUC legal framework relates to the fact
that the Competition Act aims only at the protection of competition itself and not
to the protection of individuals. Art. 18A AUC intends to protect individual
traders from unfair practices by their competitors contravening bonos mores. It
differentiates from the former Art. 2A of the Competition Act only as far as it
provides that in case of abuse, damages can be claimed and a fine up to 50.000
Euros may also be imposed. The fine can be doubled in case of recurrence. (par. 2
& 3 of Art. 18A AUC).
Private actions may also occur, although they are reportedly not common. (See
Court of First Instance of Athens decision no. 609/2003). Nowadays, civil courts
have exclusive competence in relation to the application of this provision and,
therefore, privates can only recover damages suffered only by filing a claim
before civil courts on the basis of the provisions of the AUC. According to some
scholars, all unfair competition claims based on the Law no. 146/1914 fall within

PAGE 57 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the competence of the multimember courts78, whereas others believe that the
competent court should be determined on the basis of the total value of the
claims79.
In Greek law, there is no specific statutory basis for bringing actions for damages
for infringement of competition rules, although the introduction of a specific
statutory legal basis is currently being considered by the Greek Government. The
basis for such action would be Article 914 of the Greek Civil Code ("Article 914
CC") establishing tort liability. The case law of the Civil and Administrative
Courts has established that individuals (e.g. competitors, customers, suppliers,
consumers) who have suffered injury as a result of anti-competitive behaviour
would be entitled to compensation on the basis of Article 914 CC and could bring
actions for damages before the Courts. The conditions for such action are the
unlawful character of the act committed by the author of the damage, fault, the
existence of a damage resulting from the unlawful and negligent act and a causal
link between the said act and the damage.
Here below we provide a representative example of the cases litigated before
courts after that the latter became the enforcing authority, which may considered
representative of the court practice in this field.
In 575/2010 ΜΠΡ ΑΘ (517591) the Athens Single-Member Court of First
Instance argued that cancelation of a contract may not constitute in itself an
abuse of economic dependence. In this case the parties had been renewing
annual contracts until 31.12.2007 after which date the contract was still deemed
to be valid in the absence of a signed document. Pursuant to a written agreement
of 10.3.2009, the validity of the contract was extended until 31.12.2009. The fact
that one of the parties refused to signed an additional extension of the validity on
the basis of the terms that the other party set did not constitute a “sudden”
cancelation of the contract. The continuous re-enactments of the annual
agreements did not imply that there is a continuous agreement. Since the
agreements were of annual duration and the cancelation was not deemed to be
“sudden” there was no abuse of economic dependence.
To our knowledge, there have not been many cases initiated by private parties
and very few cases have been rendered by courts on abuse of economic
dependence. In particular, no other cases specifically on art. 18a of the AUC only
were dealt with by courts. Many filed cases deal with unfair competition practices
in general and the abuse of economic dependence is dealt with only marginally. It
seems, thus, that the significance of the provision is now quite diluted.

2.2.4.7. Hungary

In Hungary, paragraphs b), c) and i) of Section 21 of the Competition Act may be

seen as provisions dealing with the issue of economic dependence or superior
bargaining position, although they do not contain expressly any mention of this
type of unilateral conduct. For instance, the refusal to establish or maintain
business relations without any justification may be seen as an abuse of dominant

78 See Marinos, Unfair Competition, 2002, p. 312.

79 See Kotsiris, Unfair Competition and Antitrust Law, 2001, p. 365-366.

PAGE 58 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

position based on the existence of economic dependence80. These provisions

were incorporated into the Hungarian Competition Act in 1990, when the
legislators kept in mind situations that could arise as a result of the shortage
economy. It is true that since then these provisions have been kept and are still
present in the Hungarian Competition Act. However, there is only a very weak
enforcement practice for these provisions81.
Paragraphs a), d) and j) of Section 21 of the Competition Act also may be seen as
tackling this type of unilateral conduct, although they do not contain expressly
any mention of this type of unilateral conduct. These provisions prevent
undertakings from fixing purchase or sales prices unfairly in business relations,
including where general contract terms and conditions are applied; stipulating
unjustified advantages by any other means; or forcing the acceptance of
detrimental terms and conditions on the other party. In addition, the rules
prevent undertaking with superior bargaining position from influencing the other
party’s business decisions for the purpose of gaining unjustified advantages, or
creating a market environment that is unreasonably disadvantageous for the
competitors. Such behaviour may be seen as an abuse of dominant position based
on the existence of superior bargaining position.
Overall, the provisions contained in the Hungarian competition Act cannot be
considered as stricter than Article 102 TFEU, since they aim at tacking instances
of exploitative, exclusionary or discrimination abuses that would normally fall
also under the scope of EU competition law. To the contrary, relevant provisions
on abuse of superior bargaining positions are found outside the Competition Act,
in particular in the 2005 Act on Trade. Paragraphs a), b) and e) of Section 7(2) of
the Act on Trade prohibit conduct such as, inter alia, unduly discriminating
against a supplier; restricting access of a supplier to marketing channels; and
imposing unfair conditions upon the supplier in connection with his business
relations with the trader or with another trader. The latter conduct may consist
e.g., in demanding the best available terms and conditions as obligatory, and
enforcing such terms and conditions with retroactive effect, i.e. compelling the
supplier to provide discounts during a specific period for a specific product only
to the trader in question, or compelling the supplier to manufacture products
under the trader's trade mark or brand name as a precondition for the marketing
of any other product of the supplier”. In addition, paragraphs c), d), g), h) and i)
of the same Section of the Act on Trade contain relevant provisions on abuse of
superior bargaining position. These rules prohibit conduct such as
 Prescribing undue risk pooling contract conditions resulting in one-sided
advantages to the trader as against the supplier, meaning in particular the
charging of expenses serving also the business interest of the trader, such
as storage, advertising, marketing and other costs to the supplier;

80 The exact wording is as follows: “It is prohibited to abuse a dominant position, in particular:
[…]
b) to restrict production, distribution or technical development to the detriment of consumers
and/or business partners;
c) to refuse to establish or maintain business relations adequate for the nature of the transaction
without any justification; […]
i) to hinder competitors from entering the market in any other unjust manner”.
81 Since the year 2000, there have been approximately 2 cases each year under these provisions.

PAGE 59 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 The unjustified amendment of contractual conditions to the detriment of

the supplier, or installing a clause permitting such possibility for the
trader;
 Asserting a threat for cancelling the contract to impel contract conditions
for lopsided advantages;
 Applying pressure upon a supplier to use other suppliers or the trader's
own supplier.
 Applying a sale price for products which are not owned by the trader below
the price invoiced as contracted, not including the prices employed for the
sale of products with some defect or for the sale of products inside of a
seven-day period before the date of expiry of their shelf life, or the
introductory prices that may be used for maximum fifteen days, or the
prices employed in a clearance sale for maximum fifteen days in any
seasonal campaign, any sales campaign due to changing models or profile,
or due to going out of business.
The Act on Trade is enforced by the Hungarian Competition Authority (GVH).
The settlement procedure, as laid down in Section 75 of the Hungarian
Competition Act, led some companies under investigation to undertake to cease
the infringing conduct, for example by reviewing their contractual practices. In
Annex VI we provide describe a number of investigations, in which the
competition authority dealt with alleged infringements of the Act on Trade. This
occurred, i.a., in cases Vj-91/2008, Vj-93/2008 and Vj-94/2008, in which the
intervention of the GVH led several retailers to review their contracts signed with
suppliers. GVH launched a proceeding against Provera Beszerzési (Purchasing)
Kft and its partners (the Hungarian Hipermarket running Cora hypermarkets,
Csemege-Match Kereskedelmi (Trading) Zrt., Profi Magyarország Kereskedelmi
Zrt.). Meanwhile proceedings were initiated against Auchan Magyarország Kft.
and Metro Kereskedelmi Kft. as well with the aim to review their contractual
practices. The GVH examined in all three cases whether the undertakings had
infringed the 2005 Act on Trade, in other words whether they had abused their
significant market power in the course of entering into the supplier contracts
applied by them.
Enforcement of the mentioned provisions – both the Competition Act and the
Act on Trade – can be also private. We refer to previous sections for details on
private enforcement.

2.2.4.8. Italy

In Italy, a rule prohibiting the abuse of economic dependence was introduced in

Article 9 of the law on industrial subcontracting, n. 192 of 1998. This legal
provision prohibits any agreement between parties in a commercial relation,
which would determine an abuse of one party’s economic dependence, defined as
an “excessive” imbalance between the duties and obligations for the parties
arising from that commercial relation. The second paragraph of the article
specifies that the abuse of economic dependence can consist in a refusal to
supply, in a sudden termination of the contractual relation, or in the imposition
of unfair or too onerous contractual conditions.

PAGE 60 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The applicability of this provision outside the domain of industrial

subcontracting has been subject to a lively debate in Italy: today, the applicability
of this rule to franchising agreements, for example, is widely acknowledged. As
confirmed by the workshop we have organized in Rome for the purposes of
drafting this Report, the rule has been interpreted over time in a way that clearly
separates cases of abuse of dominance, and cases of abuse of economic
dependence. This implies that, whenever an abuse of economic dependence is
likely to significantly affect competition in the relevant market (as defined for
antitrust purposes), then the competence is attributed to the Italian competition
authority, which addresses the conduct as an abuse of dominance tout court, and
thus without imposing any stricter standard or rule on the alleged infringer.
To the contrary, if the abuse of economic dependence cannot be construed as an
abuse of dominance, the issue is treated as a case of abus de droit, related to the
parties’ general obligation to behave correctly and in good faith during the
negotiation, conclusion and execution of the contract (as specified in the Italian
Civil Code, Articles 1175, 1176 and 1375)82. To put it shortly, abuse of economic
dependence is a competition matter only when it is also configured as an abuse of
dominance; in all other cases, it is a specific provision related to national contract
law. As a result, the Italian rule on abuse of economic dependence cannot be
included in our list of rules that are stricter than Article 102 TFEU, and would
not be affected by an extension of the convergence rule of Article 3(2) of Reg. CE
1/2003 to cases of abuse of dominance.

2.2.4.9. Latvia

In Latvia, the Competition law was amended on 13 March 2008 (Applicable as

from 16 April 2008), and a new concept of “dominant position in retail trade”
(hereinafter the “DPRT”) was introduced83. With this amendment the legislator
attempted to protect the suppliers and producers from abuse of both economic
dependence and bargaining power by the big supermarket chains. The concept of
DPRT addresses both the abuse of economic dependence of smaller suppliers or
producers as well as abuse of superior bargaining power by the retailers. The new
Section 13(2) was introduced to the Section 13 “Prohibition of the Abuse of
Dominant Position” of Latvian Competition law. The provision covers retail level
traders without specifying the sector or field of retail. Nevertheless, although not
explicitly stated in the Competition law, it is clear from the preparatory works
that the legislation aims at restricting unilateral conducts of the biggest
supermarket chains in particular.
The provisions are supposed to protect suppliers of the retailers, namely
producers and wholesalers supplying their goods for sales in supermarket chains.
The Latvian Competition Council clarified that the DPRT concept significantly
differs from the classical dominant position. Unlike what occurs for the classical
dominant position, an undertaking holding DPRT is not in the position to act
independently from its competitors or consumers but is in a position to impose

82 See i.a. Suprema Corte di Cassazione - Sezione III Civile, n. 20106 of 18 September 2009. And
see, recently, the six decisions of the Tribunale di Roma in the Logista case (Tribunale Roma,
Sentenza 17/03/2010 Meloni Giud. I. C. c. Soc. Logista Italia S.p.A.).
83 See i.a. http://infolex.lt/portal/ml/start.asp?act=legupd&lang=eng&biulid=189&srid=18&strid

=1305.

PAGE 61 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

unfair terms or payments on its suppliers. There are two preconditions for
establishing DPRT: market power or superior bargaining power (criteria for
evaluation are: market share in the relevant retail market and the purchase
amounts) of the retailer and dependence of the suppliers.
There is no particular threshold of the supplier’s turnover at which DPRT can be
presumed. When evaluating dependence of suppliers in DPRT cases where the
market power threshold is lower if compared to classical dominance, the
threshold of 22% of the supplier’s turnover defined in the Rewe/Meinl case is not
decisive and can be even lower (less than 20% in a particular case, Maxima being
the biggest buyer of the particular supplier). When the undertaking concerned
holds market power; this per se implies the existence of a risk of dependence of
suppliers. Thus, upon establishing the market power, dependence of suppliers
can be presumed. However, the position of each supplier is assessed on a case-
by-case basis.
The enforcing body is the competition authority. There is very little case law on
DPRT. For the first time proceedings were initiated in regard to conduct of
supermarket chain “Maxima Latvija” in July 2010 but a decision to close the case
was adopted (published on 5 August 2010). In that case the Competition Council
clarified that the DPRT concept significantly differs from the classical dominant
position: unlike in case of the classical dominant position, an undertaking
holding DPRT is not in the position to act independently from its competitors or
consumers but is in a position to impose unfair terms or payments on its
suppliers.
Subsequently, on 30 November, 2010 (published on 22 December, 2010) the
Competition council delivered the first infringement decision pursuant to Section
13(2) on DPRT and imposed a fine on one of the biggest super market chains
Rimi Latvia for requesting unfair discounts (unfair payment for access to the
supermarket chain Supernetto) from Latvian dairy products producer AS
Valmieras Piens. It shall be noted that so far there are only two infringement
decisions rendered in DPRT cases and both are currently challenged in the court
(not decided yet), thus settlement in a form of administrative agreement is still
possible before the cases are finally decided by the court.
Private enforcement in competition law cases is virtually non-existent in Latvia.
There have been a handful of cases in regard to Section 18 of the Competition law
violations (unfair competition cases where the former employees or competitors
have gained commercial secrets of competitors84 and so called stealing of
business cases) but no cases on abuse of economic dependence, abuse of superior
bargaining position so far have made it to the court. The two cases initiated by
the NCA for abuse of DPRT are still pending, given that both infringement
decisions have been challenged in administrative court and none of them have
been finally decided so far85. Based on our survey, with the very few private
enforcement cases, it seems that in general courts are not really familiar yet, and
are therefore rather uncomfortable with Competition law private enforcement
cases.

84One case has ended with a settlement, another is not finally decided yet.
85 As stated in the Overview of the activities and statistics of the Latvian Competition Council
in the period of 2006 – 2010, can be found in Latvian language at
http://www.kp.gov.lv/?object_id=523, first published on 27.01.2011.

PAGE 62 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.2.4.10. Portugal

In Portugal, abuse of economic dependence is forbidden by article 7 of the

Portuguese Competition Act (Law no. 18/2003 of 11 June 2003, as amended
lastly in 2008), dealing with the relative dominant position.
When this prohibition was first introduced, in article 4 of Decree-Law no. 371/93,
an objective clearly present in the mind of legislator and enforcer was to protect
small suppliers and purchasers against the increasing economic power of “large
commercial surfaces”, such as large supermarket chains.86
It should be noted that, at the time of the introduction of this prohibition in 1993,
it was clearly stated that it would limit the relevance and the scope of the
prohibitions of the other types of unilateral practices, from then on prohibited in
Decree-Law no. 370/9387
Currently, the PCA limits the scope of application of the provision to cases where
the alleged practice affect the functioning of the market or the structure of
competition there. For this reason, the provision is scarcely applied and one of
the often repeated suggestions for the upcoming revision of the Portuguese
Competition Act is that this provision be eliminated.
The provision covers all sectors and is designed to protect SMEs, especially at the
retail level. Despite the wording of the law clearly referring to relations with a
“supplier” or a “client”, the Supreme Court’s judgment of 24 April 200288
apparently indicated that economic dependence (or relative dominant position)
may exist in horizontal as well as in vertical relations.
Article 7 (abuse of an economic dependence) prescribes that “Insofar as it may
affect the functioning of the market or the structure of the competition, one or
more undertakings shall not engage in the abusive exploitation of the economic
dependence on it or them of any supplier or client on account of the absence of
an equivalent alternative”. In particular, paragraph 2 of the article refers to the
unjustified cessation (total or partial) of established commercial relationships,
with due consideration being given to prior commercial relations, the recognised
usage in that area of economic activity and the established contractual
conditions89. The judgment referred above also elaborated on the concept of an
“equivalent alternative”.

86 See 1993 Activities Report of the Conselho da Concorrência, p. 18.

87 See 1993 Activities Report of the Conselho da Concorrência, p. 18.87 see 1993 Activities
Report of the Conselho da Concorrência, p. 18. The NCA Report in question reads (our
translation): «By publishing Decree-Law 371/93, of 29 October, the legislator deprived the
regime of individual restrictive practices of some of its anti-competitive relevance, but, at the
same time, introduced the prohibition of abuse of economic dependence, thereby broadening the
scope of the control of individual behaviours of undertakings to those which do now have an
absolute dominant position». The idea is seemingly that the creation of this new figure meant
that there were now many situations of concern to the NCA which could be dealt under the abuse
of economic dependence provision, diminishing the need to resort so often to Decree-Law
370/93.
88 Supreme Court judgment of 24 April 2002 (file no. 01B4170).
89 Paragraph 3 specifies that an undertaking is understood as having “no equivalent alternative”

when: (i) The supply of the good or service in question, in particular that of distribution, is
provided by a restricted number of undertakings; and (ii) The undertaking cannot obtain
identical conditions from other commercial partners in a reasonable space of time.

PAGE 63 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

This provision was essentially inherited from the previous Competition Act
(Decree-Law no. 371/93, of 29 October)90 and it is repeatedly suggested that the
upcoming revision of the Portuguese Competition Act should eliminate this
provision.
Article 7 of the Competition Act is enforced by the Portuguese Competition
Authority, and it may be invoked in private litigation between undertakings, in
which case any civil court may be competent to apply it. This provision has been
enforced, although examples are rare. Ever since its creation in 2003, the
Portuguese Competition Authority has never adopted a decision finding an
infringement of this provision. However, two cases did discuss a possible
infringement of article 7: Fresenius Medical Care Products (20/04) and
Unibetão et al (01/06). Both were closed without a formal finding of
infringement, although in the latter case commitments were accepted from the
undertakings. Details of these cases can be found in Annex VI.
Most of the substantive provisions of Decree-Law no. 370/1993 relating to
unilateral practices restricting trade may be perceived as essentially prohibiting
abuses of a superior bargaining position, mostly for the purpose of protecting
small suppliers. The most relevant provision of this Decree-Law, in what
concerns this question, is article 4-A on “Abusive trade practices”, which
prohibits (para. 1) conduct aimed at obtaining “from a supplier prices, terms of
payment or sale or conditions for commercial cooperation that are
disproportionate in relation to its general conditions of sale”.
This provision is specifically aimed at protecting small suppliers from abusive
practices. Enforcement is the responsibility of the Competition Authority. The
legislation is often enforced (the consulted expert reported a number of litigated
cases that is higher than 100), even though the Competition Authority seems to
consider it a low priority in terms of resources availability and impact on national
welfare.91
Private enforcement of the mentioned provisions, as well as of competition law in
general, is still an exceptional occurrence in Portugal, with very few cases having
reached the courts. To our knowledge, the only case where this provision was
discussed in the context of private enforcement was the Supreme Court’s
judgment of 24 April 2002 (details provided in Annex VI). However, it should be
borne in mind that the parties had invoked an abuse of a dominant position, and
the Supreme Court discussed abuse of economic dependence, instead.
No specific settlement provisions exist for any of the restrictive practices
regulated in the Competition Act. As regards commitment decisions, the
Portuguese Competition Authority has developed a practice of accepting
commitments from undertakings: however, the binding nature of commitments
thus accepted is not entirely certain. The investigation in the Unibetão et al case

90 The first national Competition Act – Decree-Law 422/83 – did not prohibit abuse of economic
dependence. When this prohibition was first introduced, in article 4 of Decree-Law 371/93, an
objective clearly present in the mind of legislator and enforcer was to protect small suppliers and
purchasers against the increasing economic power of “large commercial surfaces”, such as large
supermarket chains. See 1993 Activities Report of the Conselho da Concorrência, p. 18.
91 See, e.g., para. 79 of the 2005 Activities Plan of the Competition Authority.

PAGE 64 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

(01/06) was closed on the basis of commitments offered by the undertakings, but
no details were made public92.

2.2.4.11. Romania

In Romania, Article 6(1)(f) of the Competition Law prohibits conducts whereby

an undertaking exploits “the economic dependence of another undertaking vis-
à-vis a similar undertaking or undertakings that does not have an alternative
solution under equivalent conditions, as well as breaking contractual relations
solely because a partner refuses to accept certain unjustified commercial
conditions”.
The provision covers all sectors and it is designed to protect trade partners.
Recent amendments raised some pending interpretative issues, but it appears
clear enough that the provision should be understood as prohibiting the
exploitation of the situation of economic dependence, namely, where non-
dominant undertakings may depend on a dominant undertaking and (1) the
former do not have an alternative solution; for example, in the case of a refusal to
supply or to continue supplying an existing business partner; or (2) a refusal to
accept unjustified commercial conditions. In practice, the provision has limited
relevance and in fact there have been fewer than 10 cases on abuse of economic
dependence to date.
To enforce the application of Art. 6 (1) (f) of the competition law concerning the
exploitative conduct of undertakings abusing their dominance, it is necessary to
give due account to the consumer’s choices as well as to undertakings with no
alternatives on the market.
The competent authority is the Romanian Competition Council. But there is also
evidence of significant litigation related to unfair competition rules, which
suggests that private enforcement of these rules is also significant93.

2.2.4.12. Slovak Republic

In the currently applicable legislation in the Slovak Republic, there are no explicit
provisions on the abuse of economic dependence or superior bargaining position.

92 See Annex VI for more details.

93 Unfair competition rules are contained in Art 4(3) RCL; an Unfair Competition Law has been
padopted already in 1991, no.11/91, and was later amended in 2007 to reflect the unfair
commercial practices directive.. See for example: Ploieşti Court of Appeals, decision no. 127 of 29
September 2008, criminal section, where the Tribunal Dâmboviţa, through a criminal sentence
no. 78, 16 March 2007 (an individual was found guilty for the commercialisation of 18 T-shirts, 3
blouses, one tracksuit, 3 sport shoes that were not original products, but wore the inscription of
the mark 'E', registered by WIPO, without any authorisation from the German corporation, E.
A.G. S.E. Sport.). See also: Bucharest Court of Appeals, IX civil and intellectual property section,
9 September 2008, S.C. E. E. J. D.D. CROAŢIA (civil sentence no. 341/19 May 2003, Bucharest
Tribunal, V civil section, case no. 5694/2002) versus S.C. N. L., S.C. N. S.R.L. and the National
Office of Inventions and Marks (alleged unfair competition acts concerned the international and
national marks 'E'. The acts were committed during the production, commercialisation and the
importing/exporting of food products by the defendant, which had as labels similar signs with
those of the plaintiff). In Annex VI we provide examples of relevant cases. Unfortunately we could
not obtain precise statistics about the number of filed cases.

PAGE 65 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Until recently, Section 3(1) of the Act No. 358/2003 Coll. on Retail Chains
defined the abuse of economic power as “a conduct of a retail chain operator vis-
à-vis its supplier by which it abuses its bargaining advantage resulting from its
economic power exercised during negotiation of a contract with the supplier
and by which it extorts to his own benefit more favourable conditions than the
conditions it would have negotiated without its bargaining advantage”. The
economic power was defined in Section 2(h) of the same Act as “a position of a
retail chain operator vis-à-vis its supplier on the basis of which the supplier is
dependent on the retail chain operator since on the relevant market it does not
have any opportunity to offer goods to other undertakings and therefore is
forced to offer the retail chain operator more advantageous contractual
conditions than it would have offered to other undertaking”.
Section 3(1) of the repealed Act No. 172/2008 Coll. provided a list of unfair trade
conditions between a customer and a supplier of groceries, if either the customer
or the supplier was economically dependent on the other. According to Section
2(b) of the Act No. 172/2008 Coll. “the economic dependence is defined as a
situation in which either the customer or the supplier enters into a contract with
the other under unfair conditions as a result of the fact that his possibilities to
enter into a contract with other undertakings on the market in the Slovak
Republic are limited due to identical or similar unfair trade conditions asserted
by other undertakings”. Section 3(6) of the repealed Act No. 172/2008 Coll.
provided that the conclusion of unfair trade conditions is prohibited.
Later, this Act was replaced by Act No. 140/2010 Coll., the so-called “Unfair
Trade in Foodstuff Act” (UFTA). The UFTA applies to all entrepreneurs as
defined under Slovak Act No. 513/1991 Coll. Commercial Code, as amended (the
“Slovak Commercial Code”). The UFTA includes an exhaustive list of more than
30 unfair contract terms. Compared to the previous legislation, the number of
contract terms labelled as unfair approximately doubled, and now includes such
terms as unilateral set-off, pecuniary consideration by supplier to the reseller for
services that have not been provided, immediate termination of contract without
giving a notice period and grounds for such termination, and other terms. It also
prohibits the sale of goods at a price below the purchase price94. All such unfair
terms are prohibited and, under Slovak law, void95. The UTFA also encourages
the adoption of ethical codes between the parties to lay down rules of a more
‘honest and transparent business’. This initiative is also within the context of
enhancing the food supply chain at the EU level.
Enforcement of the UTFA is vested in the Ministry of Agriculture (the
“Ministry”). If the investigation by the Ministry proves that unfair terms have
been agreed by the parties, it will trigger a set of consequences. In the first place,
the Ministry will order that the unfair terms be removed. Next, it will order the
party benefiting from such unfair term to return any pecuniary consideration for
goods or services provided based on such unfair term. Finally, the Ministry will

94 With the following four exemptions: 1. sale of goods when a store is closing down or when
the entire scope of business is being changed, 2. sale of goods three quarters of whose ‘best before
period’ have already expired , 3. seasonal sales, and 4. sale of damaged goods. See Turayova
(2010), Unfair Terms in Business Contracts between Resellers and Suppliers of Foodstuffs, on e-
Competitions, N 31177, available at www.concurrences.com.
95 In addition, the Unfair Terms in Foodstuff Act requires that general terms of business

(including terms and conditions for the purchase of goods, pricing, price reduction, terms of
payment, volumes of sales and marketing rules) be available to a supplier on request.

PAGE 66 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

impose on such benefiting party a fine of up to €300,000. If the Ministry does

not order the abolition of unfair conditions, the injured undertaking’s only option
is to file a petition with the court to decide on the issue. Unfortunately, there is
no available information in the Annual Statistical Reports of the Ministry of
Justice of the Slovak Republic regarding the number of cases under the previous
legislation and the state of private enforcement in this area96. Also, there were no
relevant settlement cases identified in the practice of the competent authorities,
also due to the fact that the law is very recent.
Eventually, the new Slovakia government reportedly observed that the UTFA had
to be repealed. As of 1 August 2011, the UTFA was abolished. It is still unclear
whether new rules will be adopted, which will tackle abuse of economic
dependence in retail chains.

2.2.4.13. Spain

Finally, also Spain has enacted legislation on the abuse of economic dependence,
which falls under the scope of this Report. The Spanish Act on Unfair
Competition (hereinafter the “SAUC”) - Law 3/1991 in force since 10 January
1991 - prohibits the abuse of economic dependence.
Article 16(2) SAUC considers unfair “the exploitation by an undertaking of a
situation of economic dependence when its clients or suppliers cannot find
equivalent alternatives to continue their activities”. The provision adds that
economic dependence “will be presumed when a supplier, on top of the usual
rebates and conditions, is obliged to grant regularly additional advantages that
are not granted to similar buyers”. In addition, the same Article prescribes that
it is unfair to (i) break “an established commercial relation, even partially,
without previous written notice of at least 6 months unless there has been a
breach of contract or force majeure”; and also to (ii) obtain “under threat of
breaking commercial relations, prices, payment conditions, selling conditions,
additional payments and other commercial cooperation conditions not
contained in the supply agreement”.
This provision covers all sectors and aims at the protection of SMEs against
larger undertakings.
Companies and consumers may seek injunctions against any such practice before
national courts. In addition the Spanish Competition Authority (hereinafter the
“SCA”) will be competent to deal with these cases when the alleged conduct: (i)
distorts competition97; and (ii) affects the public interest.98 The logic is that this
practice, not only harms other undertakings, but also distorts competition as a
whole, affecting the public interest. Therefore, in these cases the SCA will be
competent to impose fines.
The application of the SAUC does not require the establishment of a dominant
position, however if the infringer lacks market power, its conduct is considered to
be unlikely to distort competition and affect the public interest. Harmed

96 http://www.justice.sk/wfn.aspx?pg=r3&htm=r3/statr.htm
97 The old SADC contained a higher threshold, requiring that in order for the SCA to deal with
unfair practices, the conduct must “seriously distort competition in the market” (emphasis
added). However, there seems to be no practical consequence of such amendment.
98 Article 3 of the SADC.

PAGE 67 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

undertakings are entitled under the SAUC to seek damages before national
courts.
A special case is that of the Spanish Act on Intellectual Property, which contains a
specific provision related to the abuse of superior bargaining position in the
negotiations to license Intellectual Property Rights (“IPRs”) for cable
retransmissions. According to this provision, Section I, Chapter I of the SAIP
applies when a party, abusing its superior bargaining position, prevents the
initiation or continuation in good faith of the negotiations to authorise the
retransmission by cable, or impedes, without a valid justification, negotiations or
mediations.99 The provision is specific to the licensing of IPRs for cable
retransmission and aims at protecting licensees, notably TV channels.
National courts are competent to apply these provisions and companies are also
entitled to seek damages for abusive conduct before them.
We could not obtain evidence of settlement procedures. An example of a
commitment decision is provided in Annex VI.
Figure 3 below summarises this section graphically. Annex VI.2 and Annex VI.3
contain summary tables with more specific information regarding the rules
enacted by individual Member States on abuse of economic dependence and
abuse of superior bargaining power.
Figure
Europe3–
– Rules on abuse of economic dependence/superior
Countries
bargaining power

Rules the significantly diverge from Article 102 TFEU

Rules that slightly diverge from Article TFEU
Rules that do not overlap with Article TFEU

99 Spanish Act on Intellectual Property (“SAIP”), Article 20, Public Communication: […] 4. The
retransmission by cable as defined in the second paragraph of Article 20.2(f) within the European
Union shall follow these rules: […] g) When any of the parties, abusing its superior bargaining
position prevents the initiation or continuation in good faith of the negotiations to authorised the
retransmission by cable, or impedes, without a valid justification, the negotiations or mediations
referred in the previous paragraph, the Spanish Act for the Defence of Competition (Section I,
Chapter I) would be applicable.

PAGE 68 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.2.2 Rules on sales below cost

Besides legal provisions that focus on the abuse of economic dependence or

superior bargaining power, some European Member States have also enacted
legislation aimed at prohibiting the sale of goods below cost. Under Article 102
TFEU, selling below (certain measures of) cost is not prohibited per se even for
dominant undertakings, unless this conduct can be configured as an exclusionary
abuse, leading to anticompetitive foreclosure of competitors. In competition
cases on predatory pricing such as Akzo100 and Wanadoo101, as well as in cases on
margin squeeze – normally related to regulated sectors such as
telecommunications – the conditions that have to be met for below-cost sales to
be considered anti-competitive have been clarified. In the wake of those
decisions, the Commission Guidance on the treatment of exclusionary abuses
under EU competition law clarifies that below-cost pricing will be seen as
predatory only when it leads to the actual or likely foreclosure of as-efficient
competitors, and also leads to consumer harm. The Commission also clarified
that “normally only pricing below long-run average incremental cost (LRAIC)
is capable of foreclosing as efficient competitors from the market”, and that
when prices are below Average Avoidable Cost (AAC) the Commission will
normally conclude that the dominant undertaking is sacrificing profits.
The traditional rationale for the prohibition of sales below cost has been
explained by Colla and Lepoule (2008): “the reasons for banning below-cost
resale are generally based on two main criticisms of this practice: below-cost
resale is a predatory practice, as it seeks to eliminate a weaker competitor
(small specialized retailers) or discourage the entry of a potential competitor
into the market; by selling a product at a loss, the retailer harms the
manufacturer, undermining their image, and thus reducing the value, of their
brand”. Both propositions have been successfully criticized in the literature and
in antitrust enforcement, since below-cost sales can in most circumstances be a
sign of aggressive competition, aimed at conquering market share and customers
through legitimate (and replicable) means. Notwithstanding the widespread
agreement on the need to avoid a total ban on sales below cost, this conduct is
still prohibited per se in a number of European Member States102.

2.2.4.1. Austria

In Austria Article 5(1), no. 5, of the Austrian Cartel Act prescribes that the
forbidden abuse of a dominant position may particularly consist in an objectively
not justifiable sale of goods under its acquisition price. According to Article 5 (2),
the dominant undertaking has the burden of proof to rebut the prima facie
evidence that sales are below cost, or provide an objective justification. Such
allocation of the burden of proof differs from the general rules set forth in Reg.
CE 1/2003, and its rationale lies in the difficulty for the plaintiff to establish

100 Case 62/86 AKZO Chemie v Commission [1991] ECR I-3359.

101 Case France Télécom SA v. Commission, ECJ (First Chamber), judgment of 2 April 2009, C-
202/07 P.
102 Allain and Chambolle, 2003; Chambolle, 2003; OECD, 2006; Colla, 2006.

PAGE 69 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

existence of sales below (average variable) costs. That is why the Supreme Court
established that it suffices to show the probability of sales below costs103.
However, the provision cannot be interpreted as setting an non-rebuttable
presumption of anticompetitiveness. Quite to the contrary, so far there are no
cases where the applicant managed to prove an infringement; in all cases which
have been dealt with by the Cartel Court, the application for a finding of an
infringement of § 5 para. 1 (5) Austrian Cartel Act has been rejected.
This provision covers all sectors and is designed to protect especially SMEs from
predatory pricing strategies more effectively.
The investigating authority in this matter is the Austrian Federal Competition
Authority (FCA) and the Federal Cartel Prosecutor, which are, by law, party to
any proceeding before the Cartel Court. This amicus curiae system often serves
to provide plaintiffs with additional support. Private enforcement of these rules is
also possible104.

2.2.4.2. Belgium

In Belgium, Articles 101-102, Section 4, Chapter 4 of the Trade Practices and

Consumer Protection Act regulates sales at a loss. In a recent case, the President
of the Commercial Court of Brussels found that a parent company reduction of
80% of the ex-works original price of its pharmaceutical product was not covered
by the prohibition of sales at loss. Even if that price was less than the cost of the
production for the product concerned, it was not considered unlawful because,
first, the offer was temporary insofar as it sought to wind up stock, and second,
pricing was aligned on the price charged by competitor for the same product.
Indeed, the Belgian authorities had requested firms to reduce price of the
product concerned by at least 2% of the company’s turnover carried in Belgium.
In October 2001, the Supreme Court handed a down a relevant ruling thereof.
The Court found that sales at loss may be contrary to the prohibition of unfair
commercial practices not only when sales are made by a company that holds a
dominant position but also when a non-dominant company infringes the
requirements to act fairly on the market by seeking to attain an unlawful
purpose, i.e. exclude competitors from the market.

2.2.4.3. Bulgaria

In Bulgaria the relevant provision of the Law on the Protection of Competition

referring to the sales below cost is Article 36, paragraph 4, which belongs to
Section VII on the “Prohibition of Unfair Competition”. The said Article 36, para
4 provides for that “Prohibited shall be sales at the internal market of

103OGH 9/10/2000, 16 Ok 6/00; OGH 16/12/2002, 16 Ok 11/02.

104The relevant provision for sales below cost under Austrian Law is Article 5 (1) 5 Austrian Cartel
Act. Pursuant to the regulation, the forbidden abuse of a dominant position may particularly
consist in an objectively not justifiable sale of goods under its acquisition price. The sale below
cost was added to the non exhaustive list of abusive practices. It is also combined with a reversed
burden of proof to the extent that the market dominant undertaking has to prove either the sale
was not below cost or it was objectively justified. This provision is designed to protect especially
small and medium sized enterprises from predatory pricing strategies more effectively. The level
of enforcement is not substantial.

PAGE 70 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

considerable quantities of goods for continued period of time at prices lower

than the expenses for their production and their realisation with the purpose of
unfair attraction of customers”. The provision of Art. 36 (4) LPC refers to “sales
below cost” as part of the unfair competition rules in the law, which are
applicable to undertakings not enjoying dominant position. The enforcing
authority for sales below cost is the Bulgarian Commission on the Protection of
Competition (the “CPC”)105. Between 2008 and 2010 the CPC has issued
approximately 26 decisions in relation with sales below cost.
Also, the CPC can adopt commitment decisions, in which the companies under
investigation offer to change their practices to bring the allegedly infringing
conduct to an end. In these cases, however, the CPC can approve such
commitments and issue a decision for termination without finding a violation of
the Competition Protection Act only in relation to violations which are found to
be non-serious (i.e., the violation has no appreciable and permanent effect on
competition on the substantial part of the national market).
Private actions are possible in the framework of the general rules on tort. Private
claims to enforce competition law in the Bulgarian civil courts can be brought
either as individual actions or as class actions. The decision of the Commission
that has not been appealed or has been upheld by the courts on appeal is binding
on the civil courts when resolving a civil action brought before them. As already
stated above, and as reported by Petrov (2009), so far private enforcement
actions based on dominance abuses have been very rare. A new facilitated regime
for class actions is expected to facilitate private actions in the future106.

2.2.4.4. Czech Republic

In the Czech Republic, Amendment No. 403/2009 Coll. has repealed the former
general prohibition of re-selling at a loss (below average costs) contained in the
already mentioned Pricing Act of 1990 at Article 11(1)(e)107. Under the new rules,
only the abuse of a superior bargaining position by means of inappropriate
(unfair) pricing is forbidden (see Section 2.2.1 above)108.
The provisions of the Pricing Act significantly overlapped with the provisions on
abuse of dominance, which may also concern abusive pricing practices (such as
overpricing or predatory pricing). Nevertheless, both provisions were applied
concurrently to the same conduct, which could consequently result in two
sanctions being imposed for it (by two different authorities), as confirmed by the
Supreme Administrative Court in the case Komp no. 3/2006.
The above provisions of the Pricing Act were intended to protect all parties (both
businesses and consumers) which have a weaker economic position from
inappropriate (unfair) pricing by undertakings which have a substantially more
advantageous market position. While the Act was in force, enforcement was

105 It should be noted however that the decisions of CPC are subject to appeal before the Supreme
Administrative Court (details on the appellate process are provided for in point 18 of the Long
Questionnaire). Thus, in the cassation instance, when repeal the decision, the Supreme
Administrative Court shall decide itself the case on its merits.
106 See Petrov (2009), cit. supra note 46.
107 See Section 2.2.1 above.
108 There is no general explicit ban on abuse of a superior bargaining position, as the law refers
only generically to the “economic position” of the undertaking concerned.

PAGE 71 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

reportedly strong. As regards private litigation, the Competition Act does not
provide for any specific legal basis for bringing actions for damages as a result of
a breach of competition law. Accordingly, the legal basis for bringing such actions
under Czech law is found in the Commercial Code, Section 373109. As reported
i.a. by Bicková and Braun (2011), private antitrust enforcement is almost absent
in the Czech Republic, due to procedural problems but also a lacoìk of “trust in
the court system due to slowness and the low quality of judgments, particularly
when it comes to stating lost profit”. Nevertheless, the authors report that the
Office for the Protection of Competition has set the goal to increase the use of
private enforcement in the future110.

2.2.4.5. France

As reported by Colla (2006), below cost selling was a common practice of

retailers in France before the 1996 Loi Galland, given also that the retail sector in
France was fiercely competitive111. Even if resale below cost was prohibited, Colla
and Lapoule (2008) report that such practice was generally adopted by retailers,
because “the sales’ profit-loss threshold (SPLT) was not clearly defined and
penalties were not effective”112. This led to the decision to enact the loi Galland,
which confirmed the banning of resale at a loss, clarified the formula for
calculation of the SPLT, and reinforced the penalties applicable to infringers. The
reference price, which sets the SPLT, is set at the price stated on the invoice, plus
taxes and transport costs. The net invoice price is calculated on the basis of the
list price, and must take into account any price reductions “given at the time of
sale ... and directly related to the said operation”. The application of the Loi
Galland – as will be explained below, in Section 3 of this study – led to
undesirable consequences in terms of pricing practices, leading to price increases
and circumventing actions such as the application of discounts to group of
products purchased rather than on individual products (so-called New
Promotional Initiatives): this led to a reform that took the form of the Loi Dutreil
in 2005.
The Loi Dutreil allowed the reduction of the invoiced price, since 1 January 2006,
by the amount of commercial cooperation in excess of 20 per cent (of the
invoiced prices), and since 1 January 2007, by the amount in excess of 15 per cent
(Ferrier and Ferré, 2005, 2006). This reform allowed the retailers to lower the
price to a new SPLT (a lower minimum price), like in the example of Table I. A
fraction of these back margins is now passed-on to consumers by retailers who
reduce their final margins, in order to be more competitive. The Dutreil Act,

109 Section 373 of the Commercial Code provides that "whoever breaches a duty arising from a
contractual relationship is obliged to provide compensation for the damage caused to the other
party, unless he/she proves that such a breach was caused by circumstances excluding his/her
liability". This provision applies also beyond pure contractual relationships, as provided by
Section 757 of the Commercial Code.
110 See Dagmar Bicková and Arthur Braun, Czech Republic, The European Antitrust Review
2011, Chapter 4.
111 See Colla, E. (2006) Distorted competition: Below-cost legislation, ‘marges arrière’ and
prices in French retailing, The International Review of Retail, Distribution and Consumer
Research, 1466-4402, Volume 16, Issue 3, 2006, Pages 353 – 373
112 See Colla, E. And P. Lapoule, Banning below-cost resale in France: the impact on pricing

decisions, International Journal of Retail & Distribution Management, Vol. 36 Iss: 10, pp.746 –
758.

PAGE 72 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

allowing more competition, determined new changes in the pricing practices of

the retailers and, after the Act, pricing got increasingly important in their
marketing mix and prices on average dropped and French products gained a
more competitive position in the euro-zone countries, where French prices are
higher only than the Netherlands’ ones.
However, the Loi Dutreil only marginally solved the distortion between
manufacturers and retailers that characterized the Loi Galland. Accordingly, the
Attali Commission proposed in October 2007 to abolish the Galland/Dutreil laws
along with the banning of resale below cost in order to remove their adverse
effects on price levels and competition. The subsequent developments led to the
enactment of the Loi Chatel (Loi n° 2008-3 of 3 January 2008), “for the
development of competition in the interest of consumers”, which allows retailers
to deduct from the invoiced price the entire amount of back margins, but
maintains the double invoice obligation and the non-discrimination rule of the
general terms of sale. The new Act led to an increase of the intra-brand
competition among retailers and higher price differentiation, according to the
literature (See, i.a., Colla and Lapoule, 2008). However, retailers still cannot
negotiate manufacturers’ list prices and have an incentive to keep asking for
more marges arrières. The legislation does not allow retailers to impose on the
manufacturers an intense inter-brand competition, which could entail a stronger
price reduction.
Currently the relevant legal provisions are the following: Articles L. 420-5 and L.
442-2 to L. 442-4 of the Commercial Code prohibit resale at a price below the
purchase cost, i.e., below the invoice price plus taxes and transport costs. Article
L. 420-5 of the Commercial code only concerns final prices to end-consumers,
whereas Article L. 442-2, L. 442-3 and L. 442-4 of the Commercial code concern
both retail and wholesale prices. Articles L. 420-5, L. 442-2 to L. 442-4 of the
Commercial Code do not require a finding of dominance, nor do they require
proof of any effect on competition. Another difference is in enforcement: Article
L. 420-5 of the Commercial code is enforced by the Autorité de la concurrence or
by commercial and civil courts, whilst Article L. 442-2, L. 442-3 and L. 442-4 of
the Commercial Code are enforced exclusively by civil courts.
Exceptions to sales below cost are permitted for changes in season, style, or
upstream price levels. Smaller resellers could meet the competition of another
seller in the same area, but only if the other seller’s price was legal. And prices
could be cut for food products that were about to spoil, but the lower price must
not be advertised outside the store.
We do not have precise statistics about the number of litigated cases, but we have
good reasons to believe that such number is very high due to the policy interest
for this practice. In Annex VI we provide meaningful examples of cases on sales
below cost.

2.2.4.6. Germany

In Germany, sales below cost are regulated specifically at section 20(4) GWB. In
line with what was observed in the previous sections, it is worth recalling that
this section applies to the cases in which an undertaking deals with small and
medium-sized enterprises, which are the market operators protected by the
provision. As a result of this rule, and as recalled i.a. by Böge (2004), “powerful

PAGE 73 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

companies may not unfairly hinder their competitors or those companies

dependent on them by frequently selling their products below cost price, unless
there is an objective justification for doing so”113. This means that, as already
recalled, the rule applies also to non-dominant companies that can be defined as
being in a “relatively powerful” position. In this respect, the rule contained in the
GWB can be seen as a case of abuse of superior market power.
The authority enforcing the provision is the Bundeskartellamt. We do not have
precise statistics about the number of cases on sales below cost dealt with by the
authority. Private enforcement has been facilitated by recent amendments to the
GWB and is reportedly strong, at least for what concerns injunctive relief.
The German rule banning (unjustified) sales below cost became well-known
internationally shortly after its introduction, when the Bundeskartellamt
examined the pricing strategy of three major trading companies that sold basic
foodstuffs such as sugar, milk and flour below their respective cost prices over a
considerable period of time, without any perceptible objective justification114.
According to the Bundeskartellamt’s findings, the three trading companies had
superior market power over small and medium-sized independent food retailers
due to their companies’ sizes, market shares, and resources. As a result of these
findings, the Bundeskartellamt prohibited the pricing practice of these trading
companies. On the other hand, the three companies had engaged in a price war,
in which Wal-Mart had decided to lower its prices for sugar below those of Aldi
and Lidl, as well as below its purchase cost, thus making sugar, effectively, a loss
leader. A similar strategy was followed by Wal-Mart for margarine and milk. The
Head of the Bundeskartellamt clarified, a few years after, that the authority was
convinced that “the material benefit to consumers resulting from sales below
cost price are, at most, temporary and marginal ... temporary because market
concentration will increase after the exit of small and medium-sized
competitors ... marginal because expenses for the foodstuffs examined comprise
only a very small share of an average household’s budget, in some cases
accounting for less than one percent. At the same time, the impairment of
competition and market structures resulting from unfair hindrance of small
and medium-sized companies is permanent and appreciable”115.
One of the companies involved in this case, Wal-Mart, appealed the decision to
the Düsseldorf Court of Appeals, which ruled for Wal-Mart and reversed the
FCO’s decision. The German Supreme Court (known as the Bundesgerichtshof or
BGH) eventually reversed most of the appeal’s conclusions, deciding in favour of
the Bundeskartellamt. In doing so, the Supreme Court clarified that the
application of Section 20(IV)(2) GWB does not require either a “noticeable
impact on competition” element, or the existence of a causal link between
superior market power and pricing below cost.

113http://law.wustl.edu/wugslr/issues/volume3_2/p429Boge.pdf
114 See Decision of Sept. 1, 2000, Bundeskartellamt [BKartA] (Federal Cartel Office [FCO]),
WuW/E [BKartA] B9-85/00 (Aldi-Nord), DE-V 314; Decision of Sept. 1, 2000, FCO, WuW/E
[BKartA] B9-74/00 (Wal-Mart), DE-V 316
115 Böge (2004), at 431.

PAGE 74 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

According to the Supreme Court, if superior market power and pricing below cost
is found, the only relevant inquiry is whether the practices at issue are
“objectively justified” under the circumstances of the case116.
On 22 December 2007, the 7th amendment to the GWB has led to an even stricter
scrutiny of these practices. As explained by the Bundeskartellamt, “even the
short-term sale of food below cost price is now prohibited. Exceptions are only
possible under certain circumstances (as specified in Section 20(3) GWB), for
example if the deterioration of the goods is imminent, if the business is expected
to go bankrupt or is closing down”117. Reportedly, since this new provision came
into force, the Bundeskartellamt has had to deal with numerous complaints in
which trading companies were alleged to have violated the ban on sales below
cost price. The products mainly affected were milk products, beverages and meat
products. However, “in most of the complaints the Bundeskartellamt found no
indications which would justify the initiation of proceedings. The accused
companies were able in most cases to prove that their cost prices were below
their offer prices by presenting invoices”118. In Annex VI we provide other
meaningful cases.
Considerations on settlement procedures previously expressed in relation to the
abuse of dominant position are valid also for sales below cost. We do not have
precise information about settled cases.

2.2.4.7. Ireland

In Ireland, the Restrictive Practices (Groceries) Order 1987 (“Groceries Order”)

prohibited sales below invoice price of several grocery items. The Order was
repealed on 20 March 2006 and amendments were made to the Competition Act
to address certain business practices in the groceries sector. We will get back to
the Groceries Order in Section 4.3.3 below, where we use data from the
application of that Order to assess whether rules on sales below cost have
artificially kept prices high.

2.2.4.8. Latvia

In Latvia, there is no general prohibition of sales below cost that applies beyond
the scope of competition rules on abuse of dominance. However, in its practice
the Competition authority (Konkurences padome), within the framework of
abuse of collective dominant position case of gasoline retailers, has analysed a
few cases where it was claimed that pricing practices (more precisely, selling in
retail below the wholesale prices) could result in violation of Section 18(2) of

116 See http://www.wilmerhale.com/files/Publication/f44f800c-1128-4e4a-b92a-

61698f716b7b/Presentation/PublicationAttachment/f0f184db-64ff-4ada-841d-
d0110cdf86ee/ACF3111.pdf. In 2000, Guidelines were issued by the Bundeskartellamt on how to
interpret this legal provision. The guidelines included explanations on price and cost standards to
consider for finding of sales below cost.
117 http://www.bundeskartellamt.de/wEnglisch/download/pdf/TB_Kurzfassung_07-08-GB.pdf
118 Id.

PAGE 75 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Latvian Competition law, namely, unfair competition by means of breach of fair

dealing principles119.
Section 18(2) of the Latvian Competition law reads as follows: “Practices as a
result of which normative acts or the fair dealing principle is violated and
which have caused or could potentially cause distortion, hindrance or
restriction of competition, are regarded to be unfair competition”120. This
provision, depending on its interpretation, can prove stricter than the
corresponding scope of Article 102 TFEU.
The purpose of the quoted Section 18(2) of the Competition law according to
Competition authority is to safeguard of fair competition between equal
competitors and prohibit unfair competition. Therefore Section 18(2) should not
be interpreted as distorting or hindering effective price competition by means of
protecting those undertakings less able to compete if compared to equal
undertakings.
It should be noted that, given that sales below cost are not specifically regulated,
as there is no general and explicit prohibition for non-dominant undertakings, as
well as there is no relevant case law so far, the provision of prohibition of breach
of fair dealing principle could be stretched as covering sales below costs only on a
purely theoretical level.
Finally, it shall be considered that the Competition authority is no longer an
enforcing authority in regard to Section 18 of the Competition law violations. As
from 2009 Section 18 disputes have been adjudicated in an ordinary civil
procedure by the parties in courts of general jurisdiction. Therefore, if any,
arising cases will be privately litigated. We were unable to obtain more
information on the current level of enforcement of this provision.

2.2.4.9. Poland

In Poland, Article 15(1) of the Act of 16 April 1993 on Combating Unfair

Competition (CUC) prescribes that an act of unfair competition is the
introduction of difficulties for other entrepreneurs to access the market through,
inter alia, the sale of goods or services below their purchase cost in order to
eliminate other entrepreneurs.
Such provision aims at protecting both consumers and undertakings.
The provision prohibits, independently of undertakings’ market shares, only
those instances of sales below costs, which are aimed at eliminating competitors
from the market. Proving the latter premise is difficult and the above-mentioned
provision is – to our best knowledge – rarely applied. For these reasons, we do
not believe that Polish legislation on sales below cost can be considered stricter
than Art. 102 TFEU.

119 Case No. 460/06/05/6, decision No 71 as of 20 June 2007 on termination of proceedings. Last
viewed on 27.12.2010 at http://www.kp.gov.lv/uploaded_files/2007/DV71_2006.pdf, p.36, not
available in English.
120 Today, the NCA is no longer the enforcing authority as regards Section 18 of the Competition

law violations, which are adjudicated in ordinary civil procedures by courts of general
jurisdiction.

PAGE 76 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.2.4.10. Portugal

In Portugal, Article 3 of the Decree-Law no. 370/1993 on individual practices

restricting competition explicitly prohibits sales at a loss. This provision is the
most enforced of this Act, and forbids “to offer for sale or to sell a good to an
economic agent or to a consumer at a price below its actual purchase price,
after adding the taxes applicable to that sale and, when relevant, the charges
relating to transport”121. Paragraph 3 of the same Article clarifies that “discounts
directly related to the transaction are deemed to be quantity rebates, financial
rebates and promotional rebates, as long as they are identifiable in what
concerns the product, its quantity and the period in which they are in force”.
These provisions do not cover all sectors, as they are not applicable to a range of
products, including perishable goods, from the moment they threaten to quickly
deteriorate; goods whose commercial value has been affected, because the
situation, which dictated their need has passed, because the possibilities for their
use have been reduced, or because an important technological innovation has
occurred; goods whose re-supplies are done at a lower price, in which case the
actual price of purchase must be replaced by the price included in the new
purchase invoice; goods whose price is set by reference to the price practiced for
the same goods by another economic agent in the same field of activity, found to
be temporally and geographically in actual competition; goods sold in sales or
liquidations.
The competent authority is the Portuguese Competition Authority, which has
issued over 250 decisions to date to enforce this rule. The Competition Authority
has stated that it has “intervened … in several cases relating to sales at loss”.122
In fact, it had previously already stated that sales at loss took up most
significance, in practice, within the behaviours prohibited by Decree-Law
370/93123.
With regards to settlement procedures, we refer to the considerations expressed
in the Sections above. Private enforcement is possible under general rules on tort.
To our knowledge, there have not been any privately litigated case so far. In fact,
our national correspondent reported no evidence of existing cases.

2.2.4.11. Romania

Today, the prohibition of charging excessive or predatory pricing with the aim of
eliminating competitors, included at Article 6(1)(e) of the, Romanian competition

121 The actual purchase price is deemed to be the price mentioned on the purchase invoice, after
deducting discounts directly related to the transaction in question and which are identified on the
invoice itself or, by a reference included therein, in supply contracts or price tables and which are
identifiable at the time of the issuing of the respective invoice.
122 See the January 2010 Report, quoted above, para. 136.
123 Both in 2004 and in 2005, all the cases it decided under Decree-Law 370/93 (34 in 2004, 5 in

2005) related to sales at loss. The Authority indicated that there were 400 cases pending in
August 2004. In 2004, it decided 34 cases under Decree-Law 370/93, leading to fines totalling
€500,000, all of them relating to sales at loss. 18 of these cases were appealed. In 2005, it
decided four cases, all relating to sales at loss, leading to fines totalling €14.968 (none was
appealed). Still very little information is available from official sources of the Competition
Authority. The following summaries of cases were taken from facts described in judgments by
Courts of Appeals.

PAGE 77 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

law and applied a stand-alone concept of unfair pricing under the unfair
competition law,124 can be considered theoretically in line with Article 102(a)
TFEU. However, the provision is reportedly enforced in way that diverges from
Art. 102 TFEU, for instance, because in some cases the recoupment of losses was
required, while in other assessment remains formalistic and based on the
existence of cumulative pre-established conditions with little attention for the
analysis of the relevant market.
Second, Law no. 321/2009 regarding the sale of food products deals with pricing
strategies from all undertakings, irrespective of their market power or financial
strength, namely, to both dominant and non-dominant undertakings. Pursuant
to Article 5 of the said Law, the offerings or selling at loss shall be prohibited to
any trader, save for the legal exceptions of the Government Ordinance no.
99/2000 regarding the sales of products and services.125 Accordingly,
promotional sales as well as the sales of products as tied to packages of services
are the only excluded from the list of the prohibited acts. Within the meaning of
this ordinance, selling at loss is any selling at a price below the purchase price.
According to Article 6, any trader is prohibited to demand from its supplier not to
sell other traders the same products at a purchase price that is below or equal to
the one that he had acquired himself.
Fines for the infringement of the provision of the Law no. 321/2009 are imposed
by the Ministry of Public Finance, which may request all the necessary
documents underlying the conduct of such business relationships.
The recent Romanian Competition Commission’s (“RCC”) report126 is critical of
this provision and proposed its abrogation, mainly for two reasons: (i) because
sales at a loss may also have beneficial effects for the final consumers, even when
implemented by a dominant undertaking; and (ii) because predatory pricing is
already regulated by Article 6 RCL. However, it should be recalled that above
mentioned provision refers to all firms active in the market, irrespective of their
dominant position, and leads to the assessment of the practice on the basis of the
different criterion of the “average consumer” that may be induced or deceived by
comparative advertising, deceptive pricing, intentional predation. The RCC
however has affirmed that due to their ‘superior bargaining power’ such traders
may also exercise pressure and impose higher acquisition prices or other
financial advantages.127 This would in turn lead to a merely artificial result, and
would not contribute to the strengthening of weaker retailers.
Third, it must be recalled that according to the Romanian Accountancy law no.
82/1991, republished in 2008, OJ I-454 of 18 June 2008, the practice of selling
below costs is a form of unfair competition as dumping pricing and it may also be
interpreted and applied by the fiscal inspection authorities as amounting to tax
evasion through the declaration of duties and lower taxes to the state budget,
save for those transactions having no economic purpose.
Figure 4 below provides a graphical summary of the findings of this section. As
already explained for Figure 2 above, drawing a demarcation between
“significantly” and “slightly” diverging rules on sales below cost is far from easy,

124 RCC decision no. 237 Astral Telecom and Cablevision Romania of December 12, 2006.

126 at 7.
127 Ibid., 182.

PAGE 78 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

as the actual extent of the divergence depends on several concurring factors. In

Figure 4, we have considered that four countries (France, Germany, Portugal,
Romania) feature an explicit ban of below cost sales, which goes even beyond the
requirement of dominance. In other countries, below-cost sales are subject to a
stricter treatment than under Art. 102 TFEU, e.g. through the application of
presumptions of anticompetitiveness. These countries are Austria, Bulgaria and
Latvia. However, it should be noticed that both in Austria and in Latvia, such
negative treatment remains mostly theoretical, as it finds little or no support in
practice.

Europe –Figure 4 – National rules on sales below cost

Countries

Rules the significantly diverge from Article 102 TFEU

Rules that slightly diverge from Article TFEU

2.2.3 Tied sales

Traditionally considered as an anti-competitive practice and originally subject, in

some jurisdictions like the United States, to a per se rule, tying is now considered
as a ubiquitous practice that does not normally pose serious competitive
concerns, unless it is adopted by a dominant undertaking and is likely to produce
anticompetitive foreclosure. Accordingly, the EU Unfair Commercial Practices
Directive adopted in 2005 does not include tying in the list of unfair (i.e.,
misleading or aggressive) practices that can be challenged by EU law

PAGE 79 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

irrespective of dominance128. Given that this directive has been implemented in

all Member States and features a “maximum harmonization” approach, Member
States shall not be allowed to keep legislation in place, which bans tied or
bundled offers regardless of any reference to the market power held by the
undertaking that engaged in the allegedly abusive conduct, with the exception of
the sectors that are explicitly exempted from the application of this general rule
(financial services and immovable property)129
Under EU competition law, tying and bundling are considered common practices
intended to provide customers with better products or offerings in more cost
effective ways. However, an undertaking which is dominant in one product
market (or more) of a tie or bundle (referred to as the tying market) can harm
consumers through tying or bundling by foreclosing the market for the other
products that are part of the tie or bundle (referred to as the tied market) and,
indirectly, the tying market. In its Guidance paper on Art. 82 EC (now Article 102
TFEU), the Commission will normally take action under Article 102 TFEU where
an undertaking is dominant in the tying market and where, in addition, the
following conditions are fulfilled: (i) the tying and tied products are distinct
products, and (ii) the tying practice is likely to lead to anti-competitive
foreclosure130.
However, some EU member states still preserve legislation on tying, which
applies a per se rule or even bans combined sales regardless of whether the
undertaking that engaged in tied sales is dominant or not. Below, we briefly
describe the rules in force in the EU27. Interested readers can refer to Annex VI
of this Report for more detailed information.

2.2.4.1. Belgium

Belgium has recently replaced its general law on unfair trade practices by a new
Act. The new Trade Practices and Consumer Protection Act of 6 of April 2010
entered into force on 12 May 2010. It contains a substantial change in
comparison with the old Law as far as tying is concerned. Such practice is now
authorized. For this reason, we exclude Belgium from the group of countries
exhibiting diverging rules.
Tying was previously forbidden in Belgium that had one of the most
comprehensive bans on tied sales throughout the EU27. The main provision
prohibiting combined offers was embodied in Article 54 of the Belgian Law of 14
July 1991 on trade practices and consumer information and protection outlaws
combined offers for the consumers, with the aim of protecting consumers. The

128 Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005
concerning unfair business-to-consumer commercial practices in the internal market and
amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and 2002/65/EC of the
European Parliament and of the Council and Regulation (EC) No 2006/2004 of the European
Parliament and of the Council (‘Unfair Commercial Practices Directive’), OJ L 149, 11.6.2005, p.
22–39.
129 See Renda et al. (2009), Final Report of the Study on Tying and other potentially unfair

practices in retail financial services, a report for the European Commission DG Internal Market,
available online at http://ec.europa.eu/internal_market/consultations/docs/
2010/tying/report_en.pdf.
130 See case T-201/04 Microsoft v Commission [2007] ECR II-3601, in particular paragraphs 842,

859 to 862, 867 and 869

PAGE 80 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Article prohibited any combined offer to consumers which is made by a seller, as

well as any combined offer to consumers which is made by several sellers acting
with a common purpose131. The Article also specified that “a combined offer
exists where the acquisition, whether or not free of charge, of products, services
or other advantages, or of vouchers with which they can be acquired, is tied to
the acquisition of other, even identical, products or services.” The wording of
Article 54 thus encompassed both tying and bundling offers, and entailed a very
broad prohibition that could capture nearly every kind of cross-selling practice,
including mixed bundling. There were, however, some exceptions to such ban.132
In summary, the main exemptions to the rule of Article 54 concerned situations
in which various products form a single product or identical items at the given
conditions at all-inclusive price (Article 55), or ancillary products or services
which are offered for free (Article 56). This rule was then addressed by the
European Court of Justice in its judgment related to the VTB-VAB and Sanoma
cases133. The ECJ judgment of 23 April 2009 clearly states that combined offers
should be deemed lawful except when they can be qualified, given the specific
circumstances, as an unfair commercial practice. The consequence of the ECJ
ruling was that Belgium had to adapt its own national legislation to bring it in
line with the EU Unfair Commercial Practices Directive (UCPD) and Belgian
courts to comply with the new Law.
The new provisions on combined offers are laid down in Articles 71-72, Section 5
of the Act. A combined offer is defined by Article 2, 27° of the TPCPA as “An offer
where the acquisition of goods or services, whether or not free of charge, is tied to
the acquisition of other goods or services”. Article 71 provides that “Without
prejudice of Article 72, the combined offer to a consumer is allowed if it does not
amount to an unfair commercial practice covered by Article 84 and seq. [of the
TPCPA]”. That means that the combined offer (i) cannot be contrary to the

131 The original text of the law (in French and Dutch) can be found at the following website:
http://economie.fgov.be/protection_consumer/trade_practices/trade_law/law_on_protection_
001.pdf.
132 Article 55 allowed the combined offer of “products or services which form a whole”, and also

“identical products or services”, provided that: (a) Each product and service can be acquired
separately at the normal price in the same establishment. Thus, pure bundling and tying are not
allowed.; (b), the purchaser is informed clearly of that possibility and of the individual price of
each product and service; (c) Any price reduction granted to the purchaser of the totality of the
products or services does not exceed one third of the individual prices added together. This limits
the possibility to offer single-product conditional (loyalty) rebates. In addition, Article 56 allowed
the combined sale of products together with: (i) accessories, which the manufacturer of the
product has specifically adapted to that product and which are supplied together with that
product in order to extend or facilitate its use; (ii) the packaging or containers used for the
protection and market preparation of products, taking into account the nature and value of those
products; (iii) small products and services accepted as customary in trade, as well as the delivery,
installation, control/regulation and maintenance of the products sold; (iv) samples from the
product range of the manufacturer or supplier of the main product, provided that they are offered
in the quantities or sizes strictly necessary for an assessment of the characteristics of the product;
(v) colour photographs, stickers and other images with minimal commercial value; (vi) tickets for
legally authorised lotteries; and (vii) objects with indelible and clearly visible advertising
inscriptions, which are not found as such in the shops, provided that the cost price paid by the
supplier does not exceed 5% of the retail price of the main product or service with which they are
given away.
133 Judgment of the European Court of Justice of 23 April 2009, VTB-VAB NV v Total Belgium

NV and Galatea BVBA v Sanoma Magazines Belgium NV, joined cases C-261/07 and C- 299/07.

PAGE 81 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

requirements of professional diligence and distort or likely to distort the average

consumer’s economic behavior or (ii) be misleading or aggressive.
According to Article 72, §1 “Any combined offer to the consumer, which includes
at least one item that constitutes a financial service and which is made by an
undertaking or by different undertakings pursuing a common objective, is
forbidden.” The Article, however, contains a long list of specific practices that are
allowed134.
In terms of enforcement, private actions related to competition law are not very
developed in Belgium. Most cases are put forward on the basis of violation of the
Trade Practice and Consumer Protection Act. There is no particular provision in
the Competition Act that would serve as a basis to bring private actions before
Belgian Courts. Therefore, private parties must show that the infringement at
hand occurs through a breach of contractual obligations and/or in quantifiable
loss, i.e. tort on the basis of Article 1382 of the Belgian Civil Code135. Under the
old rules, the Court of Appeal of Brussels has several times based its reasoning
upon the prohibition of combined offers. For instance, in the Delta Lloyd Bank
case which was held a few days before entry into force of the new Law, the said
Court of Appeal ruled that the combined offer at issue (i.e a banking account
offer combined with securities investment) was deemed to be an unfair
commercial practice given that it was contrary to the requirement of professional
diligence and altered, or was likely to altered, the economical behaviour of the
targeted average consumer.136
The Supreme Court has referred to the above-mentioned exceptions to the
general prohibition so as to clear combined offers. In one particular case
regarding an appeal against the judgment of the Brussels Court of Appeal that
had allowed a combined offer in the financial services sector, the Supreme Court
discussed both the application of the old combined offer rules and the sales at

134 Article 72, §2 sets out that “By derogation to §1, the following combined offers are however
allowed: (1) financial services which form a whole; The King may, after the proposal of the
relevant ministers and the minister of Finance, set out the proposed services in the financial
sector which form a whole; (2) financial services and small goods and small services accepted
by trade practices; (3) financial services and tickets for legally authorized lotteries; (4)
financial services and objects with indelible and clearly visible advertising inscriptions, which
are not found as such in the shops, provided that the cost price, paid by the undertaking does
not exceed 10 euro, excluding VAT, or 5% of the retail price, excluding VAT, of the financial
service which they are given away. The percentage of 5% applies if the amount which
corresponds to this percentage exceeds 10 euro (5) financial services and color photographs,
stickers and other images with minimal commercial value; (6) financial services and voucher
consisting in documents that entitled, after the acquisition of a certain number of
goods/services, to be granted a free of charge offer or a price reduction when purchasing
similar goods/services, provided that this benefit is conferred by the same undertaking and
does not exceed one third of the services previously acquired. Voucher must mention the possible
limit-limit of their validity as well as the offer’s methods. When the undertaking interrupts its
offer, the consumer must benefit from the advantage offered for the portion of purchases
previously made.”
135 To claim damages, the claimant must prove (a) fault; (b) the existence of the loss
sustained incurred; and (c) the causal link between the unlawful act and the foregoing losses. As a
matter of principle, these damages are tantamount to compensation and no punitive damages can
be granted by the judge. It should be noted that beyond damages, private parties are entitled to
claim interim relief on the basis of Article 110 of the Trade Practice and Consumer Protection Act.
136 Court of Appeal of Brussels, 4 May 2010, 2008/AR/979.

PAGE 82 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

loss.137 It held that combined offers in the said sector are allowed if they fall
within the exceptions set out in the old Article 57(1)(4).138 For that purpose,
goods or services must be considered as “similar” if is included in the assortment
of the seller. The argument according to which the similarity between
goods/services that compose the combined offer should be assessed on the basis
of the intrinsic characteristics lacks merit. Second, the Court found that the
ground regarding the sales at loss was irrelevant. The combined offer provisions
enable consumers to acquire the tied good by means of voucher delivered
through the acquisition of the tying good. It results that there is no sales at loss in
this mechanism. The only price restriction related to tied product discounts is set
forth in the combined offer provisions (i.e. the benefit conferred cannot exceed
one third of the goods/services previously acquired.
The Supreme Court has also held that the combined offer is not in breach of the
Law on Trade practice when it is unknown to the potential targeted group of
consumers.139 It has also upheld the decision of the President of the Brussels
Commercial Court to suspend the application of the unlawful combined offer.
The contested offer consisted in setting a global price for various furniture that
could not be considered as forming a whole set of goods.140 In relation to
combined offers in the financial services sector, which are in principle prohibited
unless the exceptions of the now Article 72 apply, case law have established that
banking services that are offered in combination with insurance services141 or
financial instruments142 cannot be considered as a whole set of services insofar as
they do not fall within the same commercial branch. They are therefore
prohibited.

2.2.4.2. Bulgaria

Bulgaria has addressed tying and other potentially unfair practices in national
legislation. In November 2008, the Bulgarian Parliament adopted a new
Competition Act143. This law is not only an antitrust law, but also covers unfair
competition. Article 36(2) lays down the provision which prohibits the combined
offering of goods and/or services, and in particular:
“Offering or giving of an additional product to the product or service being sold,
free of charge or against the fictitious price of that other product or service shall
be forbidden, except for:
- advertising products which have insignificant value ... where there is an
explicit indication of which is the advertising company;

137 See Supreme Court, 25 January 2010, N° C.06.0025 F.

138 The relative provision in the current version of the Law is contained in Article 72(2)(6):
“financial services and equities consisting in documents that enable, after the acquisition of a
certain amount of services, to be granted a free of charge offer or a price reduction when
purchasing a similar service/goods, provided that this benefit is conferred by the same
undertaking and does not exceed one third of the services previously acquired”.
139 Supreme Court, 30 January 2001, N°990451N.
140 See Supreme Court, 20 February 2003, N° C. 010504F.
141 Court of Appeal of Brussels, 25 August 1999.
142 President of the Commercial Court of Brussels, 5 March 2008.
143 Published in State Gazette No. 102 of 28.11.2008.

PAGE 83 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

- products or services which, according to the business practices, constitute

accessories to the product or service being sold;
- products or services, which constitute a rebate for the sale of bigger
quantities”.
This provision thus prohibits tying, pure bundling and mixed bundling (only
when the bundled product is added free of charge or at a fictitious price), unless
the tied products are free samples, or accessories. The available case law suggests
that the above provision is not limited to a particular sector and applies in any
case when the interests of competitors is harmed or may be harmed. The purpose
of the provision is to regulate undertakings’ behaviour in order to guarantee a
loyal competition among them to the benefit of consumers.
The competent authority dealing with the cases of tied sales is the national
competition authority, which reportedly enforces the rule quite strongly.
Damages can be claimed before a court on the ground of Article 104, paragraph 1
of the Competition Law. We do not have precise information about litigated
cases, as cases are not publicly available144. However, it is highly likely that there
are no privately litigated cases, as private enforcement is still very much
underdeveloped.

2.2.4.3. France

France was one of the first Member States of the EU to adopt legislation to limit
the use of tying and bundling practices. Tied sales are prohibited under
restrictive trade practices, governed by Article L. 442-1145 of the Code of
Commerce , which reproduces Article L. 122-1 of the Consumer code, and under
competition law pursuant to Article L. 420-2 of the Code of Commerce as stated
above (see Section 2.2.1). These provisions cover all sectors and aim at protecting
end consumers and effective competition. The competent authority for Article L.
420-2 is the Autorité de la Concurrence; for article L. 442-1, it is possible to
claim damages before the competent civil or commercial courts. Fewer than 10
decisions have been issued by the Autorité, only one of which was issued since
the change of the law in 2008 concerning L. 442-1 of the Code of Commerce, and
none concerning L. 420-2 of the Code of Commerce.
Article L. 122-1 of the Consumer Code (which dates back to 1978 and is by
definition aimed at the protection of consumers) explicitly prohibits tied sales,
stating that “it is prohibited to refuse to sell a product, or supply a service, to a
consumer without a legitimate reason and to make the sale of a product subject
to the purchase of a minimum quantity or to the accompanying purchase of
another product or another service as well as making the provision of a service
subject to provision of another service or to the purchase of a product”146.

144 See also the Report by the International Competition Network on Tying and Bundled
Discounting, available at
http://www.internationalcompetitionnetwork.org/uploads/library/doc356.pdf. The Bulgarian
competition authority was among those that reported no private enforcement cases (see footnote
23 and accompanying text).
145 Law No 2001-1168 of 11 December 2001 Article 13 IV (2) Official Gazette of 12 December 2001)

(Law No 2003-7 of 3 January 2003 Article 50 (II) Official Gazette of 4 January 2003.
146 Text in English: http://195.83.177.9/code/liste.phtml?lang=uk&c=61&r=2133.

PAGE 84 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Article L. 122-1 addresses tying, pure bundling and exclusive agreements, but
does not cover preferential agreements and – contrary to the Belgian rule –
mixed bundling. Sales with rebates are allowed by the Consumer Code but are
nevertheless strictly regulated by other pieces of legislation as to their
advertisement, announcement or other aspects. Similar to what occurred in
Belgium (see above in this section), in France the compatibility of the ban on tied
sales with the provisions of the Unfair Commercial Practices Directive of 2005
(UCPD) has been subject to attention. On 14 May 2009, the Court of Appeal of
Paris has overturned a judgment by the Commercial Court, which found Orange
guilty of tied sale practices and unfair competition in its dispute with its
competitors Free and SFR147. The latter complained that the operator was
making subscription to its channel Orange Sports dependent on a subscription to
Orange’s broadband Internet access. Having been ordered in the first instance to
stop the practice, Orange referred before the Court of Appeal to the European
Court of Justice the judgment related to the VTB-VAB and Sanoma cases. In the
light of this jurisprudence, the parties called on the Court of Appeal to interpret
the national legislation, and more specifically Article L. 122-1 of the Consumer
Code. The Court found that the ECJ decision of 23 April 2009 was applicable to
the present dispute, and noted that Article L. 122-1 of the Consumer Code was
incompatible with the scheme instituted by the UCPD in that it prohibits tied
offers generally and preventively, regardless of any check on their unfair nature
with regard to Articles 5-9 of the UCPD. The court, contrary to the claims made
by SFR and Free, held that the mere fact of the consumer having to take out a
subscription to Orange broadband in order to have access to the Orange Sports
channel did not meet the definition of constraint, also because all operators on
the market use tied sales as a way to win customers. What was essential
according to the Court, within the meaning of the UCPD, was that the subscriber
was free to not take up the subscription, which was not contested in the present
case. Since France Orange could not be held to have infringed Article L. 122-1 of
the Consumer Code, as interpreted in the light of the 2005 Directive, the
judgment was overturned.
Moreover, the sales of services with free gifts are prohibited save for small gifts
by the virtue of L. 212-35 of the Consumer Code148.
Unfortunately it was impossible for us to collect reliable statistical information
about cases dealt with by the courts. However, our national contact reported that
after the ECJ decision in the VTB-VAB and Sanoma cases in 2009, courts have
been way more reluctant to enforce the provisions on tying in France. Before
2009, Article L. 122-1 of the Consumer Code led to frequent cases in court.

147 Court of Appeal of Paris (centre 5, chamber 5), 14 May 2009, France Telecom and Orange vs.
Free, Neuf Cegetel-SFR and LFP.
148 The text of English version reads: “All sales or offers for sale of goods or any other provision

or offer to a fixed period, to a bonus consisting of products, goods or services, if these are
identical to those forming the subject of the sale or the service provision, are prohibited. This
provision does not apply to small objects or low value services or samples”.

PAGE 85 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.2.4.4. Germany

Bans on tied sales are provided for by the Unfair Competition Act (Gesetz gegen
den unlauteren Wettbewerb, UWG), and in particular in § 4 UWG, which,
however, regulates them as unfair business conducts. In 2004 a new Unfair
Competition Act (UCA) entered into force (3 July 2004 - BGBl. I 2004 32/1414).
The purpose of the Unfair Competition Act (hereinafter the “UCA”) is to protect
competitors, consumers and other market participants against unfair commercial
practices. At the same time, it shall protect the interests of the public in
undistorted competition. § 3 of the Act provides a general clause prohibiting
unfair commercial practices. This clause is supplemented by a list of unfair acts
of competition in § 4 UCA, such as regarding unreasonably manipulating or
exploiting consumers, surreptitious advertising, sales promotions, competitions,
draws and prizes, and so on. To a great extent the new Act codifies and
systematizes the case law of the German court decided under the old Act. § 3 of
the UCA contains a general principle which prohibits: (1) acts of competition that
are (2) unfair and (3) capable of materially distorting competition (4) by harming
competitors, consumers or other market participants (5). § 4 no. 1 aims at
preventing the exploitation of a position of power in relation to the consumer by
applying pressure in a way which significantly limits the consumer’s ability to
make an informed decision149.. § 4 no. 6 of the AUC mainly aims at protecting
consumers against any undue influence on their purchasing decision by the
exploitation of their propensity for gambling. In particular, § 4 UCA also states
that such “undue influence” can be achieved through “tied sales”. However, the
provision does not prohibit combined offers by a retailer to a consumer as such;
on the contrary, this practice is generally regarded as pro-competitive150. It can
therefore does not seem to be a national standard which differs from European
law.
Competition authorities are not competent to protect the above mentioned
stakeholders against unfair competition in terms of the UCA. The claims
regarding the application of UCA are enforced in a civil law procedure. Cartel
authorities have concurrent competence only as far as a conduct under UCA is
also a conduct under GWB (e.g. the deliberate obstructing of competitors, see §
20 ACR and § 4 No. 10 UCA) 151.

2.2.4.5. Hungary

In Hungary, there are several provisions dealing with tied sales, including those
on abuse of dominant position contained in the Hungarian Competition Act. As

149 cf. Köhler/Bornkamm, Beck’sche Kurzkommentare UWG, 29th edition, UWG Article 4,
Paragraph 1.12
150 cf. Köhler/Bornkamm, Beck’sche Kurzkommentare UWG, 29th edition, UWG Article 4,

Paragraph 1.103 onwards

151 Cases of unfair competition under the UCA are within the competence of the civil courts and

the Bundesgerichtshof (BGH) as the highest civil court of final appeal in Germany. Therefore,
infringements of Art. 4 UCA are also invoked by civil persons, including the cases of “tied sales”.
The focus of the UCA is more on the consumer protection than is the focus of the ACR which aims
at first at the protection of a free market competition. Cases include: BGH, Judgement of 22
January 2009 - I ZR 31/06, BGH, Judgement of 3 March 2005 - I ZR 117/02.

PAGE 86 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

already stated above, there is no reason to believe that the provisions contained
in the Competition Act amount to rules that are stricter than Article 102 TFEU.
Accordingly, below we are going to focus our attention on the Hungarian Act on
Trade.
The Act on Trade imposes a special provision related to tied sales. Compelling the
supplier to manufacture products under the trader's trademark or brand name as
a precondition for the marketing of any other product of the supplier is
considered to be as an abuse of dominant position. As already recalled, this Act
applies to those companies whose trading activities from the previous year are in
excess of 100 billion forints.
Point e) Paragraph (2) Section 7 of the Act on Trade, while prohibiting the abuse
of superior bargaining power, clarifies that such abuse can consist in, i.a.:
“imposing unfair conditions upon the supplier in connection with his business
relations with the trader or with another trader, such as demanding the best
available terms and conditions as obligatory, and enforcing such terms and
conditions with retroactive effect, i.e. compelling the supplier to provide
discounts during a specific period for a specific product only to the trader in
question, or compelling the supplier to manufacture products under the trader's
trade mark or brand name as a precondition for the marketing of any other
product of the supplier.”152 The provision covers all sectors and aims at
protecting consumers.
The Hungarian competition authority enforces this provision, although the
number of cases is reportedly low – in 2008 the authority reported to have dealt
with very few cases on this issue153. In Annex VI we provide relevant case law. As
regards settlement provisions for cases related to prohibiting or significantly
constraining tied sales these were analysed in the previous sections. The cases Vj-
80/2004, Vj-40/20 analysed in Annex VI1 contained settlement provisions
related to tying cases. Private parties can challenge tying arrangements also in
court. However, since then there have not been any private enforcement cases to
our knowledge.

2.2.4.6. Poland

In Poland, tying and other potentially unfair practices are regulated by

competition law – the Act on competition and consumer protection – and by the
Civil Code. The Civil Code contains a catalogue of abusive contractual clauses,
which undertakings are prohibited from applying in their relations with
consumers, independently of their market power. These clauses include the
clause on tied transactions. According to Article 3853(7) of the Civil Code154,
making the conclusion, content or performance of one contract conditional upon

152 For a description of very relevant cases decided by the Hungarian Office of Economic
protection based on the Act on Trade, see Section 4.2 below.
153 See the Report of the ICN, in which the authority responded that it “has very few cases

concerning tying or bundled discounting, although the authority has now an ongoing
investigation concerning a tying arrangement in the field of broadcasting”. The Report is at
http://www.internationalcompetitionnetwork.org/uploads/questionnaires/uc%20tying/hungary%20tbd.
pdf.
154
Act of 23 April 1968 – the Civil Code; Journal of Laws of 1968 No.16, item 93 with further
amendments.

PAGE 87 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the conclusion of another contract, not directly connected with the former one, is
a prohibited contract clause155. It should be noted that the above-mentioned
provision is applicable only when the contract is not individually negotiated, as
individual negotiations of contractual terms would waive the possibility to rely on
the above-mentioned provision. The prohibition of applying the clause in
question concerns only the relationship between the undertaking and the
consumer and the undertaking’s market shares are not taken into consideration
in such cases. The provision covers all sectors and aims chiefly at protecting
consumers, rather than competition. They are applied by civil courts (individual
consumer interests) and by the Court of competition and consumer protection
(collective consumer interests). The Competition Authority is not directly in
charge of enforcing these provisions. The Authority may only initiate judicial
action before the Court of competition and consumer protection in collective
consumer interests cases.
There is also private litigation related to the above-mentioned provision,
especially in the financial services sector, which can be based on the national
competition law and on the national civil code. In particular the ruling of the
District Court in Warsaw – Court of Competition and Consumer Protection (file
no. XVII AmC 101/04) declaring contract clauses tying obtaining a (students)
loan and possessing a bank account as prohibited. Such a type of clause is now
incorporated in the Register of Prohibited Clauses. According to procedural rules,
once the District Court of Warsaw has ruled that a particular clause is unfair it is
entered in the Register and cannot be practiced.

2.2.4.7. Portugal

In Portugal, the consumer protection law (Law no. 24/96, of 31 July 1996, as
amended by the Decree-Law no. 67/2003) features a very broad provision
discouraging tying, found at Article 9(6). The provision (chiefly aimed at
protecting consumers) prohibits “to the supplier or service provider to make the
supply of goods or the rendering of a service dependent of the acquisition or the
rendering of any other goods or services.” It appears that this provision does not
have any qualifying exemptions which would make some types of combinations
possible. In this way, the provision is quite similar to Article 54 of the Belgian
Article 54 of the Belgian law on trade practices and consumer information and
protection).

155 The relevant provision in the Code reads:

“Article 3851. § 1. Provisions of a contract concluded with a consumer which have not been
individually agreed upon shall not be binding if, contrary to good practices, they set forth the
consumer’s rights and obligations in a way that is flagrantly infringing his interest (prohibited
contract terms). The above does not apply to provisions defining the main obligations of the
parties, including price or remuneration, if they were formulated in an unequivocal way. (…)
Article 3853. In the event of doubt, it shall be deemed that unfair contract terms are in
particular provisions which:
(…)
7) make the conclusion, content or performance of the contract conditional upon the
conclusion of another contract, which is not directly connected with the contract containing the
provisions under review”.

PAGE 88 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Moreover, Decree-Law no. 370/1993 on individual practices restricting trade

considers tied sales a form of refusal of sale of goods or supply of services and
prohibits them except when such tying constitutes normal commercial practice in
that activity or when objective justifications may be invoked156.
This provision covers all sectors, but is not directed at the protection of any
specific stakeholders, although the Lisbon Court of Appeals has stated that its
ultimate goal is the protection of consumers.
Civil Courts are the enforcing authority. Cases are estimated by the consulted
expert to be more than 100.

2.2.4.8. Romania

In Romania, Article 4(1) of the Romanian Law no. 321/2009 regarding the
commercialisation of foodstuffs may be interpreted as a prohibition of tying of
services in the food and agricultural sector157. Law no. 247/2010 has amended
Law no. 321/2009 and in particular it modified Art. 4 (1) which reads as follows:
“traders are prohibited to demand and cash in any payments for services which
have no direct connection with the sale transaction”.
Law no. 321/2009 aims at covering the agricultural and the food sector and aims
at protecting human health and consumers. It does not aim to address primarily
anti-competitive practices but unfair methods of competition regarding the
commercialisation of foodstuffs. This explains that any trader, i.e., both
dominant as well as non-dominant firms may fall under these severe and special
provisions if they engage in such practices. Therefore it belongs to the consumer
protection area and for this reason will not be analysed in detail in this study.

156 Article 4 (Refusal of sale of goods or supply of services) reads as follows:

1- An economic agent may not refuse the sale of goods or the supply of services to another
economic agent, according to the normal uses of the respective activity or in accordance with
applicable legal or regulatory provisions, even in the case of non essential goods or services and
even if the refusal does not damage the regular supply of the market.
2- Making the sale of goods or supply of a service conditional upon the purchase of another
good or service is considered identical to a refusal of sale.
3- The following are considered legitimate grounds to refuse sale:
a) Meeting the normal requirements of the industrial or commercial practice of the seller, in
particular the maintenance of safety stocks or own consumption needs;
b) Meeting commitments previously accepted by the seller;
c) Manifest disproportion of the order in light of the normal requirements of the purchaser
or the usual volumes of deliveries of the seller;
d) The purchaser’s incapacity, given the characteristics of the goods or services in question,
to ensure its resale under satisfactory technical conditions or to maintain an adequate after-
sales service;
e) A justified lack of trust of the seller in what concerns the purchaser’s punctuality in
meeting payment, in the case of sales on credit;
f) The existence of unpaid debts relating to previous supplies;
g) The occurrence of any other circumstance inherent to the specific conditions of the
transaction which, according to the normal uses of the respective activity, would render the sale
of the goods or supply of the services abnormally damaging to the seller.
4- It is for the seller to prove the grounds for justification mentioned in the previous number.
157 OJ I-705/2009.

PAGE 89 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The competent authority for applying the administrative fines for the
infringement of Article 4 is the National Authority for Consumer Protection
(NACP). The law is fairly recent, it is too early to report on the actual level of
enforcement. It should be noted that the practice of demanding or receiving from
a supplier the payment of services which have no direct connection with the sale
transaction is not enforced when the provisions of Article 6 of the Romanian
Competition Law (RCL) may also apply. Article 6(d) RCL contains a more similar
provision, where “making the conclusion of contracts subject to acceptance by
the other parties of supplementary obligations that have no connection with the
subject of such contracts”. This constitutes also contractual tying. Normally, a
more severe contractual tying under this special law would be the use of any
‘undue influence’ in order to demand or receive such ‘tied’ payments. Given that
the special provision does not mention them, Article 4(1) of the Law no.
321/2009 provides a parallel provision with that found in Article 6(d) RCL.
The Law. 321/2009 does not provide for specific settlement provisions. The aim
is to correct the commercial behaviour of the accused economic agent and to
enforce the legal prohibitions. The authority may also issue warnings before the
imposition of fines. However, the working procedure of the NACP also refers to
the settlement of disputes by agreement between handlers and customers.158
Individual complaints for acts of unfair competition, including consumer
protection are matters of private litigation before courts where prejudiced
consumers may lodge a complaint based on the Civil Code (Art. 1073, 1075-1706)
and demand full compensation. This also implies the proof of prejudice, damage
or harm and causality as for torts. The difficulties in proving these elements has
led so far to a fairly limited enforcement.

2.2.4.9. Other countries

Finally, in other countries tying practices are discouraged in particular sectors of

the economy, and especially in the financial services sector. This is particularly
important since, with the implementation of the Unfair Commercial Practices
Directive in the Member States, rules that ban tying practices per se are likely to
be considered inconsistent with the maximum harmonization rule of the UCPD.
However, the UCPD leaves Member States with a degree of freedom in the
financial services sector. As a matter of fact, as stated in Recital 9 and prescribed
in Article 3(9) specifically for financial services (as also for immovable property),
the UCPD is “without prejudice to the right of Member States to go beyond its
provisions to protect the economic interests of consumers”; accordingly,
“Member States may impose requirements which are more restrictive or
prescriptive than this Directive in the field which it approximates”. Countries
where specific rules are in place for the financial services sector that are more
restrictive include Cyprus159, Ireland and Slovakia. Consumer protection

158 See, the General procedure of dealing with consumers’ complaints regarding the lack of
conformity of products or services, available at
http://www.anpc.ro/anpc/anpcftp/proceduri/ProceduraGenerala.pdf.
159 There are no specific norms prohibiting or significantly constraining tied sales in Cyprus.

There is only one particular instance where tying is prohibited and this relates to mortgage loans.
According to section 37 of the Consumer Credit law, Law 39(I)/ 2001 it is prohibited for a
mortgage lender (who may according to the said law require a mortgagor to take out insurance on

PAGE 90 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

laws may also contain prohibitions of tied sales. For example, the Slovak Act No.
250/2007 Coll. “Consumer Protection Act and the amendment of the
Slovak National Council. (dated 09.05.2007, in force since 01.07.2007) applies to
every sector except those governed by professional chambers and prohibits tying
(with certain exceptions). Another example is France, where Article L 122-1 of
the Consumer Code prohibits tying. (See Annex VI for more details.)
Figure 5 below summarises the findings of this section.

Figure 5 – National rules on tying

Europe – Countries

Rules that prohibit tying

Sectoral rules (mostly financial services)

2.2.4 Unfair competition and consumer protection laws

In addition to the specific laws noted above, Member States usually have a law
prohibiting unfair competition across all, or a very broad swath, of the economy.

the mortgaged property, however, such insurance may be bought from any insurer) to impose
terms upon a mortgagor relating to insurance covering the mortgaged property which would not
have been otherwise imposed by an insurer. This rule is aimed at protecting consumers and
applies regardless of the market power held by the lender.

PAGE 91 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

These laws often apply to business-to-consumer relations, 160 but can also have
provisions applying to business-to-business relations. The Unfair Commercial
Practices Directive161 recognized that, “the laws of the Member States relating to
unfair commercial practices show marked differences which can generate
appreciable distortions of competition and obstacles to the smooth functioning
of the internal market” (Recital 3). While the Directive aimed to protect
business-to-consumer commercial practices, it noted that other unfair
competition practices, not addressed in the Directive, may hurt competitors and
business customers and left open the option of further examination of the need
for further action (Recital 8). Accordingly, although many of these rules cannot
be defined as chiefly pursuing the protection of competition, some of them are
also aimed at protecting competitors and affect the conduct of dominant (and
often also non-dominant) undertakings. Accordingly, we have decided to report
them for the sake of completeness.

2.2.4.1. Belgium

In Belgium, national law includes rules prohibiting refusal to supply that are
reportedly stricter than those relating to the application of Article 102 TFEU. In
Multipharma v Widmer (judgment of 7 January 2000), the Belgian Supreme
Court ruled that a refusal to supply is not prohibited by Belgian competition law,
cannot be prohibited under the Act of 14 July 1991 on Trade Practices and
Consumer Information and Protection (“Trade Practices Act”). However, this
does not apply to a so-called “abuse of rights” situation, which will be considered
as an unfair trade practice. The Supreme Court has indicated that an abuse of
rights exists: (i) where a supplier has no interest in the refusal (this means that
the refusal to supply was instigated with the sole intention to harm), and (ii)
where the refusal to supply would create a clear imbalance in the interests of the
parties involved (in particular when existing relationships between a supplier and
a customer are suddenly terminated). A refusal to supply that is not caught by
Article 3 of the Competition Act will thus only be prohibited as an abuse of rights

160 An example is the laws on unfair competition in force in the Czech Republic and Slovakia
where unfair competition is prohibited, and non-exhaustively defined, by Article 44 et seq. in Act
No. 513/1991 Coll., the Commercial Code (enacted in 1991, amended in the Czech Republic by Act
No. 152/2010 Coll. with effect from 01.07.2010). Examples of unfair competition include:
misleading advertising, misleading labelling, free riding on a competitor’s reputation, bribery,
comparative advertising, breach of trade secrets, and putting at risk the health of consumers and
the environment. Spain prohibits unfair competition in Act No. 3/1991 “on Unfair Competition”
(10.01.1991 amended several times, notably by Law 52/1999 of 28.12.1999 and Law 29/2009 of
30.12.2009) across the economy. The law also contains a non-exhaustive list, such as misleading
customers, denigrating competitors’ products, and illegal advertising. Bulgaria prohibits unfair
competition Section VII of the Competition Act. Examples listed there include: harming
competitors’ reputation, misleading, misleading advertising, comparative advertising, imitation,
unfair attraction of customers, prohibition of disclosure of manufacture or trade secrets. Cyprus,
in its Competition Act, prohibits the imposition of unfair trading conditions.
161 Unfair Commercial Practices Directive, Directive 2005/29/EC of the European Parliament and

of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in

the internal market and amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC
and 2002/65/EC of the European Parliament and of the Council and Regulation (EC) No
2006/2004 of the European Parliament and of the Council (‘Unfair Commercial Practices
Directive’), OJ L 149, 11.6.2005, p. 22–39.

PAGE 92 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

in exceptional circumstances162. A violation of article 3 of the Competition Act is

indeed often invoked in combination with an infringement of the 1991 Trade
Practices Act. Enforcement of the latter is a responsibility of the civil courts.
The above-mentioned findings of the Multipharma case has been further
confirmed by various case law: refusal to supply can be sanctioned either on the
basis of breach of the general EU and/or Belgian competition rules or because of
a violation of the Trade Practice Act coupled with an abuse of rights.163
In relation to the strict application of competition rules to refusal to supply, the
following findings should be highlighted:
 A dominant distributor’s refusal to supply newspapers to particular points of
sale because of profits reasons leads to an abuse of dominant position insofar
as newspapers are “unique “and they cannot be substitutable.164
 The refusal to supply of a dominant position shall not necessarily be viewed
as an abuse prohibited by competition law if there are objective reasons to it.
The refusal to supply shall only breach competition rules if such refusal is use
to exert control on the resale or is tantamount to a sanction.165
So far, Belgian Courts firmly stick to the principles established by the domestic
Supreme Court in Multipharma. Specific case law is devoted to the issue of free
riding. Settled case law is very strict towards free riding which is considered as an
unfair commercial practice upon which a company creates adverse effects on
other company’s interests. Such practice can lead to an infringement of the Trade
Practice Act irrespective of any dominant position on the market. The
Commercial Court of Namur has lined up the conditions that need to be fulfilled
to sanction free riding:166
 The provision of services/products that are copied shall be the result of
creative efforts that required significant time and financial investment. The
protected end product/services shall be sufficiently original;
 The end product/provision of services shall have an economic value;
 The person/company that copies the end product/provision of services shall
gain benefits from the efforts and investments of the seller;
 The copier shall not have carried any efforts to single out its own end
product/services with respect to that of the seller.
The case law have considered the following free riding practices as unlawful:
merely copy the end product of another company or offer very similar
products;167 copy the layout of the product in magazines168 or the slogan that is
used169 or even the general conditions.170 The Supreme Court has recalled that

162 See Geert A. Zonnekeyn, Monard D’Hulst (Kortrijk), Antitrust encyclopedia: Belgium,
November 2007.
163 See for example, Court of Appeal of Anvers, 11 October 2007; Court of Appeal of Liège, 5
February 2009 2008/RG/927 (where the Multipharma principles were applied to a prohibited
collusion between undertakings); Court of Appeal of Gand, 1st March 2010.
164 Commercial Court of Brussels, 19 October 1994, Ann. Prat. Comm & conc., 1994, p.543.
165 Court of Appeal of Brussels, 14 March 2003, A&M, 2003, p.388.
166 Commercial Court of Namur, 29 April 2009, R.R.D 2008, p.525.
167 Court of Appeal of Liège, 12 June 2008, Ann. Prat. Comm., 2008, p.470.
168 Court of Appeal of Brussels, 5 May 2009, I.R.D.I., 2009, p.266.
169 Court of Appeal of Liège, 17 January 2006, Ann. Prat. Comm., 2006, p.110.

PAGE 93 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the judge will have to assess free riding on a case-by-case basis since every
competitor is entitled to make copies.171 More specifically, the Supreme Court
held that the basic act of copying commercial offers shall be authorized unless the
copier breaches an intellectual property rights or set up its offer in contradiction
with fair commercial practice requirements.

2.2.4.2. Hungary

In Hungary, the Competition Act contains several provisions for conditional sale
practices and other potential unfair practices, which can be seen as stricter than
Article 102 TFEU. In particular, Section 2 of the Hungarian Competition Act
contains a general prohibition of unfair economic conduct, which prohibits i.a.
“to conduct economic activities in an unfair manner, in particular, in a manner
violating or jeopardizing the lawful interests of competitors, business partners
and consumers, or in a way which is in conflict with the requirements of
business integrity.” The objective of the provision is protecting competitors from
unfair competitive behaviour of their rivals. Section 2 of the Hungarian
Competition Act is not enforced by the GVH, but by courts172. Applications for
cease and desist orders and for damages with regard to unfair market practices
on the basis of section 86 of the Competition Act may be filed with the relevant
county court. The claimant may demand that the alleged violation is established
by the court, that the violation must be terminated and that continued violation
by the offender is prohibited. Claims for damages, and cease and desist orders
arising from the breach of other provisions of the Competition Act may be filed
with the courts of regular competence on the basis of the general rules of
indemnification under the Civil Code and the Civil Procedure Act. Further, civil
law disputes sometimes involve challenges to the validity of agreements which
constitute a breach of the Competition Act. The legal basis for such actions is
section 200(2) of the Civil Code which sets out that agreements concluded in
breach of legal regulations are generally null and void. In 2009, experts observed
that there had only been a few court decisions in Hungary covering private
antitrust litigation and there was no significant case law regarding claims for
damages based on breach of the competition rules on cartels and on abuse of a
dominant position. One such decision confirmed that claims for damages are
permissible if a violation of any of the provisions of the Competition Act has
occurred173. As regards the settlement procedure, remedial action, as well as
appeal process, these were have been analysed in the previous sections.

2.2.4.3. Romania

Article 6(1)(a) of the Romanian Competition Legislation prohibits the imposition

of unfair purchase or selling prices or any other unfair trading conditions and,
furthermore, the refusal to deal with certain suppliers or customers.

170 Court of Appeal of Brussels, 22 March 2005, Ing.Cons 2005, p.176.

171 Supreme Court, 29 May 2009, Pas. 2009, p.1374.
172 See Article 86(1) of the Hungarian Competition Act: “The court shall have the power to hear

and decide cases conducted against violations of the provisions contained in Articles 2 to 7.”
173 See http://www.chsh.at/fileadmin/docs/publications/Kofler-Senoner/hungary.pdf.

PAGE 94 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Besides competition law provisions, there are also other bodies of law dealing
with unfair competition practices. For instance, the Romanian Law regarding
Unfair Commercial Practices against Consumers and the Harmonisation of the
European legislation on consumer protection of December 12, 2007174 includes
the core aspects of unfair competition. It covers potentially all sectors of activities
and it is addressed to all undertakings with regard to certain business strategies
such as the marketing of products, advertising, the lack of transparency
information etc. Article 5 prohibits misleading advertising, which deceives the
consumer preventing him from making an informed and efficient choice. This
prohibition gains particular relevance not only if a monopolist deceives or is
likely to deceive an average informed consumer, but also when a non-dominant
undertaking with considerable power may be engaged in such practices. Article 6
refers to misleading actions in commercial practice such as any promotions or
sales of products if it contains false information, whilst Article 7 refers to
misleading omissions such as hiding information from consumers. Article 8
includes an aggressive form of unfair competition practices, in particular if the
average consumer’s freedom of choice is significantly impaired through
“harassment, coercion, including physical force” or by any form of “undue
influence”, i.e., the exploitation of a position of power in relation to the consumer
so as to apply pressure, even without using or threatening to use physical force in
a way which significantly limits the consumer’s ability to make an informed
choice, as detailed in Article 9.

Other potential unfair practices fall under other Romanian special provisions,
which apply to the food sector. Whenever such sector-specific legislation overlaps
with the general provisions on unfair practices, the former lex specialis precedes
the latter. We refer here to what has been already said about Law no. 321/2009
in the section on sales below cost.
An additional legislative measure is the Government Ordinance no. 99/2000
regarding commercialization and the sales of products and services on the
market, i.e., any type of marketing or sales modalities etc. Accordingly,
conditional sale is defined as conditioning the sale of one product upon the
purchase of a certain quantity of that product or the simultaneous purchase of a
different product or service. The provision of a service conditioned upon the
provision of a different service or the purchase of a different product may amount
to a conditional sale.
These rules address all individuals involved in the market, not only those
misusing or abusing their market power such as dominant undertakings or any
other firms having considerable financial strength, insofar as they engage in
intentional unfair anti-competitive practices. Such an assertion contradicts the
view that tying or conditional sales are prohibited per se, irrespective of the
market share of the firm involved in such practices.175 The provision is, indeed,
broader in scope than the unilateral conduct of dominant undertakings because
such practices can cause direct consumer harm and are based on the effects of
such practices, especially on consumers or competitors.

174Published in OJ I-899.
175Final report, Tying and other potentially unfair commercial practices in the retail financial
service sector, 181-2, available at http://ec.europa.eu/internal_market/consultations
/docs/2010/tying/report_en.pdf

PAGE 95 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

2.2.4.4. Other rules

Finally, another category of legal rules broadly aimed at protecting consumers,

rather than competition on the market, which address unilateral conduct, is
represented by rules on conditional sales practices and other potentially unfair
practices. These provisions are, to a large extent, being revised in light of the
Unfair Commercial Practices Directive, or constitute an implementing measure
that transposes the UCPD in the national legal system: in our opinion – they
would not be affected by a possible extension of the convergence rule at Article
3(2) of Regulation CE 1/2003176.
Relevant rules include the Austrian Act against Unfair Competition; Bulgaria’s
Articles 30-37 of the Unfair Competition Act; Article 44(1), of the Czech
Commercial Code; French Articles L442-1 of the commercial code (on
conditional sale practices and other unfair practices), Article L442-5 of the
Commercial Code (on resale price maintenance) and Article L446-2 (enacted in
2008 and modified twice, last in July 2010), which prohibits any contract term
that grants a producer the possibility to i.a. receive retroactive discounts, rebates
or agreements of commercial cooperation177; Germany’s Section 4 of the Unfair
Competition Act and Section § 21 of the Act on the Restraints of Competition (on
boycotts and other restrictive practices); Greek Laws 146/1914, 2251/1994 and
L149/2005; Italian Article 2598 of the Civil Code; the Spanish Act on Unfair
Competition (Law no. 3/1991, as amended, lastly in 2009), etc. These rules can
be deemed to fall outside the scope of this study, as they address mostly unfair
commercial practices and unfair competition in the form of defamation of rivals,
confusion of trademarks, misleading advertising, etc.
The figure below summarises this section. We considered that only in three
countries the national rules on unfair competition and consumer protection
might possibly amount to rules that are stricter than Art. 102 TFEU.

See above, Section 1.2.1.

176

These provisions overlap primarily with EU competition rules on vertical restraints (Article 101
177

TFEU), rather than abuse of dominance. They are enforced by civil courts.

PAGE 96 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Figure 6 – Relevant national rules on unfair competition and

consumer protection

2.3 Sectoral regulation

Besides national legal rules contained in general competition laws or other
competition-related legislation, a number of countries have taken action to
regulate unilateral conduct by dominant and non-dominant firms in a number of
specific sectors. The three most important areas in which this has occurred are:
(i) the grocery sector, where some European countries have enacted legislation
aimed at protecting both small suppliers from the big retail chains’ buyer power,
and small retailers from the competitive pressure exerted by large retailers; (ii)
the financial services sector, where conduct such as tying of certain retail
financial services (e.g. mortgage loans and payment protection insurance) has
been subject to ad hoc legislation or soft law measures such as codes of conduct;
and (iii) network industries, where a stricter approach to unilateral conduct has
been generally motivated by the need to ensure the transition from a
monopolistic market structure to a competitive one, and this led to the adoption
of ex ante regulation, which is certainly stricter than Article 102 TFEU.
Below, we describe the main rules in place in EU member states, which belong to
the groups identified above.

2.3.1 The grocery sector

Several policies are in place in EU Member states, which address specific

behaviour in the food sector. Some of these policies have been adopted in the
form of a binding law, other as soft law or even self-regulatory arrangements
such as codes of conduct. Many of these initiatives have been spurred by a

PAGE 97 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

sectoral investigation of national competition authorities. The purpose of these

rules in many countries is to protect small (and sometimes, local) suppliers from
the buyer power of large retail chains: accordingly, the provisions at hand often
impose on large buyers specific contractual obligations, specific rules on dealer
termination and/or non-discrimination obligations, aimed at avoiding that these
big players can fully exploit their bargaining strength vis-à-vis smaller
contractual parties that heavily depend on them. As a result, the protection of
competition is one of the objectives of these rules, although in most cases the
prominent goal of these legal provisions is the protection of certain competitors
or actors on the value chain (most often, SMEs). A notable example is the Irish
Restrictive Practices (Groceries) Order of 1987 (the ‘Groceries Order’), which will
not be described below as it was repealed in March 2006. We refer to this Order
here as the fact that it was repealed provided us with the opportunity of analysing
the effect of its removal through an econometric analysis, which is shown below
in Section 3.3.
We found relevant provisions in the following countries178:
 In the Czech Republic. Law no. 395/2009 “on Significant Market Power in
the Sale of Agricultural Products and Food and its Misuse” (of 09.10.2009, in
force since 01.02.2010) concerns only the sale of agricultural and food
products (see page 38 for a more detailed explanation). It was enacted to
tackle alleged abuse by retail chains of their bargaining position vis-à-vis
suppliers of agricultural products and food who are in a position of economic
dependence. Proceedings have been opened against large retail chains.179
 The German Act on Restraints of Competition, § 20(4)(1) specifically
identifies the sale of food below its cost price (as distinct from § 20(4)(3)
which applies to small or medium-sized undertakings involved in
distribution, or § 20(4)(2) which does not apply to sales that are merely
occasionally below cost price180.
 Another example is Hungary, where i) Section 7 of Act CLXIV of 2007 “on
Trade” (the “Trade Act”) applies to retailers with significant market power, ii)
Act XCV of 2009 “on the Prohibition of Unfair Trading Practices vis-à-vis the
Suppliers of Agricultural and Food Products” (the “Retailer–Supplier Act”)
applies to undertakings producing, processing, or redistributing agricultural
and food industry products without processing, and undertakings which sell
such products to end customers, and iii) Act XVI of 2003 “on the Agricultural
Market Organisation” applies to agricultural and food products.
 In Latvia, the concept, “dominant position in retail trade” (DPRT) was
introduced into the Competition Act, effective from 16.04.2008. The
prohibition of abuse of a DPRT is intended to protect small suppliers or
producers from abuse of economic dependence or abuse of superior
bargaining power by retailers. While the law does not specify which retail
sector “it was clear from the annotation to the amendments of Competition

178Some of the provisions listed below have been already discussed in previous sections. We recall
them in order to have a clearer picture of the sectoral rules in place in Member States’ legislation.
However, we did not consider them twice when analyzing the number of diverging rules per
country.

180 See page 24 above for a more detailed explanation.

PAGE 98 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

law as well as the comments of the Competition Council that the legislation
was aimed to restrict unilateral conduct of the biggest supermarket chains,
in particular”181. In 2009, the Latvian government has even requested the
Ministry of the Economy and the Competition Council to draw up proposals
to limit the influence of the major retail chains in Latvia, arguing that no
retail chain in Latvia should have more than a 10% market share.
 In Slovakia, Law No. 140/2010 on “Inappropriate Conditions in Business
Relations between Buyers and Suppliers of Groceries” (of 04.03.2010, in force
since 01.05.2010) is aimed at protecting the agricultural sector and food
industry, and avoiding misuse of retail chains’ buying power vis-à-vis
suppliers. The law is enforced by the Ministry of Agriculture, but was repealed
as of 1 August 2011.
The provisions indicated above illustrate the focus of many stricter national rules
on unilateral conduct on the retail sector, especially the supply and retail of food
products. This focus of the laws, along with the experience of the business survey,
workshops and interviews, led us to infer that enterprises active in grocery retail
and its supply as the first set of enterprises impacted by the rules which are the
subject of our study, as will be explained in Section 4 below.
Besides sectoral legislation in force, the groceries sector has been the subject of
specific sectoral investigations by competition authorities, as well as competition
cases that are relevant for the purposes of our analysis. Most investigations in the
grocery sector have been spurred by the growing concern by industry participants
over the increased potential for abuse of market dominance. Often stakeholders
have been asking for amendments in the relevant provision dealing with the
abuse of dominant position. It follows that, even if such investigations are mostly
aimed at assisting NCAs in identifying anti-competitive conduct in the grocery
sector generally, they may cover conduct that is solely caught by stricter national
rules on unilateral conduct. Thus, even though a sectoral investigation does not
constitute per se any evidence of the existence of a stricter rule than Article 102
TFEU, these initiatives by national competition authorities can contribute to the
enforcement of such rules.
In particular, the following investigations can be listed:
 In Austria an investigation by the Federal Competition Authority into buyer
power in the food chain was carried out in 2007, revealing the Austrian
grocery sector was highly concentrated; barriers to entry were high, (which
leads to a low number and limited expansion of new market entrants in the
retail and wholesale business over the last years); and there is strong evidence
of buyer power, especially in sectors with private labels and without must-
stock items.
 A similar inquiry was carried out jointly by the competition authorities in
Nordic countries Denmark, Finland, Greenland, Iceland, Norway and
Sweden. The report “Nordic food markets – a taste for competition” was
published in November 2005, looking into the competitiveness of the food
and retail markets.
One of their conclusions (p. 19) was that some of the agreements between
suppliers and retail chains may include arrangements with foreclosing and

181 Opinion of our national legal expert for Latvia. See Annex VI for more detail.

PAGE 99 OF 182
 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

other anticompetitive effects (e.g. slotting payments, marketing support). If

these agreements or practices can be shown to limit competition, they
“constitute a breach of competition rules.”
 Similarly, the Hungarian Competition Authority completed in September
2007 an enquiry into the relations between large retail chains and their
suppliers182. One year later, Hungary’s Ministry of Agriculture reportedly
proposed to sanction supermarkets and hypermarkets up to 2 billion forints
(€7.7mn) if they engaged in unfair practices against suppliers. This proposal
also aimed at defining fair practices in this sector, which reportedly would
include a ceiling for late delivery fees and limits to how much cheaper own
brands could be compared to third party brands. In addition, retail chains
would be obliged to sell at least 80% Hungarian goods as against the current
30 %.
 Another sectoral inquiry was carried out in the Netherlands, with the help
of the Bureau of Economic Policy Analysis (CPB)183. The study did not find
increasing buyer power between 1993 and 2005 by supermarkets at the
expense of suppliers’ profits from a perspective of static efficiency and low
prices for the consumer, because of increased competition among the
supermarkets and manufacturers.
 Likewise, in September 2006, the Portuguese competition authorities
undertook a study about large retailing groups in the Portuguese food sector
related to their buyer power and passing through of low prices to consumers.
The authority concluded that there was no harm to competition and
consumer welfare but that more research would be needed in specific product
markets and that case per case analysis was needed.
 A 2010 report184 by the Spanish Competition Authority on the agri-food
industry found that abuse of suppliers by retailers takes place and made some
proposals to reduce the incidence. The report made a number of
recommendations to inter alia reduce the scope for such abuse. These
recommendations include improved information provision to suppliers,
greater use of contracts perhaps even with officially approved provisions, and
codes of conduct, freely and voluntarily adopted, coupled with effective and
mandatory dispute resolution mechanisms. The report noted that the Spanish
Law on Unfair Competition (Law 3/1991 of 10 January 1991) provides that the
competition authority may intervene when alleged acts of unfair competition
distort competition and affect the public interest.
The report noted that there was a clear difference between the meanings of
competition-restricting conduct in the Competition Act and the TFEU, on the
one hand, and the Unfair Competition Act, on the other. It also noted the
desirability of developing a common approach to unfair practices deriving

182 Source: <http://www.gvh.hu/domain2/files/modules/module25/pdf/elemzesek_

gvhtanulmanyok_beszallitok_2007.pdf>
183 H. Creuse, A. Mejier, Gijsbert Zwart, H. van der Wiel, Static efficiency in Dutch supermarket

chain, CPB document nr 163, April 2008.

184 Spanish competition authority (Comisión Nacional de la Competencia) 2010. “Report on

Competition and the Agrifood Sector” (Informe Sobre Competencia y Sector Agroalimentario),
available at http://www.cncompetencia.es.

PAGE 100 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

from an imbalance of bargaining power which may significantly distort

competition in markets.
 Finally, in the United Kingdom the Competition Commission (CC) carried
out an investigation into buyer power in the food chain, which started in May
2006 and ended in April 2008185. This market investigation followed several
examinations by the Office of Fair Trading, and an appeal to the Competition
Appeal Tribunal (effectively a judicial review) by a number of aggrieved
parties. The CC’s inquiry was in fact the first into the grocery market as a
whole. Earlier CC inquiries had been confined to supermarkets alone (2000)
or to mergers between supermarkets186. The CC investigation into
supermarkets had been followed by a Supermarket Code of Practice,
established in 2001, which anyway did not lead to the expected results – as a
matter of fact, the 2006-2008 CC inquiry found many of the same abuses it
had already found in 2000. The CC proposed a new code of conduct and the
creation of an ombudsman to address the continuing problem. However, the
UK supermarket chains were strongly opposing, in 2008 and beginning 2009,
the creation of an ombudsman for suppliers. In August 2009 the Competition
Commission formally recommended that ministers appoint an ombudsman
through legislation, having failed to secure a voluntary agreement from
supermarkets to create one. In 2010, the proposal for a Groceries Market
Ombudsman Bill was tabled, accompanied by a report and an impact
assessment, but has not reached final approval at the time of writing this
report187.

2.3.2 Tying and other potentially unfair practices in retail financial

services

A recent study for the European Commission, DG Internal Market, found that 12
Member States have adopted policy initiatives addressing tying and other
potentially unfair commercial practices188. These are Belgium, Bulgaria, Cyprus,
Denmark, Finland, France, Hungary, Ireland, Poland, Portugal, Romania and
Slovakia. Of these, 8 countries have decided to take specific action to restrict
tying and cross-selling behaviour in the financial services sector. This latter
group of countries include Member States that have enacted very far-reaching
prohibitions (Belgium, France, Ireland, Portugal, Slovakia); countries that rely on
“soft-law” schemes coupled with narrower legal prohibitions (Hungary, Poland);
and one country where prohibitions are narrower in scope (Denmark).
More in detail:

185 The supply of groceries in the UK market investigation, Competition Commission, 30 April
2008. The report is available online at http://www.competition-
commission.org.uk/rep_pub/reports/2008/fulltext/538.pdf.
186 The 2000 CC’s inquiry into supermarkets concluded that supermarkets with at least an 8%

share of grocery purchases for resale from their stores, have sufficient buyer power to undertake
abusive practices that adversely affect the competitiveness of some of their suppliers and distort
competition in the supplier market—and in some cases in the retail market—for the supply of
groceries.
187 http://www.parliament.uk/briefingpapers/commons/lib/research/rp2010/RP10-021.pdf
188 Renda et al. (2009).

PAGE 101 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 In Belgium, Article 31 of the Consumer Credit Law of 12 June 1991 – which

aims at protecting consumers in their consumer credit transactions – limits
the possibility to make the conclusion of a consumer credit contract
conditional upon the conclusion of another contract189. Before the 2003
reform of the Consumer Credit Law, there was general consensus that Article
31 did not create a special regime for combined offers in the domain of
consumer credit. On the contrary, Article 31 was specifically addressed at so-
called “annexed contracts”, i.e. contracts that are imposed by the lender on
the borrower after the conclusion of a first contract, regardless of whether
they matched the definition of Article 55 of the 1991 Belgian Law on trade
practices and consumer information. After the 2003 reform, the rule
prohibits the imposition of a counterpart to the consumer, i.e. Article 31
tackles specifically preferential and exclusive agreements. The application of
Article 31 also entails a number of exceptions, which are consistent with Art
55 of the Law on trade practices and consumer information, For example, it is
allowed: (i) to sell a good together with the consumer credit contract needed
to finance the purchase; (ii) to offer credit combined with current account;
(iii) to sell insurance on the property which is bought on a credit; and (iv) to
impose an obligation to have insurance on the outstanding debt (insurance
which would cover the due amount in case of death of a borrower). As a
result, while Belgian law allows selling a good and allowing consumers to pay
in three instalments without additional charges (“3 + sans frais”), it prohibits
doing so on the condition that consumers also activate a credit card linked to
a credit account: Article 54 of the Law on trade practices and consumer
information applies to this latter offer. The same applies to the possibility of
paying in several instalments when buying at a department store, on the
condition that purchasers also activate the store’s payment card. Another
important provision as regards consumer contracts in the retail financial
services sector is Belgian Law of 4 August 1992 on mortgage credit (as
amended)190. This law – in particular Article 6.2 – provides for legal rules in
the field of payment protection insurance191. Under this law and relevant case-
law, the mortgage lender can make the mortgage rates conditional upon
having a particular type of insurance192: however this must occur through an
attached contract (contrat annexé), which can only take the form of: (i) an
insurance on the outstanding debt to cover the risk of death, aimed at
securing the repayment of credit; (ii) an insurance covering the risk of
deterioration of the property offered as collateral; or (iii) an insurance bond.
Such an offering of combined services was confirmed by the Belgian Court de

189 The text of the law could be found at (in French):

http://economie.fgov.be/protection_consumer/Credit/Credit_CD/LCC/cadre28_33bis.htm. The
law was amended as of 31 December 2010 to reflect the transposition fo the consumer credit
Directive.
190 Wet van 4 Augustus 1992 op het hypothecair krediet.
191 Payment protection insurance is an insurance product designed to cover a debt that is

currently outstanding. This debt is typically in the form of a loan or an overdraft, and is most
widely sold by banks and other credit providers as an add-on to the loan or overdraft product.
192 In its 2007 Economic Survey of Belgium, the OECD recommended that the authorities

“reconsider the regulation allowing the tying of a mortgage interest rate reduction and the
purchase of certain insurance products”. See the summary report at
http://www.oecd.org/dataoecd/51/14/38209638.pdf.

PAGE 102 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Cassation in a judgment of 30 March 2001193. The Court also ruled that only
the (forced) imposition of a particular insurer is a violation of consumers’
interests – a finding that suggests that exclusive (not preferential) agreements
in this domain are banned.
Outside the case of “attached contracts” specified in Articles 5 and 6 of the
Law on mortgage credit, Article 19 of the same law specifies that the granting
of a mortgage loan cannot be directly or indirectly conditioned to any other
contractual obligation or insurance contract or the opening of a savings
account.
These provisions are enforced by civil courts and enforcement is reportedly
strong.

 In France, in 2005 the Monetary and Financial Code was amended to

incorporate Article L312-1-2.194 This provision – chiefly aimed at preserving
the transparency of offers on the consumer market, as well as freedom of
choice for consumers – prohibits selling or offering “for sale products or
services in a bundle except when bundled products or services included in
that bundle can be purchased individually or are inseparable.” This
provision is located in the section on current accounts and deposits, in this
way prohibiting bundling of these products with other financial services. This
prohibition is subject to a de minimis rule which is set by the Minister of
Economy. Unfortunately, in the absence of administrative practice or case-
law it is difficult to see if this provision could be given a broader reading
possibly also covering other practices195.
Until 2010, the banking sector benefited from particular exceptions to the
above-mentioned prohibition. According to L. 312-9 of the Consumer Code,
the lender could offer the borrower or ask the latter to take part in a group
insurance policy that was taken out with a view to guarantee uninterrupted
payments (otherwise – payment protection insurance, PPI). However, this
provision was abolished by the Loi Lagarde of July 2010, which makes it
illegal to bundle mortgage loans together with PPI from 1 September 2010196.
The measure also tackles information deficiencies such as the cost of
insurance policies, information of coverage in case of death, illness or
unemployment. The main reason for such legislation is the fact that currently
getting a mortgage is very dependent on subscribing to an insurance policy
with the same bank. The provision is enforced by the ordinary district Courts
(tribunal d’instance). The enforcement is expected to be strong, but it is too
early to verify this empirically through an analysis of a consolidated stream of
cases.

193 The Court attached two conditions to it: (1) such joint offer should not be made compulsory;
(2) a particular insurer shall not be favoured especially if it is selected by the lender.
194 For full-length text of the document, please refer to (in French):
http://195.83.177.9/code/liste.phtml?lang=uk&c=25.
195 Chief responsibility for the enforcement of this rule is with the Bank of France.
196 See http://www.liberation.fr/economie/0101268544-credit-immobilier-les-emprunteurs-
seront-libres-de-choisir-leur-assurance.

PAGE 103 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 In Hungary, in 2006 the Financial Supervision Authority (HFSA) adopted a

Recommendation on the principles of retail crediting provision of preliminary
advice to clients and consumer protection (No. 9 of 2006 (XI. 7))197.
Paragraph 13 of this non-binding Recommendation states that “Financial
institutions shall not attach any other product or service to the contracted
loan product as being mandatory, which by its nature does not belong to the
given loan product, furthermore, the use of which cannot be justified with fair
and rational market reasons (such as bank card with an annual fee). The
financial institution should leave the use of such linked products/services to
the clients’ discretion and should not force clients to undertake unnecessary
obligations. Exceptions are when financial institutions make the use of some
linked product mandatory proportionately to and as a prerequisite of
providing certain benefits or discounts”. The document recognizes that in
certain instances tying and bundling can be objectively justified, as in the case
of clearly mentioned discounts or other benefits. Accordingly, this
recommendation challenges tying and pure bundling unless justified with fair
and rational market reasons, and allows mixed bundling with a financial
advantage to the client. Although this provision in not legally binding, the
HFSA does supervise and monitor whether the financial organizations follow
its principles, with the aim of protecting consumer and promoting the
transparency of contractual offers. However, the HFSA cannot impose fines
on service providers if the Authority gets information about such an
activity.198
 In Ireland, the 2006 Consumer Protection Code for Financial Services
generally prohibits the tied sale (and pure bundling) of financial services, but
allows mixed bundling, provided that the price of each product is clearly
indicated. Some forms of tying are explicitly allowed and regulated: these
include the joint offer of mortgage loans and current accounts; and mortgage
loans with payment protection insurance, which is also explicitly required by
the 1995 Consumer Credit Act. The latter Act also prohibits preferential and
exclusive agreements when applied by mortgage lenders. The Code is rather
strictly enforced by the Financial Regulator, mostly through inspections and
administrative sanctions procedures, which led to a total of 32 sanctions, 9
disqualifications from 1 to 5 years and a total of €9.6 million of fines in the
period between July 2006 and mid-2011199.
 In Poland, the Office of Competition and Consumer Protection (OCCP)
probed the banking sector 2008 especially as regards standard contracts
concerning mortgage loans200. As a result, based on the OCCP
recommendations, the banks concerned voluntarily decided to abandon
clauses tying obtaining a mortgage loan and possessing a bank account in a
given bank. As reported to us, a verification of the banks’ compliance is in
progress. The Polish Financial Supervision Authority was also active in this
field. In one of its cases the Authority had to deal with practices of Nordea

197 For more information please refer to (in English):

http://www.pszaf.hu/en/left_menu/regulation/pszafen_recommendations/pszafen_recommen
dations_20061204_1.html.
198 None of the HFSA cases are related to tying and bundling practices.
199 See the materials available on
http://www.centralbank.ie/regulation/processes/administrative-sanctions/Pages/default.aspx
200 See http://www.uokik.gov.pl/en/press_office/press_releases/art116.html.

PAGE 104 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Bank which bundled time deposits together with current accounts (paid
account). The authority in this case discovered that tying and bundling are
widely practiced in the retail banking sector. In particular, the Authority
voiced its concerns about:201
– tying in the banking sector (such as bank deposit with current account,
bank deposit with debit card or current account together with pension
account (so-called pension scheme).
– tying of banking and investment services (e.g., bank deposit plus
investment funds, current account plus securities account, current account
plus investment funds).
– tying of banking and insurance products (e.g., credit card with compulsory
card insurance, mortgage loan plus house/flat insurance as an obligation,
current account plus insurance available by bank internet platform).
However, to date no legal rule has been enacted to specifically address these
practices.
More in general, the OCCP questioned clauses tying mortgage loan and
possessing a bank account in a given bank only if possessing a bank account
resulted in any costs. If a bank account was cost-free, the Office did not
question such clauses.
 In Portugal, sectoral legislation prohibits making conditional the acquisition
of mortgage upon acquisition of other types of financial services. Article 9 of
the Decree-Law no. 51/2007, of 7 March (as amended by Decree-Law no.
88/2008, of 28 May) deals with tying and bundling situations in regard of
mortgage credit.202 By the virtue of Article 1, this Decree-Law is applicable to
mortgages for acquiring, constructing and completing the works in
permanent housing, or for rental and acquisition of land for construction of
housing. The combination of these two provisions ensures very wide coverage
and virtually would cover the most common mortgage types. Article 3(2)
potentially covers also mixed bundling with financial advantage (multi-
product rebates).203 Another provision is found in Decree-Law no. 144/2006,
of 31 July (as amended by Decree-Law no. 359/2007, of 2 November) which
deals with insurance issues204. The provisions are enforced by civil courts, but
there are reportedly very few cases.

201 See http://www.uokik.gov.pl/en/press_office/press_releases/art72.html.

202 The provision is worded as follows: “Credit institutions are forbidden to make the conclusion
of contracts under Article 1st of this Decree-Law dependent of the acquisition of other financial
services or products.” The translation in unofficial, for precise wording please refer to the
Portuguese version of this provision: “Às instituições de crédito está vedado fazer depender a
celebração dos contratos referidos no artigo 1.º deste decreto-lei da aquisição de outros
produtos ou serviços financeiros”.
203 “Credit institutions are forbidden to make the renegotiation of the credit dependent of the

acquisition of other financial services or products.” The provision in Portuguese is as follows: “Às
instituições de crédito está vedado fazer depender a renegociação do crédito da aquisição de
outros produtos ou serviços financeiros”.
204 This national measure sets out the conditions governing the taking up and pursuit of the

insurance mediation activities. Article 31(g) prohibits for insurance intermediaries to tie
consumers to particular insurance providers. This provision would successfully tackle also
exclusive (not preferential) dealing. In particular, insurance intermediaries are expected: “not to
impose the obligation to sign an insurance contract with a specific insurance undertaking as a

PAGE 105 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 In Slovakia, Article 3.8 of the Code of Ethics of the Slovak Banking

Association was added to the Code of Ethics with effect from 1 January 2008
after an extensive investigation of the competition authority. The article
prescribes that banks “will not tie the provision of a financial service to the
provision of another financial service”. This rule has one exemption: offering
of the financial services in a package would be allowed if it is convenient for
the customer. Accordingly, it could be possibly argued that consumers should
be duly informed that they are being offered a combination of services. This
provision potentially covers tying and pure bundling, but not mixed bundling.
Besides other things, the Code of Ethics introduces the institution of the Bank
Ombudsman, who deals with potential differences arising from failure to
observe the respective provisions of the document or in the event that the
Customer has exhausted all other available remedies.

2.3.3 Network infrastructure

The unilateral market conduct of enterprises active in network infrastructure

sectors is potentially subject to sector-specific laws, general national competition
law, and the relevant articles of the TFEU. Sector-specific rules are often stricter
than national rules on unilateral conduct that are more broadly applicable,
although such rules have often different objectives and should be cautiously
compared to Art. 102 TFEU.
National laws regulating specific sectors are found in Austria,205 Germany,206
Hungary,207 and Slovakia.208 In addition, Germany has a particular legal
provision focused on the energy sector in the competition law, §29 of the Act
against Restraints of Competition, which is a stricter rule on unilateral conduct.
In particular, it reverses the burden of demonstration and proof of
discrimination, and prohibits demanding fees unreasonably exceeding costs, for
dominant undertakings that supply either electricity or pipeline gas.209
Sector-specific stricter rules on unilateral conduct on network infrastructure
sectors can be seen within a broader context of regulatory reform in those

condition in order for the customer to gain access to another good or service supplied.” The
provision in Portuguese is as follows: “Não impor a obrigatoriedade de celebração de um
contrato de seguro com uma determinada empresa de seguros como condição de acesso do
cliente a outro bem ou serviço fornecido”.
205 In Austria, the Telecommunications Act of 19 August 2003 and Amendment of the Federal Act

on Work Inspection in the Field of Transport and the KommAustria Act (NR: GP XXII RV 128 AB
184 S. 29. BR: 6800 AB 6804 S. 700) could be seen as stricter in their practical application than
Article 102 TFEU.
206 In Germany, in addition to the GWB provision cited above, sectoral laws such as the Energy

Industry Act, the Telecommunications Act and the General Railway Act impose stricter rules on
the conduct of dominant firms in the eponymous sectors.
207 Tied sales are prohibited by specified rules for the gas and electricity sectors, and for the

telecommunication sector.
208 Slovak Law No. 610/2003 Coll. on Electronic Communications as amended (adopted

3.12.2003) prohibits tying in certain situations.

209 The reversal applies only for proceedings before the cartel authorities. According to the

Bundeskartellamt, it had 33 price abuse cases by gas suppliers in 2007/2008 but none in 2009.
By contrast, it did investigate the electric heating market. See the National Report for Germany,
“Monitoring Report,” Bundesnetzagentur, pp. 77-78 at www.energy-regulators.eu.

PAGE 106 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

sectors. Technology implied — and still largely implies — a natural monopoly210

over any given geographic area within these sectors. After competition in parts of
these sectors became feasible, and competition elsewhere was seen as resulting in
lower costs, innovation and better matching of consumer demands, there was a
general trend towards inter alia211 opening up parts of network infrastructure
sectors to competition. But the conduct of the enterprises that had inherited the
natural monopoly parts had to be controlled in order for entry and competition
to occur in newly-opened markets. It was recognized that poorly-controlled
monopolies could scupper the development of any competition and associated
benefits. The policy responses to control these monopolies ranged across a
spectrum with different scopes of application of sector-specific regulation and the
general competition law. And as experience is gained, sometimes the scope for
sector-specific regulation is expanded.
An example of stricter unilateral conduct rules in a sector-specific regulatory law
is provided by Hungary. While the Competition Authority tackles abuse of
dominance ex post under the competition law, the Energy Office (HEO) tackles
them ex ante. The Energy Act empowers the HEO to impose “special additional
obligations on e.g. selling electricity on public capacity auctions, cost-based
pricing, preparation of a bidding sample, etc. on all licensees…who…have
significant market power.”212
With respect to divergences in sector-specific economic regulation—including of
competition in the relevant sectors—the regulators themselves have a mechanism
to facilitate the development of the internal markets for gas and electricity, and
telecommunications, at least. The Council of European Energy Regulators213
inter alia cooperates with the European Commission and national competition
authorities to ensure consistent application of competition law to the energy
industry. They have produced a number of best practice guidelines, including on
access. The Body of European Regulators for Electronic Communications plays
an analogous role.

2.4 Concluding remarks: a persisting legal fragmentation

The previous sections have shown that, as far as rules on unilateral conduct are
concerned, the EU27 still exhibits a remarkable fragmentation, which can

210 A natural monopoly is roughly defined as a market which is supplied by one enterprise at
lower cost than by two or more enterprises. Examples include electricity transmission, natural gas
transmission, and rail infrastructure.
211 A number of other changes accompanied the opening to competition, depending on the

country. The network infrastructure sectors were, in many current Member States, government
departments that were transformed into operating companies and separate regulators.
Subsequently, many were privatized to varying degrees. Thus, the incentives on the enterprise
controlling the natural monopoly also changed as a part of the regulatory reform process.
212 Hungarian Energy Office (2010), “Annual Report to the European Commission,” p. 31.

www.energy-regulators.eu. The obligation to auction capacity or to engage in cost-based pricing

are somewhat analogous to remedies in abuse of dominance cases where denial of access to
essential facilities has been proven, but imposed ex ante is stricter than competition law.
213 The Council of European Energy Regulators is a preparatory body to the consultative body

established by the European Commission, the European Regulators' Group for Electricity and
Gas.

PAGE 107 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

potentially hamper the Single Market, forcing businesses wishing to engage in

cross border trade to adapt their strategy and conduct based on the rules in force
in different national territories.
In five countries – Austria, Bulgaria, Germany and Lithuania – the definition of
dominance contained in the law can be considered as significantly stricter than
the one adopted in Article 102 TFEU. Also in Austria, Estonia, Poland and
Romania there is either a reliance on stricter market share thresholds or the use
of per se rules where EU competition law would adopt an effects-based approach.
In Hungary, a specific piece of legislation introduces a concept that is akin to
dominance but differs a lot from the latter in relation to the criteria used to
define it. Looking at individual conducts, abuse of economic dependence has
been regulated in as many as twelve countries, with different modalities and
standards used, and different degrees of involvement of competition authorities.
Sales below cost are regulated in seven national jurisdictions, and have been
subject to specific sectoral rules especially in the grocery sector in the past years.
Moreover, tying has been subject to a per se prohibition with different scope and
criteria in ten Member States, which are reconsidering their laws on the basis of
the UCPD – however, eight of these countries will be able to keep these
provisions in the field of retail financial services, where the UCPD leaves room
for exceptions to the maximum harmonization rule. Finally, unfair competition
laws in at least three jurisdictions appear to be stricter than Article 102 TFEU.
All these legal provisions exhibit varying degrees of enforcement and diverge
from Article 102 TFUE in different ways. Their impact on industry players – as
will be observed in Section 4 below – is difficult to assess, and seems unevenly
distributed across industry sectors and stages of the value chain. To be sure, the
mere existence of this fragmentation has been signalled by several stakeholders
as creating obstacles and costs for those businesses wishing to engage in cross-
border operations. Those obstacles can also take the form of “chilling
competition” effects, especially where (i) the application of the national rule is
unclear; or (ii) businesses prefer to standardize their compliance programmes by
taking as reference the strictest legal system among the EU27.
Figure 7 below graphically summarises our findings on the degree of
fragmentation generated by diverging legal rules on unilateral conduct in the
EU27. In order to build this graph, we have only considered the number of
provisions that are potentially stricter than Article 102 TFEU, without accounting
for any measure of the actual extent to which each rule diverges from the EU
rule. In addition, we included in the graph also sectoral rules that provide for
criteria stricter than Art. 102 TFEU. However, it should be borne in mind that
such rules may fall within the scope of pararaph 3 of Art. 3 of Reg. CE 1/2003, i.e.
they have divergent objectives with respect to Art. 102 TFEU, and therefore
escape the rule of convergence.

Figure 7 – The fragmentation of legal rules on unilateral conduct in

the EU27

PAGE 108 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU
Europe – Countries

Based on figure 7 above, a number of countries appear to contribute significantly

to the fragmentation of legal rules on unilateral conduct. Overall, as many as 18
Member States appear as having enacted rules that appear stricter than Article
102 TFEU.
 Hungary and Germany are the countries that have the highest number of
rules that are stricter than Article 102 TFEU (7 for Hungary, 6 for Germany).
In Hungary, the divergence is the result of rules which, except for conditional
sales, do not belong to competition law but are found in other bodies of law:
in the Act on Trade that disciplines a particular form of market power on the
buyer side, the abuse of economic dependence and tying, as well as in sectoral
rules applied in the grocery sector, financial services and network industries.
In Germany the divergence is due to rules on dominance, economic
dependence, sales below cost, and tying, plus rules applicable in the grocery
sector and in network industries.
 Two countries, Romania and Slovakia, exhibits five national rules that
diverge from Art. 102 TFEU, though to a slightly lesser extent than the
countries listed above. In Romania, the most diverging rules are on
dominance, economic dependence, sales below cost, tying, and unfair
competition. Slovakia exhibits slightly diverging rules on the abuse of
economic dependence and unfair competition, as well as sectoral rules in the
grocery, financial and network industries.
 In Austria, France and Portugal there are four instances of rules that
diverge from Art. 102 TFEU. In Austria the most diverging rules are on the
definition of dominance, on economic dependence and on sales below cost. In
addition, there are sectoral rules in the network industries which follow
different criteria, although these laws transpose EU Directives into national
law and therefore should be considered in line with the relevant European
legislation. In France, divergence is due to rules on abuse of economic

PAGE 109 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

dependence, sales below cost, tying and sectoral rules on financial services.
Portuguese law diverges in the fields of economic dependence, sales below
cost, and tying.
 Bulgaria, Poland all have three types of national rules that can be
considered stricter than Art. 102 TFEU. Bulgaria diverges mostly due to
slightly stricter criteria for assessing dominance, sales below cost and tying.
Poland features slightly diverging criteria for the finding of dominance, but
also significantly diverging legislation on tying and in the field of financial
services.
 Finally, other countries feature only one or two types of diverging rules,
including Belgium, Cyprus, the Czech Republic, Estonia, Greece,
Ireland, Latvia Lithuania and Spain. The Czech Republic features
relevant national rules on economic dependence disciplined in a very
controversial legislation in the grocery sector. Similarly, in Latvia stricter
rules are provided for by competition law in relation to economic dependence
specifically at the retail level.
Table 1 below graphically summarises the findings of this section. More
information on each country and each of the rules in force is included in Annexes
VI and VII attached to this Final Report.

Table 1 – Areas where national rules are stricter than Article 102 TFEU

PAGE 110 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

3. UNILATERAL CONDUCT, BUSINESS STRATEGY AND

CONSUMER WELFARE: LITERATURE AND EVIDENCE

In this section, we explore the theoretical and methodological framework that

underlies our impact analysis of the legal rules identified in Section 2 above.
Accordingly, we discuss existing studies that have tried to assess the welfare
impact of diverging standards as well as specific rules on unilateral conduct.
Section 3.1 below discusses diverging standards, whereas Section 3.2 focuses on
rules on economic dependence/superior bargaining position. Section 3.3
discusses existing studies on sales below cost. Section 3.4 describes the existing
literature that assesses the macroeconomic impact of competition laws. This
literature review will form the basis for our subsequent analysis, in Section 4 of
this Report, of the impact of the legal rules identified in Section 2 above.

3.1 Diverging standards for defining dominance and/or

abuse
Differences in the standards applied to determine whether a firm is dominant
and whether, dominant or not, its conduct is unlawful can chill business conduct,
raise businesses’ cost of assessment and compliance, and reduce the effectiveness
of various business strategies. More generally, business conduct can be chilled as
a result of increased uncertainty regarding legal standards. If the chilled conduct
is in fact pro-competitive, the uncertainty is harmful. E.g., prosecution of
predatory pricing can inadvertently discourage low pricing as part of a lawful,
competitive, welfare-enhancing strategy. Differences in legal standards can
increase uncertainty since each jurisdiction necessarily has fewer opportunities
to clarify its own legal standards than if decisions were made under the same
standards and could more rapidly cumulate to define those standards.
The OECD policy roundtable on “Guidance to business on monopolisation and
abuse of dominance”214 provided some examples where uncertainty as to the
legal standards for dominance and for abusive conduct led to the chilling of
business conduct. Some representatives of businesses based in the European
Union stated that they must take into account European rules on unilateral
conduct in their day-to-day business even though they do not perceive
themselves as enjoying significant market power. One participant provided a
practical example, where the business he represented, GE, considered suggesting
to a customer, who had organized a “winner take all” procurement, to invite a
second participant so as to avoid his firm exceeding a certain market share
threshold. Such a move would have deprived the customer of cost-savings from
the efficiencies of a single-source policy. (pp.85-86) A different participant,
representing Michelin, said that it employed less effective supply chain
management techniques (not asking dealers for an annual supply commitment),
and did not intervene to prevent rivals from piggy-backing on certain research, so
as to avoid a risk of infringement of the dominance abuse rules.

214OECD (2007), “Guidance to Business on Monopolization/Abuse of Dominance,”

DAF/COMP(2007)43, available at www.oecd.org.

PAGE 111 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Due to the uncertainty as to the legal standards for a finding of dominance, the
management decided that the firm should behave as if it were subject to the
dominance rules when it reached a 30% share in the given market. (p. 86) A third
participant, Prof. Bloom, plead for agencies to apply the same legal standard to
the same economic effect/harm in the market, thus avoiding “abuse shopping” to
seek the one with lowest standards. She cited margin squeeze, predation, and
refusal to supply as one such area. (p. 95) The representative of GE noted that
there are other circumstances where different standards are applied to similar
situations, and pointed out that the uncertainty may lead to consumer welfare
loss. For example, a company may avoid entering a market so as not to be subject
to the tougher standard for terminating an existing supply relationship, as
compared with the standard for refusing to enter a new supply relationship (pp.
95-96).
The welfare cost of chilling business conduct enters into decision-theoretic
debates on the design of competition rules. Given that over-enforcement and
under-enforcement of competition laws can result in harmful business conduct,
including chilling of welfare-improving conduct, how should competition laws be
interpreted and enforced? Errors of over- and under-enforcement of competition
laws have been commented on at least since 1984. Easterbrook215 felt that under-
enforcement errors (false negatives) would be largely corrected by market
participants, that is, cheating would break up cartels and entry would erode
monopolies. But over-enforcement errors (false positives) would not be corrected
by market participants. This asymmetry of the costs of errors should, therefore,
be reflected in the competition rules. Literature applying decision theory, a
framework that explicitly considers the likelihood and cost of errors, to
competition rules has since blossomed. It has become apparent that not only
should the error costs of the case at hand be included, but also those generated by
changes in other firms’ behaviour (the “chilling” effect), and the cost of operating
the decision-making apparatus.
An empirical question was whether over-enforcement, the type of error with the
greatest cost according to Easterbrook, was frequent. Staff members at the US
Federal Trade Commission investigated. They surveyed judicial resolutions of
344 claims under Section 2 of the Sherman Act (prohibiting monopolization or
attempts to monopolize) between 2000 and 2007, and found that, “[O]f these,
335 [97,4%] were decided for defendants and nine [2,6%] were decided for
plaintiffs.” (Adkinson et al 2008)216 The overwhelming majority of these cases
were initiated by private plaintiffs; only about a score were initiated by federal or
state law enforcement agencies. These figures suggest that over-enforcement did
not often occur in the jurisdiction and period surveyed. They did not evaluate the
cost of operating the decision-making apparatus, including costs of defending
one’s conduct. But these figures point to an additional matter of interest to our
enquiry: The number of judicial resolutions would seem to be sufficient to ensure
that legal standards would be well-defined, but there continued to be demand for
such resolutions. Either large numbers do not guarantee legal clarity, or other
incentives drive demand for resolutions.

Easterbrook F.H. (1984), “The Limits of Antitrust,” Texas Law Review, 63(1), pp. 9-10.
215
216Adkinson, Jr. W.F, K.L. Grimm, and C. N. Bryan, “Enforcement of Section 2 of the Sherman
Act: Theory and Practice,” FTC Working Paper, 3 November 2008, footnote 141 cites several
examples.

PAGE 112 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Within a decision theoretic framework, different results may occur despite the
same legal standards. This is a result of competition authorities starting with
different prior beliefs on the costs, and the likelihoods of incurring those costs, of
intervention and non-intervention. While the information gathered during the
course of an investigation and the analytical framework and methodology may be
identical, the effect of having held different prior beliefs is to have different
perceptions about uncertainty after the information-gathering and analyses. And
this can result in different decisions as to whether and how to intervene. (Heyer
2004, p. 28)217
Evans and Padilla (2005)218 formulated explicit prior beliefs and tried to devise
decisions rules to minimize the costs of over- and under-enforcement of
unilateral conduct rules. Buccirossi, Spagnolo and Vitale (2006)219 tried to
describe the sources of error costs of over- and under-enforcement, both in the
case at hand and indirectly as a result of changes in other firms’ behaviour. They
were able to generate few solid results, but did show how errors of over- and
under-enforcement attenuate deterrence, that is, they reduce firms’ compliance
with the law. They found that claims that over-enforcement has greater welfare
effects than under-enforcement were insufficiently grounded, but found that
over-enforcement would have greater costs in a dynamic than a rent-seeking
economy. This latter concern was reflected in an expectation that the cost of
false-positives would be greater in innovation markets, where over-enforcement
may decrease incentives to develop new products and technologies, activities
which ultimately greatly increase consumer welfare. (Rill 2006)220
The second and third effects of differences in legal standards, raised business
cost of assessment and compliance, and reduced effectiveness of various business
strategies, are little studied. In principle, different standards that reduce the
geographic scope of business strategies can reduce welfare in at least three ways.
First, strategies that cannot be limited in geographic space may not be engaged in
at all. For example, if provisions in licenses of intellectual property rights cannot
be enforced everywhere, then it may be more profitable not to license certain
IPRs. Second, strategies may enjoy scale economies (the costs of market
research, design and decision-making likely increase less than linearly with
sales), and the restricted scope may not permit a full application of a given
strategy. For example, within a region reached by the same advertising media,
sales below cost may be prohibited in some parts and permitted in others. Third,
strategies may have negative spill-overs at their margins:
A shopper, upon discovering that pricing or selection are not available at one
store even though they were available, or advertised as available, at the same
brand of outlet in the adjacent jurisdiction may resist a return visit. Workshops

217 Heyer, K. (2004), “A World of Uncertainty: Economics and the Globalization of Antitrust,
EAG Discussion Paper 04-11. A modified version published in Antitrust Law Journal, 2005,
72(2), pp. 375-422.
218 Evans, D.S. and A.J. Padilla (2005), “Designing Antitrust Rules for Assessing Unilateral

Practices: A Neo-Chicago Approach,” University of Chicago Law Review, 72, Winter.

219 Buccirossi, P., G. Spagnolo and C. Vitale (2006), “The cost of inappropriate
interventions/non-interventions under Article 82,” OFT Economic Discussion Paper OFT864,
September, available at http://www.learlab.com/pdf/oft864_1158144540.pdf.
220 Rill, J.F. (2006), Remarks before the FTC and DOJ Hearings on Section 2 of the Sherman Act:

Single-Firm Conduct as Related to Competition, Session on International Issues, 12 September.

available at http://www.ftc.gov/os/sectiontwohearings/docs/Rill_statement.pdf

PAGE 113 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

involving multinational businesses should help to fill this gap in the empirical
literature.
Surveys point to continued international diversity on single-firm conduct rules.
Differences among legal standards were surveyed by the International
Competition Network (“ICN”). It surveyed its members, national competition
authorities worldwide, as to objectives of unilateral conduct laws and how
dominance or substantial market power (“SMP”) might be assessed.221
Regarding objectives, there were differences in the protection of SMEs. The
report identified two primary reasons for which unilateral conduct may be
intervened against when the undertaking is not dominant. The most commonly
cited reason was to prevent the creation of a dominant position through
anticompetitive practices. The second reason was to prohibit an abuse of
economic dependency or of economic disparity in order to create a “level playing
field” for small and medium enterprises. (p. 62) The German comment explained
that, “such protection [of SMEs] serves to maintain (or create) a diversity of
enterprises on the supply side and thus prevents anti-competitive concentration
processes at an early stage.” (p. 17) The ICN concluded that, “[R]ules designed to
protect SMEs, to protect individual freedom to participate in markets, or to
promote fairness, for example, could sometimes lead to different outcomes in
unilateral conduct cases than rules that focus only on consumer welfare and
efficiency goals.” (ICN 2007, p. 37) Regarding the assessment of dominance, all
jurisdictions responding to the questionnaire used a comprehensive set of
criteria, although market share and barriers to entry or expansion and durability
of market power were generally considered as the most important. More than
half (15) used market share to establish a rebuttable presumption of dominance
and/or a safe harbour. Twenty-eight used a behavioural definition of dominance,
whereas five—of which only one was a Member State, Latvia—used a market
share threshold (see p. 39 above). In sum, there remain differing objectives and
differing legal standards under which to assess conduct and dominance or
similar concepts.
The OECD policy roundtable on “Competition on the Merits”222 described the
various economics-based tests that are applied, in various contexts and
jurisdictions, to unilateral conduct. Most jurisdictions apply a relaxed form of the
“profit sacrifice test” (Is the profit sacrifice irrational if the conduct does not tend
to eliminate or reduce competition?) to sales below cost. (see p. 68 et ss. above)
But the analyst must choose an appropriate benchmark against which to measure
the profit sacrifice. Often a cost-based benchmark is used (chosen by reference to
the “equally efficient competitor test”), but the strict version of the test would
require calculating profits of a dominant firm engaging in the same conduct but
where it did not anti-competitively exclude or harm rivals.
This calculation can be subjective and “prone to error.” (p. 27). The “consumer
welfare balancing test” can be difficult to apply to dynamic strategies, such as

221 International Competition Network (“ICN”). 2007. “Report on the Objectives of Unilateral
Conduct Laws, Assessment of Dominance/Substantial Market Power, and State-Created
Monopolies.” Unilateral Conduct Working Group.
http://www.internationalcompetitionnetwork.org/uploads/library/doc353.pdf, visited on
15.07.2010.
222 OECD (2005b), “Competition on the Merits,” DAF/COMP(2005)27, available at
www.oecd.org.

PAGE 114 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

predatory pricing, because the timing of costs and benefits are different, and—
although not mentioned in the paper—affect different consumers. It concludes
that, “There is an inherent tension between fostering legal certainty, ease of
administration, and accuracy… Either [a form-based or effects-based] approach,
driven to excess, produces unattractive results.” (p. 10). And there is no single
economic test that is appropriate in all circumstances (ibid.)
The OECD policy roundtable on “Techniques and Evidentiary Issues in Proving
Dominance/Monopoly Power,”223 describes the methods competition authorities
and courts use to distinguish firms subject to single firm conduct provisions from
those which are not. Indirect evidence such as, “market shares, barriers to entry
and expansion, buyer power and the nature of competition in the market” are
commonly relied upon, but “Typically there is no single factor that will provide
conclusive answers.” High market shares do not conclusively prove that a firm
has substantial market power, nor do low market shares conclusively show the
opposite. (pp. 7-9) Regarding direct methods, profitability measures are not
frequently used, but can be an indicator in specific circumstances; methods to
directly measure market power are data intensive, and typically subject to
different interpretations and thus do not constitute conclusive proof(p. 9).
In sum, different legal standards can chill business conduct by increasing
uncertainty as to which conduct is lawful and unlawful, and this chilling can
harm the attainment of the purposes of competition law. In addition, legal
standards that have higher error rates than others have the effect of attenuating
the deterrence effect of the law. But there appear to be several gaps in the
empirical literature as to how significantly business conduct is chilled.
Differences in legal standards can arise as a result of different objectives of the
law, or different costs of enforcement errors and the enforcement mechanism
itself, or different prior beliefs about the size and frequency of these costs.
Further costs of differing standards, i.e., increased cost of assessment and
compliance incurred by businesses, and reduced effectiveness of various business
strategies, have been little studied.

3.2 Assessing the impact of economic dependence/superior

bargaining position
Economic dependence/superior bargaining position rules aim to restrict conduct
by firms that can “hold up” other firms, that is, that can exploit the advantages
created by, for example, relationship-specific investments.224 It is not necessary
for a firm to be in a dominant position. The concern is that firms subject to such
abuse may eventually be forced to exit the market, even though they are

223 OECD (2006), “Techniques and Evidentiary Issues in Proving Dominance/Monopoly Power,”
DAF/COMP(21006)35, available on www.oecd.org.
224 “These [“economic dependency”] laws are specifically designed to prevent post-contractual

opportunistic behaviour (i.e. associated with moral hazard situations) whereby suppliers might
become economically dependent on retail customers (e.g. through making transaction specific
investments in their trading relationship) which then opens them up to the possibility of ‘hold up’
problems whereby retailers may seek to renegotiate contracts or the basis of the relationship once
the supplier is in a relatively trapped position (i.e. retailers seeking retrospective changes).”
Dobson, P.W. and R. Chakraborty (2007), “Retailer-Processor-Distributor Interfaces,” Presented
at the “Policy for the Food Marketing Chain” Conference, Institute of Food and Resource
Economics, Copenhagen, 12 June 2007.

PAGE 115 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

sufficiently efficient to remain market participants. Historically, concern has

been directed towards the relationship between suppliers and large grocery
retailers, although in principle any intermediate goods or services market could
be of concern.
The First European Competition Forum in April 1995 discussed inter alia the
economic dependence of suppliers on large distribution groups. Diverse views
were expressed. One speaker attributed the difference in views to a difference in
market structure, asserting that Member States with concentrated groceries
markets (which he cited as France, Germany, Portugal)225 viewed economic
dependence/superior bargaining position as a problem in that sector, whereas
the others did not. Another speaker pointed out that “economic dependency”
raises no competition concerns unless there is horizontal market power, and in
any case the difficulties in definition and measurement made it difficult to
formulate legislation against it.226 (Laudati, 1995) The OECD policy roundtable
on “Buying Power of Multi-Product Retailers”227 focused on food retailers, and
reviewed loss-leading (see below) and abuse of economic dependence. In
instances of economic dependence, a fundamental difference in negotiating
power results in the retailer, for example, receiving treatment not accorded by
the manufacturer under normal competition.
Examples include “unusually easy credit conditions, non-cost justified discounts
including retroactive rebates, exclusive supply conditions, and slotting
allowances/listing fees” (p. 41). Laws prohibiting abuse of economic dependence
have resulted in few successful cases; this failure has in turn generated
increasingly restrictive rules. (OECD 1999, p. 9, 42-43) Perhaps the failure could
have been predicted, as suppliers truly dependent would be unlikely to complain,
the protection of rivals runs against the grain of modern competition law
objectives, and remedy without ossifying distribution arrangements is difficult
(pp. 42, 45).
Dobson Consulting (1999)228 prepared a large study on buyer power in groceries
in the European Union and set out the relevant economic theory and policy
analysis, and studied the specific features of the sector both with a statistical
analysis and case studies of four of the larger Member States. Focussing on the
first part, he reviewed several standard models and results.229

225 The speaker may have been misinformed: the five firm concentration ratios (CR5) for food
retailing in the cited Member States were not higher than in others. Indeed, Germany was found
to have, “a relatively healthy degree of competition.” Tables 6.5 and 7.4, and p. 162, Dobson
Consulting 1999, “Buyer power and its impact on competition in the food retail distribution sector
of the European Union,” 13.10.1999, ec.europa.eu/competition/publications/studies/bpifrs/.
226 Laudati, Laraine L. 1995. “The First European Competition Forum: Vertical Restraints,”

Competition Policy Newsletter, 1(5), summer, available at

ec.europa.eu/competition/publications/cpn/cpn1995_2.pdf.
227 OECD. 1999. “Buying Power of Multiproduct Retailers,” DAFFE/CLP(99)21, available at

www.oecd.org.
228 Dobson Consulting (1999), “Buyer power and its impact on competition in the food retail

distribution sector of the European Union,” prepared under contract IV/98/ETD/078.

229 If a monopsonist, facing competitive suppliers, can depress prices below the competitive level

then producer surplus may be reduced, but final consumers are unaffected. If supply is perfectly
elastic, though, there is no welfare loss. If the monopsonist also has monopoly power
downstream, then both upstream suppliers and final consumers are harmed. These welfare
effects may be offset if there are efficiencies of larger buyers, a sort of “natural monopsony” effect,
or if the monopsonist can perfectly price discriminate. In addition to these static welfare effects,

PAGE 116 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Dobson proposed a structured approach to the assessment of seller and buyer

power, efficiencies, and attempts to deliberately create a dependency
relationship. If there is significant buyer power, suppliers are relatively
powerless, the buyer has significant selling power, and there are no significant
efficiency gains associated with buyer power, then one may ask whether the
buyer attempts to create a dependency relationship. To help identify attempts,
Dobson proposes seeking “evidence of exclusive supply requirements, specific
custom designs or arrangements, idiosyncratic specification, etc.” (p. 21) Any
anticompetitive effects to, e.g., raise barriers to entry or mobility, or otherwise
dampen competition, would need to be weighed against potential efficiency
effects. (p. 23)
Dobson notes that the provisions on abuse of economic dependency had not been
successful, and that classifying the offence as one of unfair competition “would
indicate that the issue is really over the division of economic surplus (profits)
rather than any detrimental impact on total economic welfare (which is the
essence of concerns in competition policy).” (p. 35)
He opines that, “The extent to which "economic dependency" as a concept is
appropriate in the design of competition policy remains contentious. In so far as
dependency may result in adverse welfare effects...then it is of direct concern to
competition policy. Otherwise, it may be better tackled under unfair competition
laws - e.g. when it is simply a matter of distribution of economic rents.” (p. 36)
Dobson (2005)230 studied economic dependency in the grocery market in the
United Kingdom, largely by analysing the information made public during
various competition enquiries. He noted that an individual supplier tends to be
far more reliant on an individual retailer than vice versa, and that this asymmetry
allowed retailers to negotiate favourable terms of trade. “The critical issue is that
retailers with buyer power can credibly threaten to cause substantial damage to a
supplier’s profits if negotiations break down and its products are delisted….. All
other things being equal, this degree of dependency will be in relation to the
share of the supplier’s sales that is accounted for by the retailer.” (p. 536) He
noted that market shares were not indicative of relative economic dependency,
indeed that the UK Competition Commission had found that retailers with
(downstream) market shares of as little as 8% could have buyer power. He
pointed out that national market shares hid high local concentrations. He further
noted that the 10% purchasing price disadvantage of smaller retailers, found in
the enquiries, could force their exit, thus harming consumers by reduced store

there may be reduced dynamic welfare if lower prices prompt suppliers to reduce investment,
ultimately reducing their own efficiency, thus raising prices upstream and downstream, thereby
harming consumer welfare. (pp. 13-14) Dobson also reviews the standard model where the
upstream seller is a monopolist and the buyer simultaneously a monopsonist and, downstream, a
monopolist. Making various assumptions about the information available to the parties and their
preferences, a plausible agreement between the parties would result in a larger quantity
transacted and greater welfare than in the monopsonist-competitive supplier model. If the buyer
competes in a competitive market downstream, the level of welfare is yet higher. (pp. 14-16) He
notes that the outcomes of bargaining between sellers and buyers, when they each have some
degree of market power, will “likely [have] significant non-linearities in pricing and other
constraints in contracts. [S]uch aspects of contracts…should not necessarily be treated with
suspicion, but rather on their merits and on the relative degree of concessions and constraints on
each side.” (p. 20)
230 Dobson, P.W. (2005), “Exploiting Buyer Power: Lessons from the British Grocery Trade,”

Antitrust Law Journal 75(2), pp. 529-562.

PAGE 117 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

choice and potentially higher retail prices. Suppliers may be deterred from
investing, resulting in lower product quality, variety and innovation, thus
harming consumers in the long run. He concluded that, “As retail concentration
increases, the leading retailers will be in increasingly strong positions to exploit
the three roles they occupy for producers—as their customers (purchasing and
then reselling their goods), as their competitors (when developing own-brand
ranges), and as their suppliers (in the provision of shelf and advertising space).”
Dobson and Chakraborty (2007)231 report on developments in food retailing in
Europe, and competitive effects, and review possible policy responses. Food
retailing is increasingly concentrated; there are more multiple chain-store
retailers, greater use of large store formats, more use of information technology
that leads towards greater “demand pull,” and the emergence of pan-European
retailers. These changes have led to greater efficiency and diversity, but also may
have increased retailers’ market power, both as buyers and sellers. Further, there
may be a dynamic “virtuous circle” of growth whereby reinforcing cost and
differentiation advantages allow some retailers to grow and others to lose scale
economies, invest less, and shrink. With respect to suppliers, retailer buyer
power is essentially expressed in low purchase prices and “controlling the terms
and conditions of trade.” Under specified conditions, such buyer power harms
consumer welfare. Possible policy responses include laws against economic
dependency, but cases brought under these laws have been unsuccessful. “Codes
of practice” have been introduced in the UK and Australia.
Prohibition of price discrimination in intermediate goods markets would result in
inefficiencies. Suppliers could be permitted to cooperate in negotiations with
retailers, although this could result in competition-harming coalescing power
rather than competition-enhancing countervailing power. Laws aimed at
preserving small retailers, prohibiting sales below cost, have had inflationary
effects in Ireland and France, although more targeted prohibitions such as in
Germany may be more effective. General prohibitions of predatory pricing are
difficult to enforce—the practice is difficult to prove and enforcement may occur
too late, after the damage is done. Tight planning restrictions reduce efficiency.
Structural measures are a possible policy response. But, “There appear to be no
universal or general solutions.”
Competition authorities in the United Kingdom have studied the groceries sector
several times in recent years. Leaving aside those conducted more than a decade
ago, the UK Competition Commission conducted four inquiries into grocery
retailing between 2000 and 2008. The reports of the 2000 and 2008 inquiries
are reported here. As a general rule, they examined two of the issues on interest
in this enquiry, abuse of economic dependence and sales below cost.
The 2000 investigation was broad, originating in the perception that prices of
many consumer goods were higher in the UK than in comparable EU countries
and the US. The inquiry concluded that certain practices of supermarkets “gave
rise to a complex monopoly situation, and found that two groups of these
practices operated against the public interest.” The first group of practices

231Dobson, P. W. and R. Chakraborty (2007), “Retailer-Processor-Distributor Interfaces,”

Presented at the “Policy for the Food Marketing Chain” Conference, Institute of Food and
Resource Economics, Copenhagen, 12 June 2007.

PAGE 118 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

operating against the public interest included persistent selling of some products
below cost, which distorted competition and harmed small retailers, which in
turn harmed less mobile consumers. The UKCC recommended no remedies for
these pricing practices. The second group of practices found to be operating
against the public interest concerned the conduct of five large grocery retailers
towards their suppliers. The result was the establishment of the SCOP, which
regulates the conduct of those grocery retailers whose market share exceeds 8%.
(UKCC 2000, paras. 1.2, 1.6a, 1.7a, 1.9-12)
The fourth inquiry, in 2008, by the UKCC was similarly wide-ranging.232 The
matters considered which are relevant to this study were possible increased
buyer power on the part of the large grocery retailers (and how this may affect
suppliers and rival wholesalers), and certain pricing practices, of which below-
cost selling, and its possible distortion of competition, is relevant here. (UKCC
2008, para. 2.10)
One part of the UKCC’s examination of buyer power was to measure the
“waterbed effect.” This refers to the alleged practice of suppliers charging higher
prices to smaller wholesalers to compensate for lower prices on sales to large
retailers. They examined the pricing, volume and cost data at the SKU (Stock-
keeping unit) level for a period of up to 5 years from 29 suppliers. Together this
amounted to at least 2% of UK grocery retail sales. They found that Tesco pays,
on average, a significantly lower unit price and net price than Asda, Morrisons
and Sainsbury’s, and the four together paid 4-6% below the mean. The other
large grocery retailers paid about the mean, large wholesalers paid 2-3% above
the mean and smaller wholesalers 8-9% above the mean.
The difference in prices was attributed to price reductions for large order and the
“greater buyer power of the four largest grocery retailers, particularly Tesco,
relative to other grocery retailers and wholesalers.” However, the differences in
supplier prices did not, in themselves, have an appreciable effect on competition.
(UKCC 2008, paras. 5.19-5.23, 5.26) A “waterbed effect” was not found since
characteristics and implications of the model were seen to be absent from the UK
market. (ibid., paras. 5.42-3)
The UKCC notes that buyer power is likely to help consumers, since if
competition among retailers is effective then the savings would be passed onto
consumers. Buyer power may be a countervailing force against any upstream
supplier market power. But it may reduce suppliers’ incentives or ability to
invest, thus leading to lower capacity, quality, or innovation. Buyer power may
also spur innovation by upstream suppliers. (ibid., 9.4-5) To evaluate the extent
of buyer power, the UKCC considered the size of grocery retailers relative to their
suppliers, the prices and margins negotiated between them, the share of retail
price earned by various links in the supply chain for various primary products,233
and a review of the email between the two largest retailers and their suppliers.
(para. 9.6) The alternatives open to the supplier and the retailer during their
negotiation depend in part on size (how large as compared with alternative
trading partners?) but also on whether the retailer is vertically integrated into
wholesaling, is the “gatekeeper” to local markets, or has own-brand products.

232 The second and third were inquiries prompted by acquisitions (Safeway in 2003, Tesco in
2007).
233 Little weight was placed on this measure as it is affected by much besides buyer power. (ibid.,

para. 9.19)

PAGE 119 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

(para. 9.7) The four largest grocery retailers have a large share of total grocery
sales, and their sales are much larger than that of other retailers’ and of
suppliers. Evidence on prices and margins included a CC analysis of supplier
prices, a comparison of wholesale and retail prices,234 a survey of suppliers (in
which they identified the type of customer from which they received their lowest
and highest margins, and their relative bargaining power), and qualitative
evidence from suppliers on the factors influencing supplier pricing. (para. 9.10,
9.13-4) The UKCC concluded that, “based on the size of grocery retailers,
wholesalers and buying groups relative to suppliers, together with the evidence
on supplier pricing and margins, all large grocery retailers, wholesalers and
buying groups have buyer power in relation to at least some of their suppliers.”
(para. 9.21) In evaluating whether suppliers had been harmed, in particular
whether small suppliers had exited or innovation had slowed, the UKCC found no
evidence of declining numbers of small suppliers nor falling innovation, although
it conceded that evidence about product innovation was not straightforward.
(ibid., paras. 9.29, 9.36)
The UKCC examined the conduct of large grocery retailers towards suppliers. Of
these, “supply chain practices” are most likely to be relevant in the present study.
The concern was that retrospective adjustments to terms of supply could transfer
excessive risk or unexpected costs, or contractual terms that created moral
hazard on the part of the grocery retailer (9.44-7) The UKCC concluded that, if
unchecked, these practices would reduce suppliers’ incentives to invest and
ultimately harm consumers. (para. 9.83)
Abuse of economic dependence has largely been studied within the grocery sector
in connection with supplier-retailer relationships, although in principle the
concerns could arise with respect to any intermediate products. The concern is
that suppliers are squeezed by retailers, thus under-invest and ultimately exit.
The extent to which abuse of economic dependence is a concept fitting within the
framework of competition law is contentious, as if dependence is a matter of
distribution of economic rents then it may be better addressed elsewhere.
However, consumer welfare—a concern of competition law—can be harmed by
such buyer power if variety or innovation is reduced, or where retailers have
market power in the downstream (retail) market. The 2008 study of the groceries
sector by the UK Competition Commission found no evidence of declining
numbers of small suppliers nor slowing innovation, though conceded that
evidence on product innovation was not straightforward. The UKCC nevertheless
identified certain practices that excessively transfer risk or increase moral hazard
on the part of retailers that would reduce suppliers´ incentives to invest and
reduce consumer welfare.

234 Less weight was placed on this evidence for various reasons. Those potentially relevant for this
study include that retail and wholesale prices were not observed at the same point in time;
wholesale price used invoiced prices (which exclude discounts and overriders); use of a simple
average price across wholesalers. The econometric analysis did not allow the measurement of the
effect of size separate from the effect of scale economies on prices. (ibid., Appendix 5.3, paras.
2,11).

PAGE 120 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

3.3 Assessing the impact of sales below cost

Loss-leaders are defined as, “[I]tems sold by a retailer at sharply reduced prices,
and perhaps even below cost, in order to attract customers who, once they have
entered the store, will buy other goods at standard prices.”235 Loss leading
benefits consumers at least in the short run, but two other sets of businesses may
be harmed. Small retailers dependent on the loss-led products may be driven out
of business, but this harms consumers only if the former loss-leader reverts to
supra-competitive pricing. Retailer differentiation was not considered in the
review. Manufacturers of products vulnerable to loss-leading may under-invest in
product quality, but “Loss-leading is unlikely to increase buyer power or lead to
immediate sales reductions,” or to reduce consumers’ quality perceptions. (OECD
1999, p. 35) The conclusion was that some measures to control buyer power, such
as laws that reduce the use of price competition, e.g., by restricting loss leading,
“tend to reduce competition to the detriment of consumers while offering little or
no long term relief to threatened small stores and manufacturers.” (p. 9)
The OECD’s 2005 review of “Resale below Cost Laws and Regulations”236 found
that RBC laws likely result in higher prices to consumers because they arbitrarily
increase the costs of low-price, high volume retailers and, by reducing consumer
incentives to comparison shop, reduce competition. Even countries without RBC
laws continue to have a variety of retailers, including small local ones. RBC laws
increase incentives to vertically integrate. The equitable enforcement of RBC laws
is difficult, leading to high costs of enforcement and the risk of mis-enforcement.
If RBC laws inhibit entry by low-price, high-volume retailers, then it is relevant
that general retail price fell about 15% when such retailers entered Guam and
Hawaii.
The report referenced a number of other studies:
 The Irish competition authority reported that retail price inflation had
been higher for groceries covered by the Groceries Order (which
prohibited RBC) than for other groceries. (See below.)
 Macard (2004)237 reported that, in France, “most agree” that the Galland
law, which introduced a prohibition of sales below cost, “has been a cause
of inflation for branded products.” Retailers have dramatically increased
the sale of private-label products, at the time of publication 20 to 30% of
sales.”(p. 19) In both France and Germany, the retail sectors have been
hindered by rules restricting “large stores” or “hypermarkets.”
 EMI found no change in the rate of decrease in the number of small- and
medium-sized shops (100-400m2 and 400-1000m2, respectively), nor any
effect on the prices received by farmers and market gardeners, as a result
of a retail price war. (OECD 2005c, p. 156) The EIM study found that, “In
countries with a separate prohibition (France and Belgium), the position

235 OECD (1999), “Buying Power of Multiproduct Retailers,” DAFFE/CLP(99)21, available on

www.oecd.org. Quoting 235 Sherer, F.M. (1980), “Industrial Market Structure and Economic
Performance,” 2nd ed. Rand McNally: Chicago, p.592.
236 OECD. 2005c. “Resale Below Cost Laws and Regulations,” DAF/COMP(2005)43. available
at www.oecd.org.
237 Macard, O. (2004), “France,” in Ernst & Young Global Pricing Trends 2004 available at
http://www2.eycom.ch/publications/items/rcp_pricingtrends2004_report/en.pdf.

PAGE 121 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

of small supermarkets was under less pressure (in terms of numbers and
market share) than in the other countries. Taking the small and medium-
sized supermarkets together, the decline in this segment as far as share of
turnover is concerned is in fact greater in France than in other countries.
In countries where such a prohibition is derived from the prohibition on
the abuse of a position of economic power (Germany and Austria), the
number of small supermarkets is declining at more or less the same rate as
in countries without a prohibition, but the decline in the share of the small
supermarkets’ turnover is faster than elsewhere (especially in Germany).”
(p. 157) The effect on consumers of introducing a RBC prohibition was
unclear since the extent of RBC was unclear; retailers may simply
maintain more similar margins across items. Farmers and market
gardeners were expected to be unaffected since their selling prices are
determined by forces outside the Dutch retail market. (p. 157)
 Odabashian 1998238 surveyed milk prices in California and concluded that
RBC prohibition in California had deterred retail competition in milk,
resulting in higher prices in California than in nearby states. (p. 193)
Indeed, further investigation finds that the California Department for
Food and Agriculture publishes a list of the lowest reported lawful prices
for various dairy products.239
Various tests to detect predatory pricing, both in single and multiproduct
situations, were described in OECD’s “Predatory Foreclosure.”240 Different
measures of cost are used in different jurisdictions. Many jurisdictions take
recoupment into account in predatory pricing cases, recognizing that recoupment
can occur in a different market from that in which the predation occurred.
Pricing below an appropriate measure of cost, even by a dominant firm, is not
always predatory; it may be doing so for reasons unrelated to the exclusion of
competitors, for example as part of a promotional or inventory disposal strategy.
However, exclusion may be achieved at lower cost by raising rivals’ costs by, for
example, using regulatory mechanisms to raise entry barriers. (pp. 7-10).
“Vertical Relations in Gasoline Retailing”241 (OECD 2008) is relevant in
indicating that small retailers can respond to pricing pressure by increasing their
differentiation.
Independent petrol retailers occasionally complain of price squeezes by vertically
integrated retailers which are aimed at appropriating profits generated by the
independent retailers’ efforts and investments. (p. 7) An independent study
engaged in as a part of an investigation of such allegations found that, in fact, the
key drivers of profitability of petrol retailers were high volume and providing
ancillary services. (p. 21) And these are not static features: Where large retailers
had entered into petrol retailing, resulting in petrol retail price decreases, the
incumbents had responded by consolidating, adding retail shops applying higher

238 Odabashian, E. (1998), “ The Land of Milk and Money; San Francisco Bay Area Retailers
Charge Too Much for Milk.” Consumers Union West Coast Regional Office. May, available at
www.consumersunion.org/other/sfmilkrptwc598.htm.
239 See “Lowest Observed Retail Price for Milk” at
www.cdfa.ca.gov/dairy/pdf/.../2008/Lowest_Lawful_Retail_Jun2008.pdf.
240 OECD (2005a), “Predatory Foreclosure,” DAF/COMP(2005)14, available on www.oecd.org.
241 OECD (2008a), “Competition Policy for Vertical Relations in Gasoline Retailing,”

DAF/COMP(2008)35, available at www.oecd.org.

PAGE 122 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

mark-ups to snacks and beverages than petrol, and increasing the number of
pumps. (p. 9)
ICN 2008242 reports on a survey of national competition agencies worldwide on
the analyses they apply to predatory pricing. Predatory pricing is a two-stage
strategy whereby a dominant undertaking first charges prices so low as to induce
rivals to exit the market and later charges monopolistically high prices. Another
version of the strategy involves, for example, charging low prices to new
customers and charging high prices to those customers who have incurred high
switching costs. Virtually all competition agencies compare price to a measure of
cost, although the measure(s) used varied. Average variable cost was most
commonly cited, although average avoidable cost is gaining in popularity.
Agencies also took into account some or all of: whether losses could be recouped,
whether there was an effect on competition, whether there was predatory intent,
and a variety of defences and justifications, such as pricing to “meet the
competition” or for seasonal reasons. (ICN 2008, p. 3)
Studies cited in the United Kingdom Competition Commission’s study of the
grocery sector address these questions in the United Kingdom. Increased
concentration in grocery retailing in recent years has raised issues about the
buying power of large retail chains, and its effect on competition and consumer
welfare. Several formal investigations by the British competition authorities have
been made over the last couple of decades. These include the industry inquiries
conducted by the Monopolies and Mergers Commission in 1981,243 by the Office
of Fair Trading (OFT) in 1984 and 1999244, the “Supermarkets” and “The supply
of groceries in the UK market” investigations by the Competition Commission in
2000245 and 2008246, respectively.
The most recent, ‘The Supply of groceries in the UK market investigation’ (2008)
explored the effects of below-cost-sales but in the UK market. It did so in several
ways useful to our analysis. First, the study asked whether below-cost sales harm
rivals. They found that smaller retailers are not in direct competition with larger
retailers; instead, they are somewhat differentiated. To test the specific question
of whether conditions for predation were nonetheless present, they sent specific
questionnaires about below-cost-sales to 15 large grocery retailers. They found
that conditions for predation were not present, and, in particular, that below-cost
pricing is neither broad enough nor long enough to force the exit of smaller

242 ICN (2008), “Report on Predatory Pricing,” available at

www.internationalcompetitionnetwork.org.
243 Monopolies and Mergers Commission, Discounts to Retailers: A Report on the General Effect

on the Public Interest of the Practice of Charging Some Retailers Lower Prices than Others or
Providing Special Benefits to Some Retailers Where the Difference Cannot Be Attributed to
Savings in the Supplier’s Costs, HC311 (May 1981), available at http://www.competition-
commission.org.uk/rep_ pub/reports/181/135discounts-retailers.htm.
244 The 1999 OFT inquiry led to the referral to the Competition Commission for a full monopoly

inquiry.
245 Competition Commission, Supermarkets: A Report on the Supply of Groceries from Multiple

Stores in the United Kingdom, Cm 4842 (Oct. 2000), available at

http://www.competitioncommission. org.uk/rep_pub/reports/2000/446super.htm# full.
246 United Kingdom Competition Commission, ‘The supply of groceries in the UK market

investigation”, 30 April 2008. Available at http://www.competitioncommission.

org.uk/rep_pub/reports/2008/538grocery.htm.

PAGE 123 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

rivals.247 Further, they found that barriers to entry into small retailing were low.
(UKCC 2008, paras. 5.58-5.60)
A second part of the study to assess whether below-cost sales harm small
retailers estimated empirically the net exit rate of a specific category of retailer at
a given location and a given time as a function of the entry of a large store or a
convenience store (bearing the fascia of Tesco, M&S or Sainsbury) in the area.248
Using an existing database of the presence of retailers in several hundred
shopping centres and high streets, taking advantage of the fact that alcohol was a
product group subject to below-cost sales and that local off-licenses and
convenience stores compete in the sale of alcoholic beverages with large retailers,
they found that there was no significant effect on the rate of entry or exit on local
off-licenses, which compete in the sale of alcoholic beverages with large retailers,
within a one or two year period following entry of a large grocery store in the
area. (paras. 5.65-5.66)
Finally, the study also excluded the hypothesis that consumers would be harmed
by the above-mentioned selling policies by being misled into falsely believing that
all the prices offered by the large retailers were low. Specific research done for
the UK CC shows, on the contrary, that consumers do not normally carry lots of
product-specific price information with them and that “basket price”
comparisons are prominently available, so they are not misled. (ibid., para. 5.67)
A study of the effect of the removal of the SBC restriction in the Groceries Order
in Ireland addressed the question of whether retail prices had been actually
affected by the SBC restriction. This study, Irish Grocery Report No. 2, 9th April
2008249 used trends in grocery retail prices from 2001 to 2007 to measure the
impact of the removal of the Restrictive Practices (Groceries) Order of 1987 (the
‘Groceries Order’) in March 2006. (The Order had prohibited retailers from
selling items covered by the order below cost, where the cost benchmark was the
net invoice price.) Three sub-indices (grocery items covered by the Groceries
Order, grocery items not covered by the Groceries Order, overall grocery items)
were used.
They found that removal of the Groceries Order affected the trends in retail
prices for those items covered by the order, especially during the nine months
period following the removal of the Grocery Order. However, they also observe
that towards the beginning of 2007 (i.e. the last year of observation) “the price
trends for Grocery Order items and Non Grocery Order items appear to behave
similarly again (with both trends rising)”.
The ‘Rapport Canivet,’250 aiming to provide advice on possible reforms towards
simplifying the ‘supplier-retailer’ relationship, found that if RBC laws were

247 Questionnaires were sent to 15 large grocery retailers. Responses indicated that 10 engaged in
below-price selling. Dry groceries and alcohol were the product groups with the largest volume of
such sales. Up to 3% of total revenue consisted of below-cost sales. Durations were (8 to 25 weeks
for an individual item; for branded products the length of time is normally shorter than own-label
ones. (UK CC 2008, para. 5.53-5.54)
248 More details can be found in appendix 5.2 of UK Competition Commission (2008).
249 The (Irish) Competition Authority, “Price Trends in the Irish Retail Grocery Sector: A

Description of the Evolution of Retail Grocery Prices between 2001 and 2007”, Grocery Monitor
Report No. 2.
250 Canivet (2004), “Rapport du groupe d’experts constitué sur les rapports entre industrie et

commerce. “

PAGE 124 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

removed, price competition among retailers would be enhanced and that

anticompetitive conduct would still be caught by abuse of dominance
prohibitions. “[A]ccording to the economic theory and checked against
competition law, RBC is not necessarily anti-competitive per se, especially if one
considers that RBC cannot be exercised in the long term by firms that are non-
dominant“. At the same time, the report adds that “[nonetheless] RBC, even
when considered in the category of misleading practices, still results into the
protection of some kinds of downward distribution, e.g. in cases where big
retailers have the established practice of fixing very low (i.e. below cost) prices on
high-elasticity products” (pp. 70-74)
The Australian Competition and Consumer Commission (ACCC) studied the
effect of entry by two large grocery retailers into the sale of petrol via joint
ventures with established petrol retailers.251 The petrol outlets of each of these
joint ventures offer a 4 cents/litre discount on the price of petrol to those
customers who have purchased a minimum quantity from the supermarket or
other companies in the group. These schemes resulted in the joint ventures
gaining market share, incumbents responding by changes in geographic location
or offering their own discounts, but not accelerating their exit from the market,
and lower prices of petrol but not higher prices of groceries. To examine the price
effect, the ACCC used retail petrol prices in the five largest metropolitan areas
before and after one of the joint ventures began operating in those cities (at
three, six and twelve months). The comparator was, in each case, the import
parity indicator (Singapore price for refined petrol and the Australian/US dollar
exchange rate). The ACCC found that posted prices were lower after the entry of
the schemes by between 0.5 and 3 cents per litre, and that the schemes’ prices
were generally lower than average retail price in each city.
A series of studies have examined the effect of Wal-Mart superstores on prices or
consumer expenditure in the United States. Such stores are extremely large and
offer significantly lower prices on a wide variety of consumer products. They
provide support for the use of samples of retail prices to estimate effects on
consumers of retail practices.
 Basker (2005)252 studied the effect of the entry of Wal-Mart into
metropolitan areas on the price of ten specific, commonly-purchased
items. He excluded groceries and alcoholic beverages since most Wal-Mart
stores during the relevant period had no grocery section. He used a dataset
containing the dates when Wal-Mart stores opened, and a dataset
containing average quarterly retail prices in 165 cities of the ten items over
the period 1982-2002. He compared the price dynamics before and after
Wal-Mart entry. He found a price effect for several products, generally
with prices falling 1,5-3% in the short run and 7-13% in the long-run.
 Hausman and Leibtag (2007)253 found both a direct and an indirect price
effect of the entry of a Wal-Mart store on consumer expenditure during

251 Australian Competition and Consumer Commission (2007), “Petrol Prices and Australian
Consumers: Report of the ACCC inquiry into the price of unleaded petrol,” December.
252 Basker, E. (2005). “Selling a Cheaper Mousetrap: Wal-Mart’s Effect on Retail Prices,” Journal

of Urban Economics, Vol. 58, no. 2, September. available at www.ssrn.com.

253 Hausman, J. and E. Leibtag (2007). "Consumer Benefits from Increased Competition in

Shopping Outlets: Measuring the Effect of Wal-Mart." Journal of Applied Econometrics no. 22,
pp. 1157-1177.

PAGE 125 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

1998-2001. Using a dataset on purchases by a randomly selected set of

households across the US, they found that the rapid increase of
supercenters during the 1990s raised consumer welfare, especially among
low-income households. In particular, the average prices paid by
households decreases as more supercenters operate in a given geographic
market. Turning to details, superstores typically charge prices for food
items 5% to 48% lower than prices for the same products in supermarkets
and other conventional retail outlets. Thus, average prices in a market
decrease as more household shift food purchases towards superstores. In
addition, the authors detected an indirect price effect whereby
supermarkets lowered prices to remain competitive, further lowering
average price paid in a market. Second, using an empirical model to
estimate the total effect on consumers of the introduction of a new
shopping outlet (considered as the difference in the consumer’s
expenditure function before and after the introduction) they found that
the exact compensating variation that arises from the direct “variety
effect” of entry and expansion of supercenters, i.e., the increase in
consumer welfare as a result of the availability of a new retail outlet, is
equal to 20,2% of average food expenditure. By contrast, the indirect price
effect, the increase in consumer welfare due to incumbents lowering prices
in response to supercenter entry, is 4,8% of average food expenditure.
They found that lower-income households have a greater tendency to shop
at low-priced outlets, so benefited about 50% more than the average from
the entry of supercenters. Thus, consumer surplus is significantly
decreased when supercenter entry and expansion is limited in particular
geographic markets.
 Volpe and Lavoie (2008)254 assessed the pricing strategy of Wal-Mart
supercenters and the response of supermarkets when facing direct
competition from a supercenter, i.e., the extent to which supercenters
bring about a decrease in their rivals’ (grocery) price in 2000 and 2004.
They considered stores in the same geographic area as being in direct
competition, and devised price indexes (a weighted average price of the
goods - of a given brand, in a given department and of a given store - in a
store) built for different supermarket departments to uncover
supercenters’ pricing strategies. The authors found that the presence of
Wal-Mart supercenters results in a decrease in grocery prices of 6% to 7 %
for national brand goods at conventional supermarkets competing within
a radius of five miles; for private label goods, the price decrease is 3% to
7%. These price decreases vary according to the specific product (in total
six categories of grocery products were included in their analysis) and
also, more generally - taking into account market concentration - other
demographic variables, and store characteristics, Wal-Mart Supercenters
price their national brand and private label products 17% to 23% and 8 to
14% lower than conventional supermarkets, respectively. While they did
not consider SBC rules explicitly, and the data available to perform a
similar analysis in the European markets would be unavailable (given that
the authors directly collect data off the shelves of the stores sampled), this

254Volpe III, R.J. and N. Lavoie (2008). “The Impact of Wal-Mart Supercenters on Grocery Prices
in New England.” Review of Agricultural Economics, 30(1): 4-26.

PAGE 126 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

study nevertheless provides support for the use of retail price indices as a
measure of the consumer effect.

A series of studies have examined the effect of certain states in the United States
restricting the sale below cost of petrol.
 Anderson and Johnson (1999)255 assess empirically the impact of SBC
laws on the U.S. retail gasoline market in the years 1987 to 1992. They
conclude: “SBC laws directed specifically at the retail gasoline market have
resulted in higher margins.” They found that, “The difference between
margins at locations where a gasoline-specific (sales below cost) law is
present and where there are no SBC laws to be over 2 cents per gallon.”
The margin difference indicates that SBC laws limit downward pressure
on retail margins, thus restricting competition and raising prices for
consumers.
 Brannon and Kelly (1999)256 examined whether the State of Wisconsin’s
Unfair Sales Act – which prevents the sale of any item below cost in order
to attract business – effectively protected small, independent petrol
retailers. They compared gasoline prices in some cities in Wisconsin with
those of similar-sized markets outside the state, in the period 1996 to
1999. They found that the strengthening of the penalties for violating the
Unfair Sales Act did not bring more competition on the market.
On the contrary, the average mark-up of retail gasoline price over the
wholesale price increased in all the markets in the study. Moreover, there
is little evidence to suggest that the law saved any small, independent
petrol stations, “given that the government already drove most small
stations out of business in the past twenty years with stringent
environmental regulations.”
 Fenili and Lane (1985)257 quantitatively assessed how SBC laws affected
retail gasoline prices in 43 U.S. Standard Metropolitan Statistical Areas
(SMSAs) in 1982. Some SMSAs were located in states with SBC laws and
others were not. Their econometric analysis found that gasoline prices
were consistently higher in states with SBC laws. Annual average price
differences were 0.68 to 0.9 cents a gallon higher at the self-service pump
for regular leaded gasoline and unleaded gasoline, respectively, and 2.47
to 2.67 cents a gallon for full-serve pumps. This translated into an
additional cost to consumers of regular gasoline of more than $640
million in 1982. They noted that two massive studies of gasoline
marketing by the U.S. Department of Energy conducted in 1984 had found
no evidence of widespread predatory pricing in the retail gasoline market.
Further, “[T]he studies found that consumers are the prime beneficiaries
when retailers compete vigorously.” At the same time, the authors argued
that SBC laws were generally unenforceable, and thus have little effect.

255 Anderson, R. and R. Johnson (1999), Antitrust and sales-below-cost laws: The case of retail
gasoline, Review of Industrial Organization 14: 189-204.
256 Brannon, J. and F. Kelly (1999), Pumping up prices in Wisconsin: The effects of the unfair

sales act on retail gasoline prices in Wisconsin, Wisconsin Policy Research Institute Report.
257 Fenili, R.N. and W.C. Lane (1985), Thou shalt not cut prices! Sales-below-cost laws for gas

stations, AEI Journal on Government and Society/Regulation September/October: 31-35.

PAGE 127 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 By contrast, Skidmore and Alm (2005)258 found that, unlike most previous
work, on average gasoline prices are about a cent per gallon lower five
years after the (SBC) law is imposed. They also found that SBC laws have
an impact on the market structure, particularly translating into a greater
total number of gasoline companies. However, according to their findings,
it is not small businesses that are mostly protected by those laws, but
rather establishment with five or more employees. The methodology they
use is broader and most comprehensive than those applied in previous
studies. First, the authors used average state pricing data rather than for
cities or metropolitan areas. This particularly allows them to avoid the bias
that might originate from specificities uniquely typical of the cities or
states included in the model with respect of the rest of U.S states. Second,
they consider a time-span longer than that of most previous studies: By
collecting time series data both for the (13) states that adopted motor fuel
sales-below-cost laws, as well as on the others, during the 1982-2002
period, the authors also use variation across the states in the timing of the
adoption of these laws to investigate how these laws affected average
prices in states where they have to be implemented.
 Zimmerman (2009)259estimated the effect on retail gasoline prices of SBC
restrictions and the presence of hypermarkets selling gasoline during
1998-2002 in the United States. He constructed price indices of the
various motor fuels and exploited the fact that some states had laws
prohibiting the sale of motor fuels below cost, but others did not, and the
fact that some states had hypermarkets (who also sold gasoline) whereas
others did not. He found that the presence of hypermarkets lowered the
retail price of gasoline, albeit by a small amount, but that the presence of
SBC prohibitions reduced the size of the hypermarket effect by
approximately half.
Dobson and Waterson (2007)260 found, based on a theoretical mode, that resale
price maintenance where both suppliers and retailers are oligopolistic can aid the
coordination of retail prices, and thus raise retail prices. This is relevant in that
laws prohibiting sales below cost have an effect like resale price maintenance,
whereby manufacturers choose a variable that sets a lower limit on retail prices.
Alain and Chambolle (2005, 2009)261 provide a theoretical framework for
analysis of the effect of sales-below-cost laws in the EU. In the 2005 paper, the
authors’ analysis sustains the idea that prohibiting resale at a loss may have
inflationary effects. They show that a monopolist producer, by setting a wholesale
price where resale at a loss is banned, in effect establishes an industry-wide price
floor, thus suppresses downstream competition. In the 2009 paper, the authors
focus on legislation of countries like France, Ireland or Spain, where conditional

258 Skidmore, M., J. Peltier, and J. Alm (2005), Do state motor fuel sales below-cost laws lower
prices?, Journal of Urban Economics 57: 189-211.
259 Zimmerman, P.R. (2009), “The competitive impact of hypermarket retailers on gasoline

prices,” unpublished mimeo, available at http://mpra.ub.uni-

muenchen.de/20248/1/MPRA_paper_20248.pdf.
260 Dobson, P.W. and M. Waterson (2007), "The competition effects of industry-wide vertical

price fixing in bilateral oligopoly," International Journal of Industrial Organization, vol. 25(5),
pp. 935-962. October.
261 Allain, M-L. and C. Chambolle (2009), “Anti-competitive effects of resale-below-cost laws”,

Cahier n 2009-09, available at http://www.enseignement.polytechnique.fr/economie/.

PAGE 128 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

or deferred rebates that are not written in the invoice are excluded from the legal
minimum price threshold. They show that when, as in France or Ireland, the law
is combined with anti-discrimination rules that prevent discriminating among
retailers, this price-floor is de facto industry-wide. As in the first paper, they find
that the ban on resale below-cost enables producers to impose price-floors on
retailers. This induces higher prices on the whole range of products sold and not
only on those that would have been sold below cost.
According to an opinion issued by the French Competition Council, cited in
Gianni, Origoni, Grippo & Partners,262 retail prices charged by large retailers
increased by 1.1 per cent between 1998 and 2003, while the price of discounters
decreased. Prices of national “must carry” brands increased more than the prices
of other brands, and margins as reflected in invoices fell from 4.4 to 3.5 per cent,
whereas margins not reflected in invoices increased from 22 to 32 per cent. They
concluded that the retailers raised prices by increasing their margins, relying on
non-invoiced rebates and discounts.
Two ways of viewing this change are (1) that greater use of out of invoice
discounts reflects retailers’ buyer power, with the Loi Galland, prohibition of
sales below costs contributing to price coordination, or (2) that suppliers
facilitate the use of out of invoice discounts as it allows them to exert more
influence on retail prices.
Biscourp, Boutin and Vergé (2008)263 studied the effect of the Loi Galland,
which prohibited inter alia retailers from selling below cost in France starting in
January 1997. While below-cost sales had been prohibited before, this law clearly
defined the threshold below which sales could not be made as the invoice price,
i.e., the price paid by the retailer on delivery, but not including any end of year
rebates. According to BBV, negotiations between retailers and suppliers shifted
from “upfront margins” (rebates or discounts that are included on the invoice) to
“hidden margins” (end of year rebates and commercial cooperation). The authors
reference the producers’ association ILEC as saying that the average hidden
margin increased from 22% of net wholesale price in 1998 to 32% in 2003. Two
practices tend toward the creation of a price floor common to all retailers: The
shift towards negotiating on the basis of hidden margins rather than upfront
margins, and the practice of producers publishing “General Terms of Sale” that
are non-negotiable and non-discriminatory.
Both BBV (2008) and Allain, Chambolle, and Vergé (2008) report that AC
Nielsen looked at the prices of 1500 national branded products sold by large
retail chains. It found that retail prices increased by more than 4% during the
first two months of 1997. By contrast, the DGCCRF looked at prices of national
brands, private label, and low-price brands, and found an average retail price
increase of 0.5% over the same period. Nielsen compared the price increase with

262 Avis of the French Competition Council concerning the conditions of competition in large
retail distribution October 2004; Gianni, Origoni, Grippo & Partners, 2010, “Assessment of buyer
power in recent market investigations and mergers,” May.
263 Biscourp, Pierre, Xavier Boutin and Thibaud Vergé (2008). “The Effects of Retail Regulations

on Prices: Evidence from the Loi Galland,” Working Paper G2008/02, Institut National de la
Statistique et des Études Économiques, Direction des Études et Synthèses Économiques.

PAGE 129 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

that of the previous year (end 1995 to the end of February 1996) and found a
0.31% increase.264
Collins et al. (2001)265 studied the effect on the retail-wholesale price margin for
products covered by the Groceries Order, an Irish law prohibiting sales below
cost of certain grocery products. They found that gross margins in the category
they studied increased by 4,6 percent as a result of the imposition of the law in
1987. They find their finding accord with an Irish Central Statistics Office finding
that retailers’ margins increased from 15,8 percent to 20,1 percent over the
period 1988-1993. They conclude that the Groceries Order had had “a significant
positive influence on retail gross margins, and weakened retail competition.”
In a preliminary and unpublished paper, Chen and Rey (2009)266 analyze the
price competition among multi-product firms (retailers), with a focus on the anti-
competitive effect on small retailers of below-cost pricing. Rather than studying
the exercise of buyer power through vertical restraints, the authors show how the
large retailer can exert its seller power to exploit the smaller but more efficient
rival through loss leading.
They define loss-leading pricing as “an advertising strategy which allows retailers
to attract consumers by subsidizing some products and make profits from other
items”. Despite the fact that below-cost pricing could be seen as a way to improve
both consumers’ and social welfare by compensating consumers for their lack of
information, the authors find that “large retailers can instead use loss leading as
an exploitative device at the detriment of small retailers, without any efficiency
justification in terms of distribution cost or advertising”. Finally, the authors still
appeal cautious antitrust intervention against loss leading based on the fact that
“its harmful impact on competition depends on the concrete economic
environment.”
Prohibitions of sales below cost, outside the context of predatory pricing, are
generally aimed at the protection of small retailers in competition with large
retailers. However, a large number of studies in a variety of markets tend to show
that the effect is to raise consumer prices, thus harming consumers. The 2008
study of the groceries market by the UK Competition Commission found that
many of the arguments in support of prohibitions of sales below cost were not
borne out by the evidence. They found that small retailers were not in direct
competition with large retailers, but were instead somewhat differentiated. For
example, they found no effect on small retailer exit or entry of the entry of a large
store into the neighbourhood. They found that consumers were not misled by
loss-leading (charging low prices on certain well-publicized items) into thinking
that their entire “grocery basket” was cheaper at the loss-leading store. And they
found that below-cost pricing was not of sufficient breadth or duration as to
constitute predatory pricing. A few studies have estimated the effect on consumer
prices of sales below cost prohibitions, taking advantage of changes in the laws to
form the estimate. The estimates ranged from about 0.5% to over 4%. Another
study found that retailers changed their negotiating practices so as to better
accommodate restrictions on below cost sales.

264 LSA.fr “Loi Galland jusqu’où les prix vont-ils grimper? 6.03.1997
265 Collins, A., S. Burt and K. Oustapassidis (2001), “Below-cost legislation and retail conduct:
evidence from the Republic of Ireland,” British Food Journal 103(9), pp. 607-623.
266 Chen Z. and P. Rey (2009), “Loss leading as an exploitative practice,” preliminary note.

PAGE 130 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

3.4 Assessing the macroeconomic impact of competition

rules
The study of the macroeconomic impact of antitrust law provisions on unilateral
conduct can be firstly considered in the wider framework of studies that estimate
the cost of monopoly and ask whether antitrust contributes to social welfare. As
observed already by Renda, Riley, Rodger, Van Den Bergh, Keske, Pardolesi,
Caprile, and Camilli (2007),267 ‘There is a substantial stream of literature
discussing the contribution of antitrust enforcement as a whole to social welfare’.
An estimate of the macroeconomic impact of a possible change in the
competition rules on abuse of dominance in some Member States is thus fraught
with difficulty, as the history of the search for a macroeconomic impact of any
competition rules illustrates. Early authors such as Lerner (1937) and Harberger
(1954) significantly shaped the debate on the social cost of monopoly, and
potential ways to measure it.268 They tried to measure directly the social welfare
costs of less-than-perfect competition. Much later, other authors have tried to
measure the effect of cartels and mergers on social or consumer welfare. Nold,
Block and Sidak (1981) tried to estimate the deterrent effect of antitrust
enforcement in a single sector by measuring changes in mark-ups as a result of
antitrust actions.
Harberger (1954)269 tried to measure the deadweight losses in American
industrial sectors due to the exercise of market power. He identified a period
during which, he felt, the economy had been close to long-run equilibrium and
accounting values of capital had been close to actual values. Using accounting
data to generate figures on returns on capital in 73 sectors, and assuming a
unitary elasticity of demand and that all sectors supplied final—not
intermediate—goods, he calculated the size of deadweight loss in each industry
and summed them to estimate the deadweight losses of the economy as a whole.
It was a tenth of one percent of GDP.
Harberger’s work has subsequently been criticized. Relaxing the assumption of
unitary demand and using data from the 1960s, Cowling and Mueller (1978)
found deadweight loss to sum to 4 percent of GDP. Masson and Shaanan (1984),
using a different method, found deadweight losses of 3 percent of GDP. Second, if
one is concerned with consumer welfare, then not just the deadweight loss
“triangle” (welfare loss due to efficient transactions not occurring) but also the
“rectangle” of consumer surplus loss (loss due to higher prices) should be
measured. Cowling and Mueller (1978, 1981)270 counted expenditure on some
rent-seeking activities as welfare losses, and calculated a 13 percent of GDP loss
from imperfect competition.

267 Renda, Riley, Rodger, Van Den Bergh, Keske, Pardolesi, Caprile, and Camilli (2007), “Making
antitrust damages actions in the EU: welfare impacts and potential scenario”, final report by the
Centre for European Policy Studies (Ceps), Erasmus University Rotterdam and Luiss Guido Carli
University (LUISS), Report for the European Commission, Contract DG COMP 2006/A3/012.
268 See Lee and Brown (2005) at http://cowles.econ.yale.edu/P/cd/d15a/d1528.pdf
269 Harberger, Arnold C. 1954. “Monopoly and Resource Allocation,” The American Economic

Review, vol. 44, no. 2, May, pp. 77.87.

270 Cowling, K. and D.C. Mueller (1978), “The Social Costs of Monopoly Power,” Economic

Journal 88, pp. 727-748; (1981), “The Social Costs of Monopoly Power Revisited,” Economic
Journal 91, pp. 721-725.

PAGE 131 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Third, Harberger measured static welfare losses, but dynamic losses caused by
less competition reducing incentives to innovate are likely to be substantial.
Baker 2003271 reviews this body of work to estimate whether the benefits of
antitrust activity exceeded its costs. As a part of this, he estimated a lower bound
on the cost of anticompetitive activity on the U.S. economy. “[A]lthough any
quantitative calculus is highly speculative,” he arrives at a conclusion that, “If the
Harberger approach is taken seriously [but more realistic demand elasticities are
allowed, and consumer surplus rectangles or rent seeking and foregone
innovation are taken into account], the costs to the economy from the exercise of
market power could readily be at least 1 percent of national product…
notwithstanding the antitrust laws.”
An alternative to trying to measure losses due to all anticompetitive activities is
to estimate consumer welfare losses to well-defined classes of anticompetitive
activities. Estimates of the size of consumer overcharges due to cartels generally
estimate the difference between prices with the cartel and prices without, and
ignore deadweight losses. Levenstein and Suslow (2006)272 survey the literature.
They note that a handful of studies document a range of price increases, but that
many case studies do not posit a “but for” price, so do not estimate the effect of
the cartel. They note that very little research has been performed on the effect of
cartels on investment and innovation. Connor and Bolotova (2005)273 describes
some of the methods used to estimate a benchmark price against which cartel
overcharges can be calculated: “before and after” (assumes the non-cartel period
the market was in long-run competitive equilibrium), “yardstick” (assumes the
comparator is indeed comparable but competitive and not affected by the
cartelized market), “profitability” (assumes certain features of cost functions,
dynamism, and that other markets generate no economic profit), and
“econometric modelling” (assumes certain models of oligopoly).(pp. 18-20)
Estimates of the price rises resulting from mergers were made in Saltzman, Levy,
Hilke 1999.274 They studied mergers involving soft-drink bottlers in the United
States during 1980-1991. None of the mergers was blocked by the antitrust
agencies. Where a horizontal rival with less than a 5 percent market share was
acquired by a Coca-Cola or Pepsi bottler, the retail price of carbonated soft drinks
increased by an average of 3,5 percent; where the acquired rival had a market
share larger than 5 percent, the average price increase was 12,8 percent. (p. 122)
Vertical integration lowered prices by 4,3 percent. (p. 123) Mergers involving
“third bottlers,” i.e. those not bottling Coca-Cola or Pepsi products, resulted in
price falls of 1,2 percent if their pre-merger market shares exceeded 3,5 percent,
but price increases of 5,5 percent if they had smaller market shares. (ibid.) The
authors did not attempt to estimate the effects of mergers more broadly
throughout the economy.

271 Baker, J. (2003), “The Case for Antitrust Enforcement,” Journal of Economic Perspectives,
17(4), pp. 27-50.
272 Levenstein, M.C. and V.Y. Suslow, “What Determines Cartel Success?” Journal of Economic

Literature 44, pp. 43-95.

273 Connor, J.M. and Y. Bolotova, “Cartel Overcharges: Survey and Meta-Analysis,” unpublished
274 Saltzman, Harold, Roy Levy, and John Hilke. 1999. “Transformation and Continuity: The U.S.

Carbonated Soft Drink Bottling Industry and Antitrust Policy Since 1980.” Bureau of Economics
Staff Report, U.S. Federal Trade Commission. November.
www.ftc.gov/reports/softdrink/softdrink.pdf.

PAGE 132 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Other estimates of price rises resulting from unconsummated mergers have been
made using merger simulation. E.g., Werden 2000 predicted price rises up to 10
percent in some cities for white pan bread. Other estimates of price rises
resulting from mergers, cited in Baker 2003, are Baker and Bresnahan, 1985;
Nevo, 2000; Werden and Froeb, 1994. Peters 2003 found that merger simulation
captured most of actual post-merger price changes in six airline mergers in the
1980s.
Crandall and Winston 2003,275 in contrast to Baker 2003 in the same issue, find
that the antitrust laws have not had a significant effect on consumer welfare in
the United States. With respect to the enforcement of the law against
monopolization, they examined the consumer welfare effects of enforcement in
six “landmark” cases over the past century, and found little benefit from the
imposed remedies. They conclude that, “[N]either policymakers or economists
have yet to offer compelling evidence of marked consumer gains from antitrust
policy toward monopolization.”
The above literature has focused on estimating the social or consumer welfare
losses from monopoly power, but assumed no change in antitrust enforcement.
By contrast, Nold, Block and Sidak (1981) tried to measure the effect of anti-
cartel enforcement on changes in the mark-ups on a specific product. They
calculated the cost of a homogenous type of bread from the standard recipe and
the price of the various ingredients, in various cities. They found that increased
resources to the competition law enforcement agency and successful
prosecutions of bread cartels, which were typically local in scope, significantly
reduced mark-ups in other cities in the same region. But this effect was not found
in the period before class action lawsuits, as follow-ons to government
prosecution, became a credible threat. They concluded that the credible threat of
large damage awards deters high mark-ups.
The literature reviewed so far has taken a rather static and microeconomic
approach to measuring either the effect of monopoly power or the effectiveness of
competition law systems. But other studies have aimed to measure productive or
dynamic efficiency. That is, they aim to estimate the effect of differing
competition law and regulatory systems on total factor productivity. TFP
measures the output of an industry or economy as compared with its inputs.
Examples of studies that have traced the impact of enhanced competition on
productivity and growth are manifold.
OECD (2007) on the relationship between competition and economic
performance, summarizes the most relevant contributions in the economic
literature on large-scale effects of competition. The focus of most analyses is on
productivity and dynamic efficiency. Further to that, a study by the French
Direction Générale Du Trésor et de la Politique Économique – DGTPE (2009)276
also explored the relationship between competition in the markets for goods and
services and economic growth.

275 Crandall, R.W. and C. Winston (2003), “Does Antitrust Policy Improve Consumer Welfare ?
Assessing the Evidence,” Journal of Economic Perspectives, 17(4), pp. 3-26.
276 Bouis R., Klein Caroline, Direction Générale Du Trésor et de la Politique Économique –

DGTPE (2009), Competition and productivity gains: a sectoral analysis in the OECD countries,
No. 51 February 2009.

PAGE 133 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

According to it, ‘it is generally agreed that competition boosts productivity thanks
to “static gains” acting on the level of productivity and to “dynamic gains” that
boost its growth rate’277. Studies that have assessed the impact of competition on
productivity include Ahn (2002) and Nicoletti and Scarpetta (2003, 2005)278,
where productivity increases are categorized as superior productive efficiency
(using current inputs with less slack), better technology selection and innovation.
Aghion and Griffiths (2005)279 have identified the relationship between
competition and innovation, by depicting a U-shaped curve where excessive entry
stifles incentives to invest and exerts a lower impact on innovation. OFT 2007,
which has as a backdrop the productivity gap between the UK and its closest
international competitors (mainly France, Germany and the US), finds that low
productivity may indeed be taken as an indication of lack of competition in the
market.
Alongside with these contributions in the literature, some other studies have
modelled the effect of antitrust policy on efficiency and growth280.
Symeonidis (2003) finds strong evidence of a negative effect of collusion on labor
productivity growth. Labor productivity grew more slowly in collusive industries
than in non-collusive industries before the implementation of the 1956 cartel
legislation. Once cartels became illegal, no significant differences between
collusive and non-collusive industries existed in terms of rates of labor
productivity growth. Broadberry and Crafts (2000) find that price fixing
agreements were widespread prior to the 1956 Restrictive Practices Act and seem
to have had an adverse effect on costs and productivity. Symeonidis (2007)
examines the impact of competition on wages and productivity using a panel data
set of UK manufacturing industries over 1954-1973. Some empirical studies
concentrate on the direct impact of competition on price levels and inflation via
downward pressure on profit margins and mark-ups and changes in the
institutional structure (e.g. Neiss, 2001; Cavelaars, 2003; Przybyla and Roma;
2005). Another stream of literature directly infers the impact of competition on
economic performance and growth by looking at the relationship between
product and/or service market liberalisation and productivity.
For example, Nickell (1996) analyses a dataset of 700 British manufacturing
companies between 1972 and 1986 and finds that a 10% increase in price mark-

277 See for example OECD (2002), "Competition sur les marchés de produits et performance
économique" (Product market competition and economic performance), Perspectives
économiques, no. 72, December, pp. 189-197; European Commission (2004), "The EU Economy:
2004 Review", European Economy, no. 6; and Nicodeme and Sauner-Leroy (2007), "Product
market reforms and productivity: a review of the theoretical and empirical literature on the
transmission channels", Journal of Industry, Competition and Trade, no. 7, pp 53-72.
278 On corporate data: Nickell S. (1996), "Competition and Corporate Performance", Journal of

Political Economy, no. 104, pp 724-746; Disney R., Haskel J. and Heden Y. (2000),
"Restructuring and productivity growth in UK manufacturing", CEPR Discussion Paper no. 2463.
On macroeconomic data: Gordon, R.J. (2004), "Why Was Europe Left at the Station when
America's Productivity Locomotive Departed?", CEPR Discussion Paper, no. 4416; Nicoletti G.
and Scarpetta S. (2005), "Regulation and Economic Performance: Product Market Reforms and
Productivity in the OECD", OECD Economics Department Working Papers, no. 169, OECD
Economics Department. On sectoral data: ECB (2006), "Competition, Productivity and Prices in
the Euro Area Services Sector", Occasional Paper Series, no. 44, April.
279 Scherer F. (1967), "Market Structure and the Employment of Scientists and Engineers",

American Economic Review, no. 57, pp 524-531 had previously found the same.
280 See the survey contained in the study by the OFT, Productivity and Competition, January

2007

PAGE 134 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

ups resulted on average in a 1.3%–1.6% loss in TFP growth. Similar results were
found by Disney, Hasken and Heden (2003)281.
The impact of enhanced competition on dynamic efficiency and
innovation is more ambiguously framed in the literature, and empirical data
are sparse. The actual relationship between competition and innovation is
subject to a long-lasting debate in the economic literature, and is often referred
to the diverging views of Joseph Schumpeter (1942), who considered that more
concentrated markets favor innovation; and Kenneth Arrow (1962), who took the
opposite stance.282
Blundell, Griffith and Van Reenen (1995) investigate the negative effect of
increased market concentration on aggregate innovation by examining 375 firms
listed on London International Stock Exchange between 1972 and 1982.
Furthermore, Griffiths, Harrison and Simpson (2006) show that the introduction
of the single market in Europe increased R&D intensity by 0.9 per cent in UK
metal products industry (further associated with a 0.4 percentage increase in
TFP growth), and that in general the effect of increasing competition on
innovation is larger in countries that are closer to the global technological
frontier.
One of the papers that have looked at the impact of antitrust enforcement on
innovation is Marinova et al. (2005), where the impact of civil antitrust filings by
the DOJ on the level of innovation (measured through patent activity) is found to
have been statistically significant in the US in the period 1953-2000.283

3.4.1 Studies that analyze the effect of antitrust on GDP growth

Baker (2005) observes that “the annual welfare benefits from deterring the
exercise of market power through antitrust laws as they are enforced today could
readily exceed 1% of GDP, or $100 billion per year” (which applied to Europe,
would yield a yearly benefit of €117 billion), whereas the costs for public and
private enforcers are not likely to be greater than $2 billion per year in the
US152. This, of course, would largely guarantee that antitrust enforcement is
meritorious and ensures “value for money” from a social standpoint284.

281 See, i.a., Haskel, 1991; Nickell, 1996; Disney, Haskel and Heden, 2003; CPB, 2006; OECD,
2006; Conway et al., 2006); growth (Rey, 1997; Dutz and Hayri, 2001); or employment (Nicoletti
and Scarpetta, 2005).
282 See, i.a., R. Gilbert, Looking for Mr. Schumpeter: Where Are We in the Competition-

Innovation Debate?, in 6 Innovation Policy And The Economy 159 (Adam B. Jaffe, Josh Lerner &
Scott Stern, eds. 2006).And Baker, J. B., Beyond Schumpeter vs. Arrow: How Antitrust Fosters
Innovation (June 2007). Kauffman Foundation Conference & Seminar Research Paper No. 15.
283 Marinova, McAleer and Slottje, Antitrust Environment and Innovation, Scientometrics, Vol.

64, No. 3 (2005) 301-311.

284 Baker‘s tentative estimate is based on the observation that the traditional ways in which

economists calculate the economy-wide welfare loss from the exercise of market power – mostly
based on social welfare as in the seminal contribution by Harberger (1954) – fail to take into
account a number of factors that magnify the impact of the exercise of market power on the
economy. These include the assumption of unitary elasticity for all sectors of the economy, the
loss of consumer surplus arising from overcharges (the consumer surplus ―rectangle),
dissipation of oligopoly profits through rent-seeking, the reduction of incentives to innovate
absent vibrant competitive pressure, etc. When taking into account some of these factors, Cowling
and Mueller (1978) find that the overall social cost from the exercise of market power can be as
high as 12% of GDP.

PAGE 135 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Further to that, a legal system in which the impact of effective competition policy
on GDP growth has been measured, although with some degree of
approximation, is Australia. This is also a legal system with enhanced private
enforcement if compared with Europe, although private antitrust litigation is not
as developed as in the US. In a recent report, the OECD estimated that effective
competition policy in Australia added 2.5% to GDP, which applied to Europe
would yield a contribution in the order of €396 billion. However, this estimate
encompasses the whole competition policy implementation, which goes well
beyond the enforcement of antitrust rules and most notably includes
deregulation initiatives in a number of infrastructure sectors.

3.4.2 Studies that focus more on the impact of antitrust on

productivity growth

Borell and Tolosa (2008)285 present empirical evidence regarding the effect of
simultaneous antitrust enforcement and openness to international trade on
productivity. Their analysis strongly relies on the assumption that the decision to
enforce antitrust regulations is endogenous, thus they are interested into two
separate effects. First, the selection effect, to what extent do more productive
countries enforce antitrust? Second, the competition effect, how much antitrust
boosts productivity? The authors argue that ignoring the first effect biases the
measurement of the competition effect. They estimate that ignoring the policy
endogeneity leads to an overestimate of 18% on the effect of competition on total
factor productivity.
Buccirossi, Ciari, Duso, Spagnolo and Vitale (2009)286 find a robust positive and
significant effect with respect to the effectiveness of the application of
competition law in general. In order to do so, they estimate the impact of
competition policy on total factor productivity (TFP) growth as measured by
newly created indexes, so called Competition Policy Indicators (CPIs). Those
indexes aggregate institutional and enforcement features that the authors
consider as being key in deterring anti-competitive behaviour287. They find that,
at a disaggregated level, the effect on TFP growth is particularly strong for
specific aspects of competition policy related to its institutional set up and
antitrust activities (as distinct from merger control).
Moreover, this effect is strengthened by a good legal system, suggesting
complementarities between competition policy and the efficiency of law
enforcement institutions.

285 Borrell J.-R., Tolosa M., (2008), Endogenous antitrust: cross-country evidence on the impact
of competition-enhancing policies on productivity, Applied Economics Letters, 15, 827-831.
286 Buccirossi P., Ciari L., Duso T., Spagnolo G.and Vitale C. (2009), Competition Policy and

Productivity Growth : An empirical assessment, SP II 2009 – 12.

287 The authors generate and Aggregate CPI that summarizes all the key features of the

competition policy of a country, as well as more disaggregated ones that refer only to the features
of competition policy relative to specific behaviour (i.e. cartels, other competitive agreements and
abuse of dominance – collectively referred to as ‘antitrust’- and mergers), or only to the
‘institutional’ or the ‘enforcement’ features of a competition policy.

PAGE 136 OF 182

STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

PAGE 137 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

4. AN ASSESSMENT OF THE IMPACT OF NATIONAL RULES

THAT ARE STRICTER THAN ART. 102 TFEU

In this section, we combine the main findings of Section 2 and 3 above, and
discuss the potential impact of the national legal rules on unilateral conduct that
are stricter than Article 102 TFEU. Section 4.1 below identifies the undertakings
that appear to be most affected by these rules, while Section 4.2. assesses the
impact of the different rules. Section 4.3 discusses potential broader economic
impacts, and Section 4.4 concludes.

4.1 Indication of most affected enterprises

Our study has, based on responses, identified three sets of enterprises as
appearing to be more directly affected by stricter national rules on unilateral
conduct than are others. One set consists of those enterprises active in certain
Member States in grocery retailing, and their suppliers. These enterprises,
particularly the multinational grocery retailers, were eager to participate in the
study, suggesting the stricter rules had a significant impact on them. They were
particularly concerned with abuse of economic dependence rules, and sales below
cost rules. The second set consists of suppliers of intermediate products whose
market shares in some markets may imply that they are not in any possible “safe
harbour” with respect to presumptions about dominance.288 They were
concerned with stricter rules on the definition of dominance, and on rules on
abuse of economic dependence. These enterprises participated in the study to a
much lesser degree, so their views may be considered indicative, with need for
further confirmation. The third set consists of enterprises active in network
infrastructure such as electricity, pipeline gas, and telecommunications. They
were concerned with stricter rules aimed at their sectors, such as Section 29 in
the German Act against Restraints of Competition and competition rules within
sector-specific legislation. As with the second set, only a few of these enterprises
participated in the study, so their views, too, may be considered only indicative.
The identification of the affected enterprises used the five main sources of
information for this study. These sources were a business survey, a survey of laws
and their enforcement in each Member State, workshops and interviews of
companies and associations of companies, and existing reports and documents.
The questionnaires used to collect information from national legal experts and
businesses are attached to this report as Annexes I, II and III.
While initially the net was cast quite widely, it soon became apparent that the
vast majority of enterprises chose not to respond to our enquiries, but also that
there were three sets of enterprises that were more likely to respond to our
enquiries and thus we interpreted as being particularly affected by the stricter
national rules on unilateral conduct. First, enterprises or associations of
enterprises involved in retail trade or supply to retail trade, particularly
groceries, in certain Member States had a greater tendency to respond to our
enquiries. Of those who responded, many said that they had been affected by

288An intermediate product is one which is used in the production of other goods, that is, it is not
used by final consumers. Examples include machine tools and mining equipment.

PAGE 138 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

stricter national rules on sales below cost or abuse of economic dependence.

Second, some enterprises active in intermediate goods markets were affected by
stricter definitions of dominance, but not by the other rules that had attracted the
attention of the retailers. Third, some enterprises active in telecommunications
and energy were affected by dominance rules in either the regulatory laws or
specific provision in the competition law. Another group of undertakings, who
might be described as globally active, expressed support for convergence of the
dominance rules within the European Union, but were uninterested in being
contacted for further details.
In addition to the characterization of the sectors in which enterprises were
affected by stricter unilateral conduct rules, our research uncovered a difference
in how the national rules of one Member State affect enterprises´ conduct in
other Member States. The intermediate goods enterprises often had a single
compliance programme, so the national rules in one Member State influence the
enterprises’ conduct globally or at least across Europe. By contrast, those
involved in retail sectors said they adapted their conduct Member State-by-
Member State. There seemed not to be spillover effects of national laws to the
same degree.
These results from the business survey and workshops, and follow-up interviews,
should not be surprising in light of the legal analysis of the stricter national rules
on unilateral conduct. As highlighted below, many of these divergent laws are
specific to retail trade and its supply, or even more specific to retail trade and
supply of food.

4.1.1 The business survey and workshops

Our data has come from businesses by four routes: (i) questionnaires sent to
businesses; (ii) follow-up interviews of businesses; (iii) contacts with
associations, some of which re-broadcast our questionnaire among their
members and others which agreed to be interviewed; and (iv) workshops. Some
material is from websites of associations. We tried to build a representative
sample of businesses to which to send the questionnaires, but we experienced a
low response rate (approximately 5% out of a sample of more than 1,100
contacted undertakings) except among large grocery retailers289. In addition,

289 Both the numerator and the denominator can be criticized. Indeed, the figure of 5% is an
example of quantifying a concept better expressed qualitatively as “a low response rate.” The unit
of account is, in principle, an undertaking. But from the inception of the project, it was recognized
that SMEs would likely not have the resources to respond and that associations thereof would be
more likely to be equipped to respond. While assigning an association the weight corresponding
to the number of its members would give interviews of individual undertakings almost no weight,
assigning an association the same weight as an individual undertaking understates the weight of,
in particular in this case, SMEs. Between Scylla and Charybdis, the choice was made to give an
association a weight of one. In total, the researchers had interviews, or a questionnaire response,
or an email in response to a questionnaire, or participation at a workshop, from 55 to 60
undertakings (predominantly) and associations. For the denominator, the same issues with
respect to the weight of associations applies; in addition, the number of the membership of the
Confederation of Swedish Industry and of Assonime, respectively, contacted had to be estimated
(in the event, at 50 and 100 respectively). Among large grocery retailers, 6 were interviewed; all
but one are active across several Member States. Among suppliers to large grocery retailers, a
Europe-wide association provided an interview, as did a national association and a multinational
undertaking.

PAGE 139 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

biases were introduced into the sample since associations necessarily have sector
biases and some had geographic biases, and the workshops necessarily had
geographic biases. Overall, the relatively high response rate from large grocery
retailers—six out of arguably eight, and a related association--and, to a lesser
degree, food suppliers to grocery retailers—two associations with broad
representation, and a large multinational--suggests that the results generated by
those interviews and questionnaire responses are fairly robust. The low response
rate from most of the remaining population of millions of enterprises suggests
that those results should be treated as general indications of enterprises´
experiences which would need to be further substantiated or verified.
A questionnaire was initially sent to several hundred enterprises active across a
broad range of sectors and almost every Member States, and invitations to
workshops at two national associations, in Sweden and Italy, were issued to
broad swathes of the membership of those associations.290 A third workshop, at
the Centre for European Policy Studies, attracted more trans-European or
internationally active participants, totalling eight. Following a disappointing
response to the business survey, contact was made with a range of associations
representing enterprises or associations of enterprises, not only to re-broadcast
the questionnaire but also to interview directly. While the researchers attempted
to contact tens of associations, predominately trans-European, requests for
interviews were granted only five to ten times.
The set of enterprises to which the questionnaire was distributed was built
through a variety of methods. We first contacted the European Business Test
Panel, launched jointly by Member States and DG Internal Market and Services
and designed by Eurostat to be representative of EU businesses with respect to
size, sector and geography within EFTA. After getting data from the EBTP´s
Belgian national contact, we were dissuaded from using this mechanism and,
indeed, the website has subsequently become substantially less informative. We
then gathered names and contact information for businesses from the websites of
Member States’ ministries of commerce and national chambers of commerce and
other websites indicating the main businesses operating in the EU. The bias
introduced thereby is not known, but one can suppose that smaller firms are
under-represented as small firms would tend not incur the fixed costs of
membership in chambers of commerce. The websites had different degrees of
informativeness, so some Member States are clearly over-represented, as can be
seen from Table 2 below. To the resulting list, we ensured inclusion of those
companies that were publicly known to have recently been consulted by DG

290 The workshops were held in Stockholm in cooperation with the Confederation of Swedish
Enterprises, in Rome in cooperation with Assonime, the Association of Joint Stock Companies,
and in Brussels at the Centre for European Policy Studies. The Confederation of Swedish
Enterprises is Sweden’s largest and most influential business federation representing 50 member
organizations and 60,000 member companies with over 1.6 million employees. The
Confederation sent invitation documents and questionnaires to its legal panel, consisting mainly
of the general counsels of large listed companies, as well as ICC Sweden, and the website of the
In-House Lawyers Association. Nine companies participated at the Stockholm workshop.
Assonime , represents 600 Italian companies, of which 100 are listed in the stock exchange, thus
their membership is mainly medium and large companies operating in all sectors. The invitation
and questionnaires were initially sent only to members of the Assonime Competition Network,
but later an invitation was sent to all members. Twenty-one companies participated at the Rome
workshop. The Brussels workshop was attended by representatives of eight companies.

PAGE 140 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Competition. If the latter set of businesses has a higher rate of responsiveness,

then this biases the responses. At this stage, the sample totalled 816 firms.
Table 2. Sample of EU, Norway and Switzerland businesses

Final N. Firms in
NACE Divisions* Population
A B C D E F G H I J K L M N O P Q R S T U
→ Orig **
COUNTRY ↓ inal (Thousands)
Austria 1 2 1 1 5 1 287
Belgium 1 2 2 1 8 7 2 1 2 4 27 27 425
1 1 1 6
0
Bulgaria 6 2 8 7 258
Cyprus 1 1 0 44
Czech Republic 1 2 1 1 4 6 1 67 64 882
6 0
Denmark 1 5 1 1 8 4 212
Estonia 1 1 7 1 3 13 15 48
Finland 1 1 4 1 1 9 8 1 38 34 214
3
France 3 2 1 5 3 5 3 9 2 1 6 1 61 32 2569
2
Germany 1 3 2 1 1 4 3 1 1 1 2 70 45 1819
5 2 7
Greece 0 1 830
Hungary 0 0 589
Ireland 2 1 3 1 96
Italy 3 9 2 1 15 7 3906
Latvia 9 3 5 17 16 71
Lithuania 1 8 3 24 22 139
3
Luxembourg 1 1 0 24
Malta 5 5 5 0
Netherlands 2 1 9 1 2 1 5 21 15 540
Poland 1 1 1 3 2 1484
Portugal 1 1 6 1 21 18 867
3
Romania 6 4 1 1 1 2 1 1 3 34 14 474
5
Slovakia 5 3 3 1 1 23 9 60
1
Slovenia 0 0 100
Spain 1 2 3 3 1 2 12 16 2712
Sweden 1 3 1 1 6 4 561
United 16 1 2 1 2 5 3 1 5 6 7 1 68 29 1671
Kingdom 0
Canada 0 N/A***
United States 0 N/A***
Japan 0 N/A***
Switzerland 5 2 1 2 10 7 N/A***
Norway 1 1 1 3 6 254
TOTAL (EU) 27 23 3 4 1 1 1 2 6 7 6 1 2 2 0 0 2 1 1 0 0 80
4 6 4 3 3 0 8 5 6 0
3 2
TOTAL (ALL) 27 23 3 4 1 1 1 2 6 7 6 1 3 2 0 0 2 1 1 0 0 81 396**
4 7 5 3 3 2 9 0 6 6 **
9 2
(*) List of NACE divisions: A – Agriculture, Forestry and fishing;, B – Mining and Quarrying; C-
Manufacturing; D-Electricity, Gas, Steam and Air conditioning supply; E- Water supply, Sewerage, Waste
management and remediation activities; F-Construction; G-Wholesale and retail trade, repair of motor
vehicles and motorcycles; H- Transporting and Storage; I-Accommodation and food service activities; L-
Real estate activities; M-Professional, scientific and technical activities; N-Administrative and support
service activities; O-Public administration and defence, Compulsory social security; P-Education; Q-
Human health and social work activities; R-Arts, Entertainment and recreation; S-Other services
activities; T-Activities of household as employers, undifferentiated goods – and services – producing
activities of households for own use; U-Activities of extraterritorial organizations and bodies.
(**) Population of firms obtained by adding up number of firms that were active in 2007 (latest year for which
the observations are available) and as available in Eurostat (edb_all) in the following NACE divisions:
(B), (C), (D), (F), (G), (I), H+J), Real estate, renting and business activities (L+M).
(***) Data not available in Eurostat
(****) Includes one business characterized as France/Europe.

PAGE 141 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

An initial attempt to contact the sample resulted in significant non-response. In a

number of instances, the contact information was not correct. Although attempts
were made to update contacts, they largely failed. In other instances, the contacts
replied that they were disinterested or not concerned with the topic, or could not
respond due to a lack of internal capacity. Further investigation of some of these
contacts found that they were small, or indeed micro, enterprises. We decided
that the best way of getting a thoughtful view from them was through the
associations that represent their interests. As a result of these non-responses or
negative responses, our original sample shrank to a size of 396, as shown in a
column in the table above.
To this sample, we added 78 companies that have liaison offices in Brussels,
whether European or non-European. We felt that these companies would be
more likely to operate cross-border in the internal market, thus have a basis for
comparing different national laws, and be more likely to respond to the
questionnaire as they had already incurred expenses to participate in policy
formation. The geographic and sectoral coverage of these additional firms is
shown in Table 3. These additions to the sample likely further biased the sample
towards larger firms. In addition, 48 internet retailers and two umbrella
organizations for businesses engaged in e-trade were contacted but none agreed
to respond to the questionnaire.

PAGE 142 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Table 3 – Businesses with Brussels Liaison Offices added to sample

BRUSSELS LIASON OFFICES
NACE Divisions → A B C D E F G H I J K L M N O P Q R S T U
Tot.
COUNTRY ↓
Austria 1 1
Belgium 2 2 3 7
Bulgaria 0
Cyprus 0
Czech Republic 0
Denmark 1 1
Estonia 0
Finland 2 1 3
France 1 5 1 1 1 1 10
Germany 2 1 3 2 1 9
Greece 0
Hungary 0
Ireland 0
Italy 1 2 1 3 2 9
Latvia 0
Lithuania 0
Luxembourg 0
Malta 0
Netherlands 1 1 2
Poland 1 1
Portugal 0
Romania 0
Slovakia 0
Slovenia 0
Spain 1 1
Sweden 2 2
United Kingdom 1 1 2
Canada 1 1 2
United States 1 10 1 3 1 5 1 1 23
Japan 3 3
Switzerland 1 1
Norway 1 1
TOTAL (EU) 1 1 13 3 0 0 0 7 0 13 7 0 0 1 2 0 0 0 0 0 0 48
TOTAL (ALL) 2 2 27 3 0 0 1 10 1 19 8 0 0 1 2 0 0 0 2 0 0 78

We sent the questionnaire to this set of more than 400 businesses. In addition,
we contacted business associations representing various kinds of businesses, e.g.,
large companies, SMEs, retailers, suppliers, and industry associations
themselves, at both the national and European levels. The rationale was to
broaden our sample, by catching for example some of the actors like SMEs that
we could not directly reach with our first approach. We asked the associations to
further distribute our questionnaire to their members. But it is difficult to
estimate the additional contribution of this channel of diffusion of our
questionnaire. The response rate, in the end, was extremely low (approx. 5%).
We supplemented the business survey with interviewing those companies and
associations that either indicated an interest by responding to the questionnaire
or that we thought would be interested in the study. We conducted
predominantly telephone interviews. These interviews provided us with more
nuanced information about the effect of national rules on enterprises´ conduct.
However, they necessarily imply a limited number of interviews. Ignoring follow-
up interviews and short conversations at workshops, these totalled 25 to 30
entities, both undertakings and associations.
In addition to the channels described above, we made use of CEPS list of
corporate members. The CEPS Corporate Programme specifically concentrates

PAGE 143 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

on the interface between EU policies and business strategies. CEPS currently has
126 Corporate Members that operate in various EU members states and sectors.
Out of those, 28 companies were selected as being potentially affected by the
national rules on unilateral conduct (due to the fact that they are commercial
firms with operations in different countries, in sectors that are sufficiently
concentrated); we then contacted those companies directly. We believed that this
was a way to further increase our rate of responsiveness on the basis of the
professional trust existing between CEPS and its corporate members.
The low rate of response to our business survey can have several causes: Two of
these—possibly sending the questionnaire to inactive addresses, or addressees
confusing the questionnaire with spam or the like—should have been reduced by
repeated follow-up. Two other possible causes would suggest that the divergent
rules are not particularly costly: Enterprises may not be affected by the laws in
question, or enterprises may consider the costs of answering outweighed the
benefits. When considering whether to respond, one would assume that
enterprises, or the government affairs unit therein, would compare the
incremental costs and the incremental benefits of responding. Such costs include
opportunity costs, including addressing other issues and addressing the same
issues through other channels, such as a business association. For one
association, responding to the questionnaire would require a board meeting,
which its representative told the researchers it felt would be too costly. Some
enterprises and associations expressed to the researchers concern about
anonymity; it is conceivable that some non-respondents had a yet greater
concern in that respect and considered the risk to be too high even to respond to
the researchers. In other words, there are a number of potential costs that might
give enterprises and associations incentives to not respond to the questionnaire
or decline a request for a telephone interview.
Several responses point towards costs outweighing benefits of responding. One
participant at a workshop, encouraging others to take a more active part and to
recruit others to be interviewed, explained that, in the past, her company had not
seen the value of participating in consultations or workshops, but had recently
changed its opinion. Thus, in the past, its point of view would not have been
represented. Another addressee, an association of enterprises involved in retail
trade or its supply, explained that its process to prepare a response was quite
heavy, involving broad consultations and meetings of its board of directors, so it
would not be able to respond in a timely manner. Other addressees, who were
associations for suppliers to retailers, explained that their members would not
talk with us for fear of leaks and, ultimately, retribution by those on whom they
were economically dependent. A few of the very brief responses we received
made statements in favour of convergence but expressed no wish for follow-up.
We organized workshops in three different Member States. The first workshop
was held in Stockholm on 1 October in cooperation with the Confederation of
Swedish Enterprises, the second in Rome on 6 October in cooperation with
Assonime, the Association of Joint Stock Companies, and the third workshop in
Brussels on 28 October 2010.291 A final workshop was organized again in
Brussels on 29 November 2011.

291The workshop in Stockholm took place on the 1st October 2010, at the premises of the
Confederation of Swedish Enterprises. Ms. Widegren, the head of the international function at

PAGE 144 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Individual enterprises that responded to the questionnaire or that we

interviewed, or both, totalled approximately 20 enterprises. Another 25
enterprises made oral comments at the workshops — attended by a total of about
fifty — that were inputs into our analysis and reporting. Between 5 and 10
associations were interviewed by us.
The enterprises responding include a very large fraction (five) of the large grocery
chains headquartered in EU15 countries that entered the markets in the 2004
and 2007 accession countries, some of which also operate outside the EU, and
one large grocery chain that operates in only one Member State. They include six
of the largest companies headquartered in the EU (one is headquartered abroad)
that engage in high technology manufacturing, and in some cases consumers
products, across a range of products and worldwide. And they include four
companies providing telecommunications or energy in a number of EU Member
States, most but not all of which are historical incumbents. A handful of
approximately 4-5 companies do not fall into any of the above categories, as they
include a large predominantly foreign i.e., outside the European Union, service
provider, a large and a small food producer, and a non-grocery retailer operating
worldwide. As noted above, we had to rely on associations to present the views of
SMEs, and the associations also presented the views, both nationally and EU-
wide, of distributors/retailers and food suppliers. The workshops provided
additional information from, predominantly, intermediate goods manufacturers
and distributors/retailers of consumer products, totalling about forty. Note,
however, that participants were not compelled to speak.
The different response rates mean that different conclusions can be granted
different degrees of significance. Those regarding grocery-specific rules in the
2004-2007 accession countries are well-founded, as the response rate was very
high. Those regarding intermediate goods and infrastructure sectors are based
on a much lower response rate, so have less significance and can be regarded as
general indications.

the Swedish Competition Authority presented the case of Sweden as a case of substantial
convergence with the art. 102 TFEU. Moreover, Professor Ulf Bernitz (Department of Law,
Stockholm University) offered an overview of rules on unilateral conducts that have been applied
in Sweden since the entry into force of the Competition Act in 1993 while explaining the
characteristics of the Swedish very concentrated market. An open debate with the firms
participating to the event (in total 9) followed and concluded the event.
The workshop in Rome took place on the 6th October 2010, at the premises of ASSONIME
(associazione fra le società italiane per azioni). Also in this case we welcomed the participation of
one of the national experts at the Italian Competition Authority, Andrea Pezzoli (Direttore della
Direzione Agroalimentare e Trasporti) who explained how the Italian system has conformed with
EU competition law since the entry into force of its first competition law in 1990 (i.e. Legge
287/1990). Further to that, both Prof Roberto Pardolesi (Prof of Private Law at LUISS University,
Rome) and Prof. Cristoforo Osti (Prof. at University of Salento and LUISS University) added to
the discussion by explaining the cases of abuse of dominance and abuse of economic dependence
to see how the current jurisdiction applies to the firms, in the Italian case. Finally, an open debate
with the 21 firms participating to the workshop followed and concluded the event.
The first workshop in Brussels took place on 28 October 2010 at the premises of CEPS,
Centre for European Policy Studies. Presentation of preliminary findings both with respect to the
law and economics of the study were made by the authors. A general discussion among the eight
representatives of the firms and associations ensued.
The second workshop in Brussels took place on 29 November 2011, shortly before the
delivery of this Final Report. The discussion among the approximately twenty participants
provided some interesting last-minute insights for the drafting of this final report.

PAGE 145 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The legal and economic analyses indicated that the effects of non-convergence in
the abuse of dominance rules in the European Union would be small in
comparison with GDP and would not be felt equally by all firms. The effects of
any competition law and institutional framework are, as reported above, small,
and it must be assumed that the effect of an incremental change in competition
rules even smaller. And only a subset of firms would be affected by divergent
rules—those on the cusp of dominance per Article 102 TFEU in at least one
market or, predominantly, those involved in distribution. It is established
practice that where the subject of interest is not widespread, a survey should
focus on those populations where the subject of interest is most concentrated.
See, for example, Deaton 1997292 where he writes that:
“The purposes of the survey sometimes dictate that some groups be more
intensively sampled than others, and more often that coverage be
guaranteed for some groups. There may be an interest in investigating a
"target" group that is of particular concern, and, if members of the group
are relatively rare in the population as a whole, a simple random sample
is unlikely to include enough group members to permit analysis. Instead,
the sample is designed so that households with the relevant characteristic
have a high probability of being selected.”
He goes on to illustrate with a survey of HIV positive households in Tanzania
where, since the overall incidence of infection was low, the survey was confined
to areas where infection was known to be high.
Using prior knowledge of the population being studied to divide the population
into non-overlapping groups, i.e., to stratify, improves the efficiency of statistical
inference.
In the event, we can consider the survey to have several subpopulations, large
grocery retailers, other retailers, suppliers to grocery retailers, large network
utilities, and all other firms. The key feature is that the first subpopulation is
small, on the order of twenty, and the number of large utilities is small, too, on
the order of tens or hundreds. The other populations number in the millions,
with some 21 million firms active in the European Union, of which 3.6 million are
retailers.293
Focusing on the largest subpopulation, where the least effect of divergence would
be expected, consider the statistical significance of a single response among the
1100 firms and associations surveyed, saying that it had been affected. This is a
proportion of 0.001 if we count one association as representing one firm, and this
is the most likely estimate. Continuing the one firm association assumption, a
one-sided test to find the largest true proportion of the population that would

292 Deaton, A. 1997. The Analysis of Household Surveys. Baltimore: The Johns Hopkins
University Press.
293 The following might be considered as large grocery retailers: Ahold, Aldi, Asda (Walmart),

Auchan, Carrefour, Casino, Coop, Delhaize, Kaiser’s - Tengelmann, Lidl (Kaufland), Tesco, Edeka,
Leclerc, Intermarche, Mercadona, Metro, Migros, Morrisons, Rewe, Sainsbury’s, Systeme U,
Number of retailers overall is from Eurostat figures for 2009, source sbs_na:dt_r2. Number of
firms in wired and wireless telecommunications is about 15 000; Eurostat sbs_na_1a_se_r2.
And 36 000 in electricity, gas and water supply per Eurostat sbs_na_2a_el. “Large” utilities
would be a subset. Number of firms is from Table 1.1 in Key figures on European Business,
Eurostat 2011. at http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-ET-11-001/EN/KS-
ET-11-001-EN.PDF.

PAGE 146 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

yield such a sample 19 out of 20 times it is sampled finds that the highest true
proportion of firms that have a view on convergence and that would generate up
to one response is 0.43%, at a 95% confidence level. If the sample included 50
associations each representing on average 100 firms, the point estimate is
0.00016 and the one-sided 95% confidence proportion is 0.0775%.294

4.1.2 Affected enterprises

The legal analysis, reported in Section 2 above, identified the national laws that
contained stricter national rules on unilateral conduct. It also provided the
political background to the adoption of several of the laws. The laws were often
aimed at the relationship between retail traders and their suppliers; indeed, often
the focus was yet tighter, on retailers and suppliers of agricultural products and
food. In addition, there were some provisions aimed at network infrastructure
sectors such as electricity, pipeline gas, and telecommunications. Finally, many
Member States have laws against unfair competition, which apply to all sectors.
As a consequence of these factors taken altogether—the laws, the survey and
workshop responses, and interviews—we concluded that three sets of enterprises
seem more directly impacted by stricter national rules on unilateral conduct.
 One set are those enterprises active in grocery retailing and their suppliers in
those Member States that have stricter rules.
 The second set consists of suppliers of intermediate products where scale or
scope economies imply that they may reach the market share thresholds for
an investigation as to whether they may be dominant in a market or to be in a
position of “relative” power where abuse of economic dependence type rules
would apply.
 The third set of impacted enterprises is composed by those active in network
infrastructure sectors such as electricity, pipeline gas, and
telecommunications.
The impact of a possible convergence of national competition rules on unilateral
conduct on the third set of enterprises—involved in network infrastructure—is
not further examined below. As explained above, national economic regulation
can easily expand to displace national competition rules for these sectors, and in
any case convergence processes for economic regulation are underway.

294 A binomial distribution is used to estimate the sample properties. That is, we assumed there
were two types of firms, those that were sufficiently affected to respond, and those that were not.
If the true proportion of “affected” firms is p, and we make n enquiries, then a standard formula—
the binomial distribution—gives the probability that we will get 0, 1, 2, etc. number of “affected”
firms in our n enquiries. How would an increase in the number of firms surveyed affect the
result? The variance of the sample proportion is given by the formula, p(1-p)/n, where p is the
true proportion and n is the number of observations. Since all indications—the legal analysis, the
survey of economic studies, and the results of the existing survey—are that p is very small, the
variance is also small. (The variance is largest where p=0.5.) Additional observations would
reduce the variance, e.g., double the number of observations would half the variance. For
example, if 2200 firms were sampled and 2 replied that they were affected, then the one-sided
95% confidence limit would be 0.285% rather than the 0.43% above. But if, instead, that 2200
sample yielded 4 positive answers, then the one-sided test limit would be 0.415%.

PAGE 147 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Having identified the sets of enterprises likely to be most directly impacted by

stricter national competition rules on unilateral conduct, the impacts are
examined in the next subsections.

4.2 Impact of stricter national rules

4.2.1 Stricter definitions of dominance

The effect of stricter definitions of dominance in national rules is different from

those divergent rules aimed at specific sectors. The definitions apply throughout
an economy, potentially affecting a greater number and variety of enterprises.
The enterprises responding on this issue in the business survey and in workshops
can be characterized as makers of technologically sophisticated intermediate
products. Due to scale or scope economies, these enterprises may at some point
in their lifecycle approach the market share thresholds that trigger the legal
presumptions for dominance.
One impact of divergent legal standards is generated by companies active in the
various jurisdictions having to cope with divergence, irrespective of the content
of the divergence. The economics of product standards, explored over the past
quarter-century, can inform the wider context of this study. The TOR of this
study asked, in part, whether the benefits of adopting a particular single
standard, Article 102 TFEU, outweighed the benefits of having multiple
standards. From a broader standards perspective, relevant questions would also
include whether any single standard would provide higher welfare than multiple
standards, which single standard today provides the highest welfare, whether
new standards would emerge in response to, for example, new economics
learning or new welfare perspectives on unilateral conduct, and what paths lead
to a single standard from multiple standards. In the context of this study, all
companies must comply with Article 102 TFEU; the issue is whether they should
also be made to comply with additional standards if they are active in certain
Member States.
Different but well-defined standards have both costs and benefits as compared
with a single standard. Among the costs are those direct costs incurred by
business to conform to several standards rather than with one. In this particular
context, the costs of conforming to more than one standard include modifying
antitrust compliance programmes and providing jurisdiction-specific internal
and external legal advice. Once that advice is provided, additional costs include
that of modifying contracts and changing commercial practices from those used
to conform to a different standard, and possibly of adjusting other contracts,
commercial practices, accounting systems, and training of buyers to take into
account (mitigate) the effects of the first-round modifications. Each of these
types of costs was cited by at least one interviewee out of 25-30. However, no
interviewee stated that it had avoided or exited from any Member State as a
consequence of the divergent legal standards. Hence, among those we
interviewed, the cost of compliance was not so large as to outweigh the profits
from remaining in those markets.
There are also indirect costs of multiple but well-defined standards in the
different jurisdictions. These are the costs of foregone actions, actions that are

PAGE 148 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

within the legal rules of a jurisdiction but which are not undertaken because the
company has chosen to conform to a stricter standard. These are “business
chilling.”
On the other hand, there are significant benefits to a single, well-defined
standard across multiple jurisdictions. Those enterprises with different internal
guidelines, for example, their antitrust compliance programmes or their standard
contracts, for different jurisdictions would be able to standardize them across
their European operations. More significant, given that many enterprises have a
single compliance programme within Europe is that, with a single standard, they
do not risk refraining from conduct that would be profitable and legal in a given
jurisdictions because their internal guidelines are geared to a different standard.
In principle, a single standard would reduce uncertainty about the standard as
decisions and guidelines would cumulate more rapidly.
The OECD policy roundtable on “Guidance to business on monopolisation and
abuse of dominance”295 provided some examples where uncertainty as to the
legal standards for dominance and for abusive conduct led to the chilling of
business conduct. Some representatives of businesses based in the European
Union stated that they must take into account European rules on unilateral
conduct in their day-to-day business even though they do not perceive
themselves as enjoying significant market power. One participant provided a
practical example, where the business he represented, GE, considered suggesting
to a customer, who had organized a “winner take all” procurement, to invite a
second participant so as to avoid his firm exceeding a certain market share
threshold. Such a move would have deprived the customer of cost-savings from
the efficiencies of a single-source policy. (pp.85-86) A different participant,
representing Michelin, said that it employed less effective supply chain
management techniques (not asking dealers for an annual supply commitment),
and did not intervene to prevent rivals from piggy-backing on certain research, so
as to avoid a risk of infringement of the dominance abuse rules. Due to the
uncertainty as to the legal standards for a finding of dominance, the management
decided that the firm should behave as if it were subject to the dominance rules
when it reached a 30% share in the given market. (p. 86)
We did not uncover additional specific instances of business chilling. However,
we did find evidence of general business chilling. In particular, companies adjust
compliance programs in Europe to conform to the strictest national rules. In
other words, they do not engage in profitable, legal conduct in some jurisdictions
because their compliance programs are “tuned” to a stricter national rule.
 The representative of one intermediate goods producer reported that it has a
Europe-wide compliance programme to conform to the strictest rules. As a
result, the company does not engage in conduct which would have been lawful
in some countries. Upon questioning, the representative said that Germany is
the obvious Member State for having the strictest rules.296
 The representative of another such company reported that it has one
compliance programme for its activities globally, except for its activities in the

295 OECD (2007), “Guidance to Business on Monopolization/Abuse of Dominance,”

DAF/COMP(2007)43, available at www.oecd.org.
296 In the assessment of our team, each of these intermediate goods manufacturers operates in

concentrated sectors and might be considered dominant in some narrowly defined markets.

PAGE 149 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

United States, which is a large market for its products, where it has a
compliance programme adapted to the looser standards there.
 The representative of another such company reported that it has five
compliance programmes, each aimed to conform to the laws in, respectively,
the European Union, the United States, India, China and Russia. In Europe, it
operates on the basis that the market share presumption for dominance in
Germany is 33%. In addition, when a particularly important opportunity
presents itself, it will re-examine its compliance programme to see whether it
could be modified. The representative reported that, because it is not active in
any regulated sector or in any business-to-consumer markets, but rather in
Europe is engaged in cross-border transactions, it interpreted that (wrongly)
as ensuring coverage by EU rather than national laws. Thus, it felt that
divergences of national competition rules are not of concern. It is, however,
interested in greater international convergence.
 A major manufacturer of hi-tech products observed that there is no
“practical” justification for maintaining the current fragmentation of rules,
which increases compliance costs for companies wishing to operate in more
than one EU member state.
Intermediate goods producers whom we contacted tended to have internal
antitrust compliance programmes. Compliance programmes aim to prevent
conduct inconsistent with the law and to detect any such conduct at an early
stage, while the damage is still small.
Retailers and their suppliers appear to attempt to conform to the national laws in
each individual jurisdiction. A discussion of the retailers’ compliance costs is
provided below in section 4.2.1.2. One branded supplier to grocery retailers
throughout Europe said that the costs incurred when entering a new jurisdiction
within the EU include legal compliance costs for trading, etc. laws and the costs
of conforming with local traditions, such as whether listing fees were prohibited
or not.
The difference in the practices of retailers from those of intermediate product
makers is significant. The representative of one retailer speculated that retailers
have lower margins and so are unlikely to leave money on the table in terms of
refraining from engaging in lawful, profitable activities. This is unlikely to be the
full story, however, since it begs the question as to why it is profitable for
intermediate goods makers to “leave money on the table.” The answer may be
that retailers must conform to a plethora of national laws, rules and customs and
that there are scope economies in operating a national legal office. If one must
ensure conformity to national hygiene, land use, and labour laws, to name a few,
is the incremental cost of conforming to national competition and unfair
competition laws lower than the standalone cost?
The benefits of multiple well-defined standards include better matching, thus
better outcomes, and experience, thus mutual learning. First, different standards
may provide a better outcome by providing better matches. A different standard
may better match the objectives of a polity, or better match the market features
present in a jurisdiction. Polities differ in the relative weights they place on the
existence of SMEs (small- and medium-sized enterprises), for example, or on
consumer versus total welfare, or on static versus dynamic efficiency, or market
integration. Regarding differences among economies, as noted in the review of

PAGE 150 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

the economic literature, it has been asserted that total welfare losses from an
incrementally stricter dominance regime is higher in a more dynamic economy
than in a more static economy. More generally, however, there is no consensus
among economists as to where to draw the line on prohibited unilateral conduct
to result in the highest welfare. This leads to the second benefit of multiple well-
defined standards. Non-uniformity of standards can aid learning about the
effects of different standards and therefore the design of better standards. Of
course, there is no benefit if standards do not change in response.
Poorly defined standards, on the other hand, are costly. Several interviewees
complained about laws that were poorly drafted, poorly thought through, and
changed frequently with little lead time before they entered into force. As
reported elsewhere, each shift in the rules imposes direct costs of tens of
thousands of Euros, not counting ongoing opportunity costs of compliance.
In addition, multiple standards may in some instances represent an inadvertent
outcome: for example, it is possible that politicians and officials in one Member
State model their national rules on those in another Member State. That is, they
may seek to conform with a different national standard even if the local legal
system and system of consultation with stakeholders is different, thus producing
a different outcome.297
Convergence of stricter definitions of dominance in national competition law
with Article 102 TFEU would, in the opinion of the enterprises likely to be
affected, have a positive effect but this is difficult to calculate. Interviews with
intermediate product makers found that they aim to comply with what they
perceive to be the strictest national competition law in Europe. Convergence to
Article 102 TFEU was seen by these enterprises as beneficial. They cited reduced
business chilling, that is, refraining from engaging in profitable but legal conduct
as a result of complying with a stricter standard, but none would hazard an
estimate of the size of the effect.
The national law which was repeatedly cited by interviewed companies as
constraining, the German Act on Restraints of Competition, establishes in its text
a legal presumption, which is rebuttable. It may be the case that the actual
standards applied in the final dispositions, taking into account the types of
arguments and evidence that is used to make a determination, are not detectably
and predictably different. The comments of enterprises suggest that,
nevertheless, the legal presumptions have an effect on their conduct.
An additional benefit of convergence, savings in compliance costs from
establishing a single standard, would be small or absent in those enterprises that
have already implemented a single compliance programme for their operations
in Europe (costs for compliance are not necessarily one off-new case law, legal
provisions etc). There may, however, be enterprises for which this would be a
cost-saving.

297One interviewee, a participant in the grocery wholesaling and retailing sector, suggested that
the laws in the eastern European Member States on retailing were inspired by French laws. As
evidence for this view, he said that officials in informal discussions had used as arguments for the
new rules that they “already existed in France.”

PAGE 151 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

4.2.2 Rules on abuse of economic dependence

Firms in weaker negotiating positions, particularly small and medium-sized

enterprises, are the intended beneficiaries of the stricter rules on unilateral
conduct that prohibit abuse of “economic dependence” or “superior bargaining
power” (henceforth, “AED”). Although typical examples are suppliers to large
grocery retailing chains, some laws apply in any sector of the economy. Indeed,
while the overwhelming majority of complaints about the restrictions of AED
rules on conduct were from grocery retailers, also intermediate products
manufacturers complained that these rules meant they, too, constrained their
conduct in a way that meant they could not apply similar practices across Europe.

4.2.4.1. Views of market players

Collecting information about the incidence of AED from the enterprises likely to
be affected by the practices proved to be difficult. Enterprises likely to be affected
by AED are loath to complain since they are, by definition, economically
dependent on those imposing the harm. Even the representative of a large,
multinational, publicly traded supplier to grocery retailers characterized the
company’s position as dependent since, in any given geographic area it sold to
only two or three retailers. The company, said the representative, would not go to
court even with good cause because it feared retaliation by retailers. Others,
representing associations of enterprises that supply to grocery retailers in certain
Member States, also said that their members, even their associations acting on
behalf of members, would not complain of AED because of fear of retaliation.
An example that illustrates conduct of a sort that might be prohibited under
abuse of economic dependence legislation was provided by one contact. During
2010 a grocery chain sent a letter to its suppliers. It required the suppliers to
report information that could facilitate a cartel or tacit collusion among grocery
chains. In particular, each supplier was to inform the chain of all other large and
medium-sized grocery chains it supplied in three Member States. The
information was to be provided within a specified few days of the receipt of the
letter and, in the future, any new clients should be reported at least 12 weeks
prior to commencement of supply. If this information requirement is not met,
then the chain will impose a fee equal to 12 weeks of revenue from the chain. The
letter prefaces the information requirement with a statement to the effect that it
is important for the chain to remain perfectly informed as to the evolution of the
market and consequently of the evolution of each supplier’s portfolio of clients.
Reliability of supply did not seem to be a motivating factor, both since it was not
mentioned in the prefacing comments and because a separate clause elsewhere in
the document describes the penalties imposed if a supplier fails to deliver the
contracted volumes.
The range of other types of abuses of market power by retailers with respect to
suppliers is illustrated by proceedings against several retailers by the Hungarian
Competition Authority (GVH) under the Act on Trade (see also above, Section
2.2.1). These proceedings298 ended with commitments by the retailers to change
their contracts with suppliers in specified ways. One committed to not inserting

298Cases numbered Vj-91/2008, Vj-93/2008 and Vj-94/2008, and involving Provera Beszerzési
(Purchasing) Kft and its partners, Auchan Magyarország Kft., and Metro Kereskedelmi Kft.

PAGE 152 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

provisions regarding exclusive promotion campaigns in supplier contracts for

2009 and to use uniform letter-sizes in its contracts. The second committed to
ensuring that its supplier contracts for 2009 would allow suppliers access to
information about the stock and the extent to which their products have been
sold. The third committed to cancelling clauses its supplier contracts related to
exclusive promotion campaigns, its right to return goods to vendors without time
or quantity limits, and to the reimbursement by the supplier of discount losses
resulting from the change of suppliers or supplier programs.
A 2010 report299 by the Spanish Competition Authority on the agri-food industry
found that abuse of suppliers by retailers takes place and made some proposals to
reduce the incidence.300
The country reports at Annex VI, part 2.6 describe cases of abuse of economic
dependence or abuse of superior bargaining position in Germany. One of these,
partly upheld by the Federal Court of Justice, concerned a large grocery retailer
retroactively adjusting conditions at the moment of acquiring another trading
company.
There can be a public interest in abuses of economic dependence in the supplier-
retailer relationship. Economic theory points to no effect on consumers when
retailers have no market power: The outcome of the negotiation between retailers
and their suppliers simply shifts rent. But some of the conditions imposed on
economically dependent enterprises clearly have at least the potential to dampen
competition. The example cited above, of a retailer requiring suppliers to provide
advance market information, can easily dampen competition.301 Other conditions
imposed on economically dependent enterprises would dampen innovation, such
as when retailers are able to appropriate suppliers’ improvements in products, a
concern cited both by a supplier interviewed by us, and by MEP Sharma at the
above-cited conference.
A study to evaluate the effect of the Czech Law no. 395/2009 “on Significant
Market Power in the Sale of Agricultural Products and Food and its Misuse” (of
09.10.2009, in force since 01.02.2010) found the most frequent problems
concerned failure to observe the requirement to pay within 30 days of invoice,
below purchase price sales, and certain other contractual terms.302 This
compares with one source’s statement that the usual payment term for fresh food
vegetables and fresh meat was 90 days in his, a different Member State.
(According to the same source, retailers sell such products within a week of being
delivered.) All other things being equal, the 60 day difference could be attributed
as a benefit to small suppliers of the AED rules, although a cost to the retailers.
However, it is reasonable to expect that other contractual terms shifted to reduce
the impact of the rules.303 This would imply that the 60 day difference is an over-

299 Spanish competition authority (Comisión Nacional de la Competencia) 2010. “Report on

Competition and the Agrifood Sector” (Informe Sobre Competencia y Sector Agroalimentario),
available at http://www.cncompetencia.es.
300 See Section 2.3.1 above for a more detailed description of the Spanish sectoral investigation.
301 Requirements for successful cartelisation include an ability to reach an agreement, to monitor

compliance with the agreement, to punish deviations from the agreement, and to exclude or limit
the growth of competitors who are not members of the cartel. The information provision would
help on monitoring compliance and monitoring punishment.
302 Czech Office for the Protection of Competition, Annual Report 2010, p. 24.
303 It is reported that, following the adoption of the Loi Galland in France, negotiations shifted
from “upfront margins,” that would be taken into account in setting the price floor under the law,

PAGE 153 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

estimate of the effect of the prohibition. On the other hand, other terms too, not
so easily measured as payment delays, may also have shifted in the suppliers’
favour as a result of the law.
Another effect of the AED rules appears to harm small and medium sized
enterprises, or local enterprises. Several grocery retailers said that they had
incentives to reduce their purchases from such enterprises because the stricter
rules imposed greater legal risks. One grocery retailer said that as a result of
restrictions on quality audits in Slovakia, it has reduced its procurement there
and turned to major foreign suppliers. He went on to say that it has also reduced
the number of suppliers and variety of products, and in particular of regional
products, as a result of the complexity of negotiating the law. Another retailer,
cited above, echoed these comments, and pointed out that the change in
procurement practices harmed SMEs and, through reduced product variety,
consumers. A third grocery retailer described a process by which retailers
separated suppliers into two groups according to whether they were sufficiently
independent that the retailer could rebut a presumption of significant market
power vis-à-vis these suppliers. It said that it ensured that it accounted for no
more than 25 percent to 30 percent of any supplier’s sales in order to avoid
potential liability due to the application of rules on abuse of economic
dependence.
UGAL (the European association bringing together groups of independent
retailers in the food and non-food sectors) supports this view: “[D]istributors are
very often particularly attentive to avoid creating any dependence situation of the
manufacturer vis-à-vis them. To this end they often care not to represent more
than 30% of the turnover of each producer of private labels. It is in the interest of
the SMEs manufacturing private labels but also of the distributors not to come to
a situation of economic dependence.”304
To summarize this point, it is difficult to gather precise information about the
incidence of conduct the legislation on abuse of economic dependence is
intended to reduce. Only a handful of cases are in the public domain. The one
study of the effect of a national AED law found slow payments and sales below
purchase price as the most frequent issues addressed, the latter being an
indication of competition. In addition, AED rules appear to have had an
unintended negative effect on SMEs who supply large retailers, as retailers tend
to shift purchases away from smaller towards larger suppliers so as to reduce the
retailers’ exposure to the risk that they may be accused of abuse of economic
dependence. If SMEs provide regional specialities, or variety valued by
consumers, then consumers too are inadvertently harmed. On the other hand, as
will be described below in the assessment of the impact of convergence to Article
102 TFEU, associations representing suppliers to grocery retailers are insistent
that the AED rules provide the prospect for protection of suppliers vis-à-vis
retailers.

to “hidden margins,” that would not. Biscourp, P. X. Boutin and T. Vergé (2008), “The Effects of
Retail Regulations on Prices: Evidence from the Loi Galland,” Working paper no. G2008/02,
INSEE.
304 UGAL (Union des groupements de détaillants independents de l’Europe) (2010), Comments

on “Towards more efficient and fairer retail services in the internal market for 2020
consultation,” 10 September, at point 31. http://www.ugal.eu/document/en/10I-RMM-
UGAL.pdf.

PAGE 154 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

4.2.4.2. Costs of compliance of those whom the rules are designed to constrain

Several large grocery retailers complained that their cost of complying with abuse
of economic dependence rules was significant. They argued that they typically
develop standard contracts and a standard way of operating, which they seek to
replicate as they enter new markets. Standardization is seen as a way to keep
costs low, important when margins are low. National laws that prohibit elements
of the standard system, such as particular contract terms, require modifications.
Each modification has a cost. Many specific costs were mentioned by several
different retailers. These costs include legal advice, renegotiation of contracts,
revision of invoicing and accounting system, and retraining of staff. One grocery
retailer estimated that it cost tens of thousands of Euros on internal
coordination, legal advisers, and renegotiation of contracts to comply with one
Member State’s amendment to its law on unfair competition. The same retailer
estimated that the changes that it had to make in the terms have led to losses of
several million Euros.
One grocery retailer active in a number of Member States (and not the one cited
in the preceding paragraph) cited the following as how it modified its conduct to
comply with Czech, Slovak, and Hungarian national law changes. In no Member
State did it report a market share approaching even as much as the 8% market
share level mentioned in the United Kingdom Competition Commission´s 2000
Report as the threshold above which in the UK groceries sector, there is sufficient
buyer power that engaging in certain practices adversely affect the
competitiveness of some suppliers and distort competition in the supplier market
and in some cases the retail market.305
 Czech and Slovak Republics
o Agreements with suppliers had to be modified to be made compliant.
o Suppliers classified as either those whom the distributor could rebut a
presumption of significant market power, and those it could not. An
effect is a preference for foreign over domestic suppliers, and large
over small suppliers.
o Avoid bonuses and rebates that are prohibited under the laws.
o Modify the accounting and invoicing system to reflect requirements of
the law.
o The main problem of the [Czech Significant Market Power] Act is not
that it states additional requirements in business relations but the fact
that the wording is too general and ambivalent and therefore contrary
to the principle of legal certainty.
 Hungary
o Modify the strategy with respect to cash flow, stock finance, and
pricing policy.
o Reviewed and adjusted standard supply agreements to comply with the
Trade Act, and adjusted purchasing conditions and payment terms.

UK Competition Commission (2000), Supermarkets: A report on the supply of groceries from

305

multiple stores in the United Kingdom,” para. 1.10

PAGE 155 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

o Restrict conditions in contracts to avoid violating list of unfair trading

practices.
o In addition, it has retrained buyers, improved internal control, and
undertaken regular contact with authorities.

A third grocery retailer active in a number of Member States estimated the direct
costs of complying with one of the stricter national laws on unilateral conduct of
retailers.306
 Framework agreements with suppliers had to be modified. These were about
400 to 600 contracts, representing about 80 percent of the retailer’s annual
turnover. The legal costs for these modifications totalled about €12 000.
 The accounting system had to be modified to account for the prohibition of
many back margins. No estimate of the cost of this modification was
provided.
 The elimination of many back margins meant that purchase prices were
higher, which fed into higher consumer prices.
 Reductions in the charges for advertising that could be passed onto suppliers
resulted in reduced advertising through billboards and through flyers. These
resulted in lower income from billboard activity of about €500 000 to
€600 000 and from flyer advertising of about €500 000 to €1 million.
 In order to ensure that the retailer meets the legal requirement to pay within
30 days, taking into account bank holidays, it aims to pay within 27 days.
These resulted in losses of about €60 000 per day in which payment is made
early. Annually, these totalled €120 000 to €150 000.
A fourth large grocery retailer active in a number of Member States reported, by
contrast with the above three, that it did not have to modify its business strategy
according to which country it operated in. However, later in the interview, it
pointed out that in countries with stricter rules on economic dependence, it acted
prudently. In such situations, it asserted that large retailers bought from large
suppliers with whom no situation of economic dependence could be created.

4.2.4.3. Quantitative estimates of costs of compliance

It is not possible to provide an accurate estimate of the one-time and the on-
going costs of complying with stricter national rules on unilateral conduct. It is
likely that even the two estimates of the costs of individual retailers complying
with specific changes in specific laws are not especially accurate. But with
appropriate degrees of humility, it may be possible to arrive at estimates that are
within an order of magnitude of the true value. The assumptions are clearly
expressed, so readers can perform their own sensitivity analysis.
The two estimates of the costs to one retailer of bringing itself into compliance
with new stricter national laws on unilateral conduct by retailers can be used to

306 To avoid identifying the retailer, we do not identify the country beyond that it is one of those
listed above whose restrictions were not limited to pricing, i.e., it is one of Czech Republic,
Hungary, Latvia, or Slovakia. The retailer originated outside of the country imposing the law.

PAGE 156 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

very roughly estimate the overall direct cost of bringing retailers into compliance.
These are one-time costs; when laws change again, then costs are borne again to
bring a retailer into compliance. One estimate was “tens of thousands of Euros;”
the second estimate was €12000 for the costs of renegotiating contracts and
excluding other costs. Since it is difficult to imagine why one company´s contract
renegotiations should cost many times another´s, estimate that “tens of
thousands” is two tens, or €20000.
Assume, in one Member State, that five distributors have to modify their
contracts and internal processes to comply with a new law. Assume the cost for
one company of each cycle of adapting to new rules is €10,000 to €20,000. Then
the direct cost of five distributors understanding and complying with new rules
totals between €50,000 and €100,000 each time one Member State changes its
rules. In practice, large distributors operate in several Member States over a span
of decades. If eight Member States had stricter unilateral conduct laws, and each
Member State changed its law every five years (an assumption that seems
reasonable looking at the frequency of amendments in several Member States, as
also reported in Section 2 above), then, on average each year, distributors would
spend 1.6 times the figures calculated, or between €80,000 and €160,000, to
bring themselves into compliance by renegotiating contracts, modifying
accounting systems and the like.
The same two large grocery retailers also provided estimates of on-going annual
costs of compliance in a single Member State of “several million Euros” and
€1.12-1.75 million. The detail of one of these estimates included the lower income
from billboard and flyer advertisement sold to suppliers, and the foregone float
because suppliers must be paid earlier than they would have been. These
estimates can help to estimate the overall on-going costs of compliance. Both of
the estimates are from large retailers, of which there patently are few in any given
Member State. If five distributors in each of eight Member State incur such
losses, they sum to tens of millions of Euros, or €45-70 million, depending on
which estimate one uses. Alternative assumptions as to the number of retailers
affected or number of Member States where compliance costs are of such a
magnitude, that is, the scaling factors, would yield different figures. However,
these figures provide an order of magnitude estimate.
For comparison, if the European retail sector accounts for 4.2% of European
GDP, and EU27 GDP in 2009 was €11.787 billion,307 then European retail sector
accounts for about €495 billion. This figure is too large to constitute a
denominator to compare the estimated totals of direct and ongoing costs: The
scope of “retail” is far larger than the scope of “food” or “fast-moving consumer
goods,” and the additional costs would impact profits rather than revenues. If
“food” or FMCG constituted half of retailing, and if profit rates were 5% of
revenues, then the denominator would be about EU 24.5 billion and the level of
ongoing costs of compliance well less than a tenth of a billion Euros would be not
very significant by comparison.
An alternative way to identify a numerator is to estimate that the turnover in
2009 of eight large retailers in Eastern Europe totalled somewhat more than €50

307 Source for the 4.2% figure is the Retail market monitoring report “Towards more efficient and
fairer retail services in the internal market for 2020”
(http://ec.europa.eu/internal_market/retail/docs/monitoring_report_en.pdf. Source for the
€11.787 trillion GDP figure is Eurostat.

PAGE 157 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

billion. If profit rates are assumed to be 5%, this yields €2.5 billion. Then the
figures of €45-70 million for compliance costs, for five retailers in eight Member
States, is about 2 to 3% of profits. That is, under the various assumptions made,
compliance costs reduce profits by 2 to 3 percent.
To summarize, large retailers that are active in a number of Member States
identified their costs of bringing their operations into compliance with stricter
national rules on unilateral conduct, and their ongoing costs of compliance.
Modifications of contract terms involved costs of negotiation and lower revenues.
Personnel had to be retrained. Accounting and invoicing systems had to be
modified. Two of these large retailers estimated the monetary cost of compliance,
and these two estimates were used to scale up the costs to an EU-wide estimate.
While the assumptions are not the only ones that would be reasonable, they did
show that, with a reasonable set of assumptions, the costs of stricter national
rules on unilateral conduct on large retailers are significant compared with their
profits, but not significant when compared with European retail turnover.

4.2.4.4. Suppliers´estimates of the costs of convergence

Suppliers and associations of suppliers whom we have interviewed were divided

as to the effect of convergence of national competition rules. However, some of
those who responded as to the effect of convergence with Article 102 TFEU were
opposed, viewing such a change as “the last nail in the coffin” of suppliers. Two
were a Europe-wide and a national association of suppliers, including SMEs and
referenced below. One multinational supplier, referenced below, was in favour of
harmonization at a European level on issues related to bargaining power and
unfair competition.308 Their comments did not distinguish whether the source of
the stricter rules was a national competition law or a national unfair competition
law, but addressed convergence, or not, of the substance and of the institutions
administering or enforcing the rules on abuse of economic dependence and
market practices in which an enterprise may engage.
Supplier associations expressed opposition to convergence of stricter national
competition laws with Article 102 TFEU. One association of suppliers felt that
neither the national competition authority nor the European Commission
addressed economic abuse or abuse of buyer power by retailers towards
suppliers. The national unfair competition law provides a legal remedy, but fear
of retaliation deters suppliers and even the association from filing cases under
that law. Harmonization of the national competition law with Article 102 TFEU
would eliminate the possibility of the national competition authority bringing an
abuse of economic dependence case under the national competition rules. This
would be the “last nail in the suppliers’ coffin,” according to the association.
Suppliers typically benefit from stricter rules against abuse of economic
dependence.
By contrast, one supplier active in most Member States felt that harmonization at
a European level towards Article 102 TFEU on issues related to bargaining power

308It would be misleading to read this as a ”vote” of thousands to one among suppliers to grocery
groups. Despite many attempts via different routes, no other multinational supplier provided
input into the analysis—the researchers do not know their positions. By contrast, these
representatives of supplier associations were quick to respond to questions of grocery groups´
unilateral conduct with respect to suppliers.

PAGE 158 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

and unfair practices was crucial. The supplier felt that local initiatives such as
codes of conduct and additional laws and rules created administrative burdens
and have not been effective. For this supplier to enter a new market, e.g., a new
Member State, it must incur legal compliance costs to ensure that it complies
with local traditions, e.g., some Member States have laws on listing fees. It felt
that the current focus of competition policy on short-term consumer prices could
lead to consumer harm in the long run as suppliers have reduced incentive to
innovate in the face of retailers free-riding on their innovations. It may be the
case that this supplier was misinformed as to the effect of convergence to Article
102 TFEU on bargaining power and unfair practices legislation, as these issues
arguably lie outside the scope of Article 102 TFEU.
These views were congruent with those expressed by Mr. Alain Galaski, Director
General of the European Brands Association, at a recent conference on business
relationships in retailing. He was concerned that national rules lead to a
fragmentation of the single market. His association was agnostic in the question
of codes versus laws, but they had to be effective, pro-efficiency, create a level
playing field, and be enforced by a third party.
The desirability of a common approach towards unfair practices that are
generated by unequal bargaining positions and which significantly restrict
competition on the market, expressed in the Spanish Competition Authority’s
report cited above, is consistent with the above two comments, but would appear
to retain a national role.
The review of the economics literature on abuse of economic dependency,
reported in Section 3.2 above, noted that provisions prohibiting such abuse had
not resulted in many successful cases, which had resulted in yet more restrictive
rules.309
Determining whether abuse of economic dependence rules belong within the
scope of “competition rules” and thus would be possibly affected by removal of
the convergence exception in Reg. CE 1/2003 is not within the scope of this
study. But some suppliers, who are the intended beneficiary of these rules,
oppose convergence of national law provisions on unilateral conduct towards
Article 102 TFEU because they see such convergence as weakening, or even
eliminating, protection against abuse of economic dependence or unfair practices
deriving therefrom. Absent convergence, they see a possibility that the national
competition authority could intervene against abuse of economic dependence.
Others support harmonization of rules on abuse of economic dependence
because they wish to reduce fragmentation of the single market, thus reduce their
own legal and administrative burdens, although they differ as to whether they
prefer common national solutions or a common European solution. An effective
enforcement mechanism would reduce administrative burdens and reduce
fragmentation of the single market. It seems clear, however, that some voices in
favour of convergence do not anticipate convergence resulting in a weakening of
AED enforcement, that is, their support for convergence is based on an
assumption about convergence that may not hold.

309See OECD 1999, pp. 9, 42-43; Dobson Consulting (1999), “Buyer power and its impact on
competition in the food retail distribution sector of the European Union,” prepared under
contract IV/98/ETD/078.

PAGE 159 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

4.2.4.5. Conclusion on impact of convergence of rules on abuse of economic

dependence
If convergence would weaken or eliminate AED rules, then suppliers to retailers
felt they would be left in a much weaker position. If, on the other hand, the rules
would be harmonized but not weakened, at least some suppliers felt this would
be beneficial as it would reduce compliance and enforcement costs. Grocery
retailers favoured elimination of the AED rules, but a careful reading would
suggest that they would consider as an improvement on the current situation
even a situation in which these rules would not be eliminated, but would be
harmonized throughout the EU27. The suppliers were unable to provide an
estimate of the value of the AED rules, whereas two grocery retailers did provide
such estimates. Consequently, the net estimate of the impact of eliminating AED
rules, if that would indeed be an effect of convergence to Article 102 TFEU, is not
calculable. Counting only the estimates provided by the grocery retailers, the
costs to retailers of compliance, under reasonable assumptions (but not the only
reasonable assumptions) to scale up to an EU-wide impact, detailed above, is
somewhere around €45-70 million annually.

4.2.3 Rules on “sales below cost”

Small retailers who compete against large retailers are the intended beneficiaries
of stricter rules on unilateral conduct that prohibit “sales below cost” by non-
dominant enterprises. In addition, vertically integrated infrastructure enterprises
might be subject to stricter rules on sales below cost, but this conduct can often
be addressed also under sector-specific regulation.310.311
There are differing views as to the effectiveness of rules prohibiting sales below
cost, and thus of the impact of eliminating such rules if they were covered by
convergence to Article 102 TFEU. For example, an association of both large and
small distributors and retailers in France, where sales below cost are prohibited
except during agreed sales periods, felt that the system of tight regulation with
enforcement by the DGCCRF worked well. The main complaint was the high
overall cost, especially legal costs, of the system. By contrast, a large grocery

310 The competitors of vertically integrated network infrastructure enterprises, such as a

telecommunications company which owns part of a network and also competes with downstream
competitors to whom it sells access to the network, could also be beneficiaries of rules against
sales below cost. A discussion of whether sector-specific unilateral conduct rules would, in
practice, be affected by any convergence of national competition rules on unilateral conduct to
Article 102 TFEU is provided above at Section 2.3..
311 An association of suppliers described how its members were harmed by sales below cost. While

the beneficiaries of rules prohibiting sales below cost by non-dominant enterprises are usually
assumed to be small competitors of large retailers, this association of suppliers felt that the rules
prohibiting sales below cost provided important protection for its members. It cited an example
involving below-cost pricing by a new entrant into the market for supply of a certain product sold
in grocery stores. The low acquisition price enabled the retailer to offer the product at lower
prices to consumers. The harm to the association’s members occurred when the retailer
demanded other suppliers to match the pricing offered by the entrant. Convergence on Article
102 TFEU would make a competition law complaint against the supplier based on these facts
rather unlikely to succeed. Introductory low pricing is recognized as a legitimate part of an entry
strategy, and a new entrant into a crowded market is unlikely to succeed in excluding its
competitors by predation.

PAGE 160 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

retailer interviewed preferred an end to the prohibition of sales below cost,

arguing that it restricted competition and removal of the restrictions would lower
consumer prices. And another market participant pointed out that if large
retailers negotiate a discount compared with the acquisition price paid by large
wholesalers, the prohibition of sales below cost does not advantage the small
retailers downstream from the wholesalers vis-à-vis the large retailers. The
implication of this last interview is that small retailers cannot match the
consumer prices offered by large retailers complying with the national law
prohibiting sales below cost.
One grocery retailer welcomed removal of the prohibition of sales below cost in
France and predicted that it would lead to a reduction in prices. The
representative said that the Loi Dutreil of 2 August 2005 had loosed the
prohibition and opined that this had led to price decreases in private label and
first price products.
The representative of another grocery retailer expressed the view that the few
complaints made under the current national unfair competition law indicate that
the law works well. There would be no need for additional legislation, and in any
case it prefers voluntary codes of conduct. This retailer said that it did indeed
price below cost to introduce new products and at special times of year, but
within the national laws limiting such pricing. It noted that about 25-30% of its
total sales came from promotions.
According to UGAL, “There are not enough measures and not in all countries of
the European Union, that would prevent businesses and more particularly SMEs
from suffering from sales below cost used as a predatory tool even if they are
carried out by companies that have no dominant market position.” (UGAL 2010,
point 70) This was near the position of a retailer we interviewed, referred to
earlier, who felt that the German legislation made it difficult to prove a violation
of the sales below cost rules.
Interviewees who were large retailers suggested that it is not difficult to find ways
around prohibitions of sales below cost in some national laws, German rules
being the most cited example. One explained that the definition of sales below
cost is so simple, that it is difficult to prove a violation. By contrast, this person
said that other national laws (citing France) required a lot of creativity and
energy to comply with and estimated that half of its legal budget in France went
towards such compliance.
The economic research undertaken by the UKCC in its 2008 grocery markets
enquiry showed that, in the United Kingdom, small grocers were not affected by
below-cost sales by large retailers. It is reasonable to assume that some of the
conditions are similar in other Member States. For example, low barriers to entry
into small-scale grocery retailing make predation by large retailers an
unprofitable strategy. The persistence of small retailers can be attributed to their
not competing directly with large retailers, but being somewhat differentiated.
Unfortunately, the data to actually measure the effect on small retailers’ entry
and exit of the presence in shopping centres and high streets of large retailers is
not available for this study; in the British study, such data showed there was no
effect. (See above in the review of the economic literature for more details on the
UKCC 2008 report.)

PAGE 161 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The rules prohibiting sales below cost, aimed at protecting small retailers, appear
not to perform the protective function. Suppliers negotiate different deals with
different distributors. Large firms may well have found ways to reduce the impact
of the rules. Nevertheless, small retailers persist offering a differentiated
experience to that of the large retailers. As will be shown below in the
econometric work to quantify the effects of convergence to Article 102 TFEU,
rules prohibiting sales below cost appear to have an effect on consumer prices,
raising them by approximately 1 to 3 percent. But the rules appear not to have a
detectable effect on small retailers.
Thus, if convergence to Article 102 TFEU implies reduced enforcement of sales
below cost prohibitions beyond the scope of Article 102, then convergence is
unlikely to have a detectable effect on small retailers. However, since this
conclusion is based on a very small sample, although it is consistent with the
findings of the UKCC 2008 report, it should be seen more as an indication. With
respect to large retailers, from their interviews it appears they may benefit from
reduced legal costs of compliance. Again, not many addressed this precise
question, so this should be taken more as an indication of the likely impact.
Perhaps the most significant impact of convergence, if it would eliminate sales
below cost prohibitions outside the scope of Article 102 TFEU, would be the
effect on consumer prices in retailers subject to the rules. We estimated the
prices would fall by about 1 to 3 percent, estimates in line with others´
econometrics studies.

4.2.4 Rules on tying

The legal analysis identified some national rules on tying which were stricter
than Article 102 TFEU. However, no respondent to our survey and none we
interviewed mentioned a concern with such rules, although one network
infrastructure company mentioned that rules on conditional sales constrained its
freedom to price. Many of the tying cases described in the legal analysis
concerned telecommunications, so this may have been an area where stricter
national rules, whether in the telecommunications law or elsewhere have an
effect. But if a convergence obligation reduces the scope for stricter national
competition rules on tying, then the telecommunications regulatory regime could
easily substitute.
From a theoretical point of view, the cost of stricter national rules against tying is
the absence of pro-competitive tying. This is difficult to identify without
examining a particular market where efficient tying has been prohibited in some
markets but is pervasive in others. However, no such market was identified either
though the business survey or the legal analysis. This does not mean that no
market is affected by stricter rules on tying, but only that none was so prominent
as to have been identified. Thus, we make no estimate of the impact of
convergence of stricter national competition rules against tying but the absence
of complaints or even mention of the subject by respondents to the business
survey and interviews may indicate that the impact would be small.

PAGE 162 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

4.2.5 Impact of the divergence of stricter national rules on

relationships between retailers and suppliers

Among the comments made by grocery retailers that were active in several
Member States were those expressing dissatisfaction with divergent national
rules on the relationship between suppliers to retailers and retailers. Some of
these comments cannot be categorized as relating to a specific substantive rule,
so are reported here. Several retailers cited the laws enacted on retail trade and
in particular in the grocery sector in Eastern Europe (see Section 2.3.1 above),
with fewer citing those in France and Germany. Countries cited by at least one
retailer as having stricter national rules were Czech Republic, France, Germany,
Hungary, Lithuania, Latvia, Poland, Romania, and Slovakia. None of the
enterprises we interviewed said that they had avoided or abandoned any national
market as a result of the stricter national rules on unilateral conduct. However,
one grocery retailer said that disproportionate regulation had reduced its
incentives to make further investments.
The comments of the retailers should be read in the context of the timing of the
imposition of the stricter national rules on unilateral conduct. The retailers had
entered the eastern European markets before the stricter rules were imposed.
The rules would therefore not have influenced their own entry decisions, but
would do so for later entrants.
The more frequent citation of the laws of eastern than of western European
Member States can reflect the more recent introduction of the former than the
latter rules. For example, one respondent wrote of France, “Most of these rules
are so old and so known, that we have taken the habit to work with them, without
thinking how it could be if we were freer.”
Alternatively, the difference in frequency of citation could reflect greater
dissatisfaction with the rule-making process. One respondent complained, in
connection with the Hungarian and Slovak laws, that retailers had no
opportunity to be consulted or heard, that there were no impact studies to assess
whether the laws would be efficient, and that the laws were poorly and swiftly
drafted so that it is hard to understand the laws and apply them. He went on to
point out, more generally, that in his opinion laws cannot be copy-pasted from
one country to another without being properly embedded into the legal
environment.
The French system generated different views among the interviewees than did
the eastern European systems. Several opined that costs of the system were too
high. Retailers felt that fines were high and that the rules were actively enforced.
One mentioned that sources of the high costs included the large number of minor
provisions in the Code de Commerce and the presumption of guilt. Another
contrasted France—“over-regulated”—with Germany—“not a problem,” and said
that it expended quite a lot of energy complying with French rules. The same
retailer had roughly estimated that it expended half of its legal costs in France on
compliance with economic dependence and sales-below-cost type rules.
The association representing independent retailers, UGAL (Union des
groupements de détaillants independents de l’Europe), noted that national
regulations on unfair commercial practices are tailored to each Member State’s
circumstances. In the context of a discussion of unfair commercial practices
within the supply chain where, according to UGAL manufacturers have

PAGE 163 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

increasing market power and thus bargaining power, retailers stressed that (what
they consider to be) overly strict competition legislation would unnecessarily
weaken their economic performance of retailers and reduce the efficiency of the
whole supply chain. UGAL supports freedom of contract, opposes politically
driven interventions, and feels that, “If there are cases of abuse of market power
competition rules apply.”312 These remarks could be interpreted as referring to
stricter national rules on competition, but the association was not available for
interview. But the need for harmonization is limited by their observation that, “A
majority of operators do not have cross-border activities and therefore do not
face different legislations.” (UGAL 2010, point 20)
To summarize the above paragraphs, the concern of large retailers operating in
several Member States is not limited to the avowed intended and actual effects of
the stricter national rules on abuse of economic dependence. They are also
concerned with the rule-making process, including whether they are consulted
about proposed rules, whether the impact of the proposed rules is assessed,
whether the proposed rules fit into the existing legal framework, and whether the
rules are sufficiently understandable to be complied with by those to whom they
apply. Some were concerned with the high cost of the system of rules in certain
Member States.

4.2.6 Possible different impacts of certain national rules

Some of the laws on abuse of economic dependence in food or grocery sectors

may tend to have, as a result of the practices they prohibit and the thresholds
they set, a greater effect on foreign than domestic food retailers. Some of the
conduct that is prohibited is of the type of conduct that is engaged in by foreign
retailers, who are applying their systems from their “home” countries in a new
environment. Listing fees, for example, lower the risk and cost of carrying new
products and are partly responsible for the wide variety of products available at a
modern retailer. But they are prohibited in many stricter national laws.
In addition, some of the obligations and prohibitions in the laws in Hungary (Act
CLXIV of 2005 on Trade) and the Czech Republic (Act No. 395/2009 Coll on
significant market power in the sale of agricultural and food products and its
abuse) apply only to retailers exceeding certain turnover thresholds (HUF 100
billion and CZK 5 billion). Some of the retailers interviewed point out that the
cooperative structure of some of the domestic retailers implied that they would
not be caught by the turnover thresholds. In addition, the maximum potential
fines are proportional to the turnover stated in the annual report of the preceding
year. (See, e.g., Section 6 of the Hungarian Act on the Prohibition of Unfair
Trading Practices Relating to Agricultural and Food Products applied against
Suppliers.) Thus the maximum fine imposed on a large foreign retailer would
take into account foreign turnover, and not just that in the Member State
imposing the fine. Thus, for conduct having the same effect in the domestic
market, the maximum fine to which an enterprise is subject is not a function of
its domestic turnover, but instead of its worldwide turnover. This suggests that

312UGAL (Union des groupements de détaillants independents de l’Europe) (2010), Comments

on “Towards more efficient and fairer retail services in the internal market for 2020
consultation,” 10 September, at points 19 and 21. http://www.ugal.eu/document/en/10I-RMM-
UGAL.pdf.

PAGE 164 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

an enterprise with predominantly foreign turnover is liable for a higher

maximum fine than an enterprise with identical position in the domestic market,
but with predominantly domestic turnover.
Evidence to support the notion that practices standardized in other European
markets are restricted by the stricter national laws is provided by the three
proceedings in Hungary for abuse of economic dependence listed on the website
for 2008-2010. These were all brought against foreign owned retailers Provera
Beszerzési (Purchasing) Kft. (parent company: Louis Delhaize), Auchan
Magyarország Kft. and Metro Kereskedelmi Kft. Other large retailers operate in
the country, both foreign-owned and domestic.313
In sum, several of the characteristics of the laws appear to differently impact
foreign versus domestic retailers. Practices engaged in as part of the standardized
systems of foreign retailers are prohibited. The maximum fines are higher for
those retailers who are more active abroad; given the tendency to grow first in
one’s “home” country and then in others, this would tend to mean that domestic
retailers with the same level of sales in the Member State would be subject to
lower maximum fines. And the one available statistic on actual proceedings
supports the notion that foreign retailers are more impacted.

4.2.7 Impact of national sector specific unilateral conduct rules

Two enterprises active in inter alia energy networks cited the German Act against
Restraints of Competition Section 29, as constraining their pricing.314 The
provision applies to suppliers of electricity or pipeline gas who have, either alone
or together with others, a dominant position. The provision prohibits demanding
fees or other business terms which are less favourable than those of other public
utility companies or undertakings in comparable markets, unless objectively
justified, and shifts the burden of proof (in proceedings before the cartel
authorities) onto the enterprise. It also prohibits abuse by demanding fees which
unreasonably exceed the costs.315

313 The conduct would not appear to be unusual, based on the information publicly available in
the Hungarian Competition Authority website. In particular, the conduct with respect to suppliers
that was modified included: not inserting provisions on exclusive promotion campaigns in
contracts for 2009, the use of uniform letter sizes in contracts, providing information to suppliers
about stocks and sales of the supplier´s products, and unrestricted—in time or quantity—ability
to return unsold stocks. The investigations were launched under the Act on Trade; whether this
reflects a judgment that the Act on Competition—which is very like the analogous law in other EU
Member States—was not likely not infringed or a judgement about the relative ease of procedural
matters is unknown. If the former, then this provides support for the notion that the conduct
repaired by the commitments would be practices engaged in other EU Member States without
analogous provisions to those in the Act on Trade.
314 Other enterprises, active in other network infrastructure, cited various national legislation

already reviewed elsewhere, such as the French Commercial Code, and competition constraints in
sector specific legislation. As noted elsewhere, we do not review the latter legislation.
315 See section 2 above. Section 29 can be contrasted with Section 19(4)2 which prohibits abuse of

a dominant position by demanding “payment or other business terms which differ from those
which would very likely arise if effective competition existed; in this context, particularly the
conduct of undertakings in comparable markets where effective competition prevails shall be
taken into account.” Section 19, however, retains the original burden of proof, i.e., the
competition authority, for example, must prove a violation.

PAGE 165 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

One of the enterprises detailed its concern with Section 29. It said that although
it is too early to tell, the section being in force since 2007, it likely had two
negative effects: (1) The enterprises would constantly monitor and synchronize
prices on retail markets, and (2) the enterprises’ procurement strategies on the
wholesale market would be aligned. With respect to retail markets, the offer of a
competitively low price in one circumstance would then make it available as a
comparator by which the competition authorities could subsequently judge the
other prices offered. So rather than cut price to customers, the enterprise
maintains higher prices. The effect is like that of most favored nation clauses,
which gives firms incentives to maintain high prices because a single price cut
inevitably leads to price cuts to all customers. In addition, entry is hindered
because entrants would be discouraged from offering “low, introductory prices.”
Regarding procurement strategies, the enterprise argued that the potential
justification of retail prices is based on the average procurement costs in the
sector. (This appears to be borne out by the case summary from the national
competition authority, case B10 – 16/08 to 56/08.) Thus, utilities cannot fully
recover their procurements costs if they have deviated from their competitors’
strategy and such deviation later turns out to be more costly. The result is to
narrow the range of strategies used, so as not to risk incurring higher than
average procurement costs.
This enterprise said that Section 29 GWB had had a significant effect on its
pricing and procurement strategies, throughout the supply and value chain. But
it was unable to provide exact costs. It estimated that its costs due to the rules
totaled EUR 500.000 – 1.000.000 per annum. This can be scaled up to estimate
a cost for all of German energy companies.
If Section 29 GWB is covered by the convergence obligation to Article 102 TFEU,
then this cost would be an estimate of the impact of such convergence on energy
suppliers, appropriately scaled up. One question would be is whether consumers
would bear a cost, such as higher prices? Would competition be enhanced by
different methods that are more, or less, efficient? Or would more formal
regulation be put into place? Any answer would be speculative, but bears on the
impact of convergence if Section 29 GWB would be covered by any removal of the
convergence exception.

4.2.8 Competition and regulation

The conduct of firms active in regulated sectors is subject to the stricter of EC

competition rules and national regulation. Where national regulation allows
certain conduct, if EC competition rules prohibit it, the (stricter) latter rules
apply. Conversely, national regulation can prohibit conduct that would be
allowed under EC competition rules. National regulation can be made stricter.
Consequently, if stricter competition rules on dominant firms were eliminated,
rules having the same effect could be written into the national sector regulations.
The European Court of Justice Deutsche Telekom decision316 on 14 October 2010
reiterates the position that EU competition rules apply to conduct that is
regulated by national regulatory authorities. The Court reiterates that only if

316 Deutsche Telekom AG v European Commission, Vodafone D2 GmbH, formerly Vodafone

AG & Co. KG, formerly Arcor AG & Co. KG and Others Case C-280/08 P of 14 October 2010.

PAGE 166 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

national law requires anticompetitive conduct or precludes all scope for any
competitive conduct, can EU competition rules not apply to anticompetitive
conduct. [point 80] That is, national regulation could be so strict as to leave no
scope for competition rules to apply. It can be recalled that the dispute concerned
pricing over which the undertaking had scope for manoeuvre and not pricing
over which it had not.317
The recent history of changes in the governance of German energy network
access provides an example of national sector regulation being made stricter and
thereby replacing competition (abuse of dominance) rules.
 As a part of the liberalization of the energy sector in Germany, the 1998
amendment to the energy law provided for network access terms to be
negotiated. Dominance abuse continued, and continues, to be prohibited by
the general competition law.318
 In 2001, the Bundeskartellamt (BKartA) opened initial investigations into 23
electricity network operators on suspicion that they charged abusively
excessive fees for network use and impeded other electricity providers. The
BKartA went forward with 13 formal abuse proceedings. The investigations
were hindered in part by the parties refusing to submit the information
needed by the BKartA to assess whether the charges were abusive.319
 In 2003, the BKartA prohibited several electricity network operators from
abusing their dominant positions by demanding inter alia excessive fees for
network use. Some of these decisions were reversed, others sustained.320
 On 1 July 2005 the Energy Industry Act (EnWG) entered into force. It was
aimed at improving the weak competition that had developed over the six
years after liberalization. The Federal Network Agency regulates network
access for electricity and gas, while the competition authorities retained
responsibility for abuse control in the markets up- and downstream from the
networks, as well as cartels and mergers.321
 The regulation of network access has resulted in significant decreases in
access cost.322

317 The Suker Unie and Others v Commission judgment [Joined cases 40 to 48, 50, 54 to 56,
111, 113 and 114/73, of 16.12.1975), cited in the DT judgment at point 82, is clearer on the point
that national regulation can restrict some actions by undertakings, leaving a “residual field for
the operation of the rules of competition,” point 71.
318 Germany, Competition Law and Policy Developments July 2001-June 2002, Annual
Report by Competition Authority, points 79-83, available at oecd.org/competition.
319 Ibid., points 14, 26.
320 Germany, Competition Law and Policy Developments July 2003-June 2004, Annual
Report by Competition Authority, point 39, available at oecd.org/competition.
321 Germany, Competition Law and Policy Developments July 2004-June 2005, Annual Report by

Competition Authority, points 10-11, available at oecd.org/competition

322 Network operators had to apply to the Federal Network Agency (“FNA” ro “BNetzA”)) for

approval of network charges. The level depending on network costs. [European Regulators Group
for Electricity and Gas, National Reporting 2007-Germany, English summary, p. 30 at energy-
regulators.eu] Network charges for households fell 13% as a consequence of the regulator’s cut in
allowed network charges. [ibid., p. 9] Network access disputes handled by Federal Energy Agency
included refusal to connect at the requested network—which has implications for costs—and the
charging of costs of network extension. [ibid., p. 28.] Electricity network access charges were
reduced by the FNA in the second round of approvals. [National Reporting 2008-Germany,
English summary, p. 27] Gas network operators had to apply to the FNA for approval of network

PAGE 167 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Thus, Germany 1998-2005 illustrates the imposition of stricter national

regulation that has the effect of replacing competition rules.
It is impossible to predict whether all, or any, national legislature would react in
the same way if stricter competition rules on dominant firms were eliminated
through convergence to Article 102 TFEU. But the German experience illustrates
that national regulations and a national regulator can be created in order to apply
stricter national sectoral regulations than would be available under competition
rules, in particular Article 102 TFEU.

4.3 Assessment of the broader economic impact

This section addresses the impact of the stricter rules on unilateral conduct,
beyond those enterprises directly involved, on business, including SMEs, and
consumers. The impact on public authorities, including national competition
authorities, is also to be assessed. Finally, the impact on certain economic
variables such as competition in the internal market, competitiveness,
innovation, growth and jobs, are to be assessed.

4.3.1 The impact on businesses not involved in grocery retailing or

supply, or network infrastructure

Businesses are affected to different degrees by the different divergent national

competition law provisions on unilateral conduct. We found evidence that the
rules on abuse of economic dependence and sales below cost particularly affected
grocery retailers and their suppliers. And the stricter rules on conduct in
infrastructure sectors similarly affected only those enterprises active in those
sectors. The effects on these sets of businesses were described above, as was the
effect of stricter definitions of dominance and the likely effect of convergence to
Article 102 TFEU. However, in order to assess the impact on business throughout
the economy, the impact on businesses that operate outside of these sets of
specific businesses, must be assessed. As argued below, the logic of economic
theory leads to the finding that the effect of stricter unilateral conduct rules on
enterprises not active in the markets directly affected is limited. This implies that
the effect on them of convergence would also be limited.
The first task to assess the effect of rules on abuse of economic dependence and
rules prohibiting sales below cost is to describe those enterprises that have a
relationship to the markets at which these rules are aimed, but are not
themselves active in those markets. These are potential entrants into and
enterprises that have exited the affected markets, as well as suppliers of
complements or substitutes.

charges. In 2007, the first round was completed with an average reduction of charges of 12.1%.
[ibid., p. 15] The FNA developed principles for the appropriate calculation of contributions to
construction costs, to take effect in 2008. [ibid., p. 26] Revenue caps replaced rate approval when
the Incentive Regulation Ordinance entered into force on 6 November 2007. [ibid., p. 29] In
March 2008, the BKartA initiated proceedings against 32 undertakings for abuse of dominance in
the pricing of gas. [ibid., p. 32] In the end, the BKartA had concerns regarding 30 of the
undertakings, which were dispelled when they agreed to various forms of price cuts. [National
Reporting 2009-Germany, English summary, p. 12]

PAGE 168 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

 Potential entrants’ decision to enter the markets would be influenced by the

presence of the rules if they changed the cost of or time to entry or changed
the expected profitability of participation in the market.
 Exited enterprises’ decision to exit the markets would be influenced by the
presence of the rules if they changed the expected profitability of continued
participation in the market.
 Suppliers of complements or substitutes would be affected if the rules
changed prices in the markets subject to the rule.
These three categories are addressed in reverse order.
The rules aimed at sectors (food, retailing, network infrastructure) tend to apply
to the closest substitutes and complements, e.g., all agricultural products and
foods. Products which are not close substitutes or complements have small cross-
price elasticity.
If products have small cross-price elasticity, then even a price rise of as much as 3
percent would not have a large effect. (E.g., the cross-price elasticity between
beef and non-food is estimated as -0.06, so a 3 percent increase in the price of
beef would result in a decrease in demand for non-food of 0.18 percent. The
cross-price elasticity of demand for fruit-and-vegetable and non-food is
estimated at -0.03, so a 3 percent increase in the price of fruit-and-vegetable
would result in a decrease in demand for non-food of 0.09 percent.323) These
magnitudes indicate that suppliers of complements or substitutes not covered by
the stricter unilateral conduct rules would not be significantly affected by the
rules. They are, therefore, not considered further.
Exited enterprises can, logically, be affected only if they are still active in some
other market, and are potential entrants into the markets subject to the divergent
unilateral conduct rules. They are, therefore, not considered further other than in
their role as potential entrants.
The focus, then, is on the effects of these stricter unilateral conduct rules on
potential entrants.
Large grocery retailers that have not yet entered a given Member State but
operate in another constitute one group of potential entrants. These are the most
likely entrants into grocery retailing in other Member States, based on their
history of successfully doing just that in other Member States. But none of the
large grocery retailers we interviewed had avoided or exited a Member State as a
consequence of abuse of economic dependence or sales below cost-type rules.
There may be other potential entrants that would have entered grocery retailing
if entry costs were lower or profitability higher. Convergence to Article 102 TFEU

323 Estimates of cross-price elasticity from the Economic Research Service of the US
Department of Agriculture, www.ers.usda.gov/Data/Elasticities. These particular estimates were
from Moschini, G., et al. (1994). "Maintaining and Testing Separability in Demand Systems."
American Journal of Agricultural Economics 76 (February): 61-73. By comparison, estimates of
the cross price elasticity between beef and chicken range between 0.04 and -0.22, so a 3 percent
increase in the price of beef would result in demand for chicken changing between +0.12 percent
(increasing) or -0.66 percent (decreasing). Estimates from Menkhaus et al. (1985), "A
Reexamination of Consumer Buying Behavior for Beef, Pork, and Chicken," Western Journal of
Agricultural Economics 10(1): 116-125” in Table 4; and from Dahlgran, R. A. (1987). "Complete
Flexibility Systems and the Stationary of U.S. Meat Demands," Western Journal of Agricultural
Economics 12(2): 152-163, Table 3.

PAGE 169 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

would, in principle, reduce entry costs by reducing the costs of compliance in a

second Member State after one had entered one Member State. This would
benefit potential entrants: Some may find it profitable to enter the market, which
otherwise would not have done so. Convergence to Article 102 TFEU could
increase or decrease the profitability of participating in the market, although the
balance of opinion of large grocery retailers suggests that it would increase
profitability. This would also benefit potential entrants, for the same reason as
lower entry costs. Thus, convergence to Article 102 TFEU would likely be positive
for potential entrants. The maximum magnitude of the effect can be measured by
the cost of divergence for enterprises already in the market: These are the
quantities by which entry costs are lowered. But free entry without scale
economies would imply that all of these benefits would be competed away, or
rather consumers would benefit by increased competition.
The actual magnitude of the benefits to potential entrants would lie somewhere
between the two extremes, that is, between the cost of complying with divergence
rules and zero.
One set of potential entrants into the supply to grocery retailers could be those
enterprises that already produce food but supply product variants preferred by
some consumers that are distributed via farmers’ markets or specialist shops.
These enterprises may exist: In interviews, large grocery retailers suggested that
some suppliers who might have qualified as economically dependent and whose
products could be substituted with those supplied by larger businesses were
discontinued as suppliers. If convergence of unilateral conduct rules weakened
prohibitions of abuse of economic dependence, then such potential entrants are
less likely to enter into supply agreements with retailers. The reduction in the
value of a strategy not taken is difficult to value. The magnitude of this effect is
likely quite small.
The impact on potential entrants of divergent definitions of dominance is likely
to be very slight. If a new product has no close substitutes, then the difference
between 33.3% and a 100% market is irrelevant; determination of dominance
will be based on other factors. If a new product has close substitutes and if a
potential entrant can foreseeable reach 33.3% market share, then the market
must be fairly contestable, so such a market share would have a lesser import in
the determination of dominance.
In sum, potential entrants into the grocery retailing or network infrastructure
sectors and suppliers of complements or substitutes to those sectors are not
much impacted by convergence of unilateral conduct rules to Article 102 TFEU.
Economic theory does not identify mechanisms by which other enterprises, or
potential enterprises, could be affected by rules or changes in rules in markets
unrelated to those where these enterprises operate or potentially operate.
Therefore, we conclude that the effects on businesses not directly affected by
stricter unilateral conduct rules would not be significantly affected by
convergence to Article 102 TFEU.

4.3.2 The impact on consumers

Weakening of the abuse of economic dependence prohibition, if this were a result

of convergence of unilateral conduct national rules to Article 102 TFEU, would
likely have little effect on consumers. Economic theory predicts that, if retailers

PAGE 170 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

have no market power, then changes in negotiating power upstream have no

effect on consumer prices. It may be the case the variety and innovation suffer, if
the return to innovation by suppliers is diminished as a result of retailers’
shifting more rent in their own favour.
Change in variety and in the speed and direction of innovation cannot be
quantified, and the direction of change is unclear. Both suppliers and retailers
innovate and affect the choice of variety presented to consumers. Suppliers
complained that retailers would appropriate their innovations, and retailers
would point out that they have closer contact with consumers so are more likely
to identify innovation that consumers want. Innovation is risky: Many new
products do not survive in the market.
Economic theory suggests that more innovation results when higher profits to
innovation are more appropriable the profits are to the innovator. If the
innovator profits also from substitutes, then it has less incentives to innovate as
its sales would be cannibalized. Similarly, if it profits also from complements, it
has higher incentives to innovate. If there is “too much” competition, then the
innovator cannot capture sufficient profits from its innovation. Thus, more
imitation by retailers of supplier’s products discourages supplier innovation as
some consumers will choose a cheaper imitation in preference to the innovator’s
product. But suppliers without a broad range of products may have less incentive
to innovate as they calculate the cost of “cannibalizing” sales of existing
substitutes. By contrast, if retailers sell complements, they would have increased
incentives to innovate. If unfair competition laws protect innovators from
imitators as well as abuse of economic dependence laws do, then there would
seem to be no effect if convergence to Article 102 TFEU weakened the AED rules,
so long as the unfair competition rules on imitation remained unaffected. The
other effects—of selling complements versus selling substitutes, of dispersing
profits via upstream versus downstream competition, and the degree to which
the AED rules in fact shifted rents between suppliers and retailers—are too
complex to work out a net direction of change.
Consumers could be affected by a possible extension of the convergence rule to
national competition laws concerning unilateral conduct by changes in price, in
variety or speed and direction of innovation. The possible change in price can be
estimated from several studies that examined the effect of removal of sales below
cost prohibitions on retail prices. If convergence implied that such prohibitions
would be removed, then they serve to estimate the effect on price of convergence
in those Member States with an effective sales-below-cost prohibition.
Consumers in those Member States with effective sales-below-cost laws would
likely benefit by a decrease in prices of a few percentage points, if convergence to
Article 102 TFEU had the effect of removing these laws, with the important
exception of instances of predatory pricing.

4.3.3 An econometric analysis of the impact of sales below cost rules

We have been asked by the European Commission to provide quantitative

estimates of the effects of the various divergent unilateral conduct rules on
market actors and consumers. Quantitative estimates require economic models
that produce testable hypotheses and the appropriate data on which to perform
those tests. A substantial change in a law on unilateral conduct can form a sort of

PAGE 171 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

natural experiment; if economic models predict such a change would result in the
change in an observable statistic, then we can compare that statistic before and
after. But other changes, too, in the economic environment can generate changes
in the value of the statistic. The methodology, “difference in differences,” can
largely overcome such confounding changes324. This method takes advantage of
instances where there is a natural control group in a natural experiment. It asks
whether there is a before-and-after change in the difference between the treated
group and the control group.
For example, the treated group may be products that are subject to a law and the
control group similar products that are not subject to that law. This is the
methodology we apply to estimate the effect of one substantial change in
unilateral conduct laws in European Member States.
The Irish Groceries Order prohibited sales below invoice price of several grocery
items. Other grocery items were not subject to the Groceries Order. The
Groceries Order was removed on 20 March 2006 and amendments were made to
the Competition Act to address certain business practices in the groceries sector.
Together, these constitute a natural experiment—removal of the Groceries
Order—and a natural control group—those groceries never covered by the
Groceries Order. We asked whether prices of the covered items had been
artificially raised as a result of the Groceries Order. We found that they had been,
or more specifically, that after the Order was removed the prices of the formerly
covered items fell by 1.7 to 3.0 percent as compared with the prices of the non-
covered items. Our results largely confirm the results of the study by the Irish
Competition Authority in 2008.
We examined and rejected as unsuitable for analysis, for various reasons, a
number of other changes in national unilateral conduct laws.
 Abuse of economic dependency rules should, if they do what is on the
package, result in more or more profitable dependent suppliers. But we did
not identify any publicly available statistic that would allow us to measure
whether this result occurred: The relevant issues include sales terms well
beyond prices, and entry/exit data are unsuitably coarse. Further, many of the
most-cited laws were too recent for much data to cover the period after they
came into force to be publicly available.
 Liberalization of rules against tying, from per se to Article 102 TFEU, should
in principle result in more efficiency-enhancing, i.e., cost-reducing or buyer-
attracting, tying. It was hard to identify a statistic to measure this.
 Rules on sales below cost changed in a number of Member States over the
past decade. Economic theory generates a testable hypothesis, that prices will
be higher where sales below cost are prohibited. But to disentangle sources of
changes in prices, information about possible sources is required. This is
extremely data intensive and subject to various errors. It is far more efficient
if it is possible to identify a control group, subject to the same influences as
the studied group with the only difference being subject to, in this case, rules
on sales below cost. Such a control group was present in the Irish case, but
was absent in all the other Member States whose laws we examined. In the

324 See Annex V for a detailed explanation of the methodology and references to the relevant
literature.

PAGE 172 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

other Member States, the sales below cost rules applied throughout the
groceries or retail sector, so there was no control group whose prices could be
used for comparison.
We used the “Difference in Differences” estimator. Advantages of this approach,
as identified by, for example, Imbens and Wooldridge (2008),325 include that the
double differencing (subtracting the average gain over time of the “treated” group
from the average gain over time of the “control” group) removes biases in
comparisons between the groups in the second period that result from
permanent differences between the groups, or biases in comparisons over time in
the treated group that are the result of time trends unrelated to the treatment.326
The data we used, provided by the Central Statistical Office in Ireland, consist of
monthly consumer price indexes (CPIs) constructed specifically for the study for
the period December 2001 through November 2007. Sub-indices were
constructed, one that tracks prices for the 150 Grocery Order (“GO”) items and
one that tracks prices for the 71 Non-Grocery Order (“NGO”) items.327 Moreover,
five products categories, i.e. meat, fish, fruits, vegetables and non durable
household goods, include some items that were GO and others that were NGO.
Sub-indices for these product categories were also constructed.
The equation for the baseline model can be defined as follows:

(C1)

Where Gi is a treatment group dummy which equals 1 if the item was previously
included in the Grocery Order (GO items)328 and 0 otherwise, is a time dummy
that equals 1 in the post-treatment period, i.e. April 2006 in our case and εit is a
random disturbance.
The dependent variable is the log of ‘constructed CPI ‘for GO items and NGO
items, as defined above. The log is used so as to express the DID estimator, hence
the effect of the law, in percentage terms.

325 Imbens G.W., Wooldridge J.M. (2009), “Recent Developments in the Econometrics of
Program Evaluation”, Journal of Economic Literature, vol. 47, no. 1, pp. 5-86.
326 Another author expresses part of these ideas as, “An important virtue of the DID estimator is

that it remains consistent even if the treatment is correlated with the initial level of outcome
before the experiment.” (Muravyev 2009)
327 For a more detailed description go to Section II of the Grocery Monitor Report No.2.
328‘Grocery goods’ were defined as “food and drink sold for human consumption and households

necessaries”. As explained in the Grocery Monitor report, the definition corresponds most closely
with what households consider as groceries. Neither the definition of grocery goods used in the
1987 Order nor the one contained in the Competition (Amendment) Act 2006 (‘the Amendment
Act) were used in the report. While the former was not used since it did not include fresh produce
such as fruit, vegetables, meat and fish and the exclusion of these products would have made the
report much less meaningful, the latter was not used either because it defined ‘grocery goods’ as
“food and drink sold for human consumption that is intended to be sold as groceries”. Hence, this
definition excludes household necessaries (such as tooth brushes, shampoo and washing-up
liquid) that are normally considered to be grocery goods by final consumers. The definition of
‘grocery goods’ contained in the Competition (Amendment) Act 2006 was not used because it also
included intermediate goods that are used in the production of grocery goods for retail to
consumers.

PAGE 173 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Furthermore, we assumed that there is no selection bias in our data. Hence, we

tested the assumption that retailers did not modify their behaviour before the
change in the law (e.g. when the change in the law was announced in November
2005)329, but only after the change in the law came into effect.330
The null hypothesis we test is whether the DID estimator in equation (C1) is
significantly different from 0.
We performed a number of tests to uncover whether the experiment would yield
misleading result. One test was to ask whether there were observable
characteristics of the grocery items that could influence the dependent variable:
For the DID estimator to be unbiased, such characteristics would need to be
included in the regression model as control variables. The relative homogeneity
of the sample—all items are “grocery goods”—meant we could only identify one
way to divide the items into groups, that is, fresh and processed items, i.e.,
processed food, cleaning and cosmetic products. As expected, we did not get
significantly different estimates when controlling for these additional,
idiosyncratic characteristics.
A further robustness check was to use several different event windows, that is,
different lengths of periods before and after, respectively, the “event” of the
change in the law. As discussed in the literature, the choice of event windows can
be ambiguous.331 In our case, varying the event windows did affect the magnitude
of the measured effect, but not the sign, that is, it did not affect the direction of
the effect of the change in the law.
A final robustness check was to check whether the law had similar effects on each
product category. Separate sub-indices were constructed for those five products
categories, mentioned above, that contained both GO and NGO items. DID was
estimated for these product categories, separately, but this risked so reducing the
number of observations that the estimate becomes statistically less significant.
Multicollinearity among the explanatory variables was also tested for through a
correlation test, and they were found not to exhibit multicollinearity.
We first estimated equation (C1) above, which we define as our baseline
equation. We used a variety of event windows, ranging from 3 to 6 months before
the change in the law and from 3 to 9 months after the change in the law. (Recall
that the study by the Irish Competition Authority found that relative prices had
adjusted nine months after the change in the law.)332 As noted above, we also
estimated an “augmented” equation with control, that is, item-specific, dummy
variables, but this and other robustness checks are presented in the annex.

329 See Annual Report on Competition Policy developments in Ireland, OECD 2006.
330 In order to test this assumption, we ran regressions for which the “event” was taken as
November 2005, when the change in the law was announced and, as expected, observed no
abnormality.
331 See for example event studies in the corporate finance literature where Zhang (2007) and

Litvak (2007) find a totally negative effect of SOX on firm value while Li et al. (2008) and Jain
and Rezaee (2006) report a total positive effect with some indication that the difference in the
findings may be due to the different event windows chosen (Chhaochharia and Grinstein 2007; Li
et al. 2008).
332 Some of the results are shown in Annex V.

PAGE 174 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Table 4 below reports the results from the OLS estimation of the baseline
equation, (C1). The table is split into three columns; each column shows the
estimated coefficients obtained when using different event windows.
All of the variables are highly significant, i.e. at 1% level, and the coefficients are
comparable. In each regression conducted, including those not shown, the result
was the same: The prices for the items not covered by the Groceries Order, i.e.,
the NGO items, increased faster than the prices for the GO items after the Order
was removed. The difference in prices ranges from 1.7% to 3.0%.333 In other
words, the removal of the law prohibiting sales below cost had a clear effect.
This result is in line with the finding reported in the Irish Monitor Report. They
had found that, in the period April to December 2006, the price of GO items fell
by 1.5% while the price of NGO items rose by 2.4%.334
The negative sign of the DID estimate indicates that prices of GO items increased
less than prices of NGO prices after the Groceries Order had been removed. This
contrasts with the growth path in constructed CPIs for GO and NGO items in the
period leading up to these events, December 2001 to March 2006, the CPI GO
items increased and the CPI for NGO items decreased. Hence, the change in the
law had a clear effect.
This result and others reported in Annex V, shows that removal of the Groceries
Order, and with it the prohibition of sales below cost, resulted in lower prices for
those grocery products that had been covered by the Groceries Order. The GO
had been keeping consumer prices artificially high by 1.7 to 3.0 percent.
Econometric studies reported elsewhere have also found that broad sales below
cost prohibitions raise prices. These results together suggest that their removal
would have positive effects on consumers, who would be the main beneficiaries
of lower retail prices.

Table 4- Baseline Equation (C1)

Model Equation (C1)

(a) (b) (c)

Event window 2 Event window 3 Event window 4
ALL PRODUCTS
Jan 06 – Jun 06 Jan 06 – Sept 06 Jan 06 – Dec 06
(-3,+3 months) (-3,+6 months) (-3,+9months)

Constant 4.557*** 4.556*** 4.557***

(.006) (.005) (.005)
Treatment (β1) 0.059*** 0.059*** 0.059***
(.006) (.006) (.006)

333 For regressions where the event window was extended to nine months after the change in the
law, the price differences were even bigger – i.e. from 3.2 to 3.9% - for event windows in which we
increase both the pre- and post- event windows. Results for them are presented in the appendix.
334 See Irish Grocery Monitor Report No. 2, point 3.4 and 3.5. , p. 15.

PAGE 175 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Time (β2) 0.020*** 0.023*** 0.028***

(.008) (.006) (.006)
-0.017*** -0.022*** -0.030***
DID Estimate ( τdid)
(.004) (.007) (.007)
Observations 120 180 240
Adj. R-squared 0.570 0.479 0.424
F test 41.77 55.95 59.63
3.12 3.11 5.13
F critical value
( 3, 116) ( 3, 176) ( 3, 236)
Estimation
OLS OLS OLS
technique
Regression with robust standard errors for correction of heteroskedasticity in
parentheses
*** denotes significant at 1%; ** significant at 5%;* significant at 10%

4.3.4 The impact on public authorities

National competition authorities do not expend a significant fraction of their

resources on the enforcement of national competition rules on unilateral conduct
that diverge from Article 102 TFEU. Consequently, convergence of national
competition laws on unilateral conduct to Article 102 TFEU would have little
economic impact on these authorities.
Most national competition authorities in the European Union submit annual
reports containing inter alia data on their budgets and personnel to the
Competition Committee of the Organisation for Economic Cooperation and
Development. These provide an upper boundary of the cost of enforcing
divergent unilateral conduct rules since they provide the cost to the NCAs of the
entire area of competition policy and law.335 With one exception, annual budgets
range from €21.7 million downwards. (An annex contains the data extracted from
the annual reports for 2009 or for the period 1.07.2008-30.06.2009.)
The total cost of an NCA would be a gross over-estimate. First, of those that
breakdown their personnel according to area of the law (such as mergers,
antitrust, state aid), enforcement against anticompetitive practices accounts for
only one-third or one-half of personnel.336

335 The largest sum reported, by far, was that of France’s DGCCRF, which had an allocation to
“Competition regulation of markets” in 2009 of €80 million. By contrast, the French Autorité de
la concurrence had a budget of €19.3 million, and the Bundeskartellamt of €21.7 million for the
twelve months ended 30 June 2009. The Czech Republic Office of Economic Competition
reported a competition-related budget of €5.36 million.
336 Personnel costs account for a large fraction of total costs; only if some area involved

disproportionate use of non-personnel resources, or of particularly costly personnel, would it be

inappropriate not to attribute costs in proportion to personnel numbers. For example, France’s
Autorité de la concurrence attributes € 13.8 million of the total €19.3 million budget to personnel.
Only a few NCA provide detailed personnel numbers. In Bulgaria, 10 of the 95 experts are
devoted to unfair competition cases. The 2009 budget was about €6.3 million. If costs are
proportionate to number of experts in an area, then the cost of the unfair competition
enforcement is about €663,000. Another 14 of the 95 experts address abuse of dominance cases.

PAGE 176 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Given that this category also includes horizontal and vertical agreements,
unilateral conduct or unfair competition would account for only a part of that
fraction. Second, the incremental costs of enforcing stricter unilateral conduct
rules were generally reported as being small or nonexistent. One, France’s
Autorité de la concurrence pointed out that it brought very few cases in the area.
Another, the Irish Competition Authority estimated the cost equivalent to one
full-time case officer. The Bundeskartellamt said that it did not collect
information on norm-specific costs of investigation, and pointed out that the
legal presumptions of dominance had little impact on enforcement costs since it
had a duty to investigate every case without solely relying on legal presumptions.
The Slovak Antimonopoly Office does not have responsibility for enforcing the
laws imposing stricter unilateral conduct rules than Article 102 TFEU. The one
response pointing in the opposite direction was the Hellenic Competition
Commission, which noted that one of its arguments for the abolition of stricter
national rules on unilateral conduct was the impact of its enforcement on
caseload.
In summary, the national competition authorities themselves either estimated
that their actual additional costs were small, or pointed out that they had a duty
to investigate that did not solely rely on legal presumptions, such as market share
thresholds. Thus, we conclude that the direct economic impact on national
competition authorities of a possible extension of the convergence rule to
national competition laws concerning unilateral conduct would be small.

4.3.5 The impact on broader economic variables

The terms of reference requested an estimate of the impact of the divergent

unilateral conduct rules on macroeconomic variables such as competition in the
internal market, competitiveness, innovation, growth and jobs. A few economists
have tried to estimate the impact of the existence a competition law and policy on
GDP, but acknowledge the uncertainty of the project337. It would be
correspondingly more difficult and more uncertain to estimate the impact of
variations in unilateral conduct rules. The impact of such variations on growth,
which fundamentally results from innovation applied to resources including
labour (jobs), is unlikely to be detectable.

Raw data are from “Annual Report on Competition Policy Developments in Bulgaria 2009,”
DAF/COMP/AR(2010)5. In the Czech Republic, 38 of 126 employees work on enforcement
against anticompetitive practices and advocacy: the competition related budget was €5.36
million. (Annual Report on Competition Policy Developments in Czech Republic 2009,”
DAF/COMP(2010)12/26) In France, the DGCCRF devotes 135 employees to anticompetitive
practices and 146 to restrictive commercial practices, out of a total of 342. (Annual Report on
Competition Policy Developments in France 2009,” DAF/COMP/AR(2010)10). In Ireland, 22 of
46 employees deal with enforcement of anticompetitive practices. (Annual Report on
Competition Policy Developments in Ireland 2009, DAF/COMP(2010)12/10) In Slovakia, the
comparable figures are 24 of 73 employees. (Annual Report on Competition Policy Developments
in Slovak Republic 2009, DAF/COMP(2010)12/29)
337 For a survey, see Renda et al. (2008), at
http://ec.europa.eu/competition/antitrust/actionsdamages/files_white_paper/impact_study.pd
f, section 2.5. And also add the Study by Buccirossi et al. For DG ECFIN, now summarized in
Buccirossi, Paolo, Ciari, Lorenzo, Duso, Tomaso, Spagnolo, Giancarlo and Vitale, Cristiana,
Competition Policy and Productivity Growth: An Empirical Assessment (September 2009). CEPR
Discussion Paper No. DP7470. Available at SSRN: http://ssrn.com/abstract=1484503.

PAGE 177 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

Evidence from various studies338 indicates that prohibitions of sales below cost
reduce competition in retailing. But it is worth noting that in some cases such
prohibitions were combined with a separate restrictive law, a prohibition of
discrimination, in a way that created a retail price floor. If such prohibitions were
removed, then retail prices are estimated to fall by one to almost four percent.
And this price fall would be the result of increased price competition.
The effect of prohibitions of the abuse of economic dependence or superior
bargaining position is limited to the market for the supply of retailers if the retail
market is itself competitive. This suggests rather limited macroeconomic effects.
If the retail market is not very competitive, then the results of the upstream
bargaining are partly passed through to consumers. Innovation in food products
is suppressed if suppliers expect not to reap sufficient benefits from their
innovations. But it is not clear whether removal of AED-type prohibitions, if that
would be a result of convergence with Article 102 TFEU, would have a significant
effect on innovation in light of the existence of other laws such as those
protecting intellectual property and protecting against unfairly imitating
packaging.

4.4 Conclusions
This report is to estimate whether stricter national competition rules than those
of Article 102 of the TFEU, as permitted under Article 3 of Reg. CE 1/2003, have
an economic impact, and what would be the economic impact of their
convergence with Article 102 TFEU. A key question, which was not within the
scope of this study, was precisely which rules came within the scope of the
convergence exception of Reg. CE 1/2003. The boundary between those stricter
national laws on unilateral conduct that are inside —“competition rules”— and
outside is left to a case-by-case analysis because the boundary is defined by the
objective(s) of the national laws and of Article 102 TFEU, and these objectives are
often open to interpretation. A restrictive definition would look at ‘pure
competition law rules’ only. In this respect, it is important to note that Article
4(3) TEU obliges national authorities to avoid interfering with EU objectives,
including competition, and therefore, already at this stage, it may prohibited for a
national authority to adopt be a stricter standard in the application of the
national provisions equivalent to Article 102 TFEU. An expansive definition of
what it means to “protect competition on the market” indeed would bring very
many laws within the scope of the convergence provision. In light of the ongoing
work elsewhere to reduce barriers to the internal market, and the benefits of
advocacy for reduction of unnecessarily competition-suppressing rules,
information which arguably lie outside the scope of this study is nevertheless
reported.
Our study considered the impacts of different national rules. Stricter definitions
of dominance in some Member States, e.g. Germany, imposed costs on some
businesses, they said, but they were unable to provide estimates of those costs
and did not provide specific examples of how they adjusted their conduct.
Businesses active across Europe and the world, which might be caught by the
stricter definitions, adjusted their conduct everywhere to comply with their

338 Section 3.3 reviews many of these.

PAGE 178 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

understanding of the “special responsibility” of bearing a label of having

significant market power. Convergence with Article 102 TFEU would imply that
more enterprises would feel less constrained. Several large enterprises expressed
support for convergence to Article 102 TFEU, citing a reduction in business
chilling, that is, refraining from engaging in profitable and lawful conduct
because of a concern or uncertainty of how rules will be enforced. Many of the
enterprises expressing a view on this topic indicated that they already conform to
a single standard of conduct in Europe if not worldwide. This would indicate little
direct savings from standardization per se. Note, however, that this can be taken
only as an indication as the response rate was very low compared with the
number of enterprises expected to be affected by the stricter rules.
National rules prohibiting abuse of economic dependence or similar concepts are
a major concern of grocery retailers and their suppliers. Suppliers, whom these
rules are purported to protect, saw these rules as ineffectively enforced at
present, but were concerned that convergence would remove the prospect for
enforcement of these rules in the future by national competition authorities.
These suppliers felt they would be significantly negatively impacted, even to the
point of some going out of business, if national rules were obliged to converge to
Article 102 TFEU. This conclusion is based on interviews with a handful of
European and national associations representing suppliers. While the response
rate is very low, the associations whom we interviewed provided a consistent
view.
Large grocery retailers operating in several Member States, of which we
interviewed a relatively large fraction, found the rules prohibiting abuse of
economic dependence, or similar concepts, to significantly increase their costs of
doing business in the Member States that had implemented them. They
identified their costs of bringing their operations into compliance with stricter
national rules on unilateral conduct, and their on-going costs of compliance.
Modifications of contract terms incurred the costs of renegotiation. Personnel
had to be retrained. Accounting and invoicing systems had to be modified. In
addition to the one-time costs of adapting to new or amended legislation, the
results of the contract renegotiations had significant negative impacts on profits
as a number of costs could no longer be passed onto suppliers. Two large retailers
estimated the monetary cost of compliance and of foregone profits, and these two
estimates were used to scale up the costs to an EU-wide estimate. While the
assumptions made in order to scale up were not the only reasonable ones, they
did show that the costs of stricter national rules on unilateral conduct on large
grocery retailers are significant.
The stricter national rules on unilateral conduct had unintended consequences.
One, according to large retailers, was to make purchases from SMEs less
attractive, as retailers sought to reduce their exposure to potential claims of
abuse of economic dependence. Another, according to the large foreign retailers,
was to restrict foreign retailers’ conduct more than the conduct of domestic ones,
as foreign retailers tended to wish to apply a standard system that, for example,
had supplier contract terms that would violate the legislation. In addition, foreign
retailers perceive that they tend to be subject to larger maximum fines than
domestic retailers, as their total worldwide turnover, taking into account
turnover in foreign markets, exceeded the turnover of domestic retailers.

PAGE 179 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

The concern of large retailers operating in several Member States is not limited
to the avowed intended effects of the stricter national rules on abuse of economic
dependence.
They are also concerned with the national rule-making process, including
whether they are consulted about proposed rules, whether the impact of the
proposed rules is assessed, whether the proposed rules fit into the existing legal
framework, and whether the rules are sufficiently understandable to be complied
with by those to whom they apply. Some were concerned with the high cost of the
system of rules in certain member states, which includes bureaucracy, lengthy
proceedings and other obstacles towards speedy and proper law enforcement. .
In summary, the rules on abuse of economic dependence are costly and do not
appear to, in practice, aid the small local food suppliers they are designed to
protect. To the extent that the rules impose additional costs, they also distort
competition. Convergence with Article 102 TFEU would reduce these costs and
distortions.
Stricter national rules prohibiting sales below cost had varying degrees of
effectiveness. Some large retailers, whose pricing behaviour these rules were
intended to constrain, claimed to find some of these rules easy to circumvent and
others had adapted their business models to comply with the rules. By contrast,
others thought their competitiveness had been harmed and that market prices
were elevated as a result. The interviews of businesses thus provided only a
mixed view on the impact of these stricter national rules. The econometric
analysis we performed supported the view that retail prices are higher by one to
three percentage points for products where sales below cost is prohibited. The
result is also consistent with other econometric work performed by others, as
well as theoretical economic models.
The intended beneficiaries of sales below cost prohibitions in retail trade are
generally small retailers. But our interviews, as well as investigations performed
earlier by the UKCC in Britain, found that these prohibitions conferred little
benefit on those retailers. They tend not to compete head-to-head with large
retailers, and they tend to pay higher prices to suppliers.
Convergence of these stricter national rules on sales below cost with Article 102
TFEU may well lead to a small reduction in retail prices, according to our and
others´ econometric estimates. Small retailers, however, would lose whatever
little benefit the rules provide.
Any impact of stricter national rules against tying was difficult to identify. No
respondent mentioned a concern with such rules, although one network
infrastructure company mentioned that rules on conditional sales constrained its
freedom to price. Many of the tying cases described in the legal analysis
concerned telecommunications, so this may have been an area where stricter
national rules have an effect. But if a convergence obligation reduces the scope
for stricter national competition rules on tying, then the telecommunications
regulatory regime could easily substitute. From a theoretical point of view, the
cost of stricter national rules against tying is the absence of pro-competitive
tying. This is difficult to identify without examining a particular market where
efficient tying has been prohibited in some markets but is pervasive in others. No
such market was identified. Thus, we make no estimate of the impact of

PAGE 180 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

convergence of stricter national competition rules against tying but the absence
of complaints on the subject may indicate that it would be small.
Stricter national competition rules aimed at specific network infrastructure
sectors, such as electricity, pipeline gas, and telecommunications, were not, with
one exception, further examined in this report. The sector-specific stricter rules
on unilateral conduct can be seen within a broader context of regulatory reform
in these sectors, where the boundary between economic regulation and
competition has shifted in the past decades and continues to change at different
rates in different Member States. Several of the responses received from
enterprises active in these sectors did not distinguish stricter competition rules in
sector specific laws or in general competition laws, and beyond mentioning that
the rules restricted the freedom to price, many of the responses did not indicate
whether or how these stricter rules imposed costs on them. However, where a
respondent described how a sector-specific provision in one Member State’s
competition affected its conduct, and the associated costs, we did include it.

PAGE 181 OF 182

 STUDY ON NATIONAL RULES ON UNILATERAL CONDUCT THAT ARE STRICTER THAN ARTICLE 102 TFEU

5. MAIN FINDINGS
 There is a remarkable fragmentation of legal rules on unilateral conduct, which
affects as many as 18 Member States. Legal rules on unilateral conduct are
remarkably fragmented in the EU27, with a minority of EU Member States
featuring national legislation broadly aligned with EU competition law, and as
many as 18 countries exhibiting different types of diverging rules or standards
that are stricter than Article 102 TFEU.
 The enterprises that appear to be most affected by the current legal
fragmentation are of three types: enterprises active in certain Member States in
grocery retailing, and their suppliers; suppliers of intermediate products; and
enterprises active in network infrastructure such as electricity, pipeline gas, and
telecommunications.
 Such fragmentation can hamper the internal market, in particular by increasing
legal uncertainty and raising compliance costs for affected undertakings.
Companies wishing to engage in cross-border trade are hindered by (i) the
existence of a wide divergence in the interpretation of the concepts of dominance
and abuse thereof; (ii) the legal uncertainty surrounding the exact features of
national legislation and its enforcement; and (iii) the need to change contracts
and commercial conduct to adhere to national rules.
 Most national rules that are stricter than Article 102 TFEU do not predominantly
pursue the same objectives of Article 102 TFEU. As a matter of fact, many of
these rules are conceived to protect certain competitors, rather than the
competitive process. In some cases this is due to consolidated national traditions,
but in other cases laws are simply transplanted from other legal systems.
 Most national rules that are stricter than Article 102 TFEU appear to be harmful
for consumers. This is due to reduced product variety created by the lack of a
Single Market; but also to the fact that many of the stricter rules on unilateral
conduct are meant to protect smaller competitors, rather than the competitive
process, most often leading to price increases for end consumers. This is
particularly the case for national rules on abuse of economic
dependence/superior bargaining power, as well as rules on sales below cost,
which – as we estimated through an econometric analysis – lead to artificial price
increases of up to 3%.
 Extending the convergence rule of Article 3(2) Reg. CE 1/2003 to Article 102
TFEU would not have significant negative consequences for consumers,
businesses, and public authorities. National competition authorities would not be
significantly affected; small retailers that are meant to be protected by some rules
on sales below cost would not be put at a disadvantage; and consumers would
gain in a number of respects, as a result of the removal of existing fragmentation
in national rules.
 At the same time, extending the convergence rule of Article 3(2) Reg. CE 1/2003
to Article 102 TFEU would only partially solve the problem of fragmentation,
since national governments might retain an incentive to pursue objectives that
are inconsistent with EU competition law and the Single Market through
legislation and regulation, which is not covered by the convergence rule.

PAGE 182 OF 182