Essential Facilities Doctrine in EC and Telecommunications

ESSENTIAL FACILITIES DOCTRINE UNDER
EC COMPETITION LAW
AND
PARTICULAR IMPLICATIONS OF THE DOCTRINE
FOR
TELECOMMUNICATIONS SECTORS IN EU AND TURKEY
A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF SOCIAL SCIENCES

OF MIDDLE EAST TECHNICAL UNIVERSITY
BY
MEHMET B LAL ÜNVER
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

FOR
THE DEGREE OF MASTER OF SCIENCE
IN
THE DEPARTMENT OF EUROPEAN STUDIES
SEPTEMBER 2004
Approval of the Graduate School of Social Sciences
___________________
Director
(Prof. Dr. Sencer AYATA)
I certify that this thesis satisfies all the requirements as a thesis for the degree of
Master of Science
___________________
Head of Department
(Prof. Dr. Ali G TMEZ)
This is to certify that we have read this thesis and that in our opinion it is fully
adequate, in scope and quality, as a thesis for the degree of Master of Science.
___________________
Supervisor
(Assist. Prof. Dr. Gamze A CIO LU ÖZ)
Examining Committee Members
Assoc. Prof. Dr. Erol TAYMAZ ___________________
Assist. Prof. Dr. Gamze A CIO LU ÖZ ___________________
Assist. Prof. Dr. Gül EFEM ___________________

PLAGIARISM
I hereby declare that all information in this document has been obtained and
presented in accordance with academic rules and ethical conduct. I also declare that,
as required by these rules and conduct, I have fully cited and referenced all material
and results that are not original to this work.
Name, Last name : Mehmet Bilal ÜNVER
Signature :
iii
ABSTRACT
ESSENTIAL FACILITIES DOCTRINE

UNDER EC COMPETITION LAW
AND
PARTICULAR IMPLICATIONS OF THE DOCTRINE
FOR
TELECOMMUNICATIONS SECTORS IN EU AND TURKEY
Ünver, Mehmet Bilal
M. Sc., Department of European Studies
Supervisor: Assist. Prof. Dr. Gamze A ÇIO LU ÖZ
September 2004, 181 pages
In this study, the origin and main parameters of the Essential Facilities Doctrine are
analysed through the case-law that developed out of the application of the EC
Competition Rules. Besides putting forward the historical roots, the basic criteria and
limitations that apply to the Doctrine are elaborated so as to clarify the legal and
analytical foundations of the Doctrine in the EU context. In addition, the added value
attributed to the Doctrine in realm of competition policies pursued in network-based
industries is expounded with special emphasis on telecommunications sectors. With
this regard, the potential role of EFD against the challenging effects of ‘convergence’
phenomenon and the technological changes is discussed. At last, the effects of EFD
on the competitive dynamics of Turkish telecommunications sector which is
undergoing a liberalisation process are also examined with the accompanied Turkish
case-law.
Keywords: Essential Facilities, Refusal to Deal, Indispensability, Competition

Policies, Access, Convergence, EC Competition Rules, Competition law remedies,
Sector-Specific Regulation
iv
ÖZ
AT REKABET HUKUKU’NDA ZORUNLU UNSUR DOKTR N

VE
DOKTR N N AB VE TÜRK YE TELEKOMÜN KASYON SEKTÖRLER NE
YÖNEL K YANSIMALARI
Ünver, Mehmet Bilal
Yüksek Lisans, Avrupa Çalı maları
Tez Yöneticisi: Yrd. Doç. Dr. Gamze A ÇIO LU ÖZ
Eylül 2004, 181 sayfa
Bu çalı mada, AT Rekabet Kurallarını esas alan mahkeme içtihatları ı ı ında geli en
Zorunlu Unsur Doktrinin orijin ve temel esasları incelenmektedir. Doktrinin analitik
ve hukuksal dayanaklarının açıklı a kavu turulması amacıyla tarihi kökleri ile
Doktrine uygulanacak sınırlamalar ayrıntılı olarak ele alınmı tır. Bunun yanında,
ebeke endüstrilerinde takip edilen rekabet politikaları kapsamında Doktrinin
uygulanması ile ortaya çıkan katma de er de özellikle telekomünikasyon sektörü
kapsamında inceleme konusu yapılmaktadır. Bu çerçevede, Zorunlu Unsur
Doktrininin sektörler arası ‘yakınsama’nın sarsan etkileri ve teknolojideki
de i iklikler kar ısındaki potansiyel rolüne ili kin olarak da tartı maya yer
verilecektir. Son olarak, Türk Rekabet Kurumu’nun kararları ı ı ında Doktrinin
liberalizasyon sürecinden geçen Türk telekomünikasyon sektörünün temel
dinamiklerine olan etkileri ara tırma konusu yapılacaktır.
Anahtar Sözcükler: Zorunlu Unsur, Anla ma Yapmayı Reddetme, Kaçınılmazlık,

Rekabet Politikaları, Eri im, Yakınsama, AT Rekabet Kuralları, Rekabet Hukuku
Tedbirleri, Sektöre Özgü Düzenleme
v
DEDICATION
To my lovely wife who deserves forever love
vi
TABLE OF CONTENTS
PLAGIARISM........................................................................................................ iii
ABSTRACT ........................................................................................................... iv
ÖZ ............................................................................................................................v
DEDICATION ....................................................................................................... vi
TABLE OF CONTENTS....................................................................................... vii
CHAPTER
INTRODUCTION ...........................................................................................1
1. ESSENTIAL FACILITIES DOCTRINE IN COMPETITION LAW ............4
1.1 The Rationale and Parameters of the Essential Facilities Doctrine
(EFD) ......................................................................................................4
1.2 The US Case-Law..............................................................................8
1.2.1 Control of the Essential Facility by a Monopolist ..............14
1.2.2 A Competitor’s Inability Practically or Reasonably to
Duplicate Essential Facility........................................................17
1.2.3 The Denial of the Use of the Facility to a Competitor ........23
1.2.4 The Feasibility of Providing the Facility............................23
1.3 Assessment of the US Case-Law......................................................25
1.4 Essential Facilities Doctrine under EC Competition Law .................29
1.4.1 Main Characteristics of ‘Refusal to Deal’ Cases ................30
1.4.2 EC Case-Law on ‘refusal to deal’ ......................................32
1.4.3 Relationship between Refusal to Deal / EFD Cases and
Intellectual Property Rights........................................................37
1.4.4 The Introduction of Essential Facilities Doctrine into the EC
Competition Law .......................................................................41
1.4.5 The Attitude of the ECJ and CFI towards the Essential
Facilities Doctrine......................................................................46
1.4.6 Oscar Bronner Case: A Turning point in EFD ....................................52
vii
1.5 Assessment of the EC Case-Law......................................................58
2. IMPLICATIONS OF THE ESSENTIAL FACILITIES DOCTRINE FOR
TELECOMMUNICATIONS SECTORS........................................................61
2.1 General Overview............................................................................61
2.2 General Characteristics of Telecommunications Sectors ..................64
2.2.1 Network-Based Characteristics..........................................64
2.2.2 Network Externalities........................................................67
2.2.3 Economies of Scale ...........................................................69
2.2.4 Economies of Scope ..........................................................70
2.2.5 Economies of Density .......................................................71
2.2.6 Other Barriers to Entry ......................................................73
2.3 European Telecommunications Sector .............................................75
2.3.1 Liberalisation ....................................................................78
2.3.2 Harmonisation: ONP Directives ........................................81
2.3.3 Convergence .....................................................................85
2.4 Dual Regime in EU Telecommunications Sector: Sector-Specific and
Competition Law Rules .........................................................................88
2.5 Establishment of the EU Access Regime: Policy Objectives and
Legislative Tools ...................................................................................92
2.5.1 General Overview .............................................................92
2.5.2 Commission’s Access Notice ............................................96
2.5.3 The Rationale and Main Parameters of Applying EFD under
the EU Access Regime.............................................................103
2.5.4 The Application of EFD in Liberalisation Period: EU
Experience and Further Implications ........................................108
2.6 Recent Developments in EU after Full Liberalisation.....................113
2.6.1 Reform Process since 1999..............................................114
2.6.2 Introduction of New (2002) Regulatory Framework ........117
2.6.3 Assessment of the Recent Developments under EFD.......121
2.7 The Future Implications of EFD for Telecommunications Sectors..125
2.7.1 Convergence and Institutional Implications .....................125
viii
2.7.2 The Role of EFD within the Technological Changes .......128
2.7.3 Complementarity & Superiority of EFD ..........................131
3. IMPLICATIONS OF THE ESSENTIAL FACILITIES DOCTRINE FOR
TURKISH TELECOMMUNICATIONS SECTOR ......................................136
3.1 EFD under Turkish Competition Law ............................................136
3.2 EFD Decisions in Turkish Case-Law .............................................138
3.2.1 Eti Holding Decision .......................................................138
3.2.2 B RYAY Decision ............................................................140
3.2.3 ÇEA Decision................................................................142
3.3 EFD Cases in Field of Telecommunications...................................143
3.3.1 TTA Decision ................................................................143
3.3.2 Aria (Roaming) Decision.................................................145
3.4 Assessment of the Turkish Case-Law in light of the Community
Approach on EFD................................................................................147
3.5 Assessment of EFD Decisions in Field of Telecommunications .....148
3.6 Implications of Applying EFD for Turkish Telecommunications
Sector ..................................................................................................155
4. CONCLUSION .......................................................................................160
BIBLIOGRAPHY........................................................................................169
ix
INTRODUCTION
The purpose of this study is to determine the principles that apply to the ‘Essential
Facilities Doctrine’ under the EC Competition Law and on this basis to analyse the
particular implications of the Doctrine towards telecommunications sectors in EU
and Turkey, namely to investigate the historical and legal sources of the Doctrine,
relevant case-law and multi-dimensional impacts of the Doctrine on networked
industries with particular emphasis on telecommunications sectors.
The Essential Facilities Doctrine has its antecedents in US Antitrust Law. According
to the Doctrine, an undertaking controlling facilities which are deemed essential for
another market, abuses its dominant position, where without objective justification it
refuses access to those facilities.
Most legal systems in countries with a market economy adopt the view that firms
should be allowed to contract with whomsoever they wish. Despite this
acknowledgement, forcing a dominant undertaking in order to make available of its
own facilities to other parties is in exceptional circumstances deemed legally and
economically acceptable. Essential Facilities Doctrine, having potential to be deemed
one of these exceptions to the freedom to contract has so many impacts on the
competitive dynamics of markets with monopolistic/dominant firms. From this point
of view, the balance between exploitation of property rights and application of the
Doctrine more precisely, the debate between policies promoting economic efficiency
and those promoting more competition in the market is the leading theme of this
thesis.
The thesis covers three main sections in compliance with the title of the thesis. While
the first section deals with the nature and implementation of the Doctrine with
general repercussions of the Doctrine in jurisdictional area, the second section of the
thesis will focus on general implications of the Doctrine for networked industries,
specifically towards telecommunications sectors. In this section, European
1
telecommunications sector is taken as a basis for telecom-specific analysis regarding
the Doctrine. In the third section, relevant Turkish case-law on the Essential
Facilities Doctrine and comparative analysis between the implementations of EU and
Turkey will be discussed.
Within the first section, upon highlighting the most relevant cases in US Antitrust
Law, the stress will be placed on the EC Competition Law. After reviewing nature of
the Doctrine under EC Competition rules and the case-law, far-reaching results of the
judgments and Commission decisions will be handled meticulously. In context of the
this section, refusal to grant access to ‘essential facilities’, which is mostly deemed a
kind of ‘abuse of dominance’, is going to be analysed in the light of EC Competition
Law rules. In this framework, Doctrine-based benefits that might be added to the
traditional ‘refusal to supply’ cases will be elaborated as a key point. Accordingly,
the thesis is going to discuss how wide the Essential Facilities Doctrine should be
applied, and what limitations have to be put upon it in general terms.
In the second section of the thesis, the above-mentioned discussions will be

interrelated to the area of conflicting interests between competition law principles
and sector-specific regulations. Moreover, the ways to solve a number of problems
which concern access to the essential ‘network’ facilities and represent the most
prevailing ‘bottleneck’ cases will be discussed in perspective of the Doctrine.
Challenging effects of converging markets and future implications of the Doctrine
for sector-specific regulation as well as competition law remedies will also be
detailed therein.
Throughout the third section, it will be tried to examine the most interesting
decisions developed out of the Turkish (Competition Board) case-law and deepen the
unique parts of the decisions, thereby. In this regard, Community and Turkish case-
law on EFD will be compared so as to reach some concrete suggestions for Turkey.
The analysis then will be shifted to the current situation and future impacts of the
Doctrine for Turkish telecommunications sector.
2
Having more complexities both in technical and economic terms in comparison to
other competition law breaches, bottleneck problems or problems of access to
essential network facilities are unavoidably threatening effective competition in
relevant telecommunications markets. Even with perfect competition legislation, it is
not simple to overcome such problems where networks of the particular dominant
firms continue to exhibit natural monopoly characteristics such as scale and scope
economies, strong network externalities, etc. Regarding the bottleneck problems, the
Essential Facilities Doctrine in recent years has become a theme of central interest in
respect of both future implementation of the EC Competition Law and sector-specific
regulations in EU.
In order to evaluate the applicability of the Doctrine at the EU and national level,
legal and economic developments particularly those observed with the introduction
of the new EU Regulatory Framework will be given an important place in the thesis.
Not only structural and network-related characteristics of telecommunications sector,
but also the Community-wide regulations over this area will be handled in the thesis,
namely in the second section. The so-called developments and regulations will be
discussed in conjunction with their transformative effects on Turkish
telecommunications sector, as well. Explicitly, one of the strands of thesis will be
related to the problem of how Turkish telecommunications markets will cope with
the abovementioned bottleneck problems whilst undergoing liberalisation and
harmonisation processes along with the Acquis, in that application of the Essential
Facilities Doctrine will be the principal constituent of the discussion.
All these discussions will be circumscribed around the theme ‘third party access to
essential facilities’ and not only legal and historical but also economic and practical
consequences of the Doctrine will be tried to be expounded in a coherent manner. It
will be urged not to go beyond the scope and the purpose of the thesis outlined here
in the context of the so-called discussions.
3
CHAPTER 1
1. ESSENTIAL FACILITIES DOCTRINE IN COMPETITION

LAW
1.1 The Rationale and Parameters of the Essential Facilities Doctrine (EFD)
Competition Law serves many purposes, namely eliminating market imperfections

and harmful effects of monopolies as the main policy objectives. Other important
purposes such as promotion of consumer welfare and protection of small and
medium-sized firms could be assessed supplementary objectives contingent upon the
first ones. Main purposes of EC Competition Law could also be deducted from
Article 21 and 3 of the EC Treaty which are detailed within Articles of 81 and 82
EC2. Article 3 of the Treaty establishes that the objectives set out in Article 2 would
be achieved with the policy instruments envisaged, therein. The paragraph (g) of
Article 3 specifically refers to ‘a system ensuring that competition in the internal
market is not distorted’.
1
Article 2 EC reads as follows: “The Community shall have as its task, by establishing a common
market and an economic and monetary union and by implementing the common policies or activities,
a high level of employment and social protection, equality between men and women, sustainable and
non-inflationary growth, a high degree of competitiveness and convergence of economic performance,
a high level of protection and improvement of the quality of the environment, the raising of the
standard of living and quality of life, and economic and social cohesion and solidarity among Member
States.”
2
Article 82 EC, articulating the most common examples of abuse of dominance reads as follows:
“Any abuse by one or more undertakings of a dominant position within the common market or in a
substantial part of it shall be prohibited as incompatible with the common market insofar as it may
affect trade between Member States.
Such abuse may, in particular, consist in:
(a) directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions;
(b) limiting production, markets or technical development to the prejudice of consumers;
(c) applying dissimilar conditions to equivalent transactions with other trading parties, thereby
placing them at a competitive disadvantage;
(d) making the conclusion of contracts subject to acceptance by the other parties of supplementary
obligations which, by their nature or according to commercial usage, have no connection with the
subject of such contracts.”
4
In establishing policy objectives of EC Competition Law, historical perspective
should be embedded in legal and structural analysis. In the European context, socio-
political concerns seemingly surpassed other objectives such as those related to the
notion of ‘economic efficiency’, especially in the early days of implementation of EC
Competition Law. This was the case, though having being regarded opposite due to
the steady economic developments undergone through the integration process.
As a matter of fact, while aspiration of a peaceful and reliable supra-national system

was the foremost factor stimulating post-war European integration, understandably
the most apparent factor was economic reconstruction which initiated the member
states to co-ordinate in many fields other than economic integration. Considering the
‘integration’ imperative over which political and social concerns were predominant,
it is possible to say that, the notion of increasing ‘economic efficiency’ under EC
Competition Law was insufficiently recognised at the time when the politics of
market integration were in the ascendant.3
Having not faced with the market integration problem, the United States has become
more concerned with efficient market operation. It is argued that even the original
drafters of the Sherman Act were motivated by concerns of economic efficiency.4
When we look at the spirits of the judgments as well as their wording, an economic
theory would actually be observed as occupying a large share of the limelight in US
antitrust enforcement.5
As a corresponding matter, free market economy encompassing the ‘right to refuse to

deal’ had always a sound basis in US markets. This has been a rule of thumb unless
otherwise envisaged in an injunction or a decree of a competent authority that
requires a compulsory dealing in accordance with the US Antitrust Law. This also
3
Richard Whish, “Competition Law, Fourth edition”, 2001, Butterworths, p. 1.
4
M. H. Harz, “Dominance and Duty in the European Union: A look Through Microsoft Windows at
The Essential Facilities”, www.law.emory.edu, 1997, p. 4
5
Ibid.
5
entails another fact that a general duty to deal is unacceptable in US system.6 Not
only US, but also most legal systems with a market economy adopt the view that
firms should be allowed to contract with whomsoever they wish.
Essential Facilities Doctrine (hereinafter usually called ‘EFD’) is an important

exception to this general proposition, namely a departure from freedom to contract in
certain limited cases. That is to say, although in general, it is pro-competitive to
allow firms to keep for their own exclusive use of assets which they have acquired or
constructed, all or most developed competition laws create an exception to this
general rule.7 Explicitly, there are some exceptional circumstances in which a refusal
on the part of a dominant firm to supply goods or services can amount to an abuse of
dominant position under competition law rules.
As could be inferred here, there is a number of pre-requisites for application of

Essential Facilities Doctrine. Among others, two principal requisites of EFD would
have been met when an essential facility is owned by a dominant undertaking and a
refusal of the dominant undertaking to provide with access to the so-called facilities
takes place. Other essential requisite for application of EFD could be expressed as
the ‘investigation as to whether there is an objective justification to refuse the request
for access’. Regarding the relevant criteria, how to assess that the facility concerned
is ‘essential’ is the central pre-requisite for EFD and is therefore ought to be deemed
the core of the EFD cases.
Summarising, the obligations regarding ‘duty to deal’ or ‘duty to share essential

facilities’ arise only if the competitor cannot obtain the goods and services in
question elsewhere and cannot build or invent them itself, and unless the facility
6
Under the US Sherman Act, ‘refusal to deal’ is deemed unlawful in exceptional circumstances. In
Colgate v. U.S., the Supreme Court held that “in the absence of any purpose to create or maintain a
monopoly, the Sherman Act does not restrict the long recognised right of traders or manufactures
engaged in an entirely private business, freely to exercise its own independent discretion as to parties
with whom he will deal.” [250 U.S. 300, 307 (1919)]
7
J. T. Lang, “The Principle of Essential Facilities in European Community Competition Law – The
Position since Bronner - Notes for a lecture”, September 2000, Copenhagen, p. 2.
6
owner has legitimate business justification for the refusal.8 Given this pre-
assessment, it is possible to say that the ‘essentiality’ problem is the most
controversial problem in EFD analysis, which adds some difficulties to the
traditional ‘dominant position’ test. Beyond the definition of relevant market(s) and
determination of dominance therein, are there numerous questions to be answered
under EFD.
Another factor that makes EFD particularly important is the increase in number of
cases in which a dominant firm depends on its power related to its economies of
scope, scale and density as well as strong network externalities. In networked sectors
many dominant undertakings have networks that have been subsidised through
government expenditures, and this fact gives path for application of EFD in some
respects. Actually in Europe, critical issues regarding ‘access to essential facilities’
have arisen in connection with the liberalisation of the gas, electricity and
telecommunications industries which were in ascendant in 1990s. On the US side,
since the early 20th century in a wider spectrum has EFD found a place for itself. A
number of cases related to pipelines, power transmission networks, FM broadcasting
facilities, e-mail servers, computer reservation system for airlines, a database of
copyright invoked the Doctrine less or more.9 A remarkable feature of these cases is
presence of two markets. For instance, where respectively, CRSs (computer
reservation systems),10 wholesale electricity,11 local telecommunications network12
constitute upstream markets of an essential facility; airline passenger services, long-
distance power transmission and transmission of telecommunications services are
deemed downstream markets.13
8
Position since Bronner - Notes for a lecture”, September 2000, Copenhagen, p. 2.
9
P. Larouche, “Competition Law and Regulation in European Telecommunications”, Hart Publishing,
2000, p.177
10
Alaska Airlines, Inc. v. United Airlines , Inc., 948 F 2d 536 (9th Circ 1991)
11
Otter Tail Power Co. v. United States 410 US 366, 35 L. Ed. 2d 359, 93 S. Ct. 1022 (1973)
12
MCI Communications v. AT&T, 708 F.2d 1081, 1132-1133 (7th Cir. 1983)
13
Though prototypical formulation of EFD describes two vertically-related markets frequently called
‘upstream and downstream’, there are some counterviews over this separation which will be dealt
later. Generally saying, alike the situation in US Antitrust Law, the Essential Facilities Doctrine in EC
Competition Law is usually traced to a number of decisions of Community Courts which were
7
It is generally accepted that the Essential Facilities Doctrine was inspired from the
developments in US Antitrust Law, so it is worth starting with the discussion of the
most relevant decisions of US Courts prior to analysis of EU case-law. The pertinent
US cases would clear up the core elements of EFD and help the reader set the
differences between Article 82 EC and the Section 2 of the Sherman Act, where
somewhat different approach could be witnessed in respect of EFD.
1.2 The US Case-Law
Having a long and respected history as part of US Antitrust Law, EFD finds its origin
in the 1912 U.S. v. Terminal Railroad Association14 case. In this case, the defendant
association comprised fourteen of the twenty-four railway companies serving the city
of St. Louis. The so-called association had bought the only three possible railway
crossings across the Mississippi River. The Supreme Court held that the refusal of
the owner of a vital network such as a railway terminal, to make access available to
non-owners may “restrain (…) commerce among the States and constitutes an
attempt to monopolize commerce among the States (…)” and ordered the
Association to “provide for the admission of any existing or future railroad to joint
ownership and control of the combined terminal properties, upon a plane of equality
in respect of benefits and burdens with the present proprietary companies”.15
The Court seemed to have accepted, for purposes of its analysis, that the defendant
railroad was a monopolist in the control of an ‘essential facility’, not having called as
such, but as ‘vital network’. Court also possibly avoided from acting as a regulatory
agency, requiring the defendant pursuant to the Section 1 of the Sherman Act16 to
basically associated with the notion of ‘refusal to deal’. In context of a ‘refusal to deal’ case, the
dominant undertaking is typically present on two markets and dominant at least in the upstream
market thereby tries to exploit its dominant power on the upstream market in order to strengthen its
position on the downstream market by refusing to supply its competitors with the upstream products.
14
U.S. v. Terminal Railroad Association 224 US 383 (1912).
15
Ibid. at 515-516.
16
Section 1 of the Sherman Act, 15 USC § 1, reads as follows: Every contract, combination in the
form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the Several States
or with foreign nations, is declared to be illegal. Every person who shall make any contract or engage
in any combination or conspiracy hereby declared to be illegal shall be deemed guilty of a felony, and,
8
admit competitors to their Association rather than ordering dissolution of the
Association and/or imposing detailed obligations thereupon.
The rationale behind the Terminal Railroad judgment was extended in Associated
Press v. United States17 case. The case concerned the collaboration named
‘Associated News Organisation’ (AP) where approximately 1,200 newspapers joined
together in order for gathering, transmission, and exchange of news reports.
Membership of AP was open to all newspapers except for those competing
geographically with one of the existing members, for which more onerous
membership conditions were imposed. Regarding this situation, the Supreme Court
stated that “AP news is to be furnished to competitors of old members without
discrimination.”18 In the Court decision, it was also held that “the exclusive right to
publish news in a given field, furnished by AP and all of its members gives many
newspapers a competitive advantage over their rivals.”19 Holding that those more
onerous membership conditions violated s. 1 of the Sherman Act, the Court did not
seem to have set out a very clear rationale in its decision.20
The foundation of EFD within the single firm context has been formed with The
Otter Tail Power Co v. U.S.21 case. The Otter Tail Power Co v. U.S. is also the
foremost case where the Supreme Court has conceived of the refusal to deal as
having breached the Section 2 of the Sherman Act.22 In the context of the case, the
on conviction thereof, shall be punished by fine not exceeding $10,000,000 if a corporation, or, if any
other person, $350,000, or by imprisonment not exceeding three years, or by both said punishments,
in the discretion of the court”.
17
Associated Press v. United States 326 US 1 (1945).
18
Ibid. at 21.
19
Ibid. at 17.
20
Barry Doherty, “Just what are essential facilities?” in Common Market Law Review, 38, no. 2,
2001, p. 404. See also, P. Larouche, “Competition Law and Regulation in European
Telecommunications”, Hart Publishing, 2000, p.176, and M. H. Harz, “Dominance and Duty in the
European Union: A look Through Microsoft Windows at The Essential Facilities”,
www.law.emory.edu, 1997, p. 8-9.
21
The Otter Tail Power Co v. U.S. 410 US 366 (1973).
22
Section 2 of the Sherman Act, 15 USC § 2, reads as follows: Every person who shall monopolize, or
attempt to monopolize, or combine or conspire with any other person or persons to monopolize any
part of the trade or commerce among the several states, or with foreign nations, shall be deemed guilty
of a felony, and, on conviction thereof, shall be punished by fine not exceeding $10,000,000 if a
9
Otter Tail Company sold electricity in different towns to individual subscribers.
After the contracts expired, some towns wished to establish their own systems.
However, Otter Tail refused to sell them electricity at wholesale rates and refused to
“wheel” electricity to them. The Supreme Court stated that “The Sherman Act
requires that where facilities cannot practically be duplicated by would-be
competitors, those in possession of them must allow them to be shared on fair
terms.”23 The emphasis put on the role of intent by the Court is also eye-catching in
The Otter Tail Power Co v. U.S. judgment. In fact, the Court here decided on the
breach of the Section 2 of Sherman Act, due to the presumption of “attempt to
monopolise” on the part the defendant. The Court explicitly said that Otter Tail’s
actions “had the purpose of delaying and preventing the establishment of municipal
electric systems”24 and ordered Otter Tail to distribute power over its grid, at rates
which were compensatory. Here is revealed another form of violation of Sherman
Act apart from monopolisation, specifically being emphasised not as action but in the
form of specific intent.
Not only in the above case, but also in many other Supreme Court judgments25 is
inherent the fact that Sherman Act can be breached if the refusal to deal had been
done with the ‘intent of monopolise’. But ‘intent theory’ does not appear to have
been objectively justified so as to constitute a sound basis for distortion of
competition. According to P. Larouche, such an approach is more close to the
arguments used in EC Competition Law.26 As a factual point, in enforcement of
Antitrust Law, US Courts so far have been prone to conduct a more economics-
oriented analysis, in comparison to the Community Courts. Given the fact that US
Antitrust Law starts from the principle that refusals to deal are permissible, the end-
corporation, or, if any other person, $350,000, or by imprisonment not exceeding three years, or by
both said punishments, in the discretion of the court”
23
The Otter Tail Power Co v. U.S. 410 US 366 (1973), at 130.
24
Ibid., at 379.
25
See the following cases: Eastman Kodak Co. v. Image Technical Services Inc., 504 US 451 (1992);
Lorain Journal Co. v. US., 342 US 143 (1951); City of Anaheim, 955 F.2d at 1381; Aspen Skiing Co.
v. Aspen Highlands Skiing Corp.,472 US 585 (1985); Intergraph Corp. v. Intel Corp., 195 F.3d 13346,
1358 (Fed. Cir. 1999) (cited in P. Larouche, p. 175-178).
26
P. Larouche, “Competition Law and Regulation in European Telecommunications”, Hart
Publishing, 2000, p.176.
10
results under the s. 2 Sherman Act are able to be quite different than under Article 82
EC (ex 86).
On the other hand, the role of intent particularly emphasised in the US Antitrust Law
could be found unsurprising when considered with its main purpose. From this point
of view, one can make an observation that the language of Article 82 does not seek to
prevent dominance or monopoly, while Section 2 of the Sherman Act prohibits
monopolisation or attempted monopolisation. Under Article 82 EC, being in a
dominant position would thus not create a concern as such, but once an undertaking
has acquired a dominant position, it will be acceptably subject to scrutiny whether
for any abuse of dominance.
Going back to the Otter Tail decision, we can speak out about some conflicting
comments over the duty to deal. Some commentators suggest that Otter Tail does not
establish a general duty of deal,27 while some others conclude that for such an
exceptional circumstance, a duty to deal is imposable under EFD.28 Common critics
regarding the Otter Tail decision are related to the limits of the Court’s
jurisprudence. In this regard, it is argued that the Court should have been reluctant to
burden Otter Tail Power Co. with a detailed duty to deal as there existed a regulatory
agency to regulate terms and prices of power transmission at that time.29
In a further decision, Aspen Skiing Co. v. Aspen Highlands Skiing Corp.,30 Supreme
Court had clarified the essential facilities principles applied at Otter Tail, whereby
somewhat restrictively. The parties to this case were competitors in offering ski
27
P. Areeda, “Essential facilites: an epithet in need of limiting principles”, 58 Antitrust Law Journal,
1990, 841, no. 21, 1989, p. 844; Barry Doherty, “Just what are essential facilities?” in Common
Market Law Review, 38, no. 2, 2001, p. 401 and J. T. Soma, D. A. Forkner and B. P. Jumps, “The
Essential Facilities Doctrine in the Deregulated Telecommunication Industry” 13, Berkeley Tech. L. J.
565, 1998, p. 573.
28
M. Furse, “The ‘Essential Facilities’ Doctrine in Community Law”, European Competition Law
Review, 8, No. 8 1995, p. 470 and R. Pitofsky, D. Patterson, and J. Hooks, “The Essential Facilities
Doctrine Under U.S. Antitrust Law”, Antitrust Law Journal, Vol. 70, 2002, p. 451.
29
P. Areeda and B. Doherty Op.cit. in note 27. See also Abbott B. Lipsky, Jr. and J. Gregory Sidak,
“Essential Facilities”, Stanford Law Review, Vol. 51, No: 1187, May 1999, p. 1206.
30
Aspen Skiing Co. v. Aspen Highlands Skiing Corp 472 US 585 (1985).
11
services on Aspen’s slopes, who had a long standing revenue-sharing agreement
based on operation of a joint selling mechanism by which skiers could buy a single
ticket and ski at either resort. The defendant, Aspen Skiing Co., who was a bigger
company abandoned the agreement between themselves, refusing to accept tickets
issued by Aspen Highland Skiing Corp. for the reason of incurring great costs.
Having reached that the so-called rejection constituted a breach of Section 2 of the
Sherman Act, the Supreme Court dealt with the case within the framework of the
intent theory and business legitimacy perspective. The Court held that ‘the
monopolist did not merely reject a novel offer to participate in a cooperative venture
but had instead elected to make an important change in a pattern of discrimination
that had originated in a competitive market and had persisted for several years’.31
Although the Supreme Court did not make an explicit reference to EFD, it upheld the
decision of the Tenth Circuit Court of Appeals who described the multi-area ticket as
an ‘essential facility’ to which the defendant was denying access with the intent to
monopolize by putting the competitor ski resort out of business.32
An outcome can be inferred from the above judgment(s) that a monopolist has the
right to deny access to provide particular goods, services and facilities with its
competitors, if a legitimate business reason could be put forward as to justify the
refusal. Therefore, if there exist no monopolisation, attempt to monopolize and/or
combination or conspiracy to monopolize, refusals to deal are permissible under
Sherman Act.33 Hence, one can correlate presence of legitimate justification with
lack of monopolisation for the purpose of antitrust law. Notably to say, the fact that
monopolists have no a general duty to co-operate with its rivals could thus be
regarded as an emanation point of US Courts in their analysis as was in Aspen
decision. In this context, the reason why the Court departed from the freedom to
contract in Aspen seems to have stemmed from the absence of objective justification.
31
Aspen Skiing Co. v. Aspen Highlands Skiing Corp 472 US 585 (1985), at 603.
32
R. Pitofsky, D. Patterson, and J. Hooks, “The Essential Facilities Doctrine Under U.S. Antitrust
Law”, Antitrust Law Journal, Vol. 70, 2002, p. 448.
33
See supra note 6.
12
The most significant US case regarding essential facilities is commonly accepted as
MCI Communication v. American AT & T Co.,34 where the Seventh Court (a lower
court) set four conditions for the application of the Doctrine to unilateral refusals to
deal. Until the MCI judgment, there was no clear-cut formulation of EFD and the
phrase, ‘essential facility’ had not been used by the Supreme Court in refusal to deal
cases, which also exhibits an avoidance of the Court from the formulating clear-cut
rules of EFD.
The case concerned a dispute between MCI and AT&T, involving interconnection in
telecommunications, where the latter company has a monopoly power. Before the
Court dealing with the case, MCI brought many claims, among which the major one
was towards the extent to which AT&T allowed MCI to interconnect with its local
circuits. MCI alleged that AT&T imposed unreasonable conditions with regard to
interconnection with its local network, and accused AT&T of unlawfully refusing
multi interconnections.35 The Court of Appeals, Seventh Circuit cited EFD as the
legal basis of its decision, within the following paragraphs:36
A monopolist’s refusal to deal under these circumstances is

governed by the so-called essential facilities doctrine. Such a
refusal may be unlawful because a monopolist’s control of an
essential facility (sometimes called a ‘bottleneck’) can extend
monopoly power from one stage of production to another, and from
one market into another. Thus the antitrust laws have imposed on
firms controlling an essential facility the obligation to make the
facility available on non-discriminatory terms.
The case law sets forth four elements necessary to establish liability
under the essential facilities doctrine: (1) control of the essential
facility by a monopolist; (2) a competitor’s inability practically or
reasonably to duplicate essential facility; (3) the denial of the use
of the facility to a competitor; and (4) the feasibility of providing
the facility.
34
MCI Communications v. AT&T, 708 F.2d 1081, 1132-1133 (7th 1983).
35
Ibid., at 1132-3.
36
Ibid.
13
The most peculiar and outstanding aspect of the judgment is its bringing a four-part
test in order to verify liability under EFD. As a matter of fact, the so-called test
contributed to employ a theoretical framework for delineation of the boundaries of
the Doctrine to a great extent and gave way for further discussions challenging the
limits of the Doctrine’s applicability in jurisdictional area. Considering far-reaching
results of the so-called four-part test, it will be appropriate to discuss the conditions
set in the MCI test and their implications. Though having been developed with some
additional abusive elements, the pre-conditions for EFD set out in the MCI test are
substantially valid in essential facility cases of EC Competition Law, as well.
1.2.1 Control of the Essential Facility by a Monopolist
The ‘essential facility’ concept is generally started from the premise that the owner
of the facility has a ‘monopoly’ in the marketplace.37 A conclusion that there is no
place to interfere in a refusal to deal case unless the facility in question is owned by a
monopoly/dominant undertaking can also be deduced upon analysing the relevant
judgments.
A firm may be dominant for several reasons. The firm may have an exclusive license
constituting a legal barrier to entry. The firm may also possess an asset which is
uniquely situated geographically or has a natural characteristic which makes the asset
a natural monopoly. Finally, a firm may gain dominance because it has operated
more efficiently than its competitors.
Since the dominance is not per se38 illegal, competent authorities or courts are
usually concerned if the firm uses its dominance to deter entry of potential
competitors or to substantially lessen competition in general, at least in the European
37
The ‘dominant position’ in EC Competition Law corresponds to the term ‘monopoly’ in US
Antitrust Law with some important differences in theory and practise. In the thesis, being dominant in
relevant market rather than having monopoly is usually preferred, since the subject-matter of thesis is
mainly directed to EFD analysis on the Community Law basis.
38
Per se (illegal) means an automatic illegality on its own, not depending on any further arguments.
14
context.39 However, this inherent principle has been replaced with the
‘monopolisation’ phenomenon in US Courts. As a corresponding matter, the cases in
which the firm enjoying monopoly strengthened its position in another (downstream)
market through refusal to grant essential facilities have so far been regarded as a
breach of antitrust law in many instances.
As regards EFD cases, the foremost concern would rather be related as to whether
the competitive structure has been affected in the relevant market or not, even if an
obvious refusal to deal has taken place. Thus, it would be impossible to justify the
application of EFD if the refusal to supply had little effect on competition in the
downstream market.40 In other words, even if owner of the facility is declared as
being in a dominant position, this fact must not be sufficient for applying EFD
without considering the said competitive concern(s).
In this respect, if a firm has a dominant position in two adjacent markets, a refusal to
supply which strengthens that position would be unlawful because the refusal limits
the production of a competitor.41 However, in presence of effective competition, no
harmful effect of refusal to supply could be observed in the relevant market.
Therefore, whether there occurs elimination or lessening of competition by the
dominant firm’s refusal to grant access to essential facilities is one of the important
tests under EFD analysis.
Another discussion is related to the extent to what reliance on traditional dominance

test in conjunction with two separate markets (as downstream and upstream markets)
is viable under EFD. Leaving the EU case so as to be detailed further, whether ‘the
facility must be controlled by a monopoly’ constitutes a strict condition under MCI
test needs to be analysed.
39
However, in case of legal monopoly or a firm subjected to sector-specific regulation, access to the
essential facilities owned by the firm could be mandated under a specific form of regulation which is
convenient to be called a per se or ex ante rule.
40
Position since Bronner-Notes for a lecture”, September 2000, Copenhagen, p. 12.
41
Ibid., p. 13
15
US Courts sometimes treated the threat of downstream monopolisation as a
fundamental pre-requisite for validity of an essential facility claim.42 On the other
hand, the lower court opinions in the Aspen Skiing43 contradict the assertion that the
Essential Facilities Doctrine only applies when a monopolist firm supplies a
downstream product or service that competitors or customers must have in order to
compete. The Lower Court confronted a claim by the defendant, who “argued that ...
a duty to deal can arise only in different circumstances where, through vertical
integration, one firm has come to monopolize or control the supply of a component
necessary for production, distribution or sale of a rival’s product or service”.44 The
Court explicitly rejected such contentions: “We decline to adopt a narrow rule that
would immunise an unintegrated monopolist from antitrust liability for refusing a
competitor access of an essential facility in these circumstances. Vertical integration
is not essential to finding a violation of the antitrust laws for a refusal to deal under
the intent test.”45 Here, the Court emphasised the ‘essentiality’ rather than
‘dominance in a vertically-related market’. Further, the Court was “not convinced
that the essential touchstone of bottleneck cases is vertical integration.”46
42
See in Air Passenger Computer Reservations Sys. Antitrust Litig, 694 F. Sup. 1443, 1455 (C.D. Cal.
1988), aff’d, 948 F. 2d 536 (9th Cir. 1991). “When applying the Essential Facilities Doctrine in the
context of Section 2 of the Sherman Act, a facility should be deemed essential to the downstream
market only where control of the facility by a competitor poses a danger of monopolisation of the
downstream market.”; see also Consolidated Gas Co. v. City Gas Co., 912 F 2d 1262, 1292 (11th Cir.
1990). As explained in Consolidated Gas: “The essential facilities doctrine is designed to deal with
the danger that a monopolist in control of a scarce resource will extend its power vertically from one
level of production to another ... A facility becomes essential if, in restricting competitors’ access to
that facility, a monopolist gains a competitive advantage in another level of the market-that is, a
market downstream or upstream from the market containing the facility itself.” (cited in R. Pitofsky,
D. Patterson, J. Hooks, p. 458-461.) J. T. Lang who was the Director in the Competition Directorate
General in European Commission also states that if a dominant firm is not present on the downstream
market for which access to the facility is said to be essential, EFD does not apply, at least in
intellectual property rights cases. (cited in J. T. Lang, p.14).
43
Aspen Skiing Co. v. Aspen Highlands Skiing Corp 472 US 585 (1985).
44
Ibid., at 1518.
45
Ibid., at 1519.
46
Ibid. Similarly, an appellate court reinstated a plaintiff’s essential facility claim after it was
dismissed by trial court in Delaware & Hudson Railway Co. V. Consolidated Rail Corp., 902 F.2d
174 (2d Cir. 1990). The case involved a dispute between two rail companies. Because, the plaintiff
had a more limited track system than the defendant, it required access to the defendant’s tracks for
portions of certain shipping trips – the end portion (or ‘short haul’) – to deliver to specific
destinations. The defendant, who previously had acquiesced in such arrangements before the rival
carriers having begun competing, refused to allow access on reasonable terms. The Court found that
the plaintiff’s essential facilities claim could proceed. (cited in R. Pitofsky, D. Patterson, J. Hooks, p.
459.).
16
In view of these judgments, favourably to say that the first condition of the MCI test
must not be construed and applied as a required level of monopolistic control (i.e.
vertical integration) for the application of EFD. Reinforcing this conclusion, a
number of US courts have emphasised that the vital issue is whether plaintiff has a
competitive relationship with the alleged monopolist in the relevant market or not.47
Numerous lower court cases simply require that plaintiffs demonstrate that they are
competitors being denied access to an essential facility controlled by the defendant-
monopolist.48 This fact is verified by the Intergraph Corp. v. Intel Corp.49 judgment
where the plaintiff has not shown an adequately competitive relationship between the
parties. Hence, the competitive relationship between the parties (monopolist –
undertaking requesting access) is sufficient for satisfying the request to be valid as a
claim for EFD analysis. That is to say, in order for the criteria of a case to be
assessed under the Doctrine, an allegedly essential facility must be owned by a
monopolist / dominant undertaking, and there must be a competitive relationship, but
not a required level of dominance along the adjacent markets.
1.2.2 A Competitor’s Inability Practically or Reasonably to Duplicate Essential

Facility
The analysis of this criterion constitutes the core of the term ‘essentiality’, thereby
plays a major role in applying EFD. In advance of investigating further issues,
whether a facility is essential or not is often regarded as the most important requisite
to be clarified under the Doctrine-based analysis. Thus, this criterion (a competitor’s
inability practically or reasonably to duplicate essential facility) is vital for
application of EFD.
A facility is essential when two conditions are satisfied: Firstly, it must be examined
whether someone other than the owner of the facility in question is currently
providing the same facility. Within this initial condition, there exists a further
47
R. Pitofsky, “The Essential Facilities Doctrine Under United States Antitrust Law”, p. 23.
http://www.ftc.gov/os/comments/intelpropertycomment/pitofskyrobert.pdf
48
Ibid.
49
Intergraph Corp. v. Intel Corp., 195 F.3d 13346, 1358 (Fed. Cir. 1999).
17
implicit question with regard to the extent to what the alleged essential facility is
equivalent with other products. The second condition consists of an evaluation as to
whether the facility in question can be feasibly duplicated or not. This condition
which can be called ‘non-duplicability’ entails an economic analysis as well as a
technical one.
The first condition is relatively easy to investigate. The second one, however, would
be harder as a determination has to be made taking into account the existing
economic and technical conditions and possible changes over them. Therefore, it is
possible to say that the analysis of the second condition is highly fact-specific.
Under the first condition, the facility in question must not be available from other
sources. In the Apartment Source of Philadelphia v. Philadelphia Newspapers50
judgment, the following sentence reveals important clues in terms of interpreting the
first condition: “A facility will not be deemed essential if equivalent facilities exist or
where the benefits to be derived from access to the alleged essential facility can be
obtained from other sources.” Given this sentence of the judgment, in examining first
condition (whether someone other than the owner of the allegedly essential facility is
currently providing the same facility), a competitor must demonstrate that any other
facility is neither the same as nor equivalent with the facility which is sought access.
The same competitor has not an obligation to demonstrate that the (upstream) market
consists of the essential facility solely. In other words, definition of market does not
always overlap the delineation of essential facilities in EFD analysis.
At this juncture, a distinction between the relevant product market and the market of
essential facility has to be done. That is to say, other products which are substitutable
with the product constituting the upstream market can be different from the allegedly
essential facility. By opposite, an essential facility does not have to constitute the
relevant market with any other products deemed substitutable with itself. As a
conclusion, in order for the Doctrine to be applied in a case, the facility alleged to be
50
Apartment Source of Philadelphia v. Philadelphia Newspapers, Civ. A. No. 98-5472, 1999 WL
191649, at 7.
18
essential for operating in downstream market is not necessarily deemed the product
of the upstream market. However, a perception regarding the markets that consist of
essential facilities as ‘identical’ with those of upstream products is accepted by a
number of authors.51
Given the lack of requirement of an equivalence between the market of essential

facilities and that of upstream products in EFD cases, identification of the said
markets should be conducted cautiously. In this context, the term ‘equivalent’ used in
the sentence extracted from the above judgment must be distinguished from the
terms ‘substitutable’ or ‘interchangeable’ used for definition of markets. This is
confirmed by the second part of the above sentence, where an effort of avoiding from
definition of market is remarkable. In that part of the sentence, it is stated that unless
benefits to be derived from the allegedly essential facility can be obtained from other
sources, the facility in question is ultimately deemed ‘essential’, and there is no
relevance between being ‘essential’ of the facility and definition of markets. Therein,
the emphasis is cited upon the benefits of the facility in question and the possibility
of derivation of such benefits from another source(s). In fact, a facility will not be
deemed ‘essential’ if equivalent (but not constituting the relevant product market
together with the essential facility) facilities exist. Therefore, under EFD analysis,
there ought not to be a requirement for an overlap between the market of essential
facility and the relevant product market.52
In this respect, the proposition that cases involving essential facilities therefore
require definition of two different markets and essential facility itself constitutes one
of these markets is difficult to agree upon. The cornerstone of the ‘essentiality’ lies at
51
Antonio Capobianco, “The essential facility doctrine: similarities and differences between the
American and the European approach” in European Law Review, 26, no. 6, 2001, p. 556.
52
For instance, in telecommunications markets local loop is deemed essential facility by many courts
as well as public authorities. However, it can be substituted by wireless local loop and cable networks
to a great extent, in particular for the aim of broadband access and high-speed internet. In this context,
on the one side local loop is regarded as an essential facility, on the other side, it does not constitute a
relevant product market. In the above case, local loops constitute only a unique part of the broadband
access market, but not a separate market in its own. See below note 334 for detailed information as
regards essentiality or uniqueness of the local loop(s).
19
other legal and economic characteristics pertaining to the facility in question, rather
than those used in defining relevant markets.
Regarding the characteristics of essential facilities, both the examination that

someone other than the owner of the facility is providing the same facility at the time
of request (first condition) and investigating whether the facility in question can be
feasibly duplicated (second condition) must be negatively concluded. A number of
factors operate to satisfy these (the first and second) conditions: geographical and
topographical conditions prevent construction of alternatives; a legal license
precludes duplication; a natural monopoly exists; the unique physical characteristics
of the resource are not duplicable; the governmental characteristics of the resource
are not duplicable; the governmental regulatory environment prohibits the
construction; the existing resource satisfies the minimum efficiencies of scale; public
subsidies are necessary for construction and are lacking; a minimum market
condition exists; natural fortuity disallows the construction of an alternative; lags in
technology render the alternative infeasible or unduly expensive; or any other factor
that provides a substantial cost disincentive for the creation of a viable alternative.53
The second condition, that duplication be impractical or unreasonable, is a higher

standard when compared with being “more economic” among alternatives. This is
explicitly recognised in the Twin Labs v. Weider Health & Fitness54 judgment: “As
the word, ‘essential’ indicates, a plaintiff must show more than inconvenience, or
some economic loss; he must show that an alternative to the facility is not feasible.”
The most encountered issue in this regard is the reproductive cost which must be
enormous in an essential facility case. Here, it is insufficient for the plaintiff to show
that access to the facility is more economic than other alternatives; the mere
reduction of costs fails to make the facility essential.55 In evaluating ‘non-
53
J. T. Soma, D. A. Forkner and B. P. Jumps, “The Essential Facilities Doctrine in the Deregulated
Telecommunication Industry” 13, Berkeley Tech. L. J. 565, 1998, p. 573
54
Twin Labs v. Weider Health & Fitness, 900 F.2d 569-570 (2d Cir 1990).
55
J. Brannan, “Open Broadband: An Essential Facility Doctrine Analysis”, 1999, p.16. Available at
www.ukans.edu/-cybermom/CLJ/Broadband.htm
20
duplicability’ under EFD, costs to duplicate the facility in question are presumed so
high as to render a prohibitively high barrier before entering the relevant market.
In this context, ‘indispensability’ is not always the required threshold in order for an
asset to be deemed essential. This evaluation seems to be close with the US system,
where there are a number of precedents representing a matured application of EFD in
jurisdictional area. In a US case, Hecht v. Pro Football, Inc.56 the following quoted
statement is noticeable in this respect: “To be essential a facility need not be
indispensable; it is sufficient if duplication of the facility would be economically
infeasible and if denial of its use inflicts a severe handicap on potential market
entrants.”
In applying the threshold for non-duplicability test, there appear different attitudes in
dealing with different cases. For instance, in cases which involve government-
controlled facilities that constitute huge entry barriers such as the railway bridges,57
sports stadiums,58 electricity transmission networks,59 nationwide
60
telecommunications networks , the referred entry barriers render the test to a
medium where indispensability is not queried because of the natural monopoly
characteristics of the facility. In such cases, less strict rules enabling a wider
interpretation of the MCI test are more favourable in applying EFD. The most
peculiar character of such cases is existence of a reserved area with prohibitively
high costs and technical obstacles preventing new entries, usually typified via public
subsidization. Being a case for exemplifying the above-mentioned characteristics of
essential facilities, Hecht v. Pro Football, Inc61 judgment says something important
for practitioners. In that decision, EFD was applied to a facility (a football stadium)
not in competition with the dealers seeking access, and due to this fact the said
judgment must not be generalised on its own. However, according to Lipsky and
56
Hecht v. Pro Football, Inc. 570 F.2d 982, 992 (D.C. Cir. 1977), cert. denied, 436 U.S. 956 (197).
57
See supra note 14.
58
Op.cit. in note 56.
59
See supra note 21.
60
See supra note 34.
61
See supra note 56.
21
Sidak, the full analysis of Hecht v. Pro Football, Inc. judgment indicates that the
Doctrine and the remedy of compulsory access may make particularly good sense in
the rare circumstances of (1) a facility with clear excess capacity, and (2) public
ownership of the facility permitting availability at marginal cost.62
Considering the huge cost of such (essential) facilities together with technological
and legal impediments, it is possible to say that the courts and competent agencies
must apply EFD following a less strident interpretation in such network industries
with the aim of providing competition.
As a matter of fact, in such industries where pertinent services have been supplied on
a monopoly basis or are subject to some degree of monopoly control, there is a
situation in which dominant / monopolist undertakings hold ‘gatekeeper’ positions
while new entrants depend on gaining access to essential facilities.63 Therefore, if the
firms acting in industries such as telecommunications, gas, electricity most of which
are recently liberalised are allowed not to grant access, they would be able to control
market developments by closing the gates and re-erecting the barriers which had
been removed by the liberalisation process.64 Given this problematic situation in a
marketplace, deal with new entrants or other competitors on reasonable terms is of
particular importance when the industry in question is of a statutory or natural
monopoly, that is, even after liberalisation obligations regarding essential to facilities
are necessarily required in such sectors. In absence of such obligations, as later will
be discussed specifically for telecommunications sector, new de facto monopolies
who can perpetuate their former positions and defeat the purpose of liberalisation
could easily be created in a liberalised environment.
62
Abbott B. Lipsky, Jr. and J. Gregory Sidak, “Essential Facilities”, Stanford Law Review, Vol. 51,
No: 1187, May 1999, p. 1205.
63
N. Nicolinakos, , Access Agreements in the Telecommunications Sector-Refusal to Supply and The
Essential Facilities Doctrine under EC Competition Law”, European Competition Law Review, No:8,
1999, p. 404.
64
Ibid.
22
1.2.3 The Denial of the Use of the Facility to a Competitor
Under EFD, an essential facility claim could be taken into analysis only if the request
for access to essential facility is denied. The denial needs not to be a total denial;
rather, it is sufficient that the terms of access would be unreasonable in price, profit
margin, time obligation or other substantive criteria.65
Even all conditions of the MCI test including ‘denial of access’ are met, one cannot
conclude a per se violation of Antitrust/Competition Law. If denying access to a
competitor or a new entrant has a little impact on the competitive structure of the
relevant market, such denial may be considered negligible and granting access may
not be mandated accordingly. However, if a refusal occurs in a regulated industry,
concluding an adverse situation to regulatory scheme inter alia an anticompetitive
conduct evading price regulation or a discriminatory act detrimental to regulatory
decisions, the so-called refusal to deal gives rise to an explicit violation of Antitrust
Law and can be regarded as a per se breach of law.66
1.2.4 The Feasibility of Providing the Facility
This condition makes evident that EFD is also delimited by legitimate business
justifications beyond the tests such as essentiality, non-duplicability, elimination of
competition, etc. That is to say, after presuming that other requisites have been met,
access to essential facilities can still be inconvenient to be mandated due to
unfeasibility of providing such facilities on the part of the owner of the facilities.
Hence, a liability under EFD is not enforceable where a defendant monopolist has a
legitimate business or a technically or economically reasonable justification for
declining access to the disputed assets to its competitor. In other words, the Antitrust
Law does not require that an essential facility be shared if such sharing would be
65
Telecommunication Industry” 13, Berkeley Tech. L. J. 565, 1998, p. 576.
66
In situations where an anticompetitive refusal to grant access have taken place in contrary to
regulatory scheme in a substantive and explicit manner, the regulatory authorities might deem the so-
called refusal per se unlawful. In fact, per se rules are rarely imposed by competition authorities,
owing to their flexible statutory rules.
23
impractical or would inhibit the defendant’s ability to serve its customers
adequately.67
Determination of feasibility therefore must include an inquiry in order to verify any

business justification put forward by the defendant, such as inability to satisfy all its
customers’ requirements with the supplies available, etc. In essential facilities cases,
the justification must demonstrate that providing access would disrupt defendant’s
own business, rather than those of customers’. Therefore, a cautious evaluation of
‘determination of feasibility’ accompanied preferably with a case-by-case analysis is
required to be conducted in EFD cases.
In the context of ‘determination of feasibility’, several questions arise as to the

relevant criteria for determining under which circumstances a legitimate business
reason exists. First, what constitutes legitimate business justifications and which
criteria are used to assess ‘legitimacy’. Second, to what extent, the monopolist in
question has discretion to determine the legitimacy of a business justification.
The answers to these questions give way to different approach pertaining to different
authors and courts which consist of mainly two categorical justifications; at macro
and micro level.68 The justifications at micro level consist of factual events and
practical evidence. For instance, if a firm can demonstrate that providing access
would violate an existing regulatory scheme, a legitimate business justification
exists.69 Statutory monopoly rights over local loops between subscribers and local
exchanges, nationwide transmission networks, and similar reserved parts of
telecommunications infrastructure could be demonstrated for illustration. Or even in
absence of legal or monopoly rights, some technical impediments could be
67
R. Pitofsky, D. Patterson, and J. Hooks, “The Essential Facilities Doctrine Under U.S. Antitrust
Law”, Antitrust Law Journal, Vol. 70, 2002, p. 450.
68
See J. T. Soma, D. A. Forkner and B. P. Jumps, “The Essential Facilities Doctrine in the
Deregulated Telecommunication Industry” 13, Berkeley Tech. L., 1998, J. p. 578; J. Brannan, “Open
Broadband: An Essential Facility Doctrine Analysis”, 1999, p. 16. Available at www.ukans.edu/-
cybermom/CLJ/Broadband.htm
69
Telecommunication Industry” 13, Berkeley Tech. L.J. 565, 1998, p. 573.
24
confronted inter alia in cases where do exist a lot of dealers and scarce resources
such in allocating of radio channels (spectrum). At the micro level, the burden of
proof with regard to proving that the refusal to access to essential facility is
unjustifiable is primarily relied upon the plaintiff.70 The burden of proof then shifts
to the defendant to provide evidence establishing a legitimate business justification.71
At the macro level, there exists no firm-specific reason for justification of refusal to
grant access. Macro legitimate business justifications do not pertain to any particular
firm, but constitute ‘propositions of general policy’.72 Such justifications are
surrounded around social and public policies as well as competition policies and are
designed irrespective of practical and individual reasons. National concerns over
promotion of investments and innovations are seemingly one of the most influential
factors in shaping macro legitimate business justifications. Besides, determining
whether access to the alleged essential facilities deprives legal monopolies of their
legitimate and public rights serves as another factor in the shape of macro level
policies.
1.3 Assessment of the US Case-Law
The Essential Facilities Doctrine has a long and reasonably successful history in US
Antitrust case-law. The Doctrine can be regarded as one of the long-standing
limitations on the ‘right to refuse to deal’ in the US Antitrust Law. However, it is
hard to say that EFD is a subset of the so-called ‘refusal to deal’ notion which puts
some limitations on a monopolist’s ability to act independently of its competitors,
namely to distort competition in the market. This is why numerous US courts
established antitrust liabilities after solely finding an anti-competitive ‘refusal to
deal’ that exclude an actual or potential rival from competing, in many cases.73 That
70
Telecommunication Industry” 13, Berkeley Tech. L., 1998, J. p. 578; J. Brannan, “Open Broadband:
An Essential Facility Doctrine Analysis”, 1999, p. 16. Available at www.ukans.edu/-
cybermom/CLJ/Broadband.htm
71
Ibid.
72
Ibid.
73
R. Pitofsky, “The Essential Facilities Doctrine Under United States Antitrust Law”, p. 9. Available
at http://www.ftc.gov/os/comments/intelpropertycomment/pitofskyrobert.pdf
25
is to say, an unlawful refusal to deal could be found irrespective of EFD analysis in
the US context. This seems valid in the opposite direction, considering the cases of
monopolisation where the US Courts placed EFD at the centre of the cases, such in
MCI Communication v. American AT & T Co.74 case. Then, EFD is conceivable to be
deemed a unique Doctrine having a stand-alone basis which constitutes an exception
to the freedom to contract.
The recognition of the Supreme Court in Aspen Skiing decision that “the high value
that we have placed on the right to refuse to deal with other firms does not mean that
the right is unqualified”75 demonstrates that EFD in United States is recognised as an
exception to the ‘right to refuse to deal’ and thereby to the ‘freedom to contract’.
Having been regarded as an exception to the general principle of ‘freedom to
contract’, EFD is applied narrowly especially until the four-part ‘MCI Test’, which
was set forth by a lower court in the MCI Communication v. American AT & T Co.76
case.
The MCI Test for antitrust liability has been adopted by virtually every court to
consider an ‘essential facilities’ claim.77 Most probably due to the strictness of the
requirements set forth in the MCI Test, United States courts did not find liability so
frequently under EFD. Another reason for this attitude seemingly relates to the fact
the United States courts also suggest that antitrust liability under EFD is properly
justified when denial of access is motivated by an anticompetitive intention.78
According to R. Pitofsky, the reason why the Doctrine rarely resulted in antitrust
liability lies at courts’ analysis that primarily aim at determining whether the facility
74
See supra note 34.
75
Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 US 585 (1985), at 601.
76
See supra note 34.
77
78
See the following cases: Otter Tail Power Co. v. United States 410 US 366, 35 L. Ed. 2d 359, 93
S.Ct. 1022 (1973); City of Anaheim, 955 F.2d 1373 (9th Cir. 1992); Aspen Skiing Co. v. Aspen
Highlands Skiing Corp.,472 US 585 (1985). The intent theory pertains to the US antitrust system, and
is cited in many US courts’ judgments as the fundamental point. However, under Article 82 of the EC
Treaty, intent of a dominant undertaking is never sought in finding an abuse of dominance.
26
controlled by the defendant firm is truly essential to competition or not.79 In order to
prove this view, he cites some court decisions, one of which reads as follows: “A
facility controlled by a single firm will be considered “essential” only, if control of
the facility carriers with the power to eliminate competition ...”80 Here exists no
emphasis relating to an anti-competitive conduct (action) that is framed with clear-
cut rules. Rather, ‘the power to eliminate competition’ is cited in this and many other
court decisions in reaching a violation of Sherman Act.81 Numerous US courts have
held that a refusal to deal coupled with an anti-competitive intent may support a
finding of antitrust liability even absent proof that the withheld input constitutes an
‘essential facility’.82
From this point of view, though having a long and respected history, EFD seems far
from being considered as an integral component or a stable principle of US Antitrust
Law. In essence, several objections are raised by many commentators regarding the
scope and boundaries of EFD in the US context.83 Primarily, while the EFD cases are
generally individualised so as to involve only unlawful attempts to create or extend
monopoly power, invoking the Doctrine in such cases would be unnecessary because
Section 2 (of the Sherman Act) already addresses such activities without requiring
resort to EFD. Confirming this objection with another respect, it is sometimes
asserted that application of the Doctrine may inhibit the incentives to create
innovative and cost-reducing facilities.84
79
80
City of Anaheim, 955 F.2d 1373 at 1381 (9th Cir. 1992) (cited in R. Pitofsky).
81
Similarly, in numerous cases, the Supreme Court reached its decisions upon considering
defendants’ power of eliminating competition rather than depending on concrete actions. See the
following cases: Alaska Airlines, Inc. v. United Airlines , Inc., 948 F 2d 536 (9th Circ 1991); Twin
Labs v. Weider Health & Fitness, 900 F.2d 569-570 (2d Cir 1990).
82
Op.cit in note 79, p. 9.
83
See, P. Areeda, “Essential facilites: an epithet in need of limiting principles”, 58 Antitrust Law
Journal (1990), 841, , no. 21, 1989, p. 841-51; Antonio Capobianco, “The essential facility doctrine:
similarities and differences between the American and the European approach” in European Law
Review, 26, no. 6, 2001, p.556. and T. F. Cotter, “Intellectual property and the essential facilities
doctrine”, The Antitrust Bulletine, Vol. XLIV, No:1, 1999, p. 233.
84
Prof. Areeda, one of the pioneers of the critical approach, drew a more analytical picture
establishing supplementary restrictions to be attached to the Doctrine, in one of his articles. According
to him, the following principles should be followed:
27
On the other hand, there are some counterviews such as those of M. Hirsh and G.
Richeimer, advancing that the idea of ‘limiting principles’ does more harm than good
if the principles limit EFD out of existence, even for circumstances in which those
principles urge antitrust enforcement.85
In this picture of conflicting views, some commentators on the one side disagree with
the application of the Doctrine on a particular set of facts (such as MCI test), some
others on the other side premise that forcing a monopolist to deal in exceptional
circumstances is properly justifiable under EFD.
At this juncture, necessarily saying, the ‘essentiality’ and the ‘duplicability’ tests
envisaged under US case-law accompanied with the welfare-enhancing goals of US
antitrust policy contributed to a reasonably and clearly defined basis for EFD. In fact,
practitioners must apply the Doctrine to a case in consistent with the widely accepted
principles of antitrust policies, which in essence aim at limiting the discretion of
dominant undertakings in order to preserve and enhance competition.
As it can be inferred from the suggestions of P. Areeda,86 if the criterion ‘elimination

of competition’ is strictly added to the four-part MCI test and when existence of the
so-called criterion is correctly verified in a case, likely harmful effects of EFD would
• There is no general duty to share. Compulsory access, if it exists at all, should be very exceptional.
• A company’s facility is ‘essential’ only when it is both
(i) critical to the plaintiff’s competitive vitality; and
(ii) the plaintiff is essential for competition in the market-place.
‘Critical to the plaintiff’s competitive vitality’ means that the plaintiff can not compete effectively
without it and that duplication or practical alternatives are not available.
• No one should be forced to deal unless doing so likely substantially to improve competition in the
market-place by reducing price or by increasing output. Such an improvement would be unlikely, in
particular, when it would chill desirable activity. This is, of course, a very important point in the case
of intellectual property rights.
• Even when all these conditions are satisfied, denial of access is never per se unlawful; legitimate
business purpose may justify not sharing a facility with third parties.
• The monopolist’s intention is irrelevant because every firm that denies its facilities to rivals does so
to limit competition with itself and increase its profits. (P. Areeda, “Essential facilites: an epithet in
need of limiting principles”, 58 Antitrust Law Journal (1990), 841, no. 21, 1989, p. 852-3).
85
Merril Hirsh and Gabriela A. Richeimer, “The essential Facilities Doctrine: Keeping the Word
‘Epithet’ from Becoming One”, p. 42. Available at http://www.rdblaw.com/News/PressReleases/-
Essentialfacilities.pdf
86
P. Areeda, “Essential facilites: an epithet in need of limiting principles”, 58 Antitrust Law Journal
(1990), 841, no. 21, 1989, p. 853.
28
most probably be minimised. In any way, when compared with the EC case-law, it is
possible to say that, a more characterised and a well-defined EFD approach has been
developed in United States owing to its evolution from the very beginning.
1.4 Essential Facilities Doctrine under EC Competition Law
In the context of EC Competition Law, the Essential Facilities Doctrine was

developed through Commission decisions primarily under Article 82 (ex-86) of EC
Treaty. In contrast to the U.S. case-law, the debate over whether there is ever a duty
to deal has not been echoed so much in the EC case law.87 This stems from the fact
that there is a broad area in Community rules where the suppliers have a duty to deal
with their competitors. The Community case-law prior to the introduction of EFD
into EC Competition Law makes it obvious that there are a lot of decisions
underpinning the duty to supply (both competitors and customers) wide spectrum. In
EC Competition Law, EFD therefore, - unlike the US system - is not an exception to
the legal system, but is generally construed as a special application of the legal
scheme ‘duty to deal’. J. Temple Lang gives a variety of examples to rationalise the
duty to supply, which originate from the decisions of Community Courts.88 For the
sake of illustration, pertinent types of abusive practises which contradict the ‘duty to
supply’ are quoted here from his article:
• dominant companies may not discriminate if the discrimination has significant

effects on competition,
• dominant companies may not refuse to supply competitors or customers if the
refusal has significant effects on competition,
• dominant companies may not increase or extend their dominance in the same
markets or use their power in one market to monopolise another,
• dominant owners of intellectual property rights commit an abuse only if they do
something more than merely exercise those rights to prevent the monopoly given by
them being infringed,
87
2001, p. 404.
29
• dominant companies may not selectively treat customers or competitors with
which they deal less favourably to discourage or penalise competition,
• dominant companies may not make their willingness to supply conditional an
acceptance of restrictive undertakings.
As could be realised from the above list, refusal to supply instances are judged under
Article 82 (ex-86) of the EC Treaty as a type of abuse of dominance. However, there
are some Commission decisions which evidence that Article 81 of the EC Treaty is
also convenient to be a legal basis for ‘duty to deal’ with competitors or consumers.89
In one decision of the Commission, Article 86 (ex-90) has been referred
exceptionally, as well.90 In this regard, the European Court of Justice has seemingly
interpreted Article 82 of the EC Treaty in a rather broad manner in order to resort
widely in refusal to deal/essential facilities cases.
Given the fact that evolution of EFD in EC Competition Law is associated with the
‘refusal to deal’ cases, the most important refusal to deal cases which are more
central to the Doctrine in the European context will be focused on below.
1.4.1 Main Characteristics of ‘Refusal to Deal’ Cases
Article 82 EC articulates the four basic types of abuse of dominant position,91 and
did not mention about refusal by a dominant undertaking to supply a consumer or
88
J. T. Lang, “Defining legitimate competition: companies’ duties to supply competitors, and access
to essential facilities”, Fordham Corporate Law Institute, International Law and Policy, 1994, p. 447.
89
Decision of 12 December 1991, Case IV/M.102, TNT/Canada Post, DBP Postdienst, La Poste, PTT
Post & Sweden Post [1991] OJ C 322/19, Decision 78/72 of 21 December 1977, Spices [1978] OJ L
53/20, Decision 93/405 of 23 December 1992, Schöller [1993] OJ L 183/1, Decision 93/406 of 23
December, Langnese [1993] OJ L 183/19. The last two decisions were brought before the CFI
(Judgments of 8 June 1995, Cases T-7 and T-9/93, Langnese-Iglo GmbH & Co. KG v. Commission
[1995] ECR II-1533 and 1611), which upheld them a far as they are relevant here. The ECJ dismissed
the appeal against the first CFI judgment (Judgment of 1 October 1998, Case C-279/95, Langnese-
Iglo GmbH v. Commission,) (cited in P. Larouche, p. 179)
90
Decision 94/119 of 21 December 1993, Port of Rφdby [1993] OJ L 55/52. (cited in P. Larouche, p.
183)
91
See supra note 2.
30
competitor. The so-called four types of abuse constitute the major examples under
Article 82 and are construed so as to cover ‘unlawful refusal to deal’ by the courts.92
However, pursuant to Article 82(b), an abuse of dominant position may be in the

form of ‘limiting production, markets or technical development to the prejudice of
consumers’ and with this perspective, unlawful refusal to deal is able to be associated
with the so-called indent of Article 82.
The legal basis for a finding that Article 82 has been infringed by a ‘refusal to
supply’, may also be Article 82(c) which prohibits discrimination through “applying
dissimilar conditions to equivalent transactions with other trading parties, thereby
placing them at a competitive disadvantage”. However mostly encountered types of
‘refusal to deal’ do not always overlap the abusive conducts under Article 82 (b) and
(c) of the EC Treaty. This fact is manifested under the Commission’s Access Notice
where the Commission sees three relevant scenarios:93
(a) a refusal to grant access for the purposes of a service where another operator has
been given access by the access provider to operate on that services market;
(b) a refusal to grant access for the purposes of a service where no other operator has
been given access by the access provider to operate on that services market;
(c) a withdrawal of access from an existing customer.
Among these scenarios, the second one (b) does not make up a traditional abuse of
dominance, i.e. a discriminatory conduct. It is a unique scenario under which ‘refusal
to supply’ and thereby ‘essential facility’ cases could arise in a distinct manner. From
this point of view, it is possible to say that ‘refusal to supply’ has thoroughly been
92
After an analysis of Article 82, the four examples listed therein are convenient to be considered
non-exhaustive, enabling increase of the range of abusive practices.
93
Commission’s Access Notice on Application of Competition Rules to Access Agreements in the
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at 84.
31
developed as a distinct kind of abuse of dominant position under the enforcement of
Article 82.
In ‘refusal to deal’ cases, there are two related markets involved: the market for the
supply of access to whatever is in question (upstream market), and the market for the
goods or services for the production of which access is needed (downstream
market).94 For instance, a car manufacturer must have access to engines, and if there
were a sole supplier of efficient engines then, this supplier would absolutely be in
dominant position and the dominant supplier possessing the engines on its own might
be forced to provide access to its engines with the car manufacturers. Here, the
market for which access to engines is needed (presumably, car manufacturing
market) is called ‘downstream market’ and the market for the supply of engines is
called ‘upstream market’.
1.4.2 EC Case-Law on ‘refusal to deal’
The leading case on unilateral refusal to deal in EC Competition Law is Commercial

Solvents v. Commission95 where the Court of Justice condemned a dominant firm for
its refusal to supply a downstream competitor with the raw material which is needed
to perform its productive activities. Commercial Solvents is also the leading Article
82 case on refusal to supply downstream competitors by a vertically integrated
firm.96
In this case, Commercial Solvents held a dominant position on the market for amino-
butanol, a raw product used in the manufacture of ethambutol, an anti-tuberculosis
drug. Zoja was an Italian producer of the ethambutol and was dependant upon
supplies of amino-butanol, the dominant supplier of which was Commercial
Solvents. Following a change in company policy, ICI, who is the subsidiary of
Commercial Solvents began to manufacture ethambutol, and thereafter supplies to
94
Position since Bronner-Notes for a lecture”, September 2000, Copenhagen, p. 5.
95
Joined Cases 6,7/73, Commercial Solvents v. Commission [1974] ECR 223.
96
The Essential Facilities”, www.law.emory.edu, 1997, p. 14.
32
Zoja were cut. The Commission decided that Commercial Solvents had abused its
dominant position by such a refusal to supply, and on appeal, the ECJ upheld the
Commission decision in the following words:97
However, an undertaking being in dominant position as regards the
production of raw material and therefore able to control the supply to
manufacturers of derivatives, cannot, just because it decides to start
manufacturing these derivatives (in competition with its former
customers) act in such a way as to eliminate their competition which
in the case in question, would amount to eliminating one of the
principal manufacturers of ethambutol in the common market. Since
such conduct is contrary to the objectives expressed in Article 3(f)
[now3(g)] of the Treaty and set out in greater detail in Articles 85 and
86 [now 81 and 82], it follows that an undertaking which has a
dominant position in the market in raw materials and which, with the
object of reserving such raw material for manufacturing its own
derivatives, refuses to supply a customer, which is itself a
manufacturer of these derivatives, and therefore risks eliminating all
competition on the part of customer, is abusing its dominant position
within the meaning of Article 86 (now 82).
The anti-competitive aspect of the Commercial Solvents’ behaviour was particularly

clear where the refusal to supply would exclude the only serious competitor that ICI
would face in the downstream market.98 At this juncture, it is possible to say that
refusal to supply a competitor in a downstream market constitutes one of the
infringements of Article 82 if its effects are of potential to eliminate all competition
therein.
After a couple of years, in United Brands,99 the Court of Justice stated that a
dominant firm can not stop supplying a long standing client if the orders placed by
such a customer are in no way out of the ordinary. Therein, a Danish ripener and
distributor (Olesen) had been buying bananas from several suppliers, including
United Brands. After Olesen began to promote the products of a rival supplier
(“Dole” Bananas) and helped to advertise it, United Brands cut off deliveries of
97
Joined Cases 6,7/73, Commercial Solvents v. Commission [1974] ECR 223., at 25.
98
Richard Whish, “Competition Law, Fourth edition”, Butterworths, 2001, p. 611.
99
Case 27/76, United Brands v. Commission, [1978] ECR 207.
33
bananas with its “Chiquita” brand to Olesen. The Commission imposed a fine to
United Brands for its refusal to supply Olesen. The Court agreed with the
Commission and found that the unilateral refusal to deal in question was an abusive
practise, establishing the following statement:100
In view of these conflicting arguments it is advisable to assert

positively from the outset that an undertaking in a dominant
position for the purpose of marketing a product ... cannot stop
supplying a long standing customer who abides by regular
commercial practice, if the orders placed by that customer are in no
way out of the ordinary.
It is noteworthy here that ECJ in United Brands, focused on the conduct of a

dominant company as it affected its customer in downstream markets, regardless of
whether or not the dominant company is vertically integrated. In addition, from the
Court’s statement, it can be concluded that a dominant firm may stop supplies to a
competitor if there exist legitimate business reasons.
Both in Commercial Solvents and United Brands judgments, did not arise any
discussion as to whether raw materials or banana supplies from the owners were
essential for the dealers. The ECJ, most probably having the concern of enhancing
competition within the Community-wide market in its mind, put an end to the anti-
competitive conducts of dominant undertakings, in both cases. Though having many
common characteristics, the language in United Brands was in J. Temple Lang’s
view, less sweeping than Commercial Solvents and thus the duty to supply a
customer or distributor may be seen less rigid than the duty to supply a competitor.101
However, subsequent cases related to EFD have been logically built upon these
judgments with further different arguments.
Telemarketing102 is another judgment in which the ECJ confirmed that an

undertaking holding a dominant position with regard to the production and/or supply
of certain products that are necessary to compete in another market may not, without
100
Case 27/76, United Brands v. Commission, [1978] ECR 207, at 182.
101
J. T. Lang, Defining legitimate competition: companies’ duties to supply competitors, and access to
essential facilities, Fordham Corporate Law Institute, International Law and Policy, 1994, p. 449.
102
Case 311/84, Cenbtre belge d’études du marché, [1985] ECR 3261.
34
any objective justification, refuse to supply those products. In Telemarketing, RTL
(the defendant) was a television broadcaster having a dominant position on the
market for advertisements directed at the French-speaking community in Belgium.
RTL would only accept advertisements for telemarketing103 provided that one of its
subsidiaries got the contract to answer viewers’ calls. The Court held that this was an
abuse of dominant position by reserving an ancillary activity (telemarketing services)
for itself without any objective necessity:
Commercial Solvents also applies to the case of an undertaking

holding a dominant position on the market in a service which is
indispensable for the activities of another undertaking on another
market. If ... telemarketing activities constitute a separate market
from that of the chosen advertising medium, although closely
associated with it, ... to subject the sale of broadcasting time to the
condition that the telephone lines of an advertising agent belonging
to the same group as the television station should be used amounts
in practice to a refusal to supply the services of that station to any
other telemarketing undertaking. If, further, that refusal is not
justified by technical or commercial requirements relating to the
nature of the television, but is intended to reserve to the agent any
telemarketing operation broadcast by the said station, with the
possibility of eliminating all competition from another undertaking,
such conduct amounts to an abuse prohibited by Article 86 [now
82], provided that the other conditions of that article are satisfied.
It must therefore be held in answer to the second question that an

abuse within the meaning of Article 86 [now 82] is committed
where, without any objective necessity, an undertaking holding a
dominant position on a particular market reserves to itself or to an
undertaking belonging to the same group an ancillary activity
which might be carried out by another undertaking as part of its
activities on a neighbouring but separate market, with the
possibility of eliminating all competition from such undertaking.
103
‘Telemarketing’ involves giving the telemarketing company’s telephone number in television
advertisements, enabling viewers to call a particular telephone number, to place orders or get
information concerning the products advertised.
35
Telemarketing is outstanding with its some different characteristics among refusal to
deal cases. Firstly, the case concerns ‘access’ to broadcasting service of RTL which
the Court deemed indispensable for another kind of service rather than a contract of
‘supplying a good’ breach of which results in a classical refusal to deal case.
Secondly, Telemarketing is the leading case in the area of ‘extending monopoly’
among the refusal to deal cases.104 Here does the RTL’s attempt for reserving
telemarketing activities for its subsidiary have an explicit effect to conclude such a
conclusion. Such type of a refusal which is more ‘structural and less behavioural’
makes this case more different than others.105 Thirdly, whereas Commercial Solvents
and United Brands involved purely private parties, Telemarketing also involved
exclusive rights over TV broadcasting granted to RTL, so that RTL was invested
with some public authority, and therefore could be thought to be under a strict duty
to behave fairly and without discrimination towards third parties in relation to those
exclusive rights.106
From the refusal to deal cases respectively Commercial Solvents, United Brands and
Telemarketing, the following inferences can be made.107 First of all, in these cases
the ECJ made clear that the refusal to supply an already existing competitor (or a
customer) who decides to market a competing product amounts to an abuse of a
dominant position when the refusal is not objectively justified. Another common
point in these cases is that they involve practices by which a dominant company in
one market is using its power in such a way in order to strengthen its position and at
the same time, to eliminate competition in a related market.
104
The distinction between this “extension of monopoly” case and a “refusal to deal” case is clearer in
RTT v. GB-Inno-BM. This concerned a supermarket chain prosecuted for selling telephone handsets
which had not been approved by RTT, the Belgian telecommunications administration, as required in
Belgian law. RTT had a legal monopoly in operating the public telecommunications network, and also
laid down the technical specifications for equipment itself. Thus, RTT held a dominant position and
was not entitled to extend its monopoly to an ancillary market without any objective necessity. Here,
RTT was extending its monopoly from operating the network to selling telephone equipment, but it
was not refusing to sell anything. (cited in B. Doherty, p. 413)
105
Publishing, 2000, p. 170.
106
Ibid.
107
N. Nicolinakos, “Access Agreements in the Telecommunications Sector-Refusal to Supply and The
Essential Facilities Doctrine under EC Competition Law”, European Competition Law Review, No. 8,
1999, p. 400.
36
1.4.3 Relationship between Refusal to Deal / EFD Cases and Intellectual
Property Rights
Along with the aim of encouraging competition in many fields of Community-wide

services and remove the trans-national borders between Member States, the
Commission made a great effort to accelerate liberalisation policies. In this regard,
removal of special and exclusive rights in a wide range of markets which formerly
represented reserved areas became one of the decisive policy tools in 1990s. The
Commission decisions regarding refuse to deal / EFD cases made it obvious that a
broad area on duty to supply including compulsory licensing occupied an important
place in Community Law.
The Magill108 judgment has a distinctly different character among the line of ‘refusal
to deal’ cases, particularly for its implications to ‘compulsory licensing’ scheme. In
Magill, three television broadcasters (companies) in Ireland (RTE, ITP, BBC,) each
had been publishing weekly listing magazines, giving their own programme details
for more than a few days in advance. Other publications such as daily newspapers
were entitled to reproduce the listings, but these licenses only included programme
details for a day or two in advance. A publisher (Magill TV Guide) wished to publish
the listings of the three television broadcasters in UK and Ireland in a single weekly
publication. The broadcasters refused to release their programme listings, claiming
that the programme listings were copyrighted under Irish and UK Law.
The Commission found that the three television companies had abused their
individual dominant positions with regard to their own TV listings by refusing to
make those listings available to Magill. This finding of abuse appears in resemblance
to the line of cases, Renault and Volvo v Erik Veng; where the possibility of
compulsory licensing had been introduced under Article 82.109 Upon concluding an
108
Joined Cases C-241 & 242/91P, RTE and ITP v. Commission, [1995] ECR I-743.
109
In both cases, Renault (Case 53/87, CICRA v. Renault [1988] ECR 6039) and Volvo v Erik Veng
Renault (Case 238/87, Volvo AB v. Eric Veng, (UK) Ltd. [1988] ECR 6211) a car manufacturer held
intellectual property rights over car body parts and refused to license other manufacturers to make
copies, even in exchange for a reasonable price. The Court adopted a compromise in those cases in
which unfair prices for the parts in question were deemed abusive but the refusal of the owner of
intellectual rights (i.e. design of the model) did not amount to an abuse. (cited in B. Doherty, p. 406-
7).
37
abuse of dominance, the Commission required three television companies to supply
advance information in order to enable comprehensive weekly TV guides to be
published. The Commission’s decision was brought before the CFI and the ECJ, both
of which upheld the so-called decision. In upholding the Commission’s decision, the
ECJ identified three reasons. Firstly, there was a potential consumer demand for a
single weekly multi-channel magazine, which the television companies were not
meeting. This reason was conferred as follows:110
The broadcasters’ refusal to provide basic information by relying

on national copyright provisions thus prevented the appearance of
a new product, a comprehensive weekly guide to television
programs, which the appellants did not offer and for which there
was a potential consumer demand. Such refusal constitutes an
abuse under heading (b) of the second paragraph of Article 86 of
the Treaty.
Secondly, there was no objective justification for their refusal.111 Thirdly, the Court
held that “the appellants, by their conduct, reserved to themselves the secondary
market of the weekly television guides by excluding all competition on that market
[with reference to Commercial Solvents], since they denied access to the basic
information which is the raw material indispensable for the compilation of such a
guide.”112
A dominant undertaking’s ‘reserving to itself the downstream market’ having been

emphasised many times in preceding decisions, the most prominent and debatable
aspects of the Magill decision are different from this point. The most controversial
point discussed among commentators has concentrated on application of the Doctrine
to the area of ‘intellectual property rights’.113
110
Joined Cases C-241 & 242/91P, RTE and ITP v. Commission, [1995] ECR I-743, at 54.
111
Ibid., at 55.
112
Ibid., at 56.
113
See Romano Subiotto, “The Right to Deal with Whom One Pleases under EEC Competition Law –
A Small Contribution to a Necessary Debate”, 6 ECLR, 1992, p. 238-240 and Richard Whish,
“Competition Law, Fourth edition”, Butterworths, 2001, p. 700.
38
According to Subiotto, the Magill decision represents an extension of the existing
case law and confirms that Article 82 forms a sufficient legal basis to impose on
dominant undertakings a general duty to supply, such in cases of sharing proprietary
assets with new customers, granting copyrighted materials to third parties, etc.114
Criticising the decision with respect to licensing copyrights, he points out three
issues: According to him, the Commission’s arguments reveal that television listings
did not deserve to be protected by copyright since the owners of the listings
(broadcasters) seemingly had a natural monopoly over the preparation of television
listings; but in spite of this rationale an inevitable conclusion that the broadcasters
had abused their dominant position by refusing to permit Magill to use their
television listings is unacceptable.115 Under Subiotto’s second concern raised
towards the Commission’s approach, the decision is criticised because it seems to go
against the fundamental assumption made by the EC law regarding the copyright
protection, that is, in the absence of harmonisation, Member States remain free to
determine the subject-matter that they deem fit to be protected by intellectual
property rights.116 Thirdly, according to him, the Commission’s approach would
introduce legal uncertainty since the Commission would then arrogate property unto
itself the role of determining the cases in which intellectual property right protection
is or is not justified and, consequently the cases in which the owner of intellectual
property rights would be under an obligation to share its assets with third parties in
order to enable them to compete with it.117
In respect of Subiotto’s concerns, it will be appropriate to state that since the

intellectual property legislation is concerned with the creation and commercial
exploitation of some legal proprietary rights which constitute an exclusive area, one
should be cautious when reconciling competition law rules with intellectual property
rights.
114
Romano Subiotto, “The Right to Deal with Whom One Pleases under EEC Competition Law – A
Small Contribution to a Necessary Debate”, 6 ECLR, 1992, p. 242. See also Christopher Stothers,
“Refusal to Supply as Abuse of a Dominant position: Essential Facilities in the European Union”,
European Competition Law Review, 22, no. 7, 2001, p. 262.
115
Romano Subiotto, “The Right to Deal with Whom One Pleases under EEC Competition Law – A
Small Contribution to a Necessary Debate”, 6 ECLR, 1992, p. 239.
116
Ibid.
39
Though involving many controversial sides, the Magill judgment brought forth some
contributions to the line of refusal to deal / EFD cases. For instance, in Oscar
Bronner decision which will be expounded in following parts, the ECJ specifically
referred to four criteria that were specified in Magill judgment: the information
sought by Magill was indispensable for carrying on the business in question (the
publishing of a general television guide); such refusal prevented the appearance of a
new product for which there was a potential consumer demand; there were no
objective justifications for the refusal to supply; and the refusal would exclude all
competition in the secondary market of television guides.118 Among the referred
points in the Magill judgment, the first two ones are noticeable for their implications
to EFD. As a matter of fact, the term ‘indispensability’ cited in Oscar Bronner is
deemed equivalent with and sometimes more decisive than the term ‘essentiality’ in
EFD analysis.
As a further development related to the core of the Magill decision, the conviction
that dominant broadcasters’ refusal to provide basic information by relying on
national copyright provisions would either prevent the appearance of a new product
or cause exclusion of competition in the secondary market had its repercussions in
the Commission’s Access Notice, as well. The paragraph 90 of the Access Notice
reads as follows:119
The Commission must ensure that the control over facilities

enjoyed by incumbent operators is not used to hamper the
development of a competitive telecommunications environment. A
company which is dominant on a market for services and which
commits an abuse contrary to Article 86 [now 82] on that market
may be required, in order to put an end to the abuse, to supply
access to its facility to one or more competitors on that market. In
particular, a company may abuse its dominant position if by its
actions it prevents the emergence of a new product or service.
117
Ibid.
118
Joined Cases C-241 & 242/91P, RTE and ITP v. Commission, [1995] ECR I-743., at 40. In Oscar
Bronner, the ECJ stressed the exceptional face of Magill decision, in this regard.
119
40
Similarly, ‘the emergence of a potential new service or product’ has been set forth as
one of the cumulative conditions in order to impose a competition law remedy,
namely to order an access obligation under the subsequent paragraph of the Access
Notice.120
Whereas the emphasis is put on new products or services within such words, to what
degree the ‘emergence of a potential new service or product’ will be taken as a
precondition for a duty to deal under EFD is quite vague. Considering the fact that it
does not seem possible to specify in detail the meaning of the so-called clause
(emergence of a potential new service or product), such a precondition could
constitute a problematic situation in the dynamic and fast-moving sectors such as
telecommunications. In essence, while encouraging new entrants who can provide
new products or services to compete as many services as possible, the Commission
would not wish this to lead to inefficiencies which would be harmful to the market in
the long term.121
In view of this assessment, reasoning of the Magill decision must be optimised such
as not to chill desire for innovation and investments and surrounding circumstances
around the emergence of a new service or product must be delineated more clearly.
1.4.4 The Introduction of Essential Facilities Doctrine into the EC Competition

Law
Unlike the situation in US, the institutions delegated with the powers to enforce the
competition rules in EC Law has until recent years, invoked EFD respectively in a
broad manner, among which Commission has been the forerunner one.122 Mostly, an
implicit expression was used for finding abuses of dominance under EFD to date and
120
121
N. Nicolinakos, , Access Agreements in the Telecommunications Sector-Refusal to Supply and
The Essential Facilities Doctrine under EC Competition Law”, European Competition Law Review,
No:8, 1999, p. 408.
122
See Barry Doherty, “Just what are essential facilities?” in Common Market Law Review, 38, no. 2,
2001, p. 397, and P. Nihoul and P. Rodford, EU Electronic Communications Law, Oxford University
Press, 2004, p. 471-2.
41
the proponents of EFD have often depended on those findings for proving that EFD
is a part of the Community Law.
Two of the initial cases where the Commission resorted to EFD are London
European/Sabena123 and British Midland/Aer Lingus124 decisions both of which did
not include a reference to EFD. The London European/Sabena case concerned access
to Sabena’s computerised reservation system, owing to which Sabena occupied a
dominant position in the Belgian market for computerised air travel reservations. In
the case, Sabena refused to list London European’s flights in its computer reservation
system (CRS) if the latter did not raise fares on its Brussels-Luton route or accepted
to procure groundhandling services from Sabena. The Commission stated that it is a
misuse to impose a higher tariff level to a competitor as a precondition to get access
to the reservation system. In Commission’s statements, access to Sabena’s CRS was
of “capital importance ... for all companies seeking to operate competitively on the
Belgian market.” The Commission condemned Sabena’s misconduct, presumably
considering that the CRS constituted an essential facility for air travelling services.
In British Midland/Aer Lingus, Aer Lingus refused to interline with British Midland
when the latter began to compete with the former on the Heathrow-Dublin route. The
Commission found that Aer Lingus had abused its dominant position by terminating
its interline agreement with British Midland. Moreover, the participation of Aer
Lingus at IATA conferences on tariffs, not allowing British Midland to participate in
the IATA system on interlining and tariffs, constituted an infringement of Article 85
[now 81] but not 86 [now 82] of EC Treaty. Interlining125 facilities should on the
one hand be made available to new entrants particularly in the initial stage of their
business, on the other hand be limited to the time frame which is objectively
necessary for a competitor to become established in the market.126 In its 22nd report,
123
Decision 88/589 of 4 November 1988, London European/Sabena [1988] OJ L 317/47.
124
Decision 92/213 of 26 February 1992, British Midland/Aer Lingus [1992] OJ L 96/34.
125
Interlining is a standard facility essentially based on IATA agreement pursuant to which most of
the airlines authorise other airline companies or travel agents to sell their services through a single
ticket.
126
D. Glasl, “Essential Facilities Doctrine in EC Antitrust Law: A Contribution to the Curent Debate”,
European Competition Law Review. No: 6, 1994, p. 309.
42
the Commission made it clear that this decision was taken within the context of a
period when ‘the European air transport industry was being liberalised’ and argued
that ‘airlines making use of the new opportunities for competition should be given a
fair chance to develop and sustain their challenge to established carriers’.127
In aftermath of these decisions related to air transport sector, Commission had taken
highly important steps in two cases that concerned the port of Holyhead (in Wales).
The decisions taken by the Commission regarding these cases [B&I Line128 and Sea
Containers129] condemned a dominant port owner for granting other competitors
access to the port facilities with terms less favourable than those which it gave to
their own services. In these cases, the port in question was owned and managed by
Stena Sealink Ports who was a subsidiary of Stena Line AB, and another subsidiary
of Stena Line, Stena Sealink Line was operating a ferry service between Holyhead
and Dublin.
In B&I Line,130 B&I complained that Stena Line’s ferry schedule was such that its
own ferries were forced to suspend loading or unloading when a Sealink ferry passed
in the narrow harbour channel. B&I’s complaint was triggered by a new timetable
requiring B&I to interrupt its operations more often than before. Having suffered
from this situation, B&I asked the Commission for interim measures to suspend the
new timetable before it could come into effect.
The other Commission decision cited above is Sea Containers131 the facts of which
are similar to the B&I Line. Sea Containers desired to initiate a new ferry service
between Holyhead and Dublin and though having been able to do so, could not begin
its activities due to the delays and difficulties imposed by Stena Sealink Line (a
subsidiary of Stena Line). Stena Line was therefore able to initiate its own ferry
127
Mark Furse, “The Essential Facilities Doctrine in Community Law”, European Competition Law
Review. Vol. 16, No: 8, 1995, p. 471.
128
Decision of 11 June 1992, B&I Line plc/Sealink Harbours Ltd. [1992] 5 CMLR 255.
129
Decision 94/19 of 21 December 1993, Sea Containers/Stena Sealink [1994] OJ L 15/8.
130
Op.cit. in note 128.
131
Decision 94/19 of 21 December 1993, Sea Containers/Stena Sealink [1994] OJ L 15/8.
43
service in advance of Sea Containers, consequently accruing to itself the economic
interests thereof. According to the Commission,132
… an undertaking which occupies a dominant position in the

provision of an essential facility and itself uses that facility (i.e., a
facility or infrastructure, without access to which competitors
cannot provide services to their customers), and which refuses other
companies access to that facility without objective justification or
grants access to competitors only on terms less favourable than
those which it gives its own services, infringes Article 86 if the
other conditions of that Article are met. An undertaking in a
dominant position may not discriminate in favour of its own
activities in a related market. The owner of an essential facility
which uses its power in one market in order to protect or strengthen
its position in another related market, in particular, by refusing to
grant access to a competitor, or by granting access on less
favourable terms than those of its own services, and thus imposing
a competitive disadvantage on its competitor, infringes Article 86.
The Commission’s decision in Sea Containers is remarkable on two points: Firstly, it

is the first Commission decision where an essential facility theory had been clearly
articulated. Secondly, this decision can be read so as to stand for the proposition that
a firm which owns an essential facility, and which both controls that facility as well
as competes in downstream markets using that facility, has a heightened duty to deal
on a non-discriminatory basis with downstream competitors.133 The so-called
heightened duty seemed to pose a test stricter than those used in both previous
Community and the US Antitrust cases.134 In this respect, undertakings controlling a
bottleneck might be considered to be ‘super-dominant’, implying that they have a
higher responsibility than the obligations attaching to ‘merely’ dominant firms.135 In
M. H. Harz’s view, the Commission in reaching such a conclusion, was particularly
132
Decision 94/19 of 21 December 1993, Sea Containers/Stena Sealink [1994] OJ L 15/8, at 16-17.
133
The Essential Facilities”, www.law.emory.edu, 1997, p. 15.
134
Review. Vol. 16, No: 8, 1995, p. 472.
135
Richard Whish, “Competition Law, Fourth edition”, Butterworths, 2001, p. 617.
44
influenced by the dual role occupied by Stena Line, being on the one hand an
essential facility owner and on the other hand a downstream competitor.136
In later years, EFD was further used in a different sequence of cases which
concerned railway services. In this sequence of cases, the most prominent one where
the right of access to rail services on a non-discriminatory basis was recognised is
European Night Services.137 The case was related to an agreement between a number
of railway companies planning to run services through the Channel Tunnel via a joint
venture called ENS (European Night Services). Under the agreement which was
brought before the Commission for exemption, ENS with its parent companies,
would be in charge of the sleeper cars as well as distribution of tickets, while the
parents would provide access to infrastructure and traction (locomotive and crew).138
Given that the rail services provided by the parents are ‘necessary’ for rail transport
operators, the Commission granted an exemption for joint venture but attached a
condition.139 The parents of ENS were obliged to grant ENS’s competitors with “the
same necessary rail services as they have agreed to supply to ENS ... on the same
technical and financial terms” according to the decision.140 The parties challenged
this condition before the Court of First Instance, while having not challenged the
existence of an essential facility therein. Leaving further stages of the case to be
detailed later, Larouche’s original views are worth noting, here.
Larouche makes a distinction between EFD cases depending upon whether

characteristics of the abuse reveal either behavioural or structural elements.141 For
instance, in B&I Line and Sea Containers cases are there behavioural elements, i.e.
refusal to supply or a discriminatory treatment. In fact, in the so-called cases, the
Stena Line, port owner was either refusing grant access to its downstream competitor
136
Review. Vol. 16, No: 8, 1995, p. 472.
137
Decision 94/663 of 21 September 1994, European Night Services [1994] OJ L 15/8.
138
Ibid., at 11.
139
Ibid., at 46 and 80.
140
Ibid.
141
Publishing, 2000, p. 183-187.
45
to the port facilites or granting access to those facilities with terms less favourable
than those used for their own services. However according to P. Larouche, the
condition imposed in rail services cases, i.e. European Night Services is more in the
nature of structural relief, arising because of the source of concern than a response to
any anti-competitive behaviour.142
This distinction is important particularly for distinguishing the essential facilities

cases from refusal to deal cases. More explicitly, refusal to deal cases in general
include behavioural elements, i.e. discrimination, discontinuation of supplies to
existing customers; while EFD cases usually exhibit structural character involving
facility sharing issues that stem from technical and structural imperatives. At this
juncture, the instances where there is no other effective remedy than EFD for
opening ‘essential’ resources to competition are of importance to reveal differences
of EFD cases. That is to say, refusal to access to an essential facility can generate
distinct problems where there usually occurs an insuperable entry barrier or a
bottleneck situation at the infrastructural level, and in such cases, EFD is more
capable with its structural remedies, comparing to other competition law tools, i.e.
rulings on non-discrimination, tying prohibitions.
1.4.5 The Attitude of the ECJ and CFI towards the Essential Facilities Doctrine
The Commission is the first European institution to consciously introduce EFD into
EC Competition law. However, Community Courts have in general been reluctant to
explicitly use the term “essential facility” as the legal basis in their judgments.
Nevertheless, an observation can be made that in some decisions of ECJ and CFI are
there important clues recalling EFD. In this regard, Tiercé Ladbroke143 is deserved to
be noted as the leading judgment of the CFI, where EFD has been recognised.
Implying the existence of EFD as a part of Community Law, the Court of First
Instance in Tiercé Ladbroke took a cautious stance towards the application of the
Doctrine. In this case, Tiercé Ladbroke was the owner of betting shops in Belgium,
142
143
Case T-504/93, Tiercé Ladbroke SA v. Commission [1997] ECR II-923.
46
operating on horse races held in Belgium, France and the UK. It sought to improve
its coverage of French races by broadcasting TV pictures thereof and requested the
copyright holders to provide itself with TV pictures and sound commentaries. Faced
with a refusal from the right-holders, Ladbroke complained to the Commission on
the ground that the racing associations (copyright holders) were collectively
dominant over the supply of films of French horse-races and were required,
according to the judgment in Magill144 to grant a license. The Commission rejected
Ladbroke’s arguments, and the Commission decision was upheld by CFI. CFI agreed
that the market was the Belgian market for sound and pictures.145 The Court did not
seriously consider whether the copyright holders were dominant over the supply of
films of races to Belgium or the market should have been confined to transmissions
of French races, considering that that market was ancillary to the national betting
markets.146
CFI held that since the copyright holders had not yet granted any licences for
Belgium, it was impossible for Ladbroke to argue that the refusal constituted a
discrimination.147 In this regard, the Court also distinguished the preceding cases,
Commercial Solvents,148 Telemarketing149 and London European/Sabena150 on the
ground that the refusal in T. Ladbroke came from an undertaking which was not
present on the same (downstream) market as the undertaking requesting the product
in question.151 In addition, CFI specifically refused to apply Magill due to the fact
that the refusal to supply television listings prevented Magill from entering the
downstream market for comprehensive television guides; however in the existing
144
See supra note in 110.
145
Case T-504/93, Tiercé Ladbroke SA v. Commission [1997] ECR II-923, at 81-89 (product market)
and 102-108 (geographic market).
146
However; Valentina Korah, who analysed the T. Labroke case in an article with a detailed manner,
states that the agreement alleged between the race course associations not to grant a license to
Ladbroke might have restricted potential competition in Belgium, and should have been examined by
the Commission. (See V. Korah, “The Ladbroke Saga”, E.C.L.R. No: 3, 1998, p. 169-176)
147
Case T-504/93, Tiercé Ladbroke SA v. Commission [1997] ECR II-923, at 123-124.
148
149
150
151
Op.cit. in note 147, at 133.
47
case, the refusal to supply did not prevent Ladbroke from being present on the
betting market. The Court held that,152
… the refusal to supply the applicant could not fall within the
prohibition laid down by Article 86 (now Article 82) unless it
concerned a product or service which was either essential for the
exercise of the activity in question, in that there was no real or
potential substitute, or was a new product whose introduction might
be prevented, despite specific, constant and regular potential
demand on the part of the consumers.
Taking the above excerpt into account, the most prominent point of the CFI’s
decision in T. Ladbroke might be seen in close relation to the ‘essential’ character of
the facility. Explicitly, in the decision, it is queried if the facility in question is either
essential for performing the existing activities or exhibit unique characteristics that
pertain to a new product or service. Though having not applied the Doctrine as the
necessary conditions were not fulfilled, the CFI agreed that essential resources must
be shared in its ruling.153 Given this point of view, this decision is conveniently to be
called as a milestone that paved way for further development of the Doctrine in EC
case-law.
When the European Night Services154 case is re-examined in light of the above
statements, one can see that the CFI has applied Magill155 and T. Ladbroke156 to that
case. Annulling the Commission’s decision mentioned above, CFI pointed out -
among others - three points in European Night Services. Firstly, The Court found that
the Commission had mistaken by advancing that ENS was a “transport operator”
under Directive 91/440 and stated that ENS, being a railway undertaking for
passenger transport was not a transport operator.157 Secondly, the Court concluded
152
Case T-504/93, Tiercé Ladbroke SA v. Commission [1997] ECR II-923, at 123-124, at 131.
153
P. Nihoul and P. Rodford, EU Electronic Communications Law, Oxford University Press, 2004, p.
479.
154
Joined Cases T-374, 375, 384 & 388/94, European Night Services v. Commission, [1998] ECR II-
3141. The decision of the Commission is discussed supra, p. 45.
155
156
157
Op.cit. in note 154, at 185-187.
48
that traction (the supply of locomotives, crew) could not be an essential facility for
some reasons that will be discussed below, whereas the Commission adversely had
found that traction was an essential facility. In this regard, non-discrimination
obligation imposed on transport operators under Directive 91/440 was considered
invalid for ENS.158 Thirdly, since Directive 91/440 guarantees non-discriminatory
access to infrastructure for railway undertakings and international groupings (i.e.,
associations between railway undertakings), there is no need to include conditions
based on EFD in a competition law decision.159
As P. Larouche points out,160 the Commission intends to use EFD beyond the
traditional line of cases involving anti-competitive behaviour, as it did in European
Night Services by imposing additional obligations, i.e. non-discrimination in
agreements. CFI’s annulment was also a sound decision in this perspective, since
there was a clear additional obligation in Commission decision, which extends
competition law remedies to impractical ex ante obligation(s). Even for the
liberalisation purposes, construing EFD in such a way overriding sector-specific
regulations is absolutely detrimental to the diligence of the EU regulatory framework
in the relevant sector.161
On the other hand, in European Night Services, the judgment of CFI revealed
important consequences regarding EFD:162
A product or service cannot be considered necessary or essential

unless there is no real or potential substitute .
Consequently, with regard to an agreement ... which falls within

Article 85 (1) [now 81 (1)] of the Treaty, the Court considers that
158
3141, at 212-217.
159
Ibid., at 221.
160
161
The main reason for such an overriding problem is the existence of EU regulatory framework in
that field, that is, Directive 91/440 guarantees non-discriminatory access to infrastructure for railway
undertakings and international groupings, and such provisions envisaged under the so-called Directive
would satisfy the access seekers in actual terms.
162
49
neither the parent undertakings nor the joint venture ... may be
regarded as being in possession of infrastructure, products or
services which are ‘necessary’ or ‘essential’ for entry to the
relevant market unless such infrastructure, products or services are
not ‘interchangeable’ and unless, by reason of their special
characteristics – in particular the prohibitive cost of and/or time
reasonably required for reproducing them – there are no viable
alternatives available to potential competitors of the joint venture,
which are thereby excluded from the market.
Although the CFI cited Ladbroke and Magill in its above statement, it went further
than the Ladbroke formulation, itself a considerable gloss on Magill.163 In order for a
product or a service to be regarded as ‘essential’, two cumulative conditions must
exist according to the above paragraph of the European Night Services judgment.
First of these conditions is ‘non-interchangeability’, and the second one is
‘unavailability of a viable alternative’. According to the first condition which might
be deemed the main thrust of EFD, existence of an essential facility is related to the
question as to whether the alleged essential facility is interchangeable or not, in that
are inherent some important conclusions.
At the first glance, the term ‘interchangeable’ seems suitable to be considered in

conjunction with the core element(s) used in defining relevant markets. But as
mentioned above, a product is not per se ‘essential’ even if it constitutes the relevant
product market, and as a corresponding matter under EFD, there is no overlap
requirement between the market for provision of essential facility and the relevant
product market.164
When European Night Services is examined, it would be seen that the Commission
has determined the downstream market so broad as to cover all means of passenger
transportation and advanced that the rail services provided at the upstream level are
‘necessary’ (deemed essential facility) for rail transport operators.165 On the other
163
2001, p. 412.
164
The relationship between EFD and definition of relevant market(s) is discussed supra p, 19-20.
165
3141, at 295. In P. Larouche’s establishment, the Commission in European Night Services decision
50
hand, CFI made the implication that even if those markets exist, economic realities
exhibit that rail services are interchangeable with other modes of transportation.166 At
this juncture, for a better understanding of CFI decision, examination should go
beyond the technical features of the alleged essential facility to look its economic
position.167 Here CFI implied that other means of transportation than rail services are
convenient for operation not in technical terms but also in terms of economic
viability, then the transport operators can choose whichever they prefer among the
transportation means in order to carry out their activities.
In general terms, CFI by referring ‘interchangeability’, would mean that when taking
the relevant end-user (downstream) market into account, lack of access to the facility
would affect competition on the relevant market, since such a facility is necessary in
order for competitors to be active on that market.168 In other words, practically, any
competitor in that situation must be in a bottleneck position without the alleged
essential facility. That is to say, what the term ‘interchangeability’ calls in European
Night Services is different from the role of ‘interchangeable’ products in the context
of market definition. Herein, the prevailing purpose of using the so-called term
seems to refer the economic situations of competing firms, namely, economic
viability of firms in presence of different options.
As to the second condition of European Night Services judgment, in order for a

facility to be essential, we are faced an investigation for the availability of a viable
alternative to the existing facility. Here is stressed a situation where duplication of
the alleged essential facility is impossible (i.e. for considerably huge costs or
technical obstacles). That is to say, in assessment of the CFI, in order for a facility to
be essential, it must be non-duplicable by reason of its special characteristics. What’s
more, those special characteristics of the facility must be valid for any other
determined at the outset that the relevant market was intermodal including all means of transportation
(thereafter the CFI upheld this determination) whereas it could have determined an intramodal market
only covering rail transportation. (See P. Larouche, “Competition Law and Regulation in European
Telecommunications”, Hart Publishing, 2000, p. 191.)
166
Hence, traction (locomotive and crew) is not able to be deemed an essential facility in light of the
‘interchangeability’ phenomenon.
167
51
competitor situated in such a condition that is deprived of access. In this perspective,
in order to determine whether the product or service in question is a viable
alternative (non-duplicable) or not, any competing firm in the same position as that
of the claiming party must be taken into account objectively.
1.4.6 Oscar Bronner Case: A Turning point in EFD
Two months after the European Night Services decision of the CFI, the ECJ ruled on
EFD in Oscar Bronner v Mediaprint,169 with a preliminary judgment. When
compared with other antecedents of the Doctrine, Oscar Bronner is paid a greater
attention for its role of confining EFD into highly exceptional circumstances. After a
brief history of the case, a detailed analysis of the Court ruling will be given below.
The defendant (Mediaprint Zeitungs- und Zeischiftverlag GmbH & Co. KG) was the
leading publisher of daily newspapers and magazines, having dominance in the
Austrian market for daily newspapers. It had established a nation-wide distribution
network for the delivery of its publications in early morning of each day. The
plaintiff (Oscar Bronner GmbH & Co. KG) was a rival publisher of a local
newspaper with respectively a much smaller market share. Having advanced its
inability to afford a new delivery system, Oscar Bronner asked to be included in the
so-called distribution system and offered to pay a reasonable fee. When the
Mediaprint rejected, Oscar Bronner brought the case before the Austrian Court which
referred a question to the ECJ. Before the EJC, arguments of the plaintiff were
primarily based on EFD.
Advocate General Jacobs delivered a detailed opinion, referring the case-law on

refusal to deal and framed the conditions for applying EFD therein.170 Upon
examination of relevant decisions both in US and EC case-law, A. G. Jacobs
divulged his thoughts regarding essential facility theory as follows:
168
Ibid., p. 193.
169
Case C-7/97, Oscar Bronner GmBH & Co KG and Others v. Mediaprint Zeitungs-und
Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791, [1999] 4 CMLR 112.
170
Opinion of A.G. Jacobs in Case C-7/97 [1998] ECR I-7791, [1999] 4 CMLR 112 at 35-53.
52
The Commission considers that refusal of access to an essential
facility to a competitor can or itself be an abuse even in the
absence of the other factors, such as tying of sales, discrimination
vis-a-vis another independent competitor, discontinuation of
supplies to existing customers or deliberate action to damage a
competitor (although it may be noted that in many of the cases
with which it has dealt such additional factors are to a greater or
lesser extent present)
This statement particularly establishes that a free-standing Essential Facilities

Doctrine could exist. However, Advocate General Jacobs implied that the duty to
share essential facilities must be delimited, attaching some pre-conditions:171
…for example where duplication of the facility is impossible or

extremely difficult owing to physical, geographical or legal
constraints or is highly undesirable for reasons of public policy.
It is not sufficient that the undertaking’s control over a facility
should give it a competitive advantage. I do not rule out the
possibility that the cost of duplicating a facility might alone
constitute an insuperable barrier to entry … if the cost of
duplicating the facility alone is the barrier to entry, it must be
such as to deter any prudent undertaking from entering market …
Furthermore, Advocate General Jacobs states that the dominant undertaking must
have a “genuine strangehold on the related market” in order for an essential facility
to be shared or granted access.172 From the Opinion of Advocate General Jacobs, one
can reach the conclusion that application of EFD depends on two major conditions:
First of these conditions is regarding the position of any competitor who would enter
in a market. According to this condition, a facility is essential if without it there
would be an insuperable barrier to entry for competitors and competitors would
suffer that such a deprivation would preclude them to operate in a fair competition.
The second condition is whether the owner of the essential facility has a stranglehold
or not, revealing a sui generis character. The term ‘stranglehold’ seemingly
corresponds to the term ‘bottleneck’, which is used in describing a difficult position
171
Opinion of A.G. Jacobs in Case C-7/97 [1998] ECR I-7791, [1999] 4 CMLR 112, at 65.
172
Ibid.
53
that a new entrant or a competitor faces when competing with the incumbent
operator173 already entered in the market and possessed the key technology,
infrastructure or services.
Regarding the Oscar Bronner v Mediaprint case, Advocate General Jacobs

concluded that there was no duty to grant access to the newspaper delivery system on
the part of the dominant publisher, because new entrants had other alternatives and
were capable to compete without access to the existing delivery system.174
The ECJ ruled in line with the Advocate General’s Opinion.175 Though having not
referred to EFD, the Court in an implicit manner tried to provide restrictive
conditions for application of the Doctrine, which once more exhibits a real inference
for the existence of EFD in Community Law. In its judgment, the ECJ stated that the
Austrian Court initially would determine whether there existed a separate market for
the home-delivery of newspapers or a larger market including other distribution
methods as the relevant market.176 In conjunction with this point, the Court
underlined the importance of an appropriate definition of relevant market in order to
find whether there is an abuse of dominant position or not.177 ECJ has also drawn the
attention towards whether or not the refusal to access eliminates all competition in
the relevant market. ECJ in its assessment cited a number of specific criteria which
reminds the European Night Services178 decision:179
“Therefore ... it would still be necessary, for the Magill180

judgment to be effectively relied upon ... not only that the refusal
173
The term ‘incumbent’ is used in such a way to encompass both dominant undertakings and
monopoly firms, in the thesis.
174
Opinion of A.G. Jacobs in Case C-7/97 [1998] ECR I-7791, [1999] 4 CMLR 112, at 68.
175
P. Nihoul, who criticises harshly the restrictive approach of Oscar Bronner dedicates total
responsibility of the judgment to A. G. Jacobs, who wrote the opinion in Oscar Bronner case. (See P.
Nihoul and P. Rodford, EU Electronic Communications Law, Oxford University Press, 2004, p. 482)
176
Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791, [1999] 4 CMLR 112, at 32-36.
177
Ibid.
178
179
180
The Court referred the Magill judgment for its speaking out the exceptional circumstances that
pertain to EFD. Having seemingly accepted EFD, the Court in Oscar Bronner deals to confine the
54
of the service comprised in home delivery be likely to eliminate
all competition in the daily newspaper market on the part of the
person requesting the service and that such refusal be incapable
of being objectively justified, but also that the service in itself be
indispensable to carrying on that person’s business, inasmuch as
there is no actual or potential substitute in existence for that
home-delivery scheme.
The above statement stipulates three issues that apply to essential facilities cases. In
this framework, in order for a refusal to grant access to be unlawful the following
conditions must be met:
1) The refusal must be likely to eliminate all competition in the relevant market on
the requesting party,
2) The refusal must be incapable to be objectively justified,
3) The facility in question must be indispensable in order for business of the

requesting person to be carried on (inasmuch as there is “no actual or potential
substitute in existence.”)
The last criterion seems to be more objective when compared with those set out in
precedents under the application of EFD. Necessarily saying, in spite of its bearing
an uncertainty, the term ‘essential’ has been used frequently in preceding cases and
uncertain face of EFD which stems from the so-called term has been to a great extent
removed with the clear-cut rules of the Oscar Bronner judgment.
In resemblance to the European Night Services decision, the Court has dealt with the
‘non-substitutability’ (that corresponds to ‘non-interchangeability’ used in European
Night Services) and ‘indispensability’ (that corresponds to ‘non-availability of a
Doctrine to certain exceptional conditions as did in Magill. In Court’s view, Magill was an
exceptional case for four reasons:
“In Magill, ... without that information, the person wishing to produce such a guide would find it
impossible to publish it and offer it for sale (paragraph 53), the fact that such refusal prevented the
appearance of a new product for which there was a potential consumer demand (paragraph 54), the
fact that it was not justified by objective considerations (paragraph 55), and that it was likely to
exclude all competition in the secondary market of television guides (paragraph 56).” [See C-7/97
[1998] ECR I-7791, [1999] 4 CMLR 112 (supra note 110), at 53-56]
55
viable alternative’ used in European Night Services)’ phenomena.181 The Court
thereafter pointed out that postal delivery and kiosk sales were substitutable options,
“even though they may be less advantageous for the distribution of certain
newspapers”.182 Moreover, even if the existing distribution network had been
insufficiently substitutable with postal delivery and kiosk sales, and a specialised
delivery system was required, the Court would have been able to consider that a new
delivery system was feasible (a viable alternative) for the plaintiff to create (alone or
with others).183 That is to say, even unless the alleged essential facility was
substitutable, the second step whether access to the facility in question was
indispensable (non-availability of a viable alternative to that facility) should have
been ensured in the case. The two-fold cumulative requirements for applying EFD is
quite similar with the pattern of the European Night Services.
Upon conferring the abovementioned ‘indispensability’ test as combined with the

‘non-substitutability’ test, the Court refined the so-called tests within the following
paragraphs:184
It should be emphasised in that respect, in order to demonstrate

that the creation of such a system is not a realistic potential
alternative and that access to the existing system is therefore
indispensable, it is not enough to argue that it is not economically
viable by reason of the small circulation of the daily newspaper
or newspapers to be distributed.
For such access to be capable of being regarded as indispensable,

it would be necessary at the very least to establish, as the
Advocate General has pointed out at point 68 of his Opinion, that
181
See supra note in 162. See also P. Larouche, “Competition Law and Regulation in European
Telecommunications”, Hart Publishing, 2000, p. 194-6.
182
Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791, [1999] 4 CMLR 112, at 43.
183
In Oscar Bronner, the Court held that use of Mediaprint’s home delivery system was not
indispensable, since there were other means of distributing daily newspapers (such as through shops,
kiosks and by post); furthermore, there were no technical, legal or economic obstacles that made
impossible to establish home delivery systems of their own for other publishers of daily newspapers.
[See Case C-7/97, Oscar Bronner GmBH & Co KG and Others v. Mediaprint Zeitungs-und
Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791, [1999] 4 CMLR 112, at 44]
184
Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791, [1999] 4 CMLR 112, at 45-46.
56
it is not economically viable to create a second home-delivery
scheme for the distribution of daily newspapers with a circulation
comparable to that of the daily newspapers distributed by the
existing scheme.
These two paragraphs demonstrate that it is not easy to prove that a product or
service is indispensable in a case, after the abovementioned tests which were brought
out by T. Ladbroke and European Night Services and matured by Oscar Bronner. As
a matter of fact, the blank points in the criteria attributed to the term ‘essentiality’
which have not been resolved within the prior Court decisions have forcefully been
fit up with the two paragraphs quoted above.
Although a number of questions remain open, the above two paragraphs are worth
being appreciated and discussed for bringing an economic perspective which was
unfamiliar in the context of the preceding case-law on EFD. According to the criteria
set under the above paragraphs the relevant market must be such that only one firm is
economically viable and, hence, it is impossible for two firms to operate
simultaneously in the market unless at least one of them is unprofitable.185 This view
is preferable to any other approach when the first paragraph is considered together
with the latter one. As Bergman suggests whilst the first paragraph says that “ ... it is
not enough to argue that it is not economically viable by reason of the small
circulation of the daily newspaper or newspapers to be distributed.”, the
interpretation that a competing publisher must lack the ability to duplicate the home-
delivery scheme even if it reaches the same circulation as does Mediaprint seems
improper.186 In fact, the part of the sentence “economically viable to create a second
home-delivery scheme for the distribution of daily newspapers with a circulation
comparable to that of the daily newspapers distributed by the existing scheme.” most
probably takes the form of ‘comparable market share’ (i.e. half of the market)
indicating the economic background for EFD (not to be applied).187 That is to say, if
there are two oligopolists competing effectively in a market, then one need not refer
185
Mats A. Bergman, “The Bronner Case-A Turning Point for The Essential facilities Doctrine?”,
European Competition Law Review, No: 2, 2000, p. 61.
186
Ibid.
187
Bergman elaborates economic analysis of the Oscar Bronner decision in his article with a detailed
manner. (See op.cit in note 185., p. 59-63)
57
EFD and compulsory third party access, because the two competing firms most
probably reached the same market shares via their respective already constructed
networks. Contrarily, two firms possessing symmetric market shares but asymmetric
infrastructures can not be considered in a marketplace where there is a serious need
for access from one firm to another.
As a corresponding matter, the Essential Facilities Doctrine would only be applicable

to markets in which two or more firms can never be economically viable on their
own unless the Doctrine is applied.188 At this juncture, the Doctrine as Bergman
states is applicable if a symmetric duopoly with two vertically integrated firms is not
economically viable in a relevant market.189 Such markets correspond to the markets
where natural monopolies existed even though two or more firms entered the market
afterwards. However, this approach is far from the approach adopted in most
essential facilities cases which were deprived of an economic analysis. Moreover, an
over zealous approach for duty to share the essential facilities has been followed
during the precedents since Commercial Solvents in EC case-law. In order to remove
any inconsistency, a proper market analysis must be conducted prior to imposition of
an obligation of ‘third party access’ to an owner of (allegedly) essential facility. Such
a market analysis which indicates whether potential competition prevails in the
market or not is indeed indispensable in any bottleneck problem in order to clarify
what EFD calls for each case.
1.5 Assessment of the EC Case-Law
Among the controversial points discussed above, it is inferable from the duty to
deal/essential facilities cases, in particular from the Oscar Bronner judgment that
ECJ recognised EFD as a part of the European Competition Law. However, during
such cases, the Community Courts (CFI and ECJ) consistently noted that more
188
Mats A. Bergman, “The Bronner Case-A Turning Point for The Essential facilities Doctrine?”,
European Competition Law Review, No: 2, 2000, p. 61. However, the straightforward reactions of
many commentators differ from Bergman’s view under a strict interpretation of the decision. After
strict reading of the decision it is usually accepted that, for the EJC, access may be ordered only if no
alternative distribution channel may be created for a newspaper of circulation equal to that owned by
the controlling undertaking. (See P. Nihoul and P. Rodford, EU Electronic Communications Law,
Oxford University Press, 2004, p.481)
189
Ibid.
58
caution regarding the application of the Doctrine must be devoted. In this context,
ECJ and CFI provided a significant contribution in defining and clarifying the
principles that apply to EFD.
In such dealing for refining the surrounding conditions around EFD, ‘enhancement
of competition’ must be premised as the foremost criteria to be taken into account.
Other criteria such as essentiality, indispensability, etc. are to be deemed technical
tools, which are sometimes needless when there is no competitive concern. As a
matter of fact, if a ‘refusal to deal’ investigation is adequate in a competition law
case, then a more technical EFD analysis can be ruled out therein. Correspondingly,
all relevant cases must be remedied in the context of competition law principles,
following an economics-oriented analysis. Such an approach dedicating much more
attention towards competition law principles accompanied with an economic analysis
would help the convergence of the legal attitude of the US and EC Courts, evidently.
The examination as to whether the duty to deal contributes enhancement of

competition must be the primary concern in both ‘refusal to deal’ and ‘essential
facilities’ cases. The distinguishing aspect of the essential facilities cases is attributed
to the ‘essentiality’ term which has been reinforced by the ‘indispensability’ and
‘non-substitutability’ tests introduced with Magill,190 T. Ladbroke,191 European Night
Services,192 and Oscar Bronner193 decisions. In fact, before Oscar Bronner, EFD has
matured to an appreciable level and has been harmonised with the line of the
competition law principles to some extent. Having dealt with the remaining questions
under EFD, Oscar Bronner is thought of the closing case within the series of
essential facilities cases. In essence, the difficulties surrounding the question
‘whether the facility in question is essential or not’ have been removed substantially,
after Oscar Bronner.
In this regard, the concerns over investment initiatives and protection of property
rights with regard to application of EFD have been eliminated, leaving behind little
190
191
192
193
59
reservations. In so far as the strict rules of the Oscar Bronner are applied, one need
not actually concern about the legality of EFD under EC Community Law.
It can be drawn from the above line of case-law that the Courts made a considerable
effort in order to assess EFD cases under the competition law tools, i.e. market
definition, test for dominance, etc. This effort is observable in many decisions such
as Sea Containers,194 European Sabena,195 and Telemarketing196 where the Court
referred the term ‘essential’. This fact also helps doubters be convinced about the
rationalisation and application of the Doctrine under competition law.
Another notable point can be inferred from the EC case-law is that EFD has proved
to be a very effective instrument for liberalising monopolistic services and opening
reserved markets to competition. Companies that are subject to regulation or run
under government management for many times own facilities which are deemed
essential in respective markets. Hence, EFD is of an important potential to contribute
fostering competition in such markets, when applied in a clear and consistent
manner.
A remaining question regarding EFD relates to the question as to whether the

Doctrine must be abandoned or maintained in a liberalised environment or a
competitive marketplace. A related issue comes after with ‘overlap’ problems
between (general) competition law rules and sector-specific regulation in area of
‘access to essential resources’. That is to say, after liberalisation, it could be asked if
sector-specific rules must be confined or replaced with competition rules in dealing
with access issues. Furthermore, ‘what are the roles of the national regulatory
authorities and competition authorities in this regard’ are the other questions to be
answered under the Doctrine. These open questions will be tried to be answered in
the second Section of the thesis depending on the legal analysis of the Doctrine made
here, with a special emphasis on telecommunications sector.
194
195
196
60
CHAPTER 2
2. IMPLICATIONS OF THE ESSENTIAL FACILITIES

DOCTRINE FOR TELECOMMUNICATIONS SECTORS
2.1 General Overview
Industries such as electricity, telecommunications, gas, railways are typical examples

of network industries in the sense that a substantial part of the products they produce
consists of transport from one destination to another via a network. Almost all the
network industries have combination of multiple segments some of which are
operated by incumbent undertakings who often perform as de facto monopolies,
while some others are operated in a competitive sense. Competitive and non-
competitive segments of the network industries are illustrated in the following table:
Table 1 (Source: Competition in Telecommunications, Jean-Jacques Laffont and Jean Tirole, The
MIT Press, Fourth Ed., 2002, p. 98.)
Industry Non-competitive Competitive

Segment∗ Segment
Telecommunications Local Loop Long-Distance;

International
Electricity Transmission grid Generation
Gas Pipelines Extraction
Rail Transportation Tracks, stations Passenger and freight

services
Postal Services Local delivery network Consolidations,

presort bureaus, etc.
∗
Non-competitive segments of networks are potentially able to be deemed as an essential facility.
61
Telecommunications is a prominent example of network industries, which exhibit
many different characteristics either monopolistic or competitive. Whereas many
telecommunications services such as long distance, international callings have been
offered by a number of competing firms, there have not existed sufficiently
competitive markets at the wholesale (network) level, since the very beginning.
During a long time, not only some segments, but all telecommunications networks
and services (the entire sector) were regarded as ‘natural monopoly’ by
governments.197 Gradually, governments came to realise that not all segments of the
telecommunications sector exhibited characteristics of a natural monopoly while at
the same time technological developments reduced costs of services and increases in
demand required new service offerings such as voice messaging, video on demand or
Internet services.
Given the market imperfections and the risks to competition, most governments have
taken the decision to intervene directly in the sector in order to guarantee access to
‘essential facilities’ and networks controlled by the incumbent operators, to mitigate
network externalities and large sunk costs, and to prevent anticompetitive
behaviour.198
Consequently, telecommunications markets have been subjected to both sector-

specific regulations and competition law remedies. On the one hand, comprising
various access measures for achieving effective competition in a non-liberalised
environment, on the other hand envisaging removal of the legal barriers such as
special and exclusive rights progressively, wide-ranging reforms were pursued by
governmental policies in telecommunications sectors.
197
Natural monopoly means a situation in which any amount of output is always produced more
cheaply by a single firm: the cost of production is lowest when one firm serves the entire market. If
the technology of a telecommunications network exhibits natural monopoly owing to the economies
of scope, scale, etc. then a single firm presumably construct and operate that network at a lower cost
than can two or more firms. Avoiding duplication of facilities particularly duplication of the fixed
costs of the network system, has been an important component of the ‘natural monopoly’ argument
for access regulation. The argument is that since costs are minimised by not duplicating transmission
facilities, competent authorities should bar entry of competing firms. [cited in Daniel F. Spulber,
“Handbook of Telecommunications Economics”, Elsevier Science B. V., 2002, Amsterdam, The
Netherlands, (Ed. by M. Cave, Sumit K. Majumdar, I. Vogelsang), p.486-7.]
198
ITU (International Telecommunications Union), Competition Policy in Telecommunications,
Document: CPT/04, Geneva, 20 - 22 November 2002, p. 32.
62
Among the so-called reforms, liberalisation of sector has well advanced almost all
over the world. However, telecommunications policies after liberalisation are
difficult to be said as efficacious as liberalisation policies. Even after liberalisation,
there have appeared so many monopolistic structures perpetuating entry barriers in
favour of incumbent operators. Though legal obstacles having been removed,
bottlenecks problems have still prevailed and de jure monopolies have simply been
replaced with their new de facto equivalents.
In the context of European telecommunications sector, it became apparent that the

intensity and rapidity of changes in this sector necessitated a more active application
of competition law and taking further steps towards effective competition at EU and
national level. The remedies designed for ensuring efficient access to essential
facilities have become the forerunner measures within the so-called steps.
Particularly in 1990s, the Essential Facilities Doctrine has gained a key role in
achievement of fair and reasonable access to infrastructure facilities deemed ‘natural
monopoly’. Many policy makers believed that duplication of essential facilities
would be inefficient and incumbents’ networks would rather be mandated for third
party access either under EFD.
Both at the EU and national level, main characteristics of telecommunications sector,

such as vertically integrated markets, non-duplicability of networks, advantages of
state-ownership raises important questions to be solved. One of those questions
regarding which telecommunications network resources constitute essential facilities
has a great practical importance. This is so because, the delineation of essential
network facilities is of direct relevance in shaping the access regulations and
determining the duties of dominant operators. Too narrow a definition could impede
competition by preventing competitors from being able to obtain necessary network
components on reasonable terms. Too broad definition can stimulate uneconomic
entry or provide insufficient incentives for competitors to invest in and develop
alternative network infrastructure.
Both the applicability and the future implications of EFD in EU as well as Turkish
telecommunications sectors will be discussed in this Section of the thesis. Priorly,
63
general characteristics of telecommunications sector and Community-wide
regulations within the EU context will be examined.
2.2 General Characteristics of Telecommunications Sectors
The last decade saw unprecedented changes in the global telecommunications

industry. Numerous state-owned telecommunications operators were privatized and a
wave of policies arranging liberalisation programmes swept the world. Despite
liberalisation trends; strong network externalities that are related to desire of
customers to make and receive calls in a network, large sunk costs involved in the
construction of ‘essential facilities’ such as local access networks, the long legacy of
statutory public monopoly, vertically integrated markets, economies of scale and
scope and the benefits reaped through the established networks such as wide
subscriber base, have all favoured the incumbent operators. These characteristics of
telecommunications sectors provide the incumbent operators with major advantages
comparing to new entrants and other competing operators.
2.2.1 Network-Based Characteristics
Primarily, the key question “What distinguishes a networked industry from other
industries” must be answered, before deepening the core of telecommunications.
Briefly, networks are able to be defined as a collection of smaller entities connected
with one another in order to function (at least part of the time) as a large entity.199
Suppliers and customers of firms are significant parts of networks. In this regard, all
relevant players in the networks consider networks crucial to their success and think
themselves as a part of a larger, independent whole.200
P.H. Longstaff classifies network industries as generally falling into three categories:
transportation, communications, and energy.201 According to him,
199
P. H. Longstaff, “Networked Industries: Patterns in Development, Operation, and Regulation”,
Program on Information Resources Policy, Center for Information Policy Research, Harvard
University, 2000, p. 3. Available at http://www.pirp.harvard.edu
200
Ibid., p.6
201
Ibid.
64
telecommunications system is a two-way (point-to-point) network over which any
sender or receiver can reach any other sender or receiver with access to the
network.202 (See below, Figure 1) Therein, traffic is allowed to move in both
directions in the network hierarchy. However one-way networks such as Cable TV
networks203 are organised as point-to-multipoint networks that typically do not send
traffic two ways. (See below, Figure 2)
Figure 1: Point-to-Point Network
202
University, 2000, p. 60. Available at http://www.pirp.harvard.edu Under P. H. Longstaff’s premises,
most point-to-point networks include many point-to-multipoint hubs, where all points are not directly
connected but are instead connected through a central controlling mechanism. Hubs become more
critical to the operation of the system, because they switch traffic from one line to another and for that
reason can become bottleneck.
203
However, Cable TV networks have the potential of being upgraded to two-way networks and when
upgraded for enabling two-way access, these networks are able to be used for (voice) telephony. That
is explicitly to say, Cable TV networks historically have been used to transmit ‘content’, i.e. television
services to the home, and usually have only one-way (access) capability. But once cable modems are
set in place, they are able to provide two-way access.
65
Figure 2: Point-to-Multipoint Network
(Source: P. H. Longstaff, “Networked Industries: Patterns in Development, Operation, and

Regulation”, Program on Information Resources Policy, Center for Information Policy Research,
Harvard University, 2000, p. 4.)
Other examples for point-to-multipoint networks include broadcast, energy networks

(gas and electric) and computer systems which have a central processing unit.
Modern networks usually have many connections at many levels, some within the
network (for example, two lower level hubs of the network that have direction
connections to each other) and some connect to other hubs through an outside
network (such as airports connected by bus routes.204
When telecommunications and telecommunications networks are the subject-matter,

we face a hierarchical structure, including a wide range of services. Simply,
telecommunications can be defined as the service of enabling electronic transfers of
information from one point to another or from one point to multiple points.205
Telecommunications involves two fundamental components: the transmission of a
204
University, 2000, p. 5. Available at http://www.pirp.harvard.edu Intensive interaction between
different networks creates multi-functional capabilities and stimulates innovative services. Emerging
interactive applications and innovations accelerate the ‘convergence’ between neighbouring markets
such as ‘broadcasting’ and ‘telecommunications’, removing the traditional boundaries between these
markets. The phenomena ‘convergence’ is going to be analysed in following parts.
205
Ibid., p. 60.
66
signal between two distinct points, and a switching function which selects a specific
transmission path for the desired communication.206 More specifically, in a
telecommunications network, calls go from customer premises equipment (terminals)
to a local area, it may be routed through several layers of higher level switching
centers before getting to the receiver’s local hub.207 The ‘local access network’
connects the customers to the national and international networks and thus
constitutes the most significant part of telecommunications networks.
2.2.2 Network Externalities
When telecommunications networks are considered, all players in a marketplace act

interdependently due to the hierarchical nature of the network. Thus, in order for a
telecommunications network to function effectively, a high degree of co-operation is
required from all parties involved. Investment in one part of the network creates
potential benefits across the whole network and similarly, blockages and deficiencies
in one part of the network can create bottlenecks, increased cost and reduced revenue
in other parts of the network.208 The so-called interdependence of the network
components is observed in the form of ‘network effects’ or ‘network externalities’ in
networked industries and is most frequently seen within the context of
‘interconnection’ of telecommunications networks.
‘Interconnection’ can be defined as the commercial and technical arrangements

under which service providers can connect their equipment, network and services to
enable customers to have access to the customers, services and networks of other
service providers.209 Interconnection is regarded as a special form of ‘access’, which
206
William W. Sharkey, “Handbook of Telecommunications Economics”, Elsevier Science B. V.,
2002, Amsterdam, The Netherlands, (Ed. by Mc. Cave, Sumit K. Majumdar, I. Vogelsang), p. 181
207
Ibid.
208
William H. Melody, “Telecom Reform: Principles, Policies and Regulatory Practices”, (2001),
p.49. Available at http://www.itu.int/industry overview/.
209
ITU(International Telecommunications Union), “TREG Interconnection self-training modules:
Introducing Interconnection”, 2003, p.1. http://www.itu.int/ITUD/treg/selftraining/module1.asp
In the Access Directive 2002/19/EC, interconnection is defined as “the physical and logical linking of
public communications networks used by the same or a different undertaking in order to allow the
users of one undertaking to communicate with users of the same or another undertaking, or to access
services provided by another undertaking.” When operators agree in order to provide interconnection
between their own networks, this agreement is called ‘interconnection agreement’, which is an
67
takes place not between a network provider and a user, but between two ‘equal’
party, i.e. two network providers.
Interconnection has been at the heart of the telecommunications regulation, since the
beginning of the development of this sector. Central to interconnection agreements is
the fact that interconnecting networks yield positive ‘network externalities’ in which
the value of the network to each customer increases as the number of customers
increases.210 Therefore, the total value of a customer joining the network depends on
not only the private benefits but the external benefits of being able to send and
receive call(s) from any other parties within the network.
The dashed line in the figure below illustrates interconnection which allows
customers on each network to contact each other as if they are in the same network.
SA1 SB1 SA1

SA5 SB4 SA5
A B A
SA2 SA2
SA4 SB3 SA4
SA3 SB2 SA3

(a) (b)
Figure 3: Interconnection
(Source: Jonas Holm, “Regulating Network Access Prices under Uncertainty and Increasing
Competition: The Case of Telecommunications and Local Loop Unbundling in the EU”, 2000,
University of Copenghagen, Institute of Economics, p. 4.)
The larger traffic has the interconnecting networks between each other, the more
network externalities emerge. This explains why interconnection is crucial between
point-to-point (two-way) networks. From this point of view, in case of a dispute in
agreement that stipulates the obligations and responsibilities of each contracting party with regard to
interconnection.
210
Ibid., p. 2.
68
reaching an interconnection agreement, always the larger network has more
advantage, comparing to smaller network. This picture remains the same even in lack
of interconnection between networks, because the value of subscribing to the larger
network is higher than the value of subscribing to the smaller one.
With network externalities, adding a new customer to a network increases the surplus
of other subscribers who are able to call and be called by the new customer, and
therefore affects not only customers’ demand for the service but also their demand
for subscription.211 However, network externalities described above represent direct
network effects, in that consumer utility directly depends on the market size,
independently of price system.212 There also exist indirect network effects that are
generated indirectly via market mechanisms such as economies of scale, scope and
density.
2.2.3 Economies of Scale
Economies of scale reflects the opportunities for reduced unit costs with increased
output.213 Economies of scale provides efficiency advantages with firms using this
opportunity and new entrants may find it difficult to compete with such firms who
perform with large scale production.
Owing to the great economies of scale inherent in a market, smaller companies

would tend to merge into larger units to remain competitive.214 Economies of scope
are sometimes lessened with administrative costs that are attributed to large
211
John-Hee Hahn, “Nonlinear Pricing of Telecomunications with Call and Network Externalities”,
November 2001, p.2. Available at http://www.keele.ac.uk./depts/ec/web/wpapers. The so-called
article of John-Hee Hahn deals with call externalities as well as network externalities, where the ‘call
externality’ is described as the benefit of incoming calls to a subscriber who does not have to pay for
the calls, usually the party being called. John-Hee Hahn gives the following example for
demonstrating this kind of externality. In mobile communications, an important reason to subscribe to
a network is to be able to be reached by others. Some subscribers (i.e. students) often use their mobile
phones for receiving incoming calls only without making any outgoing calls. For outgoing calls, they
tend to use an alternative service, say a fixed-link (public) telephone.
212
Ibid.
213
214
Ibid.
69
production units, which are presumably related to increasing co-ordination of
activities in a large organisation. Although degree of economies of scale differ
according to features of the facilities or services, particularly in ‘local access
networks’ is confronted considerable economies of scale.
Whereas new entrants face great economies of scale at ‘local access’ networks, the
same firms are most times able to access to some other segments of the sector more
freely due to less economies of scale. Markets for provision of terminals and
handsets can be shown for sector segments exhibiting less economies of scale which
include small independent suppliers competing with large operators in the supply of
many kinds of terminals.215 This is so, because state-owned operators have lost most
of the terminal market to competitors, in many countries. Nevertheless, it is possible
to say that for provision of many telecommunications services, economies of scale
remains a significant barrier to new entries in many countries, especially in countries
undergoing liberalisation process, such as Turkey.
2.2.4 Economies of Scope
A firm which produces telecommunications (i.e. toll and local) services is said to
enjoy economies of scope, if it can produce these services at lower cost than would
occur if each service were produced separately by a stand-alone firm.216 Economies
of scope are thus defined as cost savings related to supplying a number of different
services by the same firm.217 Economies of scope can be a barrier against smaller
telecommunications operators only supplying a limited range of services.
Morten Falch describes ‘economies of scope’ as a theme having three different

types:218
215
216
Melvyn A. Fuss, “Handbook of Telecommunications Economics”, Elsevier Science B. V., 2002,
Amsterdam, The Netherlands, (Ed. by Mc. Cave, Sumit K. Majumdar, I. Vogelsang), p. 153-154.
217
218
70
• economies of horizontal integration (i.e. telephony and data);
• economies of vertical integration within the network; (i.e. local and long
distance); voice and value added network services (i.e. call forwarding); and,
• economies of vertical integration beyond the network (i.e. information production

and distribution).
Before liberalisation was completed in EU countries, monopoly firms had been

producing all telecommunications services, themselves. Liberalisation accelerated
development of new services and the pace of technological changes in
telecommunications sector. Technological developments interrelated with increasing
‘digitalisation’ have improved technical possibilities for vertical separation of
different service elements as well as integration of transmission of different services
in the same network.219 As a consequence, decreasing costs of transmission give
more flexibility for the location of value added services such as voice-mail and other
intelligent network services, and all these facts give path for increasing each type of
economies of scope.220
2.2.5 Economies of Density
Economies of density is related to the fact that network costs per connection
decreases with increasing density of connections, the primary reason for which is
shorter access lines and better capacity utilisation of the network.221 Economies of
density resemble economies of scale with the difference that the latter is related to
the number of customers, whereas the former is related to density of customers – the
number of customers within a given area.222
219
220
Ibid.
221
Cable TV networks are worth noting here. The major cost of establishing such networks is to dig
down the cables. Therefore, it is extremely expensive to deliver cable TV services to the first customer
at a given road, as it is needed to dig up the entire road. But as soon as the main cable is in place, the
cost of connecting more houses to the cable TV network amounts to laying down a cable from the
house to the road. (cited in Jonas Holm, p. 5)
222
Jonas Holm, “Regulating Network Access Prices under Uncertainty and Increasing Competition:
The Case of Telecommunications and Local Loop Unbundling in the EU”, 2000, University of
Copenghagen, Institute of Economics (MSc Thesis), p. 5.
71
Established economies of density make very difficult for new entrants to compete
with incumbent operators in a relevant market, where local networks with a high
penetration already have been established. This is of course particularly important in
the residential market, where the revenue per customer is much lower than in the
business sector.223
A new entrant may either build its infrastructure on its own or access to the networks
of the incumbent operator. In some instances, a network can be duplicated on top of
other types of infrastructures i.e. local, cable TV or electric power networks. But
even in these cases considerable investments must be made before sufficient
economies of density in supplying interactive network services can be achieved.224
From a regulator’s point of view, the entry barriers created through economies of
density as well as by economies of scope and scale can be surmounted either by
supporting the building of alternative infrastructures or resorting to access measures
in order to ensure competitors fair access to existing network facilities. The first
solution may promote competition in the long run, whereas it may be a costly
solution, as it entails duplication of network facilities by incurring huge expenditures.
Considering the fact that such a duplication is seriously uneconomic and inefficient,
it would be more preferable to provide new entrants with access to the incumbents’
facilities on reasonable terms.
Aiming at achievement of a pro-competitive environment, the Essential Facilities

Doctrine could be invoked in sector segments where significant economies of
density, scale and scope constitute a considerable entry barrier. However, even in
invoking EFD, one must be cautious and take the ‘efficiency’ problem into account
with the possible short & long term conflicts. That is to say, EFD should be applied
in more or less strict conditions after investigating how to promote the competition
under the parameters cited above.
223
224
Ibid.
72
2.2.6 Other Barriers to Entry
General characteristics of telecommunications networks provide strong advantages to

well-established incumbent operators. Such advantages stem not only from the nature
of network industries, but especially from the historical and governmental
foundations of incumbent operators. As a matter of fact, many established incumbent
operators have a long history of providing local access services at subsidised rates.
Economies of scale, scope and density are all created upon these advantages and
such strong advantages take the form of entry barriers for many times. Most
prominent types of such entry barriers (other than those stated above) are able to be
classified as follows: 225
• Government restrictions such as monopoly franchises or restrictive licensing

practices,
• High fixed/capital costs,
• Intellectual property rights such as copyright and patent protection (which may
affect the availability to a competing supplier of key inputs or outputs),
• Control over network architecture and development of network standards
• Existence of a high degree of customer inertia
• Vertically integration alongside the upstream and down stream markets
• Application of cross-subsidised rates (between different service areas)
Here must be done a distinction between barriers to entry. Economies of scale, scope
and density are inherent advantages already established from the scratch, which are
able to be called ‘natural’ or ‘structural’ barriers to entry. Other advantages of
incumbent operators given above are to a great extent, created upon economies of
scale, scope and density, while those advantages also give rise to establishment of
economies of scale, scope and density. There is a thus positive feedback effect
225
H. Intven, J. Oliver, E. Sepulveda, “Telecommunications Regulation Handbook”, 2000, Published
by McCarthy Tetraultz, p. 5/11 Available at http://www.infodev.org/projects/314regulationhandbook/
73
between the so-called natural advantages (economies of scope, density, and scale)
and unnatural ones (monopoly franchises, customer base, etc.). In addition, the lastly
cited two entry barriers (vertically integration and cross-subsidisation) consist of
‘behavioural’ elements such as abusive practices that an incumbent operator engage
in order to prevent new entries. That is to say, the so-called (two) barriers to entry are
not naturally existed in the market and have strategic elements.
When dynamic nature of telecommunications markets is considered, entry barriers

are potentially able to vary from time to time and from place to place. Barriers to
entry in telecommunications sector are related to both size and lack of flexibility in
investments.226 In this regard, required level of investments against the demand for
different services is potentially deemed to exclude new entrants in the
telecommunications markets. This situation which requires considerable financial
strength constitutes an important entry barrier. But of more important is the lack of
flexibility in investments already made.227 Investments in telecommunications sector
are for a large part called ‘sunk costs’ – investments in relatively long-lived assets
ear-marked for a specific activity.228 That is to say, once the investment is made it
will be very difficult to leave the market without major losses.
EFD has elements of both natural monopoly and barriers to entry.229 As avoiding
duplication of network facilities is deemed the primary concern under ‘natural
monopoly’ notion, many policy makers are of the opinion that such a duplication in
telecommunications sector would bear undesirable impacts in efficiency terms. Also,
by virtue of their cost and difficulty of duplication, essential facilities resemble
certain forms of barriers to entry.230 In fact, an entrant can not presumably duplicate
existing network facilities that are alleged ‘essential’ economically because, while an
226
227
A fixed network cannot be moved and can only serve communication between certain specific
locations. Many other industries can sell their products at different markets without major changes in a
production equipment. However, in telecommunications this is not possible.
228
229
Daniel F. Spulber, “Handbook of Telecommunications Economics”, Elsevier Science B. V., 2002,
Amsterdam, The Netherlands, (Ed. by M. Cave, Sumit K. Majumdar, I. Vogelsang), p. 181.
230
Ibid.
74
incumbent has already incurred high amounts of fixed costs, new entrants have to
make an irreversible investment with all accompanying economic risks. Moreover,
new entrants’ costs of duplicating allegedly essential facilities usually surpasses
those of incumbent operator.
2.3 European Telecommunications Sector
History and development of EU telecommunications sector is convenient to be

examined within three phases. Although there appear some different approach, the
said three phases are widely accepted as respectively liberalisation, harmonisation
and convergence. These phases will be expounded below, as they are important
achievements of the European Union in the global environment as well as in terms of
development of information society.
Essentially, technological developments (including convergence) and rising demand

for telecommunications services contributed to give increased significance to the
telecommunications sector, both economically and socially.231 As a consequence, a
serious growth in economic activities in telecommunications and related sectors such
as broadcasting, information technologies and multi-media services is confronted in
the European context. Correspondingly, EC law and policy in field of
telecommunications have been given a distinct role via a set of fundamental
principles.232
As the economy is globalising, demand for telecommunications is growing massively

and firms and individuals increasingly need to communicate with other firms and
individuals. At the same time, technological advances over the past thirty years
231
European Commission, 1987 Green Paper (Towards a Dynamic European Economy: Green Paper
on the development of a Common Market for Telecommunications Services and Equipment),
COM(87)290, Presentation at 1-3.
232
EC Telecommunications policy would be much like EC environmental policy, for instance, which
is given a distinctive content and significance through a set of fundamental principles defined at EC
level and anchored in the EC Treaty itself. The basis principles of EC environmental law are set out in
Article of 174 EC, which states at para. 2 that “Community policy on the environment shall aim at a
high level of protection taking into account the diversity of situations in the various regions of the
Community. It shall be based on the precautionary principle and on the principles that preventive
action should be taken, that environmental damage should as a priority be rectified at source and that
the polluter should pay.” (cited in P. Larouche, “Competition Law and Regulation in European
Telecommunications”, Hart Publishing, 2000, Introduction.)
75
progressively enabled telecommunications operators to meet such consumer
demands. Innovations such as fibre optics, digitalisation and packet-switching
changed completely not only the technical, but also the economic environment of
telecommunications.233 Whereas most commentators used to agree that
telecommunications was a natural monopoly, this traditional approach tended to be
turned aside along with the abovementioned changes having taken place. The
combination of these factors led the Commission to issue, in 1987, its Green Paper
(on the development of a Common Market for Telecommunications Services and
Equipment) which envisaged a number of changes in EU telecommunications policy
towards progressive liberalisation. The Commission’s Green Paper set out to develop
a Community-wide program for action in this area that was based, generally, upon
two complementary main themes. The first involved a process of liberalisation,
which sought to create fully liberalised markets in telecommunications sector. The
second involved a process of harmonisation of the conditions for the operation of
telecommunications networks during and after the liberalisation period.
The fundamental basis for adoption of directives with the purpose of ensuring EU
policy objectives was Article 86(3)234 of EC Treaty that is integrated with Article 95
EC. On the one hand, Article 86 entrusts to the Commission a specific obligation of
surveillance and of eliminating competition distortions and/or restrictions. This
Article applies more precisely when the Commission acts in its function as the
guardian of the Treaty to put an end infringement(s) of competition rules. On the
other hand, Article 95 has a function of harmonisation in order to abolish barriers
233
Publishing, 2000, Introduction.
234
Article 86 EC reads as follows:
1. In the case of public undertakings and undertakings to which Member States grant special
and exclusive rights, Member States shall neither enact nor maintain in force any measure
contrary to the rules contained in this Treaty, in particular to those rules provided for in Art.
12 and Art. s 81 to 89.
2. Undertakings entrusted with the operation of services of general economic interest or having
the character of a revenue-producing monopoly shall be subject to the rules contained in this
Treaty, in particular to the rules on competition, in so far as the application of such rules does
not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The
development of trade must not be affected to such an extent as would be contrary to the
interests of the Community.
3. The Commission shall ensure the application of the provisions of this Art. and shall, where
necessary, address appropriate directives or decisions to member states.
76
resulting from a divergence of national legislation or regulations. From this vantage
point of view, Articles referred above (Article 86, 95) are complementary in nature
and are not to be deemed substitutable between each other.
On the basis of the so-called Articles, the Commission put forward an action
programme for implementation of the 1987 Green Paper, which the Council agreed
by a Resolution of 30 June 1988.235 The Community positions under this action
programme can be outlined as follows:236
1. Member States may leave telecommunications infrastructure under monopoly,

and must preserve network integrity in any event,
2. Amongst services, only public voice telephony may be left under monopoly,
3. Other services must be liberalised,
4. Community-wide interoperability must be achieved through harmonised,
5. An Open Network Provision (ONP) framework must be put in place to regulate

the relationship between monopoly infrastructure providers and competitive service
providers (including trans-border interconnect and access)
6. Terminal equipment must be liberalised,
7. Regulatory and operational functions of the PTOs (public telephone operator)

must be separated,
8. Competition law must be applied to PTOs (public telephone operator), in

particular as regards cross-subsidisation,
9. Competition law must be applied to new service providers as well,
10. The Common Commercial Policy must be applied to telecommunications,

competition law must be applied to international telecommunications.
235
Council Resolution of 30 June 1988 on the development of the common market for
telecommunications services and equipment up to 1992 [1988] OJ C 257/1.
236
Publishing, 2000, p. 4-7.
77
As could be seen, EU telecommunications sector was envisaged to be liberalised
according to the Community positions stated above. In order to ensure that the
operation of the networks which was left under monopoly not affect the competitive
services, some precautionary measures were needed to be adopted at that time. That
is to say, some questions remained open, since there was no obligation to liberalise
infrastructure (See above 1), while services were in principle liberalised. These
questions with regard to the so-called complicated situation had been tried to be
smoothed out with the adoption of the Open Network Provision (ONP) Directives
afterwards. Below will be explained the historical processes until and after full
liberalisation as well as technical characteristics of the ONP Directives, respectively.
2.3.1 Liberalisation
As pointed out above, the liberalisation of EU telecommunications markets has taken

place largely upon the implementation of Article 86(3) with. Aiming at adopting
liberalised telecommunications markets, the Commission interpreted Article 86(3) in
a broader sense and abandoned the traditional view which was surrounded only by
‘management of special and exclusive rights’.237 This expansionist interpretation
reveals the ambition of the Commission for rapid development of
telecommunications sectors along with liberalisation. In confirming and clarifying its
powers under Article 86(3), the Court of Justice considerably strengthened the
Commission’s position in the liberalisation process, and fears that an excessive use
of the so-called Article might upset the institutional balance between the Community
and the Member states proved to be unfounded.238
237
238
J. Braun and R. Capito, “EC Competition and Telecommunications Law” (International
Competition Law Series, Vol. 6), Kluwer Law International, 2002 (Edited by C. Koenig, A. Bartosch,
J. Braun), 2002, p. 55. After the adoption of the Commission Directives as regards liberalisation under
Article 86(3), some Member States challenged such directives before the ECJ. However, the ECJ
upheld the validity of those directives in its decisions, i.e. France v Commission and Spain v
Commission. In the former judgment, the ECJ upheld the Commission’s authority to require Member
States under Article 86(3) to abolish exclusive rights regarding technical equipment. In the latter
judgment, the Court similarly held that the Commission has the power on the basis of 86(3) to adopt a
directive laying down general rules that specify Member States’ obligations under the Treaty.
78
According to H. Ungerer, EC telecommunications liberalisation developed mainly as
a consequence of three factors.239 Firstly, by the end of the eighties, the growing
digitalisation of European telecommunications networks began to transform them
into multipurpose information infrastructures. The opportunities offered by
telecommunications networks and services started to extend into markets
substantially beyond the traditional telephone service, such as markets for value-
added-services i.e. Internet services.
As a result, the access to the traditional monopoly networks in the

telecommunications sectors became a major issue in all EU Member States, and there
was a growing conviction that without a loosening of monopoly rights, it could
neither be assured that new markets could develop nor that the new services offered
could be made available to consumers. Secondly, in British Telecommunications
judgment,240 European Court of Justice confirmed that EC Competition Rules are
applicable to the telecommunications sector. Thirdly, the impact of developments in
the United States, in particular the AT&T divestiture consent decree and the resulting
transformation of the US market, began to be felt in Europe. At the same time, the
progressive liberalisation of the telecommunications sector and the privatisation of
British Telecom in the United Kingdom since 1982 made Europe more receptive to
the concept of market deregulation.
These and similar driving factors gave path for the Commission to set forth a
comprehensive policy framework regarding liberalisation of EU telecommunications
sector. The Green Paper 1987 has been the leading transformative tool on the part of
the Commission and following the adoption of the Green Paper, many Directives has
been enacted and implemented.
The first step following Green Paper in liberalisation process is the adoption by the
Commission of the Terminal Equipment Directive 1988.241 The 1988 Terminal
239
Herbert Ungerer, “Access Issues under EU Regulation and Anti-Trust Law - The Case of the
Telecommunications and Internet Markets”, July 2000, Research Paper, WCFIA Fellows Program
1999/2000, Harvard University (Weatherhead Center for International Affairs), p.10.
240
Commission Decision 8/861/EEC, OJ L 360/36.
241
Commission Directive 88/301/EEC of 16 May 1988 on competition in the markets in
telecommunications terminal equipment, OJ 1988 L 131/73.
79
Equipment Directive obliged Member States to remove special or exclusive rights
relating to the importation, marketing, connection, bringing into service and
maintenance of telecommunication terminal equipment.242 The Terminal Equipment
Directive also obliged Member States to create an independent (regulatory) body
responsible for drawing-up specifications for, and subsequently monitoring, a type-
approval process for competitively-supplied equipment.243
The liberalisation of services continued with the enactment of the Services

Directive.244 It provides for removal of special and exclusive rights granted by
member states for the supply of all telecommunications services other than voice
telephony.245 But, in doing so, the Directive implicitly permitted the continuation of
monopoly with respect to telex, mobile, paging and satellite services.246 Member
States were repeatedly after the Terminal Equipment Directive, required to ensure
that regulatory functions within the framework of the Directive 90/388/EEC be
carried out by bodies independent of the telecommunications organisations.247
The above-mentioned exceptions to the liberalisation scheme set out under both
Terminal Equipment and Services Directive have been seemingly emerged pursuant
to the derogatory provision of Article 86 (2). Given the fact that one could not
mention about fully competitive services in an environment equipped with such
protectionist tools, the Commission took action for removing all special and
exclusive rights, at last.
242
Commission Directive 88/301/EEC of 16 May 1988 on competition in the markets in
telecommunications terminal equipment, OJ 1988 L 131/73, Art. 2 and 3.
243
Ibid., Art. 6.
244
Commission Directive 90/388/EEC of 28 June 1990 on competition in the markets for
telecommunications services, OJ 1990 L 192/10.
245
Ibid., Art. 2.
246
Ibid., Art. 1. The monopolies over the telecommunications services which continued to be deemed
‘reserved service’ after the Services Directive were lifted through legislative amendments via the
adoption of further directives such as Satellite Directive (94/46/EC), Cable Directive (95/51/EC), and
Mobile Directive (96/2/EC). In other words, with the adoption of these three Directives, special and
exclusive rights over cable, satellite and mobile communications services were totally removed.
Therefore, the so-called Directives (on cable, satellite and mobile services) that were enacted in a
sequence can be seen as logical extensions of the original Services Directive.
247
Ibid., Art. 7.
80
In this regard, the Commission took its final step towards liberalisation in 1996 with
the adoption of the Full Competition Directive.248 This Directive required Member
States to ensure that any remaining restrictions on services competition as well as
deployment of alternative infrastructure be removed by 1 January 1998.249 Full
Competition Directive also specified that remaining restrictions on the use of
‘alternative infrastructure’ should be lifted by 1 July 1996.250
2.3.2 Harmonisation: ONP Directives
The liberalisation process was complemented and reinforced by a series of Directives

designed in order for gradual harmonisation of Member States regulations on a
Community-wide basis. That is explicitly to say, Harmonisation (ONP) Directives
essentially aimed at standardisation of the national measures as regards the
conditions for access to and use of public telecommunication networks and services.
Harmonisation (ONP) Directives followed to a large extent the principles set out in
the liberalisation directives.251 However, this relationship is predominantly valid in
the opposite direction, that is, a harmony exists between harmonisation and
liberalisation directives. According to P. Larouche, liberalisation directives, i.e.
Terminal Equipment and Services Directives (Directives 88/301 and 90/388) contain
rules of precedence that seem to give priority to harmonisation (ONP) directives.252
248
Commission Directive 96/19/EC of 13 March 1996 amending Directive 90/388/EEC with regard to
the implementation of full competition in telecommunications markets, OJ L 74.
249
Ibid., Art. 1.
250
Ibid. The term ‘alternative infrastructure’ refers to any telecommunications infrastructure owned by
some other party than the local telecommunications operator, which then constitutes an alternative to
the public telecommunications infrastructure.
251
P. Nihoul, “Convergence in European Telecommunications: A case study on the relationship
between regulation and competition (law)” IJCLP (International Journal of Communications Law and
Policy), Issue.2, Winter 1998-99, p. 24-28. Available on the IJCLP Website at http://www.digital-
law.net/IJCLP/index.html.
Here, the distinction must be established that whereas liberalisation directives were adopted by the
Commission pursuant to Art. 86(3), harmonisation directives were adopted by the European
Parliament (after 1993) and the Council pursuant to Art. 95 EC.
252
Publishing, 2000, p. 66. Full Competition Directive (Directive 96/19/EC) sets out a general principle
concerning the relationship between the harmonisation and liberalisation measures:
“The establishment of procedures at national level concerning licensing, interconnection, universal
service, numbering and rights of way is without prejudice to the harmonisation of the latter by
81
This chain generally revealed as follows: when a liberalisation directive, i.e. Services
Directive was adopted, this Directive was followed by relevant ONP Directives.
Consonantly, when a harmonisation directive, i.e. 95/62/EC (Voice Telephony)
Directive was enacted, the Commission undertook to review its (liberalisation)
directive for ensuring consistency. Then, following adoption of a liberalisation
directive, a new ONP regime considering the newly liberalised environment was to
be set out by the European Council and Parliament, accordingly.
Harmonisation measures which deal with highly important policy objectives at EU

level, have been carried out under the concept “Open Network Provision”. Open
Network Provision (ONP) is defined in the Open Network Provision Framework
Directive as ‘the harmonisation of conditions for open and efficient access to and use
of public telecommunications networks and, where applicable, public
253
telecommunications services’. The ONP Framework Directive was adopted at the
same time as the Commission’s Services Directive, aiming at realisation of
harmonisation objectives as well as facilitation of liberalisation process in an
effective manner. The ONP Framework Directive provided that access to public
telecommunications networks and already-liberalised public telecommunications
services would be provided on the basis of non-discriminatory, objective and
transparent conditions published in an appropriate manner.254 It stipulated that access
could be denied only on the basis of “essential requirements” which were also
specified in the liberalisation directives, i.e. Services Directive.255 The principles
defined in the ONP Framework Directive were later refined and applied to various
appropriate European Parliament and Council legislative instruments, in particular in the framework
of open network provision (ONP).”
253
Council Directive 90/387/EEC of 27 June 1990 on the establishment of the internal market for
telecommunications services through the implementation of open network provision, OJ L 192/1, Art.
2.
254
Ibid., Art. 1 and 3.
255
Council Directive 90/387/EEC of 27 June 1990 on the establishment of the internal market for
telecommunications services through the implementation of open network provision, OJ L 192/1, Art.
3. Article 4(1) of the Service Directive required the conditions of access to telecommunications
networks to be objective, non-discriminatory and public. The Directive permitted Member States to
make supply of services as well as the operation of networks subject to relevant procedures such as
licensing conditions. However, such conditions are only permissible to the extent that they are aimed
at securing compliance with ‘essential requirements’ which are defined as the “non-economic reasons
in the general interest which may cause a Member State to restrict access to the public
telecommunications network or public telecommunications services”.
82
telecommunications services, i.e. leased lines, voice telephony and interconnection
with further ONP Directives.
One of the following ONP Directives is concerned with leased lines256 and the other
one is related to voice telephony.257 The ONP Leased Lines Directive required each
Member State to ensure that users within the Member State have access to a
minimum set of analogue and digital leased lines with harmonised technical features
from at least one organisation (in practical terms, such organisations have been the
incumbent operators enjoying the exclusive and special rights) in that State and
stipulated that, until effective competition has been achieved, prices for leased lines
must be cost-oriented, non-discriminatory and transparent.258
Second of the further ONP Directives, ONP Voice Telephony Directive259 was
adopted in 1995. ONP Voice Telephony Directive laid down the minimum
requirements pertaining to access to the fixed public telephone networks. In this
regard, the Directive required Member States to ensure that public pay telephones be
provided so as to meet the reasonable needs of users as well as to ensure that users
have access to operator assistance and emergency and directory enquiry services.260
Whereas the ONP Voice Telephony (95/62/EC) and Leased Lines (92/44/EEC)
Directives aimed at harmonisation measures relating to specific telecommunications
256
Council Directive 92/44/EEC of 5 June 1992 on the application of open network provision to
leased lines, OJ L 165/27.
257
Council Directive 95/62/EC of European Parliament and of the Council of 13 December 1995 on
the application of open network provision to voice telephony, OJ L 131/6.
258
Op.cit in note 256, Art. 7 and 10. The ONP Leased Lines Directive was amended in 1997 so that
requirements which had originally been applicable only to bodies with ‘special or exclusive rights’
became applicable to any organisation (firm) with ‘significant market power’ (SMP) in the supply of
leased lines within a particular geographic area in a Member State. SMP was presumed to exist where
an operator had 25% or more of the relevant market.
259
Council Directive 95/62/EC of European Parliament and of the Council of 13 December 1995 on
the application of open network provision to voice telephony, OJ L 131/6. Subsequently, this
Directive was replaced with the Directive 98/10/EC of European Parliament and of the Council of 26
February 1998 on the application of open network provision to voice telephony and universal service
in a competitive environment, OJ L 101.
260
95/62/EC (ONP Voice Telephony) Directive prescribed some basic rules for the provision of
universal service, as well. The concept ‘universal service’ could be summarised as “the obligation to
provide access to the public telephone network and to deliver affordable telephone service to all users
reasonably requesting it.” (cited in the ONP Voice Telephony Directive.)
83
services; the Interconnection Directive261 was adopted in order for interconnection of
networks and interoperability of services through the application of ONP. Thereby,
the main concern of the so-called Directive was the harmonisation of conditions for
open and efficient interconnection of and access to the public telecommunications
networks and publicly available telecommunications services. According to Article 4
of the Interconnection Directive, telecommunications operators shall have a right and
an obligation to negotiate interconnection. Moreover, organisations with significant
market power (SMP) were obliged to meet all reasonable requests for access to their
networks.262 Pursuant to the Interconnection Directive, the role of the national
regulatory authorities is determined as monitoring and - if necessary - interfering
with the interconnections agreements in order for ensuring the ONP objectives in
each Member State.263
From this point of view, the harmonisation (ONP) measures have a thoroughly
reinforcing as well as a transformative effect in ensuring competitive
telecommunications services and markets at EU/national level. Given the fact that in
a liberalised environment end-users as well as potential competitors could be affected
in lack of harmonisation measures, one can easily understand the vitality of such
measures. In this sense, the ONP framework could be deemed the cornerstone for
telecommunications policies, because it is concerned with establishment of
fundamental principles which have served as a blueprint for the future of Member
States.
261
Directive 97/33/EC of the Parliament and the Council of 30 June 1997 on interconnection in
telecommunications with regard to ensuring universal service ad interoperability through application
of the principles of Open Network Provision (ONP), OJ No L 199/32.
262
Before the Interconnection Directive having provided SMP criteria, the organisations adressed for
(access) obligations were the incumbent operators enjoying special and exclusive rights in the sector.
Subsequent to entry into force of the Interconnection Directive, the SMP criteria have been given a
central role for access obligations. According to Article 4 of the Interconnection Directive, operators
having SMP were required to meet all reasonable requests for access to their networks, including
access at points other than network termination points offered to the majority of users.
263
Directive 97/33/EC of the Parliament and the Council of 30 June 1997 on interconnection in
telecommunications with regard to ensuring universal service ad interoperability through application
of the principles of Open Network Provision (ONP), OJ No L 199/32, Art. 9.
84
2.3.3 Convergence
Being widely accepted as having the meaning of ‘coming and meeting together’
literally, convergence does not have a common definition in field of electronic
communications. But it is generally acknowledged that the traditional boundaries
between ‘broadcasting’ and ‘telecommunications’ are blurred by convergence. In
Convergence Green Paper,264 it is stated that “the term ‘convergence’ eludes precise
definition, but it is most commonly expressed as:
- the ability of different network platforms to carry essentially similar kinds

of services, or
- the coming together of consumer devices such as the telephone, television

and personal computer”
Convergence between telecommunications and media is the most paradigmatic

change in rapid evolution of converging markets. How the convergence between
telecommunications and media takes place is able to be explained as follows.265
Traditionally, these two sectors are operated in their respective isolation, from a
technological, industrial, commercial and legal standpoint. The model of
telecommunications was point-to-point communications on a two way switched
network, with its own technology, its own firms, its own services and its own
framework. Conversely, audiovisual media was based on point-to-multipoint (one-
way) networks, with wireless technology, firms dealing with media only and specific
services (radio and television) regulated under a specific legal framework.
Convergence in this case means that the so-called clear boundaries between these
two sectors are becoming blurred. (See the figures below.)
264
Green Paper on the Convergence of the Telecommunications, Media, and Information Technology
Sectors, and the Implications for Regulation Towards an Information Society, COM(97) 623 (3
December 1997), p. 8. Available at http://europa.eu.int/ISPO/infosoc/telecompolicy/en/comm-en.htm
265
Publishing, 2000, p. 204-206.
85
Figure 4: Before convergence
Figure 5: After convergence
(Source: Conference on “EU Telecommunications Regulations & Law”, Brussels, 25th & 26th June
2002)
As is seen in the figures above, converging markets are undergoing dramatic changes
towards a distinct platform including multi-functional networks, removing
boundaries, wide-ranging problems all over the neighbouring sectors.266 Although in
the past, the commercial separation of telecommunications and broadcasting has
266
The ability of different network platforms to carry essentially similar kinds of services is the most
prevailing side of convergence. For instance, the ‘audiovisual sector’ convergence is characterised by
digital television, Internet and many other services which are emerging somewhere between two.
Digital television offers a new range of choice and interactivity and has the ability to be transmitted
and understood by computers and therefore links the global reach of computers with content
distribution by broadcasters.
86
been mirrored with separate regulatory authorities; with convergence, the
commercial distinctions are being eroded and the rationale for multiple regulations is
being questioned.
All the current and possible problems regarding ‘convergence’ phenomenon with
their future implications were analysed in the Commission’s ‘Green Paper on the
Convergence’.267 Therein, the Commission suggested three possible methods for
regulating converging markets.268
(i) remaining with the current approach, i.e. separate regulatory framework for
telecommunications and media, extended to new converged activities as the case
may require,
(ii) developing a new framework for converged activities alongside the existing ones
and,
(iii) fusing all existing framework into a single new ‘converged’ framework
Final consultation over these possible solutions led to the conclusion that the
regulation of infrastructure should be separated from that of content, and the sector-
specific regulation should be phased out as markets become more competitive and
can be left to competition law alone.269 The consultations of the Commission gave
way to the transformation of the existing Directives and introduction of a new
Regulatory Package, which is going to be detailed later.270
267
Green Paper on the Convergence of the Telecommunications, Media, and Information Technology
Sectors, and the Implications for Regulation Towards an Information Society, COM(97) 623 (3
December 1997), p. 8. Available at http://europa.eu.int/ISPO/infosoc/telecompolicy/en/commen.htm.
268
Ibid. These methodological questions at EU level have found their respective answers in the
‘Results of the Public Consultation on the Green Paper on the Convergence of the
Telecommunications, Media and Information Technology Sectors’ (10 March 1999, (COM) 99 108).
269
Op.cit. in note 267, p. 49.
270
Here, suffice to say that the convergence of the telecommunications, broadcasting and information
technology sectors means that all transmission networks and services should be covered and operated
under a single regulatory framework. This transformation has taken place to a great extent with the
introduction of the 2002 EU Regulatory Framework at the EU level.
87
2.4 Dual Regime in EU Telecommunications Sector: Sector-Specific and
Competition Law Rules
Like other network industries such as the postal, energy and rail transportation,
telecommunications sector has always been and still is a heavily regulated sector.
The EU telecommunications sector has also been characterised by an intensively
interventionist policy, in that a more sophisticated application of general competition
law played a crucial role. The need for competition law has evidently been echoed in
the 1987 Green Paper, which envisaged a policy package that became the core of the
EC telecommunications policy in 1990s.271
In the course of implementing EU telecommunications policy, application of EC

Competition Law has been of primary importance from the scratch. In addition,
especially with the acceleration of the liberalisation process, a comprehensive sector-
specific regulation has also been developed, giving rise to a dual regime at
EC/national level.272
Notably, since 1990s many ‘networked’ sectors had undergone many regulatory
reforms depicting common characteristics, which are still under agenda of many
countries particularly of EU candidate countries such as Turkey, Romania, etc. These
different regulatory reforms were resulted in sector-specific regimes that are
distinctly different from traditional competition rules.
In the context of a dual regime, telecommunications and related sectors such as

information technology, Internet and multi-media services are controlled on the one
hand through sector-specific regulation and on the other hand via general
competition law rules. Within the European dual regime; the Directives, Regulations,
Recommendations and Notices constitute the major EU sector-specific measures,
271
European Commission, 1987 Green Paper (Towards a Dynamic European Economy: Green Paper
on the development of a Common Market for Telecommunications Services and Equipment), COM
(87)290, p.16-17, 184-5.
272
An exception to the dual regime rule in telecommunications sector is New Zealand, where national
regulatory authorities have been abolished in favour of competition authorities.
88
whereas the Commission’s Decisions under the oversight of the ECJ comprise the
core of the EC Competition Law.273
The most advantageous and determinative aspect of general competition law rules is
that they apply entirely sector independent. This makes them more flexible in
comparison to sector-specific rules which include more complex and detailed
regulatory principles that apply exclusively to one sector.274 However, by and large,
regulatory authorities have wider control rights than competition authorities, given
the fact that competition law rules challenge the lawfulness of conduct, while
regulatory authorities engage in detailed regulation of wholesale and retail prices,
profit sharing, investments, etc.275 Besides, regulatory authorities are more at ease
with quantitative evidence, which they often use to set very detailed regulations, as
in the case of cost-based pricing rules. In contrast, competition authorities are in
shortage of detailed data, being usually more at ease with cases based on qualitative
evidence (price discrimination, price fixing, vertical restraints, etc.).276
273
As regards application of EC Competition rules, on the one hand, the European Commission is
empowered to apply Articles 81-89 EC and to take measures at the EU level (under the control of the
European Court of Justice) on the other hand, national competition authorities have the authority to
enforce the competition law rules, subject to the Community procedures, among which the Regulation
17 (Council Regulation No 1, implementing Articles 81 and 82 of the Treaty, OJ 13, 21.2.1962 and
subsequent Notices) is deemed to be the main guideline. However, as regards sector-specific rules, EU
model is based on a decentralised approach under which telecommunications regulation is carried out
by the national regulatory authorities in accordance with the European Regulatory Framework(s),
without a Community-wide regulatory body.
274
Both the competition law rules and the sector-specific rules aim at ensuring competitive markets
and eliminating market failures. From this point of view, the broader concept ‘competition policy’ is
used so as to encompass all types of government policies inter alia sector-specific rules, competition
law principles, etc. which are directed to enhance competition in field of telecommunications.
In economic terms, both sector-specific rules and competition law principles are based on common
welfare foundations such as allocative, productive and distributional efficiencies. According to the
first one (allocative efficiency), resources must be allocated so as to produce the maximum benefits to
consumers, that is the economy must maximise allocative benefits. As to the productive efficiency, the
resources must be produced at the minimum cost so that they can be released to satisfy other demands,
that is the economy must maximise productive efficiency. In the context of distributional efficiency,
the resources must be distributed to maximise distributional efficiency. In the light of these objectives,
on the one hand (usually) ex ante and more detailed remedies are invoked under sector-specific
regulation, on the other hand (usually) ex post and more flexible (sector independent) measures are
applied under competition law.
275
Jean-Jacques Laffont and Jean Tirole, “Competition in Telecommunications”, The MIT Press,
(Fourth Ed.), 2002, p. 277.
276
Ibid., p. 278.
89
As widely accepted, sector-specific rules have a transitional character and are
designed to ensure that the telecommunications markets would be more competitive.
In this respect, after telecommunications markets have become more competitive, the
Community’s sector-specific regulation may thus be phased out and the sector may
solely governed by the Treaty’s competition law regime.277 This is illustrated within
the following figure:
Figure 6: Transition from heavy-handed (sector-specific) regulation to reliance on competition law

(Source: Conference on “EU Telecommunications Regulations & Law”, Brussels, 25th & 26th June
2002)
However one might have doubts as to whether general competition law rules will
replace sector-specific rules which are more complicated by nature. Since this
discussion needs a detailed analysis, one must improve any argument advancing
replacement of one type of measure with another, in a cautious manner.
In Telia-Telenor278 case, European Commission drew the line between the two types
of remedies (sector-specific rules and competition law rules), where the Commission
pointed out that a merger control procedure yielding evidence for the creation or
277
J. Braun and R. Capito, “EC Competition and Telecommunications Law” (International
Competition Law Series, Vol. 6), Kluwer Law International, 2002 (Edited by C.Koenig, A. Bartosch,
J. Braun), p. 64.
278
Case No. IV/M.1439, Telia/Telenor (1999) OJ [2001] L 40/1, [2001] 4 CMLR 1226.
90
strengthening of a dominant market position can not be overruled by regulatory
control.279 At the same time, the European Commission required the two companies
(Telia AB from Sweden and Telenor AS from Norway) to unbundle the local loops
in their countries, and granted the parties a conditional clearance for merging.280
Similarly in Vodafone /Mannesman,281 the Commission only cleared the merger after
the parties submitted commitments to de-merge Orange Plc and to give other mobile
operators access to their inter-operator roaming tariffs and wholesale services.
These two cases exhibit a new form of regulation under general competition law
rules which exclude sector-specific regulation to some extent. Here, we do face ex
ante remedies under merger control as a non-typical competition law measure, in
contrast to general nature of general competition law rules which basically include ex
post measures. These (hybrid) measures take the form of either structural remedies
that stimulate network competition or behavioural remedies which aim at ensuring
reasonable access to key inputs such as content, local loop or set top boxes, under the
merger.282
Such combinative usage reveals that the Commission perceives ex post and ex ante
remedies as complementary tools in order for elimination of market failures in
telecommunications markets.283 Hence, it is possible to say that even if sector-
279
Georg Koopman, “Competition Policies and Telecommunications Regimes”, p. 19-20. Available at
http://www.hwwa.de/Projekte/Forsch_schwerpunkte/FS/Hande/Publikationen/Koopman%20edit.pdf
280
The intended merger between Telia AB and Telenor AS was ultimately not realised although it was
conditionally permitted. It was the first case to be considered under the EU’s Merger Regulation
involving the merger of two incumbent national telecommunications operators in EU. However, the
Commission decision was debated with regard to the question whether the Commission should use
merger control to advance its regulatory agenda. Such debate was developed for the reason that Telia-
Telenor decision was rendered more than a year ahead of the Regulation 2887/2000 concerning
unbundled access to the local loop.
281
Case No. M. 1795 (2000) Vodafone/Mannesmann, IP/00/373 of 12 April 2000.
282
Damien Geradin and J. Gregory Sidak, Seminar on “European and American Approaches to
Antitrust Remedies and the Institutional Design of Regulation in Telecommunications”, 22 January
2004, Maastricht, The European Institute of Public Administration (EIPA), p. 17.
283
This seems so, because though being created for ensuring common objectives in general,
competition law and sector-specific rules contain different types of measures which complement each
other. More specifically, telecommunications regulations include (i) removal of barriers in order to
promote new entry into respective markets and to define the conditions of entry, (ii) determination of
procedures for number allocation, number portability, dial parity, and radio-electric spectrum
allocation, (iii) setting forth access and interconnection conditions and prices, (iv) designation of price
91
specific regulations have reached their aims and were superseded by competition law
rules, ex ante remedies are still going to be applied in telecommunications sector.
2.5 Establishment of the EU Access Regime: Policy Objectives and Legislative

Tools
2.5.1 General Overview
In most countries, a combination of general competition law and sector-specific rules

is generally deemed a mechanism improving the development of telecommunications
sector. In parallel with the so-called mechanism, a theoretical paradigm relying upon
a distinct perspective is sometimes used. According to that paradigm, two levels of
competition, ‘service-based competition’ and ‘facilities-based competition’ are
invoked in order for policy arrangements.
When the facilities of the incumbent operators are granted for access or shared with
third parties, one can say about the existence of an access regime. An access regime
is based on determination of terms, conditions and charges to be applied between the
access provider and the access seeker. In this context, such terms, conditions and
prices could not be always determined by the parties, particularly in case of dispute.
Not only in case of dispute, but also in general terms, access conditions must not be
left to the parties at all, because of the imbalance between the parties.
Considering the asymmetrical powers (imbalance) between incumbent operators and

new entrants, the public authorities usually determine prices and other conditions of
an access regime in favour of new entrants and urge to facilitate new entries via some
policy tools. That is to say, competent authorities, namely competition authorities
and regulatory authorities generally follow a ‘service-based competition’, which
encourages access to essential infrastructure, under familiar conditions. Provision of
services on fair and reasonable conditions rather than constructing networks is
and quality standards for telecommunications services, including universal service. Meanwhile
competition law rules essentially include prohibition of (i) anticompetitive agreements between
undertakings, (ii) abusive behaviours of dominant undertakings, (iii) mergers and acquisitions which
affect competition seriously.
92
embodied under service-competition strategies. However, such a strategy is not
permanently desired, because public policies do or at least ought to concern
investments and facility buildings, in the mid or long-term. Then, a political attitude
favouring network duplication rather than access to existing networks constitutes a
‘facility-based’ strategy. That is to say, when the new entrant prefers to build its own
facility instead of sharing with or access to the facilities of incumbent operator, there
exist facility-based competition. Facilities-based competition takes place when
access to existing network components is discouraged by the competent authorities
or is not preferred by operators. With a simplified account, the entrants’ incentives to
build their own facilities depend on the difference between the expected profit flows
from facility-based competition and service-based competition.284
Competition policies in telecommunications sector aim at promoting both service-

based and facility-based competition. Having regard to the disadvantages of
duplication of access networks and advantages related to emergence of competitive
services, national regulatory authorities usually prefer to adopt service-based
competition in the short term. In this regard, regulatory authorities as well as
competition authorities have become so familiar with mandatory access to network
facilities duplication of which is impossible or unfeasible in economic terms. As a
consequence, the notion of ‘access to essential facilities’ has become a theme of
central interest in respect of application of both EC competition law and European
telecommunications policies, in recent years.
Granting third party access to essential network facilities and the competitive
measures for this end figured on the agenda of EU telecommunications sector step-
by-step. First step in this process was the British Telecommunications285 case, where
the ECJ confirmed that the EC competition law is applicable to telecommunications
sector. In this case, the Commission found that British Telecommunications had
abused its dominant position by taking action to prevent certain private message-
284
M. Bourreau, P. Dogan, “Service-based vs. Facility-based Competition in Local Access
Networks”, Telecommunications Policy, 2001, Vol. 25, Issue. 3, p. 2.
285
93
forwarding agencies from offering a new type of service called ‘international telex
service’. Upholding the Commission’s decision, the ECJ stated that “the employment
of new technologies that accelerate the transmission of messages constitutes
technological progress in conformity with the public interest and can not be regarded
per se an abuse”.286
As value-added services287 such as internet access, electronic mail, voice mail and
online databases were progressively liberalised in EU, access to essential network
facilities started to become a recurrent theme and a central issue in the
telecommunications, media, and information technology markets.288 As a matter of
fact, one of the reasons why the Commission drives for a policy encouraging new
entry is that much of the growth in the telecommunications sector is due to the
expansion of value-added services, offered by new independent service providers.289
In order for the provision of competitive services not to be affected, the Commission
emphasised importance of the rapid development and use of Internet services,
drawing the necessary measures ensuring access to the bottleneck facilities for the
so-called policy reasons.290
Alongside the progressive liberalisation and harmonisation measures carried out in

1990s which are convenient to be deemed crucial steps towards open access regime,
competition law became more important than before. This is so, because the coherent
application of the EC Competition Law rules became more needed in order for
achieving internal market objectives in a liberalised environment, where small and
286
287
Value-added telecommunications services generally mean the telecommunication services which
employ computer processing applications that provide the users with additional, different or
restructured messages or involve users interaction with stored message.
288
Herbert Ungerer, Competition Workshop on “Ensuring Efficient Access to Bottleneck Network
Facilities: The Case of Telecommunications in the European Union”, 13 November 1998, Florence, p.
6.
289
J. Kalliala, “Market Definition Under the EC Competition Law in the field of Voice Telephony”,
2000, PILC Student Paper, Brussels, p. 36
290
At that time, access to the Internet in Europe was substantially more expensive than in United
States and most possibly the Commission had taken this fact into account when adopting measures in
order to facilitate new entries.
94
medium-sized operators, myriad new services and a great many (potential) strategic
anti-competitive behaviours are much more involved.
Throughout these developments, mandated access has gained a key role, via which
new entrants are enabled to provide new services or the existing services as
alternative service providers. As a corresponding matter, the issue of access and
interconnection began to dominate attention in the application of EC Competition
Law as a prelude to full liberalisation of telecommunications sector with the 1996
Full Competition Directive.291 Under the ONP regime, national regulatory authorities
were established in all Member States to deal with access issues, and national
regulatory regimes began to accompany enforcement of competition rules at national
level. The rapid establishment of European harmonised access and interconnection
regime together with the properly functioning competition law rules led to an
effective opening of core segments of the telecommunications network
infrastructure, and allowed rapid development of competition in both long distance
and international services, and in the long-distance network backbone.292
Regarding these developments, the Commission needed to issue a Notice concerning

application of EC Competition rules in telecommunications sector.293 In this regard,
291
6.
292
Ibid., p. 24. In telecommunications networks do exist three distinguishable segments: local loop,
which is the network component between subscribers and the point(s) of interconnection at local
exchanges; long distance, which is the network of cables and switching equipment that connects the
local exchanges to higher levels of exchange known as transit exchanges; and international, namely
the network of cables and related switching equipment which leads traffic from the international
gateway and hence out of the country and to public telephone operators in other countries. (cited in J.
Kalliala, p. 36) All of these segments are connected with each other by access and interconnection
arrangements. From the demand point of view, not only long-distance and international carriers but
also companies running private networks such as Internet Service Providers need access to and
possibly interconnection with the network of the incumbent operator in order to be able to exchange
traffic originating or terminating in the country concerned.
293
It is noticeable that effective enforcement of competition policies was achieved by the combination
of binding Directives with non-binding Recommendations, Notices, etc. that are called ‘soft
legislation’. As being one of the most important soft legislative measures, the Access Notice (1998)
aimed at not establishing strictly binding measures, but ensuring certainty with regard to the dual
regime concerning access agreements in telecommunications sector. Prior to the Access Notice,
European Commission in 1991, issued a Guideline that is entitled ‘Commission Guideline on the
application of EEC competition rules in the telecommunications sector (91/C 233/02),02), OJ C 233,
6.9.1991. The Commission has also issued two Recommendations on Interconnection Pricing in order
95
“Commission Notice on the Application of the Competition Rules to Access
Agreements in the Telecommunications Sector” was published in August 1998. The
purpose of the Notice was to set access principles stemming from EC Competition
Law in order to create greater market certainty and more stable conditions for
investment and commercial initiative in the telecommunications and multimedia
sectors, to define and clarify the relationship between competition law and sector-
specific legislation, and to explain how competition rules will be applied in a
consistent way across the converging sectors involved in the provision of new
multimedia services, and in particular to access issues and gateways in this
context.294
2.5.2 Commission’s Access Notice
In 1990s, European telecommunications policies were prompted to move forward

with the soft legislation such as Recommendations, Notices, etc. which was issued by
the Commission with the aim to ensure a stable and harmonised implementation of
the Directives. The Access Notice is one of the most important pieces of such
legislation, having been issued in order to guide operators with regard to access
agreements. Given the fact that access agreements are playing the key role within the
telecommunications sector, the Access Notice should be considered as a statement of
policy regarding competition policies in European telecommunications sector.295
This is so, because at the time of issue of the Access Notice, a fully liberalised and a
sufficiently competitive sector had not been gained and the consequences of full
liberalisation could not have been estimated entirely by the practitioners.
for establishment of price ranges for interconnection rates across the EU, based on the ‘best practice’
of the three Member States which have the lowest interconnection rates at the time of the issue of the
Recommendation. (Commission Recommendation of 8 April 1998 on interconnection in a liberalised
telecommunications market (98/322/EC) and Commission Recommendation of 8 January 1998 on
interconnection in a liberalised telecommunications market (98/195/EC).
294
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, Preface.
295
Mark Naftel, “Does the European Commission’s Telecommunications Access Notice Send the
Correct Economic Signals to the Market?”, Phoenix Center for Advanced Legal and Economic Public
Policy, January 1999, p. 1. Available at www.phoenix-center.org
96
Access Notice therefore concentrated on the scenarios of distortions and/or
restrictions of competition in telecommunications sector. In specific terms, the
concerns of competition breaches by the means of access agreements were the main
impulse for issuing Access Notice. Furthermore, the Essential Facilities Doctrine
found its most explicit formulation in the Access Notice, which was drawn from a
broad range of Commission decisions and Court judgments.296
The Commission’s 1998 Access Notice refers the ‘essential facility’ concept both in
defining the boundaries of dominance and describing an abuse of dominance. It is
visible that whilst explaining both ‘dominant position’ and ‘abuse of dominant
position’, the Commission devotes a primary role to EFD. According to the Access
Notice,297
A company controlling the access to an essential facility enjoys a

dominant position within the meaning of Article 86 (now 82).
Conversely, a company may enjoy a dominant position pursuant
to Article 86 (now 82) without controlling an essential facility.
In addition, a company according to the Access Notice may abuse its dominant, if by
its actions, it prevents the emergence of a new product or service.298 In respect of
demonstrating the general attitude of the Commission to the Doctrine, the paragraph
68 of the Access Notice deserves to be quoted, here:
In the telecommunications sector, the concept of ‘essential

facilities’ will in many cases be of relevance in determining the
duties of dominant telecommunications operators. The
expression ‘essential facility’ is used to describe a facility or
296
1999/2000, Harvard University (Weatherhead Center for International Affairs), p. 15.
297
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at. 69.
298
Ibid., at 90. This paragraph is reminiscent of the ECJ judgment in Magill, where one of the reasons
for advancing an abuse by the Court was as follows: “…thus prevented the appearance of a new
product, a comprehensive weekly guide to television programmes, which the appellants did not offer
and for which there was a potential consumer demand.” See supra note 110.
97
infrastructure which is essential for reaching customers and/or
enabling competitors to carry on their business, and which can
not be replicated by any reasonable means.299
It is not hard to deduct from the above paragraphs that, at least with respect to access
agreements in telecommunications sector, the Commission explains dominance on
the basis of the main thrusts of the concept of essential facilities.
After the so-called explanation, the Commission cites five conditions for an
obligation to grant access (to an essential facility) to be imposed: 300
(a) Access to the facility in question is generally essential in

order for companies to compete on that related market,
(b) There is sufficient capacity available to provide access,
(c) The owner of the facility fails to satisfy demands on an
existing service or product market, blocks the emergence of
a potential new service or product market,
(d) The company requiring access is prepared to pay the
reasonable and non-discriminatory price and will otherwise
in all respects accept non-discriminatory access terms and
conditions,
(e) There is no objective justification for the refusal to provide
access
Within the Access Notice, ‘access’ concept has been given a broader meaning when
compared to the ‘access to essential facilities’. According to the paragraph 71 of the
Notice, ‘access’ is related to a range of situations including the availability of leased
299
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at 68. The paragraph 68 of
the Access Notice expressly refers to the definition included in the “Additional commitments on
regulatory principles by the European Communities and their Member States” (“Regulatory Annex”
or “Reference Paper”) in the context of the World Trade Organisation (WTO) Basic
Telecommunications Agreement, which defines ‘essential facilities’ in the following manner:
“Essential facilities mean facilities of a public telecommunications transport network and
service that:
(a) are exclusively or predominantly provided by a single or limited number of suppliers;
and
(b) cannot feasibly be economically or technically substituted in order to provide a service.”
300
Ibid., at 91.
98
lines301 and enabling a service provider to build up its own network and
interconnection in the strict sense namely, interconnecting two telecommunications
networks, for instance mobile and fixed.302
Subsequent to dealing with the concept of access, the Commission expresses its post-
liberalisation concerns with regard to access in the following statements:303
In relation to access it is probable that the incumbent operator

will remain dominant for some time after the legal liberalisation
has taken place. The incumbent operator which controls the
facilities, is often also the largest service provider, and it has in
the past not needed to distinguish between the conveyance of
telecommunications services and the provision of these services
to end-users. Traditionally, an operator who is also a service
provider has not required its downstream market operating arm to
pay for access, and therefore it has not been easy to calculate the
revenue to be allocated to the facility. In a case where an operator
is providing both access and services it is necessary to separate
so far as possible the revenues as the basis for the calculation of
the company’s share of whichever market is involved.
Upon these findings, the Access Notice offers the prohibitive tools in order for
removing competition breaches, specifically in the context of access. Paragraph 83 of
the Access Notice sets forth such rules as follows:
A refusal to give access may be prohibited under Article 86 (now

Article 82) if the refusal is made by a company which is
dominant because of its control of facilities, as incumbent TOs
(telecommunications operators) will usually be for the
foreseeable future. A refusal may have the effect of hindering the
301
For the purposes of Directive 92/44/EEC, leased lines are defined as:
“telecommunications facilities which provide for transparent transmission capacity between network
termination points and which do not include on-demand switching (switching functions which the user
can control as part of the leased line provision).” (Article 2 of the amended Directive 92/44/EEC.)
302
303
Ibid.
99
maintenance of the degree of competition still existing in the
market or the growth of that competition.
A refusal will only be abusive if it has exploitative or anti-

competitive effects. Service markets in the telecommunications
sector will initially have few competitive players and refusals
will therefore generally affect competition on those markets. In
all cases of refusal, any justification will be closely examined to
304
determine whether it is objective.
The most remarkable point here is the established criterion as to whether a refusal to
access is abusive or not in a case. The so-called criterion is seemingly well-defined
as ‘having exploitative or anti-competitive effects’. According to the Access Notice,
a refusal to grant access will not be deemed ‘abusive’ by the Commission, if it does
not bear exploitative or anti-competitive effects. However, the first paragraph of the
above excerpt reveals per se elements when deciding to an infringement of an
abusive practise under Article 82.
Within the context of the Access Notice, does also exist an articulation for
distinguishing the telecommunications markets as ‘downstream’ and ‘upstream’, in
accordance with the EC competition case-law.305 Within this perspective, the former
corresponds to a market where services are offered to end-users, whereas the latter
correspond to a market where access is offered for the facilities necessary to provide
304
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at 83. In the subsequent (84th)
paragraph of the Access Notice have been envisaged three potential types of ‘refusal to grant access’
as follows:
(a) a refusal to grant access for the purposes of a service where another operator has been given
access by the access provider to operate to that services market;
(b) a refusal to grant access for the purposes of a service where no other operator has been given
access by the access provider to operate on that services market;
(c) a withdrawal of access from an existing customer”
305
Article 45 of the Access Notice, under the sub-heading ‘Relevant Product Market’, makes clear the
distinction between downstream and upstream markets as follows: “It is clear, therefore, that in
telecommunications sector there are at least two types of relevant markets to consider - that of a
service to be provided to end users and that of access to those facilities necessary to provide that
service to end users (information, physical network, etc.). In the context of any particular case, it will
be necessary to define the relevant access and services markets, such as interconnection to the public
telecommunications network, and provision of public voice telephony services, respectively.”
100
the services to end-users. In the Access Notice, the Commission recognised that
given the pace of technological change in the telecommunications sector, any attempt
to define particular product markets (in the Notice) would run the risk of rapidly
becoming inaccurate or relevant.306
In light of the above explanations it is possible to say that the Access Notice
enlightened the way after liberalisation by setting out important guidelines aiming at
ensuring pro-competitive markets. Among those guidelines, the implications for
application of EFD in telecommunications sectors occupy an important place.
However, the requirements laid down in the Access Notice have been largely
criticised for being framed under an over-zealous and over-interventionist
approach.307 As a matter of fact, the Access Notice regards it to be sufficient if the
owner of the infrastructure either fails to satisfy demand on an existing market, or
blocks the emergence of a potential new product, or impedes competition on an
existing or potential market.308
Considering the restrictive approach of the ECJ portrayed in its Oscar Bronner
judgment, the Commission’s approach regarding EFD in the Access Notice exhibits
a quite contrast to the attitude of the ECJ in Oscar Bronner. From this point of view,
the Access Notice does not seem to have taken the Court Rulings into sufficient
account, by somehow interpreting EFD in a broad manner. Such a perspective is
seemingly framed in line with the philosophy of the Commission decisions which
306
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at 47. In contrast to the case
within the Access Notice, the relevant product markets are precisely defined in the last
Recommendation of the Commission concerning market definitions (Commission Recommendation
of 11 February 2003 on relevant product and service markets within the electronic communications
networks and services, O.J. 8.5.2003 L 114/45). In the Annex of the so-called (last) Recommendation,
seven markets at the retail level, and eleven markets at the wholesale level are specified in accordance
with the new EU Regulatory Framework.
307
See Andres Bartosch, “EC Competition and Telecommunications Law” (International Competition
Law Series, Vol. 6), Kluwer Law International, 2002 (Edited by C.Koenig, A. Bartosch, J. Braun), p.
136. and N. Nikolinakos, “Access Agreements in the Telecommunications Sector-Refusal to Suply
and The Essential Facilities Doctrine under EC Competition Law”, 1999, European Competition Law
Review, No:8, p. 404.
308
Andres Bartosch, “EC Competition and Telecommunications Law” (International Competition
Law Series, Vol. 6), Kluwer Law International, 2002 (Edited by C.Koenig, A. Bartosch, J. Braun), p.
149.
101
reveal a clear ‘service-based competition’ approach. According to P. Nihoul and P.
Rodford it appears that the positions adopted by the Commission in Sea
Containers/Stena Sealink309 and the Access Notice are identical in substance.310
‘Essential character of the facility’, which was primarily brought into the case-law by
the so-called decision, was maintained by the Access Notice, in their opinion.
Indeed, it seems that Access Notice has a vision just the same as Commission’s
attitude in antitrust cases, particularly for the concurrent policy objectives.
On the other hand, though including some quotations from the decisions of the
Community Courts, no detailed guidance for assessment of such judgments was
provided by the Notice. In spite of existence of a number of provisions regarding
EFD and related criteria that remind preceding Court rulings, many access issues
have not been clarified in the Access Notice. For instance, in the Notice does exist
neither a clarification concerning the clear-cut boundaries of ‘objective justification’
nor an analytical explanation relating to the application of EFD on a Community-
wide basis.
With regard to access policies, the Access Notice is able to be seen as either a policy
guideline or an over-interventionist legislative tool in order for prevention of
potential refusal(s) to grant access to essential facilities. The new (2002) Regulatory
Framework which will be analysed below, sets out more flexible and less
interventionist provisions regarding access to essential facilities, envisaging a
progressive transition from ex ante sector-specific regulation to ex post competition
law remedies in the sector. Interestingly saying, there is a remarkable contrast
between the philosophies of the Access Notice and the 2002 Regulatory Package.
As the past experience has shown, the markets exhibit dynamic growth in terms of
new entry, investment and development of services following full liberalisation. In
particular during liberalisation period, EFD has potential to cope with bottleneck
309
310
P. Nihoul and P. Rodford, EU Electronic Communications Law, Oxford University Press, 2004, p.
475.
102
cases and does offer more effective solutions in comparison to sector-specific rules.
Thus, in transition periods from monopolies to liberalised markets or in sectors
where the market data reveal a non-competitive structure, EFD could be invoked as
the key policy tool as the Access Notice envisages.
2.5.3 The Rationale and Main Parameters of Applying EFD under the EU
Access Regime
In order to introduce competition in networked sectors, entrants need to be granted

access to the incumbent’s network facilities, the reason for which is the inherent fact
that these sectors are characterised by a significant degree of economies of scale,
scope and density. Not only economies of scale, scope and density but also other
entry barriers such as behavioural hindrances make compulsory access inevitable.
That is explicitly to say, for many times ‘negotiated access’ does not yield
satisfactory conclusions, and ‘mandated access’ is called forth for securing
competition between the market players. In telecommunications sector, such access
is furthermore required because entrants, in order to offer a competing product need
the ability to terminate calls on the incumbent’s network.311 In other words, every
operator has a monopoly over its network, and when another operator wants its
subscribers to reach the subscribers of the concerned operator, that operator has no
option other than call termination on the concerned operator’s network. Thus, access
to another operator’s network (usually in the form of interconnection) is crucial for
interaction between customers and for ensuring competitive services.
Aiming at ensuring efficient access through networks, many policy tools have been
developed so far. Among the measures of ‘mandated access’, EFD has emerged one
of the most influential methods for granting new entrants access to incumbent’s
essential network facilities. In parallel with the relationship between mandated access
and EFD, the interplay between competition law and sector-specific rules in the
telecommunications sector is most prominent, where access to ‘essential facilities’ is
311
Jonas Holm, “Regulating Network Access Prices under Uncertainty and Increasing Competition:
The Case of Telecommunications and Local Loop Unbundling in the EU”, 2000, University of
Copenhagen, Institute of Economics (MSc Thesis), p. 107.
103
concerned.312 In essence, all the telecommunications sectors including those of EU
and Turkey are well-suited for application of EFD, since all telecommunications
services and networks, by nature, exhibit almost common entry barriers and similar
access problems.313
Firstly, duplication of telecommunications infrastructure is seriously uneconomic,

especially from the point of new entrants’ views. Critical bottlenecks (essential
network facilities) do exist where new entrants need access to them in order to
compete with the incumbent operators. Such network facilities are prohibitively
expensive, which occasionally prevent or deter new entrants from entering the
market. Ownership of ‘local telephone networks’ may be used, for instance, to deter
or delay the entry of competitors, in the field of information and communications
services which rely on access to and use of the local network for the provision their
own products.314 From this point of view, cost structures of networks often constitute
a bottleneck situation and thereby an insuperable barrier when incumbent operators
are not forced to deal with their competitors under EFD.
Secondly, the rationale behind competition policies in the telecommunications sector

lies at the factual point that the monopoly control still dominates a lot of network
segments in the sector, even after abandonment of special and exclusive rights.
Monopolies or dominant undertakings mainly produce their outputs in the upstream
market(s) and sell their products in the downstream market(s) in a more efficient
manner. The ‘efficiency’ advantages generated through the strong positions of the
incumbent in the two adjacent markets so many times thwart the competition in the
downstream markets. That is to say, new entrants hardly perform their activities in
the downstream markets against the vertically integrated position of incumbents.
Both competition authorities and national regulatory authorities try to prevent the
312
Georg Koopman, “Competition Policies and Telecommunications Regimes”, p. 36. Available at
http://www.hwwa.de/Projekte/Forsch_schwerpunkte/FS/Hande/Publikationen/Koopman%20edit.pdf
313
In this part of the thesis, the analysis on entry barriers and access problems is predominantly done
independently of country-specific evaluations. The specific implications of EFD for Turkish
telecommunications sector is going to be discussed in the 3rd Chapter.
314
Op.cit in note 312.
104
dominant operators from strengthening their positions in adjacent markets via some
tools, among which EFD has proved itself through the case-law successfully.
Thirdly, incentives for dominant operators to engage in anticompetitive behaviours

may ultimately derive from the presence of network externalities.315 Given the fact
that in the presence of positive network externalities, the value of a product supplied
by a dominant operator increases when an additional consumer is included to the
network of the product, such network externalities that are peculiar to networked
industries constitute a basis for anti-competitive behaviours. In fact, many
established incumbent operators have a long history of providing public telephony
services at subsidised rates, which provides themselves with strong network
externalities.316 As a corresponding matter, the competitive structure in the sector
might be distorted through established network externalities in absence of any pro-
competitive measures. In this context, EFD is deserved to be deemed an effective
tool to neutralise such distortions with its balancing role between the asymmetrical
powers of market players.
Fourthly, an incumbent operator can use its control over bottlenecks to increase a
competitor’s costs and make its service less attractive to customers.317 In other
words, an incumbent operator can increase the competitors’ costs through increasing
the prices of its own network facilities or somehow refuse to share its infrastructure
with other operators at competitive and fairly established prices. In both situations,
there occurs an absolute need on the part of competing firms to access to the existing
network components for carrying out their activities. In absence of compulsory
access by regulatory tools or EFD, new entrants could be in a bottleneck situation to
compete and may find it very difficult to persuade customers to switch from an
incumbent operator that has served them for many years. Here suffice to say that,
315
Georg Koopman, “Competition Policies and Telecommunications Regimes”, p. 10. Available at
http://www.hwwa.de/Projekte/Forsch_schwerpunkte/FS/Hande/Publikationen/Koopman%20edit.pdf,
316
H. Intven., J. Oliver, E. Sepulveda, (2000), “Telecommunications Regulation Handbook”,
Published by McCarthy Tetraultz, (5th Section), p. 9. Available at http://www.infodev.org/projects/-
314regulationhandbook/
317
Ibid.
105
EFD by its adjustable nature to any bottleneck situation, would be more effective in
solving access problems, comparing to sector-specific rules.318
In the light of the above explanations, it is possible to say that structural deficiencies
and strong advantages inherently existed in favour of incumbent operators make
telecommunications markets quite unique and subject to EFD. When the Doctrine is
reminded, we face four main elements. These are respectively, control of an
(essential) facility by the dominant undertaking, inability of other competitors to
duplicate the facility practically, denial of use of the facility by the facility owner
(with substantial harm to competition) and the absence of a legitimate business
justification for denial of access.
According to I. Vogelsang and B. M. Mitchell, the notion of essential facilities is

closely related to that of (normative) monopoly for the reason that the second
element of the Doctrine spells out natural monopoly.319 In their opinions, the
difference between an essential facility (bottleneck) and a natural monopoly is that
the former refers to an input for which a market may not exist or may not be
meaningfully defined, whereas the latter refers to a market or a set of markets.
In fact, market definition in bottleneck cases is not an easy phase to be tackled before
making a decision concerning abuse of dominance. While the upstream / downstream
market pattern may be present, there may not be any market in the casual sense of the
word at the level of the (essential) facility.320 Whereas two easily identifiable markets
are present in the classical (vertical relationship) cases, i.e. Commercial Solvents,321
318
The adjustability of EFD to converging markets and technological changes is going to be discussed
in the following parts of the thesis.
319
I. Vogelsang and B. M. Mitchell, “Telecommunications Competition: The Last Ten Miles”, 1997,
The MIT Press and the AEI Press, Washington, D. C., p. 56.
320
Publishing, 2000, p. 205. Regarding essential facilities cases, Larouche advises caution in market
definitions, considering that a finding that the essential facility constitutes a relevant market might be
misleading. Pointing out that many allegedly essential facilities could not be individualised and be
offered on a stand-alone basis, he emphasises that the established criteria for market definition are of
very limited help and finding that there is a relevant market for access schemes goes beyond market
definition and into market structuring.
321
See supra note 97.
106
United Brands,322 the established criteria for market definition (the normative
establishment of two strictly separate markets depending on some tests, i.e. demand-
side & supply-side substitutability) are not so reliable in the cases where a part of
network rather than a good or service is alleged to be an ‘essential facility’. With
respect to the cases Magill,323 Oscar Bronner,324 and European Night Services,325
where the markets for access (to essential facility) are respectively programming
information, home-delivery network, and tracks of railway undertakings and their
locomotive engines, the subject-matter of the case(s) is not the same as that of the
former (vertical relationship) cases. In these latter cases, as P. Larouche points out,326
the markets do not consist of trading goods or services, but rather constitute either a
unique network or a part of a huge infrastructure such as rail tracks, airports,
telecommunications which are not so convenient to be assessed under the classical
tests of market definition.327
As a matter of fact, network industries such as telecommunications are distinguished

by the fact that their services have a geographic component involving the use of the
network.328 Here, one is confronted a market, boundaries of which are drawn not
through the consumer preferences or relevant quantitative / qualitative tests but rather
with regulatory and geographic constraints. That is to say, the essential network
facilities are subject to interventions through competition policies, and in such an
intervention based on the ‘essentiality’ of the facility, market definition sometimes
becomes meaningless. Then, uniqueness of the essential resources is placed at the
centre of the essential facilities cases.
322
See supra note 99.
323
See supra note 110.
324
See supra note 169.
325
See supra note 154.
326
327
This is so, because the criteria with regard to assessment for definition of a relevant market mainly
depend on the ‘interchangeability’ phenomenon which is affiliated with the tests of ‘demand-side
substitutability’ and of ‘supply-side substitutability’. See Richard Whish, “Competition Law, Fourth
edition”, Butterworths, 2001, p. 22-36, and J. Kalliala, “Market Definition Under the EC Competition
Law in the field of Voice Telephony”, 2000, PILC Student Paper, Brussels, p. 13.
328
Op.cit in note 326.
107
Besides difficulties relating to ‘market definition’, ‘grounds for intervention’ in
essential facilities cases differ in comparison to classical cases.329 While in the
former cases, the remedy is easy to identify as it usually takes the form of resuming
the trade of goods / services, in the latter cases, the remedies generally consist of
duty to share with or access to essential sources towards third parties. Thus, more
structural remedies in essential facilities cases usually take place whilst more
behavioural ones seem to be relevant in classical (refusal to deal) cases.
In this framework, especially when access to one of the essential resources such as
local loops of incumbents, rights of way, poles and conducts and other unique
components of communications networks, etc is the subject-matter of a case,
‘essentiality’ test would become the key argument in applying EFD. Moreover, in
essential facilities cases, ‘dominance’ becomes far less meaningful, and is replaced
by the notion of ‘essentiality’.330
2.5.4 The Application of EFD in Liberalisation Period: EU Experience and

Further Implications
The replacement of ‘dominance’ with the ‘essentiality’ within the picture cited above
demonstrates that there is an equivalence between EFD and the economic concept of
‘natural monopoly’. Inherent in the concept of an ‘essential facility’ is the premise
that the owner of that facility possesses monopoly power.331 The term ‘facility’ itself
connotes an integrated physical structure or large capital asset with the degree of cost
advantage or a unique character that usually confers monopoly power and market
control by virtue of its superiority for its intended purposes.332 In this regard,
329
Publishing, 2000, p. 205, p. 209.
330
Ibid., p. 207. Larouche cites the example of the local loop for illustration. According to this
example, if a service provider controls 5% of the local loops in a local access zone such a city, at first
glance it would not appear to be in a dominant position. Yet the competitive concerns surrounding the
local loop (due to the fact that provision of some certain services to the end-user is impossible without
access to the local loop) could initiate the competent authorities to require the owner of the 5% of the
local loop for granting access to other parties.
331
Abbott B. Lipsky, Jr. and J. Gregory Sidak, “Essential Facilities”, Stanford Law Review, Vol. 51,
No: 1187, May 1999, p. 1211.
332
Ibid.
108
‘essentiality’ means that the facility in question is a unique product, service or
infrastructure which any competitor inevitably needs for carrying out its activities.
From this point of view, such a unique facility is generally accepted to have been
possessed by monopolists; namely by firms having either de jure or de facto
monopoly.
As a matter of fact, in particular geographic areas, the incumbent operator may have
a degree of market power which goes beyond simple dominance, and may extend
through what is sometimes called super-dominance to de facto monopoly.333 In such
circumstances, refusal to grant access to essential network facilities, i.e. denying
(unbundled) access to the local loop may constitute an abuse of the dominant
position; potentially, it has the effect of eliminating a competitor’s ability to compete
in downstream markets with the owner of the (essential) facility.334 However,
irrespective of market power, there could also appear some competitive concerns in
related market activities, if the facility requested is a key input such as content, local
loop or set top boxes. That is to say, if such a facility exhibits an ‘essential’
character, one can presume that EFD is applicable even though the said facility is
owned by a small and medium-sized firm.335
333
Martin Cave and Luigi Prosperetti, “European Telecommunications Infrastructures”, Oxford
University Press and the Oxford Review of Economic Policy, Vol. 17, No. 3, 2001, p. 429.
334
Ibid. The owner of the allegedly essential facility in such a case is normally the incumbent operator
who has the monopoly over access to the local access network. Herein, unbundled access to the ‘local
loop’ means the provision of access to local exchanges at an unbundled basis, which enable the third
parties to compete at the same conditions with the incumbent operator. The essential character of the
local loop stems from its distinct role in fixed telecommunications infrastructure. The local loop is the
physical component of the fixed public telephone network connecting the network termination point at
the subscriber’s premises to the main distribution frame (at the local exchange), and this feature of the
local loop makes it the crucial part of the network in terms of direct access to the end-user. In this
perspective, the local loop may be deemed an essential (network) facility, when its ‘essential’
character and the potential substantial harms to competition in case of denying access are considered
together.
335
However, such an application seems improper under the competition law principles, because if a
facility is not owned by a dominant undertaking, an abuse of dominance under EFD can not be found
in a case. That is to say, EFD is an argument that contributes to answer the question ‘whether an
abusive practice has been conducted by the concerned firm or not’. On the other hand, a version of
EFD is presumably applicable via sector-specific regulation, namely, in an implicit manner and
dedicated context (to a specific area), rather than by relying upon a clearly defined market and a
(abusive) practice of a market power. For instance, in case of competitive concerns, some access
obligations related to a specific area of regulation can be imposed on operators not enjoying a
dominant position. Pertaining to some specific hubs of telecommunications, i.e. local loops, points of
109
Given the role of EFD, the applicability of EFD has gained more importance during
the liberalisation period of EU. In particular, with the acceleration of legislative steps
towards full liberalisation, there appeared a number of cases exhibiting the close
relation between ‘essential facilities’ and ‘legal monopolies’.
Influence of EFD in EU liberalisation period is remarkably seen in Atlas336 case

which culminated with an exemption decision regarding a telecommunications
alliance. In Atlas, the Commission focused the ‘non-discrimination’ principle and
specifically referred FT and DT’s potential power to supply ‘building blocks’ for a
number of services in a discriminatory basis. The services in question were defined
as “access to the PSTN, the ISDN and to other essential facilities, and also ...
reserved services”.337 At the time of the decision, PSTN, ISDN and other services
which were ordered to be granted access were under the scope of legal monopoly,
and were deemed an essential facility by the Commission. The Commission held:338
However, even when all telecommunications facilities and

services are non-reserved, FT and DT will at least for a number
of years remain indispensable suppliers of building blocks for the
relevant services in France and Germany. Given that FT and DT
are shareholders of Atlas, it is essential for the safeguarding of
fair competition between Atlas and other existing and future
telecommunications service providers to eliminate the risk that
the former might be granted more favourable treatment regarding
... such facilities and services which remain an essential facility
after full and effective liberalisation of telecommunications
infrastructure and services in France and Germany.
Atlas is a distinct decision regarding two points. First, the Commission seemingly
took action regarding the fear that FT and DT who are the shareholders of Atlas
would discriminate in providing the building blocks. Then, Commission made its
access or interconnection, etc. some ex ante obligations can be imposed by regulatory authorities,
regardless of the market share of the operator concerned.
336
Atlas and Phonenix/GlobalOne [1996] OJ L 239/23.
337
Ibid., at 53.
110
decision by referring a possibility of discrimination and presumably having regard to
the concern of preventing a potential discrimination. Second, the main reason for the
Commission’s decision seems to have been the ‘essentiality’ of the building blocks
which were offered under monopoly at that time. In the decision, as P. Larouche
points out, an explicit link between legal monopolies and essential facilities is
established by the Commission.339
The Commission in this and other decisions surrounded by liberalisation aims such
as London European/Sabena340 and British Midland/Aer Lingus341demonstrated its
aspiration for ensuring that creation of super dominants succeeding the former
monopolies be prevented. In this framework, the application of EFD seems to have
been given a central role in liberalisation period. The objectives targeted by the
liberalisation makes such an application evidently necessary. Particularly in absence
of alternative measures, EFD is of potential to be used as an effective tool in a wide
range of bottleneck situations during liberalisation periods. The experience in the EU
telecommunications sector can be shown as a quite successful example for
application of EFD during the transformation period from monopoly to competitive
markets.
The main reason for rapidly evolution of EFD in the EU context relates to the
inherited reserved area under the control of incumbent (monopoly) firms. More
explicitly, until the beginning of liberalised services and networks, lack of alternative
networks dominated the sector. The paragraph 64 of the Access Notice which reads
as “The development of effective competition from alternative network providers
with adequate capacity and geographic reach will take time.” also confirms this
factual point.342
338
Ibid., at 34.
339
340
See supra note 123.
341
See supra note 124.
342
111
While other competing firms rely upon access to incumbents’ essential network
facilities, incumbent operators hold a ‘gatekeeper’ position. However, it must not be
expected that this fact would be changed just after the liberalisation. This is why
market players unavoidably need access to the existing networks particularly due to
the economic constraints for a long time. The statements of H. Ungerer reveal this
fact:
In the fixed network field, the new entrants are faced with a
situation where the incumbents hold fixed network assets built
over one hundred years of monopoly. None of the new entrants
can, in the short term, build parallel networks in the local loop
which could rival these assets worth 200-300 billions of euros of
investment. 343
For the sake of illustrating such a bottleneck situation, one can consider emergence
of a variety of different service providers after liberalisation. A lot of Internet Service
Providers (ISPs), international operators and other telecommunications service
providers after being awarded a license, enter the market and use the same network
access lines and local switches to reach subscribers. In this situation, in order for new
entrants not to be faced with a bottleneck problem, mandating access to the
incumbent’s essential network facilities is crucial for viability of new entrants in the
market.
As a result of rapidly completed liberalisation of EU telecommunications sector,

many new firms would have faced a difficult situation where no other option than
access to incumbent’s networks was economically viable. However, such a fateful
situation has not been taken place in the EU context, thanks to the Commission’s
efforts. The Doctrine’s success, in this regard illustrates the importance the
Commission attaches to the acceleration of liberalisation.
343
H. Ungerer, “The arrival of competition in European telecommunications”, 3rd European Forum
on the Law of Telecommunications, Information Technologies and Multimedia: Towards a Common
Framework, Luxembourg, June 19, 1998, p. 7-8.
112
In order to reap the benefits of liberalisation, many commentators affirm the vitality
of the Doctrine within the transition periods towards full liberalisation.344 To sum up,
the history and development of EFD to date reveals that EFD has the potential to
remedy the bottleneck situations with regard to ‘third party access’, during
liberalisation periods.
2.6 Recent Developments in EU after Full Liberalisation
In telecommunications policy of the EU, a dual system has so far been continued to
exist and to an appreciable extent has also been successful in tackling access and
interconnection issues. In this context, dual system at EU/national level has ensured
efficient access to the incumbents’ networks particularly with regard to access and
interconnection regime, inter alia in unbundling and co-location issues, elimination
of anti-competitive conducts as regards access and interconnection, etc.
In securing such an efficient dual (access) regime, sector-specific regulation which

incorporates predominantly ex-ante obligations played a comparatively preceding
role. Except for merger and acquisition controls, imposition of ex ante obligations is
quite rare under competition law remedies.345 Most probably due to this fact, the
Commission acknowledged in the Access Notice that EC Competition Law rules are
not sufficient to remedy all the problems in the telecommunications sector.346 From
this point of view, sector-specific regulations in the form of Directives, Regulations,
etc. have stimulated the sector during and after liberalisation in the EU, remarkably.
344
See N. Nikolinakos, “Access Agreements in the Telecommunications Sector-Refusal to Supply and
The Essential Facilities Doctrine under EC Competition Law”, 1999, European Competition Law
Review, No: 8, p. 399-411; J. T. Lang, “Defining legitimate competition: companies’ duties to supply
competitors, and access to essential facilities, Fordham Corporate Law Institute, International Law
and Policy”, 1994, p. 437-524 and Antonio Capobianco, “The essential facility doctrine: similarities
and differences between the American and the European approach”, European Law Review, 26, no. 6,
2001, p. 548-564.
345
Under competition law analysis, the respective steps in general are, definition of relevant market,
identification of dominance, and investigation as to whether dominant undertaking’s behaviour
constitutes an abuse of dominance or not. The sanctions and actual remedies of general competition
law are imposable according to the conclusion of that investigation.
346
The paragraph 14 of the Access Notice reads as follows:
“Community competition rules are not sufficient to remedy all of the various problems in the
telecommunications sector. NRAs (national regulatory authorities) therefore have a significantly
wider ambit and a significant and far-reaching role in the regulation of the sector. It should also be
noted that as a matter of Community law, the NRAs must be independent.”
113
Through such regulations, many reserved areas have undertaken a considerable
diffusion of competition.
While competition has expanded rapidly and significantly into some sector segments,
i.e. long distance and international telephony, some other segments such as local
access networks largely remained being operated and managed by incumbent
operators. Thus, in many countries have remained various questions concerning how
to foster competition in local telecommunications markets after liberalisation. These
questions continued to be open for some time on the part of practitioners as well as
policy makers at the EU level. On the other hand, as the time was passing the gap
between EU and USA was widening in terms of broadband services, particularly
high-speed internet applications.
One of the remaining issues was related to the debate on consequences of

convergence including transformation of the regulatory framework. In this context, at
late nineties the premise that all transmission networks and services should be
covered by a single framework echoed widely and the Community Institutions,
mainly Commission commenced to take action towards this end. Below will be
explained the relevant Commission’s efforts and legislative steps, respectively.
2.6.1 Reform Process since 1999
Given the short-term requirements to speed up large-scale deployment of Internet at

affordable rates in EU, and to open up development towards high-speed multimedia
Internet applications, the Commission chose to pursue a wide-ranging reform, taking
the convergence phenomena into account.347
Within this framework, in November 1999, the Commission launched a review of the
existing regulatory framework with the publication of the Communications
Review.348 The Commission Review presented a series of policy proposals
347
348
“Towards a new Framework for Electronic Communications Infrastructure and associated
services”, COM(1999)539.
114
introducing significant changes to the 1998 Package,349 main reasons of which can be
specified as ‘adaptation of existing legislation towards convergence’, ‘introduction of
wide-spread Internet applications’, and ‘elimination of de facto monopolies after full
liberalisation’, etc.
As the initial step, the Commission issued a Recommendation,350 where the

Commission recommended the implementation of full unbundling by Member
States, on 31 December 2000 at least. In this regard, the Commission set out detailed
guidance to assist national regulatory authorities on fair regulation of unbundled
access to local loop, in particular with regard to pricing issues, technical conditions
and consultation procedures.
Under the unbundling scheme, the Commission took a further step and proposed a
Regulation for unbundled access to the local loop in accordance with the Lisbon
Summit.351 Notably, after a soft regulation, i.e. Recommendation, the Commission
issued a Regulation which is an instrument rarely used at the Community level for
being directly applicable in the Member States without transposition.352
349
The Open Network Provision (ONP) measures constitute the so-called “1998 package” of
legislation, which were issued during the transition period along the way for a fully harmonised and a
liberalised EU market in field of telecommunications.
350
“Commission Recommendation on Unbundled Access to the Local Loop”, C(2000)1059, 26 April
2000. http://www.europa.eu.int/information_society/topics/telecoms/regulatory/maindocs/comgreen/-
index_en.htm
351
Proposal for a Regulation of the European Parliament and of the Council on unbundled access to
the local loop, adopted 12 July 2000 Com(2000)394. http://www.europa.eu.int/information_society/-
topics/telecoms/regulatory/new_rf/index_en.htm#ull
The conclusions of the European Council of Lisbon of 23 and 24 March 2000 note that, for Europe to
fully seize the growth and job potential of the digital, knowledge-based economy, businesses and
citizens must have access to an inexpensive, world-class communications infrastructure and a wide
range of services. The Member States, together with the Commission, are called upon to work towards
introducing greater competition in local access network before the end of 2000 and unbundling the
local loop in order to help bring about a substantial reduction in the costs of using the Internet in the
Lisbon Summit. (cited in the Regulation 2000/2887, Preface)
352
The reason for preferring such a harsh measure is presumably stemming from the perception that
the development of Internet in Europe is insufficient in comparison to USA. In the Preface of the
Regulation 2887/2000, the legal justification for issuing a Regulation concerning unbundled access to
the local loop is divulged as follows:
“In accordance with the subsidiarity as set out in Article 5 of the Treaty, the objective of achieving a
harmonised framework for unbundled access to the local loop in order to enable the competitive
provision of an inexpensive, world-class communications infrastructure and a wide range of services
for all businesses and citizens in the Community cannot be achieved by the Member States in a secure,
harmonised and timely manner and can therefore be better achieved by the Community. In accordance
115
In fact, for reaching regulatory objectives, unbundling policies have been preferably
put into effect by EU Institutions, while there are alternative access technologies
such as cable networks, broadband wireless local loop, satellite technologies, etc.
(See the Table 2.)
Table 2: Access Technologies (Source: Chris Doyle, (2000), “Local Loop Unbundling and
Regulatory Risk”, Journal of Network Industries, Vol. 1, p. 41)
Access Platform Technologies

Twisted Copper Pair Analogue modems, ISDN (Integrated Services Digital
Network), xDSL (generic family of Digital Subscriber
Lines capable of expanding capacity on copper wire)
Optical Fibre High bandwidth services direct to home or kerb
Cable Television Cable modems allowing bi-directional traffic
Wireless Cellular (GSM), UMTS (Universal Mobile
Telecommunications System – broadband cellular),
Narrowband wireless (DECT, Digital Enhanced Cordless
Telephone), Broadband wireless (LMDS, Local
Multipoint Distribution Service)
Satellite VSATs (Very Small Apperture Terminals), LEOs (Low
Earth Orbiting Satellites)
Given the so-called access networks, there principally exist a number of alternatives
other than using the incumbent’s local loop, among which, cable television networks
would be deemed the most prominent access platform. However, this is not the case,
because cable networks require technical adaptation in order to be used for
telephony.353 Other alternative networks include electricity cables entering the
subscriber’s premises and radio links; but since these forms of access are not
commercially developed, they do not offer an immediately available competitive
alternative for the bulk of telephone users.354
with the principle of proportionality as set out in that Article, the provisions of this Regulation do not
go beyond what is necessary in order to achieve this objective for that purpose.”
353
Exceptionally, in UK, cable networks were built out not only with the usual co-axial cable for the
supply of television services, but with an additional twisted copper pair with the intention that it might
be used for telephony. There is no other equally successful example in EU as regards cable TV
connections, especially in the context of using cable network for a wide range of services such as
cable telephony, internet, etc.
354
J. Kalliala, “Market Definition Under the EC Competition Law in the field of Voice Telephony”,
2000, PILC Student Paper, Brussels, p. 39. For a comparative analysis of the access technologies in
context of their costs, speed and ubiquity, see Chris Doyle, “Local Loop Unbundling and Regulatory
Risk”, Journal of Network Industries, 2000, Vol. 1, p. 33-54; M. Bourreau and P. Dogan, “Service-
116
One of the main reasons behind the regulatory preference in favour of mandatory
unbundling policies is the fact that, the only networks which have been developed
nation-wide in each of the Member States were local loops (local
telecommunications networks) at that time. In fact, alternative infrastructures (i.e.
cable television, satellite, wireless local loops) do not generally offer the same
functionality and ubiquity, and this made local telecommunications networks to a
significant degree, non-substitutable against other access technologies. In view of
these reasons, local loop unbundling was deemed the most effective way allowing
new entrants to compete with incumbent operators under the same conditions in
offering high-speed Internet and multimedia applications based on broadband
technologies and was initiated by the EU policy makers immediately after
liberalisation.
2.6.2 Introduction of New (2002) Regulatory Framework
In response to the conclusions of the European Council of Lisbon of 23-24 March

2000 and the results of the public consultation on the 1999 Communications Review
of the Electronic Communications Sector, the Commission proposed in July 2000 a
package of measures for a new regulatory framework for electronic communications
networks and services. The new package consists of five European Parliament and
Council Directives under Article 95, one Commission Directive adopted under
Article 86 and one Commission Decision on a regulatory framework for radio
spectrum. The following legislative documents are included in the so-called package
(2002 Regulatory Framework):
1- Directive (2002/21/EC) on a Common Regulatory Framework, OJ L 108,

24.4.2002.
2- Directive (2002/19/EC) on Access, OJ L 108, 24.4.2002.
3- Directive (2002/20/EC) on Authorisation, OJ L 108, 24.4.2002.
based vs. Facility-based Competition in Local Access Networks”, Telecommunications Policy, 2001,
Vol. 25, Issue. 3, p. 167-184 and Lixia Li, “Local Loop Unbundling: International Experiences and
Implications for China”, December 2002, University of Strathclyde, (MSc Thesis)
117
4- Directive (2002/22/EC) on Universal Service and Users’ Rights, OJ L 108,
24.4.2002.
5- Directive (2002/77/EC) on Competition in the Markets for Electronic
Communication Services, OJ L 249/21 16.09.2002.
6- Directive (2002/58/EC) on Data Protection and Privacy, OJ L 201/37, 31.07.2002.
7- Decision 676/2002/EC on a Regulatory Framework for Radio Spectrum Policy in
the European Community, OJ L108/1, 24.04.2002.
The new legislation aims at creating a single framework for all electronic
communications networks and services. At the Global Internet Summit which was
held in 24 May 2000, Commissioner Liikanen defined the major goals expected from
the new Package.355 These are (i) simplification and clarification of the existing
framework aiming at bringing the number of regulatory measures down to 6 from 20
(Directives), (ii) introduction of greater flexibility in the framework, (iii) adaptation
of the 1998 Package in the light of technological developments and convergence of
markets and (iv) introduction of greater competition in particular within the least
competitive segment of telecommunications network, the local loop of incumbent
operator.
Given the so-called policy objectives, the new Framework, aiming at adaptation to
the changing needs of electronic communications sectors in the Internet age,
consolidated the existing legal measures under six Directives, under which all
transmission networks and services are covered under a single regulatory framework.
This framework does not cover the content of services delivered over electronic
communications networks.356
355
356
The term ‘telecommunications’ which was used in the 1998 Package has been replaced with the
one ‘electronic communications’ in the new (2002) Regulatory Framework so as to include all
transmission networks and services under the new Framework. Behind this change was inherent the
aim of introducing the principle ‘technology neutrality’ under the new (2002) Regulatory Framework.
118
In order to answer the conditions of the growing convergence of telecommunications
and broadband markets and the resulting requirement for a more flexible framework,
predominance of the sector-specific rules has been set aside. That is to say, the new
Framework is built on the main thrusts of competition law instead of the detailed ex
ante rules which were the core of the 1998 Package. This key change has been
spelled out in the Preface of the Framework Directive (Directive 2002/21/EC)
clearly:357
It is essential that ex ante regulatory obligations should only be

imposed where there is not effective competition, i.e. in markets
where there are one or more undertakings with significant market
power, and where national and Community competition law
remedies are not sufficient to address the problem.
The most prominent and supportive fact revealing the movement towards greater
reliance on the competition law is the transformation of the concept ‘Significant
Market Power-SMP.’ According to the old (1998) Framework, operators possessing
a market share more than 25% in a relevant market were presumed having SMP and
the concept of SMP played a major role under the asymmetrical regulation, where
the access obligations used to be imposed to the operators having SMP. However, the
concept of SMP in the new Package is redefined in line with the competition law
principles, namely in accordance with the definition of ‘dominant position’.358
357
Directive 2002/21/EC on a common regulatory framework for electronic communications
networks and services, 2002 O.J L 108/33. (Framework Directive), Preface (at 27) Superseding role of
competition law and the increasing need to a more flexible framework is rationalised in the paragraph
25 of the Preface of the Framework Directive, as follows:
“There is a need for ex ante obligations in certain circumstances in order to ensure the development of
a competitive market. The definition of significant market power in the Directive 97/33/EC of the
European Parliament and of the Council of 30 June 1997 on interconnection in telecommunications
with regard to ensuring universal service and interoperability through application of the principles of
open network provision (ONP) has proved effective in the initial stages of market opening as the
threshold for ex ante obligations, but now needs to be adapted to suit more complex and dynamic
markets. For this reason, the definition used in this Directive is equivalent to the concept of
dominance as defined in the case law of the Court of Justice and the Court of First Instance of the
European Communities.”
358
Article 14 of the Directive 2002/21/EC (Framework Directive) reads as follows: “An undertaking
shall be deemed to have significant market power if, either individually or jointly with others, it
enjoys a position equivalent to dominance, that is to say a position of economic strength affording it
the power to behave to an appreciable extent independently of competitors, customers and ultimately
consumers.” This definition is almost the same as the ‘dominant position’ definition of which was
119
In line with the new definition of SMP, the new Regulatory Framework has adopted
two procedures concerning ‘market analysis’ and ‘market definition’.359 The
rationale of such an adoption is inherent in the competition law approach adopted by
the 2002 Framework. This is why imposition, maintenance, or withdrawal of
regulatory obligations such as mandatory access and/or unbundling is dependant
upon the competitive analysis to be carried out in accordance with the Framework
Directive under the new regime. Pursuant to Articles relevant to the market analysis
and market definition included by the Framework Directive, the Commission issued
a Guideline360 and a Recommendation361 according to which, relevant markets are to
be defined in accordance with the principles of EC Competition Law.362
established in the United Brands Judgment. (Case 27/76, United Brands v. Commission, [1978] ECR
207; See supra note 100).
In the new Framework, it is important to stress that the existence of a dominant position cannot be
established on the sole basis of large market shares, in contrast to the old regime. According to the
new regime, national regulatory authorities should undertake a thorough and overall analysis of the
relevant market before coming to a conclusion that the undertaking concerned enjoys significant
market power (SMP). In assessing SMP, the following criteria are used to measure the power of an
undertaking to behave to an appreciable extent independently of its competitors, customers and
consumers. These criteria include amongst others:
- overall size of the undertaking,
- control of infrastructure not easily duplicated,
- technological advantages or superiority,
- absence of or low countervailing buying power,
- easy or privileged access to capital markets/financial resources,
- product/service diversification (e.g. bundled products or services),
- economies of scale,
- economies of scope,
- vertical integration,
- a highly developed distribution and sales network,
- absence of potential competition,
- barriers to expansion.
[cited in the Commission guidelines on market analysis and the assessment of significant market
power under the Community regulatory framework for electronic communications networks and
services (2002/C 165/03) ]
359
networks and services, 2002 O.J L 108/33. (Framework Directive), Articles 15 and 16.
360
Commission guidelines on market analysis and the assessment of significant market power under
the Community regulatory framework for electronic communications networks and services (2002/C
165/03).
361
Commission Recommendation of 11 February 2003 on relevant product and service markets within
the electronic communications sector susceptible to ex ante regulation in accordance with Directive
2002/21/EC of the European Parliament and of the Council on a common regulatory framework for
electronic communications networks ad services, O.J. 8.5.2003 L 114/45.
362
The market definitions in electronic communications sector were made by the Commission
according to the Commission Recommendation of 11 February 2003 on relevant product and service
markets. In the Annex of the so-called Recommendation, seven markets at the retail level, and eleven
120
Summarising, sector specific rules has a subsidiary role under the new regime where
ex ante obligations are imposable only if effective competition does not prevail in the
relevant market. In other words, under the new regime does appear no asymmetrical
regulation without a detailed market analysis.
2.6.3 Assessment of the Recent Developments under EFD
The recent developments in EU have important implications with regard to

application of the Essential Facilities Doctrine. Having evolved through the case-law,
the Doctrine has culminated in the Oscar Bronner363 case with important results that
appealed to many practitioners and commentators. The appealing face of the
Doctrine seems to have also influenced the Community Institutions.
As expounded in the first Section, the Oscar Bronner judgment draw a mile-stone in
application of EFD, by bringing somehow restrictive principles such as
‘indispensability’ and ‘non-substitutability’ tests. After the combination of these tests
with the further test ‘non-duplicability’ cited in the so-called judgment which has
already been inherent in the US case-law, the Doctrine was strictly characterised and
became more reliable. Since then, the main themes of the Doctrine have been
incorporated as a part of the EC Competition Law. In some of the soft legislation
particularly in the Access Notice, important criteria, i.e. the conditions for abuse of
dominance and access obligations, are correlated with the Doctrine. In essence, the
Access Notice regarded EFD as the central measure within the competition law tools
with regard to third party access.
Whereas some soft legislation provides some details with regard to the ‘essential
facilities’ notion, the binding measures, i.e. Directives, Regulations include implied
remedies instead. Even the so-called binding provisions reveal the fact that some of
the network components are of potential to be deemed ‘essential facility’, and
markets at the wholesale level are determined in accordance with the Directive 2002/21/EC
(Framework Directive).
363
121
thereby must be covered under access regulation. 2002 Regulatory Framework has a
number of examples in somewhat hidden way, in this regard.
For instance, within the framework of Article 12 of the Access Directive, national
regulatory authorities are empowered to impose obligations on operators to meet
reasonable requests for access to, and use of specific network elements and
associated facilities. For such an access obligation, a situation should be arisen where
the national regulatory authority considers that denial of access or unreasonable
terms and conditions having a similar effect, (i) would hinder the emergence of a
sustainable competitive at the retail level, or (ii) would not be in the end-user’s
interest.
According to the so-called Article, within the framework of access obligations,

operators may be required inter alia:364
(a) to give third parties access to specified network elements and/or

facilities, including unbundled access to the local loop;
…
(c) not to withdraw access to facilities already granted;
…
(e) to grant open access to technical interfaces, protocols or other
key technologies that are indispensable for the interoperability of
services or virtual network services;
…
(h) to provide access to operational support systems or similar
software systems necessary to ensure fair competition in the
provision of services;
(i) to interconnect networks or network facilities.
When the abovementioned obligation “granting open access to technical interfaces,

protocols or other key technologies that are indispensable for the interoperability of
services or virtual network services” is considered under EFD, an actual equivalence
364
Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access
to, and interconnection of, electronic communications networks and associated facilities (Access
Directive), Art. 12.
122
could be inferred between the Doctrine and the core of this requirement. However,
any specific form of access is not given precedence in the context of mandatory
obligations. This gives a large room for application of EFD in both implicit [in
specific forms of access (ex ante) obligations] and explicit [through typical (ex post)
enforcement of the Doctrine] manner.
Given the general understanding of the new (2002) Regulatory Framework, one can
say that ex post obligations via explicit application of EFD have superior role over
sector-specific obligations. Affirming such a perception, the superiority of ex post
obligations is emphasised both in the Access Directive and the Framework
Directive.365 The Access Directive with the aim of optimisation of level of access
obligations, devoted particular emphasis to the market analysis and cited the
progressive transition from the detailed ex ante regulation to the less detailed ex post
obligations, consonantly. It is worth quoting here the parts of the so-called Directive
that are most relevant to EFD:366
Mandating access to network infrastructure can be justified as a

means of increasing competition, but national regulatory
authorities need to balance the rights of an infrastructure owner
to exploit its infrastructure forits own benefit, and the rights of
other service providers to access facilities that are essential for
the provision of competing services. Where obligations are
imposed on operators that require them to meet reasonable
requests for access to and use of networks elements and
associated facilities, such requests should only be refused on the
basis of objective criteria such as technical feasibility or the need
to maintain network integrity ... The imposition by national
regulatory authorities of mandated access that increases
365
Since the responsibility of enforcement of both Access and Framework Directives is given to the
national regulatory authorities and not to competition authorities; as regards electronic
communications sector, the regulatory authorities are fully competent in imposing any (either ex post
or ex ante) access obligations. Although competition authorities are equipped with the power to
investigate abusive practices and ‘refusal to grant access’ cases, regulatory authorities under Article 8-
12 of the Access Directive, are also empowered to remedy such cases via its respective tools specified
therein.
366
networks and services, 2002 O.J L 108/33. (Framework Directive), at 19 (Preface)
123
competition in the short-term should not reduce incentives for
competitors to invest in alternative facilities that will secure more
competition in the long-term...
As could be seen, mandatory access scheme is transformed within the new

Framework and is superseded with a type of ‘exceptional’ intervention aiming at
removing the market distortions which are temporary in nature. The aim is to reduce
ex ante sector-specific rules progressively as competition in the relevant market
develops. When the competition becomes sufficiently effective in the marketplace,
then mandating access to third parties in the form of mandatory unbundling, co-
location, etc. would become quite limited. Thereby the application of EFD will most
probably be given way in a wider range, as an ex post measure by nature. This is so
because, along the technological and market developments within the
telecommunications sector, no one could predict the extent to what the network
components should be mandated for third party access. This fact entrusts a crucial
role to EFD in the Internet Age, owing to its flexible nature. Thanks to the flexibility
of the Doctrine, some network elements which are not currently deemed as an
‘essential facility’ could be opened to third party access by application of the
Doctrine, in the future. In fact, there is always the possibility of arising of new
bottlenecks (essential facilities) as a result of technological developments. And, such
an occurrence may require an intervention where the new bottlenecks are not able to
be defined in a precise form in advance. All these possible developments would
make EFD more effective under the competition policies.
As a conclusion, though the Community Courts have not made any decision as
regards application of EFD in field of electronic communications, under the new EU
legislation is clearly recognisable an implicit adoption of the Doctrine which bear
new appearance(s) of EFD in future. As will be discussed below, important signals
are identifiable for the future envisioning of the Doctrine in not only
telecommunications but also other neighbouring markets.
124
2.7 The Future Implications of EFD for Telecommunications Sectors
2.7.1 Convergence and Institutional Implications
As mentioned above, telecommunications markets have been undergoing a phase of

rapid convergence with neighbouring markets, particularly since mid-1990s. The
broadcasting, telecommunications and information technology markets are rapidly
converging towards a single multi-media market in which TV operators supply voice
telephony, telecommunications companies supply video images, and where the
Internet is delivering both basic voice telephony and moving pictures on a
commercial basis.367
While attention is generally concentrated on the convergence between

telecommunications and broadcasting, the impact of convergence is easily seen also
in other fields, i.e. between fixed-mobile communications, finance and
368
telecommunications, etc. At this juncture, in terms of access to the essential
network facilities, two major consequences seem to emerge:369
First, new types of service providers will require new types of resources and access
to new types of bottlenecks and bottleneck holders, ranging from sophisticated
network resources to access to set-top boxes, conditional access systems, navigator
software, AIPs (Application Programme Interfaces), and content rights,
Second, convergence threatens to outpace existing sector-specific regimes.

As the neighbouring markets converge rapidly, an increasing demand for high-speed
Internet through new technologies such as DSL, upgraded cable television networks,
etc. is witnessed. Within such fundamental changes, notably to say that the Internet is
367
Cowie, C. and C. T. Marsden (1998), “Convergence, Competition and Regulation”, International
Journal of Communications Law Policy, p. 1. Available at http://www.ijclp.org
368
15.
369
Ibid.
125
at the very centre of crossing field of the converging markets.370 The Internet has the
potential to develop into the link between the current networks and the digital
delivery systems of the future.371 In fact, having a global and a pervasive nature, the
Internet is of the central importance for building the Information society, as well.
Given the threatening face of the ‘convergence’ phenomenon, converging markets

and services such as digital TV and other multi-media services will probably bring
out major challenges for the existing regulatory regimes. The remarks of Herbert
Ungerer have important implications about the relationship between the convergence
and the future developments:372
Convergence is driving infrastructure provision but it is also

defining the future bottlenecks (essential facilities)…whether the
band-width (broadband) explosion will happen, will entirely
depend on the competitive conditions…Convergence can now
not mean the creation of new super-monopolies – and this danger
is very real. It is the immediate question underlying most current
competition cases in the area.
Though not being new in telecommunications sector, emergence of new bottlenecks

would be much more faced after a unique multi-media sector has evolved out of the
converging markets. As Ungerer states, such an emergence is directly related to the
competitive conditions in the market. In other words, a fateful situation occurs when
a super monopoly possessing big market shares in each of the converging markets
emerges and follows new strategies detrimental to new entrants and consumers.
Convergence must not be deemed a bad fate through the future complexities, alone.
That is to say, if a super monopoly begins to operate in the converged marketplace
and an increase in competition breaches is observed, one might then, mention about a
case of bottleneck situation emerged out of convergence.
370
See supra, p. 87 (Figure 5)
371
Herbert Ungerer, “Competition in Telecommunications - the Regulators’ Challenge”, Asia
Telecom 97 Forum, Singapore, 10.06.1997, p. 5.
372
Ibid.
126
In case of super monopolies which have emerged out of the distorted competition in
a converged environment, the question of institutional design of competition policies
comes to the agenda of public authorities. In such a competitive failure, it is widely
argued that sector-specific regulation may lack flexibility and lead to technological
by-pass and regulatory obsolescence in a market as dynamic as digital television.373
At this juncture, traditional approaches with regard to regulation of network
industries might be challenged within the context of rapid technological
developments.
In fact, considering the global nature of new electronic communications services

such as World Wide Web and the growing face of convergence, it is unlikely that
any detailed sector-specific regime for regulating access could satisfactorily ensure
the regulatory objectives. Given the fact that sector-specific regulation is thought to
be imposed to ensure a competitive and dynamic environment; competition law
remedies must be recalled instead of sector-specific rules in the event of a converged
marketplace that has arisen in the intersection area of competitive and dynamic
sectors. That is to say, sector-specific rules must be progressively withdrawn after a
competitive structure has been emerged and would rather be replaced with
competition law rules. Considering that EFD is placed at the centre of the access
measures under competition policies, all the markets and competition breaches might
be assessed via a sophisticated interpretation of EFD in a converged marketplace,
then.374 In the light of above statements, EFD would be the most convenient
competition law tool in a converged marketplace, owing to its adjustability to
electronic communications markets which involve steadily changing needs.
373
Herbert Ungerer, “Competition in Telecommunications - the Regulators’ Challenge”, Asia
Telecom 97 Forum, Singapore, 10.06.1997, p. 18.
374
According to Ungerer, the very concept of private sector self-regulation of the Internet will make
strict application of competition law indispensable for securing competitive access markets in order to
prevent the emergence of new bottleneck (essential facility) holders at the level of the global
telecommunications market. (See Herbert Ungerer, Competition Workshop on “Ensuring Efficient
Access to Bottleneck Network Facilities: The Case of Telecommunications in the European Union”,
13 November 1998, Florence, p. 23.)
127
2.7.2 The Role of EFD within the Technological Changes
In parallel with the technological and market developments which generally

culminate with increase of technological bottlenecks, emergence of innovations and
deployment of new technologies would naturally be a topic of agenda in competitive
regimes. As Doyle points out, technological innovation is one of the two main
drivers of convergence in global trends.375 It should be noted that if neighbouring
markets converge without adequate investment, in other words, when a convergence
occurs in an inefficient manner, more complicated and unidentifiable anti-
competitive behaviours may be witnessed in the marketplace. That is to say, unless
investors are encouraged sufficiently, there appear undoubtedly competitive concerns
that would arise from monopoly control(s) of bottleneck (essential) facilities after
convergence which has evolved with the increase of technological opportunities.
At the EU level, a remarkable point in this regard is that the physical networks in
most EU countries were constructed in the era of national monopoly, and in the
aftermath of the removal of monopolies, a limited investment is undertaken in
telecommunications sectors. In order to compensate lack of investment and foster
competition at whole (infrastructural) level, building new networks must be
encouraged, in actual terms. In this respect, when service-based competition reaches
to an appreciable level, EFD must be interpreted narrowly as having been portrayed
in the Oscar Bronner judgment. This is so, because the over-zealous application of
the Doctrine has the potential seriously to undermine the incentive for firms to
innovate.376
The Commission made it clear that it favours this approach in field of

telecommunications. The Commission stated in its Communication Document on
Cable TV Review that “From a competition policy point of view, convergence must
be built on the development of a broadband base of pro-competitive infrastructures
375
Chris Doyle, “Local Loop Unbundling and Regulatory Risk”, Journal of Network Industries, 2000,
Vol. 1, p. 39. Available at http://www.cdoyle.com/papers/llurisk.pdf Other main driver of
convergence is ‘liberalisation’, according to C. Doyle.
376
A. Overd and B. Bishop, “Essential Facilities: The Rising Tide”, European Competition Law
Review, 1998, No: 4, p. 185
128
of telecommunications and cable TV networks.”377 Similarly, in Article 8 of the
Framework Directive, ‘promotion of competition in the provision of electronic
communications networks’ is specified as a policy objective of the new Regulatory
Framework, and in the same Article, ‘encouraging efficient investment in
infrastructure and promoting innovation’ is envisaged as one of the ways to reach
this end.378
However, we are now faced a sharp decrease of investment in comparison to the

years preceding full liberalisation, particularly in the context of fixed telephony
networks. Under service-based competition, encouragement of new entries by the
Commission through mandating access has led to a bias towards duplication of
networks. Considering the fact that investment in local telecommunications networks
would be unattractive, many investors turned their face to mobile communications. It
is actually eye-catching that mobile technology is generally regarded as an authentic
success of European industry when compared with USA.379
On the other hand, the incumbents acting in the local telecommunications markets
beginned to face the competition of mobile and cable telephony as competitive
substitutes to the local loops. Necessarily saying, cable TV is participating in the
competition nearby mobile technology, with the advantage of being capable to
provide multiple services including voice, data and video and cost advantages.380 At
the local access networks (level), there are also a variety of technology options other
377
Communication from the Commission on the Consultation on the Green Paper on the
Liberalisation of Telecommunications Infrastructure and Cable Television Networks, COM(95)158,
03.05.95., p.23. Available at http://europa.eu.int/ISPO/infosoc/legreg/16bar3c.html
378
networks and services, 2002 O.J L 108/33. (Framework Directive), Art. 8.
379
University Press and the Oxford Review of Economic Policy, Vol.17, No.3, 2001, p. 425.
380
However, in order for co-axial cables which are used in performing cable TV services to extend to
other areas, i.e. cable telephony, internet, etc. a technical adaptation (upgrade) is necessary via
additional twisted copper pair.
129
than cable TV, which reveal different characteristics and cost structures. (See Table
2, p. 118)381
In confrontation of the increase of alternative technologies, i.e. cable and satellite

networks, local loop unbundling becomes more substitutable, especially in terms of
broadband (high-speed) Internet applications. Thus, the perception that the local
loops are essential facilities would be discounted against this picture.
Correspondingly, the superiority of unbundled access to the local loop in provision of
high-speed Internet might be challenged in time. In near future, non-substitutable and
indispensable character of the incumbent’s local loops would therefore be deemed
unfounded.
Given the advantages and disadvantages of other access technologies, arguments

alleging the ‘essentiality’ of local loop seem to be less meaningful, in terms of EFD.
Another point that evades application of EFD in context of local access networks is
related to the increase of network providers after liberalisation. Because of scale
economies, alternative access infrastructure will typically able to be provided by new
entrants without any need for regulatory support in a range of areas.382 In Europe,
there already exist many alternative access providers, among which cable television
operators occupy an important place, and operators using other access technologies
run comparatively less.383 When the discussions on the ‘non-substitutability’ and
‘indispensability’ themes are reminded within the context of technological
developments and radical changes after liberalisation of EU telecommunications
381
Currently, most consumers can connect to the Internet from their home or small business premises
through dial-up (narrowband) access at very low speeds. In the narrowband Internet access market,
local loops may seem to be in the dominant position, because narrowband modems (with access speed
up to 56Kbps) are the mostly adopted technology at present. Aiming at promoting advanced Internet
and broadband services, different countries established different regulatory strategies among which,
unbundling policies became the forerunner policy in the EU. (See supra, p. 117-118) However in the
broadband Internet access, local loop based technology such as DSL and leased line (those with speed
not more the 2Mbps) is challenged by other alternative broadband access technologies such as cable
TV, fixed wireless access, etc. From that point of view, in particular, the ability to offer competitive
video, voice, and high-speed data services of cable TV networks after an upgrading process has
attracted non-incumbent operators to enter the cable television market.
382
Chris Doyle, “Local Loop Unbundling and Regulatory Risk”, Journal of Network Industries, 2000,
Vol. 1, p. 41. Available at http://www.cdoyle.com/papers/llurisk.pdf
383
Ibid.
130
sector, the strong face of EFD seems to be turned over. Such an interpretation may be
done in line with recent developments; however this must be verified along with the
recent legislative steps. When EFD is assessed under the new regulatory framework,
somewhat a different conclusion could be inferred as will be mentioned below.
From this vantage point of view, two points are remarkable with regard to the
Essential Facilities Doctrine. First, within the technological changes, the rapid
increase of essential (bottleneck) facilities enables EFD to be used in a more
consistent and wide-ranging way, thanks to the flexibility of the Doctrine. Second,
while the access technologies are figured on the agenda of the regulatory policies,
increasing alternative technologies gain importance and invoking EFD becomes
relatively needless. In spite of such a prediction, EFD would rather be deemed
necessary for dealing with future bottlenecks, since most of the bottleneck situations
and facilities are not easily identifiable for the foreseeable future, and EFD is
adjustable to every technological change actually.
2.7.3 Complementarity & Superiority of EFD
Most of the debate so far has been in the EU preoccupied with how best to open
incumbents’ networks to third parties.384 On the one side, everyone acknowledged
the problems concerning third party access, and sought possible ways in order to
resolve such problems. On the other side, compensation of insufficiency in
investments and innovations comprised a macro policy problem to be tackled in time.
Ideally, regulatory policies harmonise service-based and facility-based competitive
strategies in an efficient manner. Therefore, in policy procurement, one ought to
strike a balance between the incentives for deployment of innovative facilities and
the obligations for access to already established networks.
Against the policy concerns regarding lack of investments, facility-based

competition, increase of access technologies, etc. it can be supposed that the
enforcement of EFD must be handled in a ‘complementary’ manner. Such an
384
University Press and the Oxford Review of Economic Policy, Vol.17, No. 3, 2001, p. 416.
131
approach also seems to be adopted in the Commission Guideline (2002/C 165/03)
which was issued in accordance with the Framework Directive. This Commission
Guideline includes important clues as regards the application of EFD under the new
Framework. In the paragraph 81 of the so-called Guideline, emphasis is cited to the
complementarity of the Doctrine:385
... In particular, the doctrine of ‘essential facilities’ is

complementary to existing general obligations imposed on
dominant undertaking, such as the obligation not to discriminate
among customers and has been applied in cases under Article 82
in exceptional circumstances, such as where the refusal to supply
or to grant access to third parties would limit or prevent the
emergence of new markets, or new products, contrary to Article
82(b) of the Treaty.
Under the aspirations of stimulating high internet penetration and the development of
e-commerce and interactive applications, determined regulatory steps had been taken
to date by the EU Institutions. In this regard, the incumbents’ local loop(s) were
required to be granted for access on reasonable terms and conditions, in December
2000. However, as a result of liberalisation and growing convergence, there emerged
a number of technological substitutes which paved way for new and cheap services
in telecommunications sector.
Given the fundamental changes such as removed borders between markets, increased
technologies for access and high-speed innovations, envisioning EFD in a cautious
manner seems much more important than before. In renewing face of EFD, simple
but effective criteria rather than technically detailed conditions must be emphasised.
At this juncture, the key issue for the competent authorities, to resolve in EFD cases
would simply become whether usage of alternative technologies or facilities is
commercially or technically feasible.386 Unless the alternative facilities are not
385
Commission guidelines on market analysis and the assessment of significant market power under
the Community regulatory framework for electronic communications networks and services (2002/C
165/03), at 81.
386
University Press and the Oxford Review of Economic Policy, Vol.17, No. 3, 2001, p. 417. Inherent in
132
technically or economically feasible, then reliance on EFD would not be consistent
with the core of the EC competition law principles.
The second line of justifications that evade application of EFD lies at macro policy
reasons such as enhancing investment on infrastructure, promoting access
technologies, protection of intellectual rights, etc. Through such reasons, facility-
based competition might be initiated so as to take precedence over service-based
competition by some policy makers. From the scratch, EU Institutions favoured
service-competition consciously. In fact, the EU Institutions have made most of their
decisions on the straight way of the service-based competition, which is
predominantly inspired with notion of development of competition in services, rather
than infrastructure. However, one can receive some important signals regarding lack
of adequate investments within the 2002 EU Regulatory Framework, that give
priority to facility-based competition in order to remove the imbalance already
established in favour of service-based competition in the sector.
From that point of view, one can say that the Oscar Bronner judgment is proved after
the recent changes. This seems so, because in the 43rd paragraph of the so-called
judgment, does exist one of the main thrusts of EFD, making the distinction between
the ‘indispensability’ and ‘being less advantageous’.387 According to this distinction,
in order for a facility to be essential, establishing that the facility in question is less
advantageous among alternatives is not sufficient at all. In fact, the indispensability
and non-substitutability tests brought by the Oscar Bronner makes clear that EFD is
not easily applicable to all ‘refusal to deal’ cases.388
Considering the market and technological developments together with the Oscar
Bronner judgment, some conclusions are note-worthy here. Primarily, it is visible
such a ‘feasibility’ analysis is the examination of harmful effects to competition arisen out of not
granting access. This is why in any EFD case, the primary concern is related to the competitive
impacts of the refusal of the essential facility owner.
387
388
It is commonly accepted that since Oscar Bronner decision, it is hard to demonstrate that an
infrastructure or a facility is ‘essential’ for the purpose of applying EFD. See supra, p. 52-58.
133
that 2002 EU Regulatory Framework adopted more flexible provisions related to
mandating third party access. When the Access Directive is considered with other
legal measures under the new Package, one can conclude that a less interventionist
policy is encouraged in the new Framework. Under this policy, facility-based
competition is seemingly given way as an alternative regulatory policy against
service-based competition which has dominated the EU telecommunications policies
so far.
Such an approach that envisages measures to restore the balance between the
incentives to build new networks and to use existing networks re-shaped EFD, as
well. As mentioned above, many provisions of the 2002 EU Regulatory Framework
give path for the application of the Doctrine. However, the Doctrine seems to have
been envisioned on a ‘complementary’ basis under the new Framework.
Though having not been referred clearly under the new Framework, one can extract
some provisions related to EFD from the Access and Framework Directives that
constitute the core of the new Framework. Notably saying, the philosophy and the
regulatory tools of the new Directives are in compliance with EFD. Hereby, it must
be reminded that the new Framework rendered equivalence between ‘dominant
position’ and ‘significant market power’, and in determining access obligations
competition law-based terms and obligations are for many times cited in the new
regime. Among tools envisaged for mandating third party access, EFD is remarkable
from this vantage point of view. That is to say, not only competition authorities but
also national regulatory authorities could refer EFD in conferring access principles,
because it lies at the area of intersection between sector-specific regulation and
general competition law, and yield effective solutions. Therefore, it is possible to say
that in future, the concept of ‘essential facilities’ will in many cases be of direct
relevance in determining the duties of dominant operators in telecommunications
sectors.
In addition, as emphasised above, converging markets would require the flexible

nature of the Doctrine due to possibility of emergence of new bottlenecks, and this
134
fact might elevate EFD to superior level among competition policies. Furthermore, in
this century that is involved with fast-moving changes such as Internet telephony,
digital television, and other multi-media services, the need for the Doctrine would
become more evident and urgent so as to supersede the application of sector-specific
rules. Unforeseen changes in technology and steady convergence of markets reveal
that network architecture in electronic communications sector become more
complicated and require less stringent definition(s) of bottlenecks. As expounded
above, sector-specific rules are designed according to the distinct features of each
sector and do not have potential alone, to keep track of the rapid technological
changes and increasing bottlenecks. Given the fact that EFD is quite flexible and
adjustable to high-speed technologies and converging markets, it should be stated
that the Doctrine is exactly capable to cope with bottleneck situations in future
telecommunications sectors.
As a conclusion, the extension of competition law to telecommunications sector puts

the Essential Facilities Doctrine in a ‘key’ situation at EU/national level. Actually,
the Essential Facilities Doctrine represents the most effective tool of competition law
in determination of access rules. It is obvious that access scheme will remain to lie at
the core of the competitive measures in the field of telecommunications, and all these
facts strengthens the superiority of the Doctrine in future telecommunications
sectors.
135
3CHAPTER 3
3. IMPLICATIONS OF THE ESSENTIAL FACILITIES

DOCTRINE FOR TURKISH TELECOMMUNICATIONS
SECTOR
3.1 EFD under Turkish Competition Law
Like the case in many European countries, the application of competition law is an
integral part of free market economy in Turkey. Considering that competition policy
focuses on the aim of realisation of two main purposes, namely, prevention of anti-
competitive agreements and prohibition of abuse of dominant position, it is possible
to say that the Turkish system conferred in the relevant legislation aligns with
competition policy objectives in general. Articles 167 and 172 of the Constitution,389
establish the fact that the economic choice in Turkey is free market economy that
depend on some protectionist rules in fields of anti-trust and consumer rights.
In order to ensure both the implementation of the abovementioned Article of the

Constitution and the harmonisation of the EC Competition Rules, the Law on the
Protection of Competition No. 4054 was enacted in 1994.390 After examining the
definition, legal boundaries and the given examples of ‘abuse of dominance’ together
with other prohibited practices, one can say that the Law on the Protection of
Competition (hereinafter sometimes called ‘Turkish Competition Act’) largely
389
Article 167 of the Constitution of the Republic of Turkey reads as follows: “The State takes
necessary measures for ensuring and improving the sound and regular functioning of the markets for
money, loan, capital, goods and services; it prevents, in markets, de facto monopolisation and
cartelisation or those which shall arise by agreement”. On the other hand, Article 172 of the
Constitution reads as follows: “Since the protection of consumers is only possible in free market
economy, conditions for free competition should be ensured, and cartels and monopolies should be
prevented in compliance with the requirements of this market”
390
Here it is particularly useful to refer the obligations under the Association Agreement (Ankara
Agreement) of September 12, 1963 between Turkey and EEC. Article 16 of the Association
Agreement requires that the principles conferred in the provisions of the Rome Treaty concerning
competition, tax and the alignment of legislation be applicable within the association relationship.
136
depends on Articles 81 and 82 of the Rome Treaty. In fact, Turkish Competition Act
identifies the prohibited practices, in particular agreements restricting competition
(set forth in Article 4) and abusive practices (set forth in Article 6) almost the same
as the referred Articles of the Rome Treaty do.391
For purposes of implementation of Article 6 (Article entitled ‘Abuse of Dominant

Position’), it is crucial to determine whether an undertaking holds a dominant
position in the Turkish context. Article 3 of the Competition Act defines dominant
position as “any position enjoyed in a certain market by one or more enterprises by
virtue of which, those enterprises have the power to act independently of their
competitors and purchasers in determining economic parameters such as the amount
of production or distribution, price and supply”. This definition is similar with the
‘dominant position’ definition of the United Brands judgment, which was also
referred in other Community Court judgments and Commission decisions.392 Another
related issue that is not specified in the Turkish Competition Act is the procedure
regarding ‘market definition’. Market definition is a well-known, important step to
reach to an abuse of dominant position in a case, and due to this fact it is detailed in
391
For the reason that the subject-matter of this part relates to Article 6 of the Law on Protection of
Competition, that Article is worth quoting here:
“Abuse of Dominant Position
Article 6- Any abuse, by one or more enterprises acting alone or by means of agreements or
practices, of a dominant position in a market for goods and services within the whole or part of the
territory of the State, is unlawful and prohibited.
Abusive practices are, in particular, as follows :
a) To prevent, directly or indirectly, other enterprises in its area of commercial activities or
practices which aim to impede the activities of the competitors in the market;
b) To make discrimination, directly or indirectly, by way of imposing dissimilar conditions
for equivalent and same rights and obligations to the purchasers who have equivalent position;
c) To make the conclusion of contracts subject to the acceptance of restrictions concerning
resale conditions such as the purchase of other goods and services or acceptance by the intermediary
purchasers to display other goods and services or maintenance of a minimum resale price;
d) Practices which aim to distort competition in a market for goods and services by means
of taking financial, technological and commercial advantages created by the dominant position in
another market;
e) To restrict production, marketing or technical development thereby causing a
disadvantage for the consumers.
392
In the United Brands Judgment (Case 27/76, United Brands v. Commission, [1978] ECR 207.), the
dominant position is defined as follows: “A position of economic strength enjoyed by an undertaking
which enable it to prevent effective competition being maintained in the relevant market by affording
it the power to behave, to an appreciable extent, independently of its competitors, customers and
ultimately consumers.”
137
the Communiqué No.1997/1 which was issued in accordance with the Law on
Protection of Competition.
Having close relation to Article 82 of the Rome Treaty in the context of the
principles concerning abuse of dominant position; Turkish Competition Act does not
confer an explicit statement prohibiting a ‘refusal to deal’ with competitors or
consumers.393 Similarly, under the Turkish Competition Act does not exist a
prohibitive rule on refusal to grant access to ‘essential facilities’. However, there are
a number of cases that illustrate the applicability of the Essential Facilities Doctrine
under implementation of Article 6.
3.2 EFD Decisions in Turkish Case-Law
3.2.1 Eti Holding Decision394
The case surrounds the relationship between legal monopoly and essential facilities
under competition law. Eti Holding is the sole operator in field of extracting,
gathering, refining, manufacturing and marketing of boron products in the Republic
of Turkey, in accordance with the Law numbered 2840. Enforcement of this Law has
been entrusted to Eti Holding under a legal monopoly. Eti Holding is also a world-
wide operator, controlling the share of %65 of the world’s reserves.
Ceyta A. ., a competing firm in the downstream (derivative) market, requested

colemanite ore, a kind of boron raw material from Eti Holding in order to produce
393
In terms of scope and purpose, Turkish Competition Act is in harmony with the Community Rules.
Referring ‘abuse of dominance’ rather than ‘monopolisation’, and adopting the principle that being
dominant in a relevant market is not per se unlawful reveal the fact that Turkish system depend on the
EC Competition Law principles. However, the ‘compensation of three times of the damage or the
profit’ rule set forth under the second paragraph of Article 58th which reads as: “In cases where
damages arise from an agreement, or a decision or from the heavy negligence of the parties, the judge
may, upon the request of the parties who have damages, decide on a compensation three times of the
actual damages or three times of the profit gained or likely to be gained by the parties who caused
damages.” seems to be adapted from the Antitrust Law of US.
394
Competition Board’s Decision dated 21/12/2000 and numbered 00-50/533-295.
138
derivative colemanite. Eti Holding rejected to supply the requested raw material, and
consequently Ceyta A. . brought the case before the Turkish Competition
Authority. The Authority did not find an infringement of the Competition Act,
holding that there did not appear a duty to deal in the colemanite market due to the
legal constraints therein. In its decision, the Competition Board designated the
‘derivative colemanite’ market as the relevant market, implying that existence of the
legal monopoly in the colemanite market justifies the ‘refusal to deal’, ironically. In
reaching such a conclusion, the Board identified the essential facility with
“unavailability of an alternative other than the inevitably needed facility that is
owned by the dominant undertaking or impossibility of duplicating a new source
economically and reasonably, in order to carry out a transaction” and pointed out that
colemanite is the essential facility for production of final products such as glass,
ceramic, etc.
Eti Holding decision is debatable from some respects, one of which relates to the
interpretation of the Law numbered 2840 and Law on Protection of Competition.
Scope of the Turkish Competition Act has been drawn in Article 2 of the so-called
Act, and the focus is directed on “any undertaking operating or affecting the goods
and services markets within the territory of the Republic of Turkey”. On the other
hand, Article 3 of Turkish Competition Act clarifies what an undertaking is for the
purpose of Turkish competition law regime. According to the said Article, an
undertaking is defined as “any natural or legal person who produces, markets or sells
goods and services and who forms an economic whole, capable of acting
independently in the market”.
Thus, any undertaking - either public or private - carrying out an economic operation
and/or affecting the goods and services markets within the territory of the Republic
of Turkey falls under the scope of the Turkish Competition Act. In spite of the
quoted statutory provisions, the Board refrained from challenging the refusal to
supply of a public undertaking in Eti Holding decision. Although the reason for such
a refraining stems from the existence of the legal monopoly, Turkish Competition
Act is applicable to all undertakings irrespective of the nature of the undertaking
139
concerned. Therefore, Eti Holding decision is quite controversial for its interpreting
the current legislation.
On the hand, there seem controversial points in market definition and reconciliation
of essential facility with the relevant products, within the scope of decision. In the
decision, a distinction between downstream and upstream markets has not been made
clearly, whereas the case is quite suitable for such a two-fold determination of
markets. The unclear approach on market definition is perpetuated in delineation of
essential facility, as well. On the one hand ‘colemanite’ was specified as the essential
facility, on the other hand ‘derivative’ but not raw colemanite was accepted as the
product that constitutes the relevant market. In this context, a structural or
operational link has not been established for the ultimate decision, namely neither a
relationship between the market products and essential facility nor a link between
essential facility and refusal to supply has been set forth in the decision.
However, the Competition Authority seemed to have affirmed the validity of the
notion of ‘essential facilities’ in Eti Holding decision. Though an analysis has not
thoroughly been done in terms of the criteria that apply to EFD, a distinct perspective
giving way to the enforcement of the Doctrine could be observed within this
decision.
3.2.2 B RYAY Decision395
B RYAY case illustrates a typical ‘refusal to deal’ case. In the case, holding a joint
dominant position in newspaper and magazine distribution market, BBD and
YAYSAT aimed at imposing more onerous terms and conditions to the competing
newspapers and magazines, and for this end founded B RYAY. BBD, YAYSAT and
their joint venture B RYAY reached a consensus to determine the distribution
conditions, commission fees and other charges to be applied to other distribution
companies within the Main Contract of B RYAY. Thereafter, BBD and YAYSAT
forced their dealers not to sell newspapers and magazines distributed by other
395
140
distribution companies, made some amendments to the applied distribution contracts,
refused to renew existing contracts with some publication owners and transferred to
B RYAY the distribution activities of publications already distributed under the
existing contracts. Against his picture, it was decided to open an investigation to
establish whether there existed an infringement of competition in the context of
Articles 4 and 6 of Turkish Competition Act.
The most prevailing characteristic of the newspaper and magazine distribution
market is the existence of oligopoly between three undertakings that is strengthened
through horizontal as well as vertical integration(s) within the markets for
distribution (upstream) and publication (downstream). Prohibitively expensive costs
for duplicating a distribution network and vertical integration between the already
established companies made new companies’ situation quite difficult in comparison
to the incumbent firms.
In the case, the Competition Board was confronted with both an explicit refusal to
deal and a discriminatory act that cause harmful effects to competition in the market.
Thereby, the Board, finding that Article 4/a,b and Article 6/a,d were infringed with
the so-called activities, imposed fines on BBD, YAYSAT and B RYAY and obliged
all municipality kiosks to sell all publications provided by all distribution
companies, thus made it available for consumers to find any newspaper and
magazine at final sales point.
The prohibitively expensive cost of duplicating a distribution network cited in the

decision reminds the Oscar Bronner judgment. In fact, there appears a similarity
between the nature(s) of the requested facility in the said two decisions. However, in
B RYAY case was held a joint dominance by three undertakings whereas Mediaprint
was holding dominant position on itself in Oscar Bronner case. Another difference
seems to emerge in relation to the ‘essential’ character of the facility in question. In
Oscar Bronner, the distribution network was deemed substitutable with other means
of distribution such as post, kiosk, etc. while any alternative way is hardly seen, on
the part of the requesting party in B RYAY decision. As a matter of fact, due to the
anti-competitive agreement signed between the dominant distributing companies in
141
B RYAY case, other companies would have been in a ‘genuine strangehold’. This is
so, because municipality kiosks were not able to sell already distributed newspapers
and magazines under the said agreement, and could not renew their long-standing
agreements according to the previous conditions. Hence, the conditions in B RYAY
seem harsher, comparing to those in Oscar Bronner, and the Board decision
regarding B RYAY case therefore has a sound basis in terms of criteria that apply to
refusal to deal / EFD cases.
3.2.3 ÇEA Decision396
ÇEA is the last dated decision with regard to application of EFD in Turkish case-
law. In the case, ÇEA was the sole operator which was given concession to
generate, transmit and distribute power within a specific area of southern Turkey.
When Enerjisa, a firm acting in the market for power generation requested ÇEA to
transmit the power needed for generation, ÇEA did not accept to make an
agreement for transmission and did not provide interconnection for transmitting
power. On the other side, ÇEA refused to another undertaking named Toros which
was operating as a power supplier on the basis of auto-generation. The said refusal
had taken place with prejudice to the agreement made between ÇEA and Toros in
the year of 1999.
When the so-called undertakings claimed that the refusals of ÇEA constituted a
competition infringement under the Law 4054; the Board, in its decision, first held
that ÇEA was the dominant undertaking in the market for power transmission. After
then, the Board pointed out that electricity is a non-storable product and for this
reason, interconnection for power transmission and distribution during the process of
generation is crucial for undertakings. After all, on the ground of the uniqueness of
the power transmission, ÇEA was held as the controller of the ‘essential facility’ in
the downstream (power transmission) market. In other words, the infrastructure for
power transmission was deemed a facility ‘essential’ to power generation in the case.
Following the abovementioned assessment, the Board found that ÇEA abused its
396
142
dominant position by delaying to meet the Enerjisa’s requests for power transmission
and thwarting the transmission of power on the part of Toros and by preventing the
so-called undertakings from access to the essential facility in the market for power
transmission.
The most remarkable points of the decision are reference to the essential facility and
definition of the relevant market(s). Unlike the usual distinction in essential facilities
cases, the decision conferred that the dominant undertaking abused its dominance in
the upstream market which was defined as the market for provision of power
generation. However, in general, owner of the essential facilities possesses
monopoly/dominance at least in the upstream market and abusive practise usually
take places in the downstream market. However, in this case, an adverse situation is
confronted, where the dominant supplier controlling the essential facility in the
downstream market (market for provision of power transmission) abused its
dominant position in the upstream market. Regarding the ‘essential’ character of the
transmission infrastructure which was deemed ‘essential facility’ in the decision,
relevant points will be discussed in relation to other EFD cases in field of
telecommunications.
3.3 EFD Cases in Field of Telecommunications
3.3.1 TTA Decision397
TTA Decision is highly important in terms of introduction of EFD into Turkish

Competition Law. In the case, Türk Telekom (TTA ) acting as the legal monopoly in
carrying out (fixed) voice telephone services as well as establishment and operation
of all telecommunications infrastructure was accused of abusing its dominant
position for several reasons. Accusations concentrated on charging its corporate users
below the rates which it applied to its competitors, in field of providing broadband
397
143
internet access services. Not having an explicit EFD character, the case involved
tariff-related abusive practices in particular, vertical price squeezing.398
Holding that Türk Telekom was dominant in the market involving the necessary
infrastructures for the provision of broadband/narrowband internet access service, the
Board found that Türk Telekom infringed Article 6 of the Turkish Competition Act,
by abusing its dominant position through determining the charges of services
provided under the name of TTNet to users, below the charges which it applied to
competing undertakings in the same market. Besides, in the Board decision, some
citations were made to EFD through references to the terms of the usage of the
monopoly rights by TTA . After referring the philosophy and the basic criteria that
apply to EFD, the position of TTA was examined in terms of the Doctrine in a
rather superficial manner. As a conclusion, since TTA has the control on access to
end-users over cable-television and local telecommunications networks (which
exhibit natural monopoly characteristics according to the decision) and it is
considered unpredictably impossible to duplicate such networks; the services which
TTA solely provides were deemed convenient to be called ‘essential facility’ in the
decision.
Another remarkable point expounded in the decision relates to the relationship

between the Turkish Competition Authority and Telecommunications Authority
(sector-specific regulator). After quoting the relevant Articles of both Turkish
Competition Act and Telecommunications Legislation (Law No. 406 and 2813), it
was construed that the powers of each Authority are uniquely designed and due to
this fact, differ from each other in terms of scope, purpose and implementation. In
398
Vertical price squeezing is a particular type of anti-competitive conduct that may be engaged in by
incumbent operators acting in two or more ‘vertical’ markets, i.e. downstream and upstream markets.
This form of conduct can occur when an operator with market power controls certain services that are
key inputs for competitors in downstream markets, and where those same key inputs are used by the
operator or its affiliates to compete in the same downstream market. If the incumbent decided to
engage in vertical price squeezing, it could increase the price to competitors for the upstream input
(i.e. dedicated local circuit rates in the context of TTA case) – while leaving its downstream prices
the same (i.e. prices for its own dedicated internet access services in the context of TTA case). The
effect would be to reduce or eliminate the profits of competitors and their profits would consequently
be squeezed. To increase the squeezing effect, the incumbent could also reduce its downstream prices
(for internet access).
144
defining the respective powers of the said Authorities, the rule lex posterior derogate
priori (the law dated later instead of the law dated earlier applies to the case) was
deemed inapplicable owing to both wording and the spirit of the
Telecommunications Legislation. Moreover, it was not considered possible under the
current legislation, that the Telecommunications Authority can investigate and
conclude (with necessary remedies) the cases that involve breach of competition in
telecommunications sector.
3.3.2 Aria (Roaming) Decision399
This decision constitutes the core of the case-law regarding EFD under Turkish
Competition Law. The case relates to ‘refusal to grant access’ to their own respective
networks by two GSM 900 operators (Turkcell and Telsim) towards a GSM 1800
operator (Aria) and seems to be totally built upon EFD.
In the case, there were three players, among which Turkcell and Telsim were granted
GSM 900 licenses in 1998. Some years later in early 2001, Aria was awarded a GSM
1800 license and commenced to operate, however needed access to the existing
networks of Turkcell and Telsim via making a roaming agreement,400 since their
parallel networks covered almost all the country and Aria’s coverage was so limited
at that time. However, Turkcell and Telsim refused to open their networks to Aria, in
other words did not accept the conditions offered by Aria for roaming agreement.
In the decision of the Competition Board, the criteria set forth in US and EC case-
law regarding EFD, respectively control of an (essential) facility by a dominant
undertaking, inability of other competitors to duplicate the facility practically or
399
400
For the purpose of the Law No. 406, Roaming means “inter-systems conveyance which provides
operation of services of an operator through the equipment of clients of another operator or which
provides interconnection to another system, provided that certain technical compatibility exists”
Roaming is a typical access agreement signed between (wireless, i.e. GSM) network operators or
service providers to allow access by one service provider’s customers to the network or services of
another service provider located outside the service area of the first service provider.
145
reasonably, denial of use of the facility by the dominant undertaking (with substantial
harm to competition), and the absence of a legitimate business justification for denial
are referred and thereafter it is investigated if the so-called criteria have existed in the
case or not.
The most important argument taken into account within the investigation was
regarding the determination of access charges by Turkcell and Telsim at the very
high rates which could potentially exclude Aria from competing and financing its
investments whilst carrying out its services. The justifications of the respondent
operators (Turkcell and Telsim) which consisted in technical and economic
constraints were not considered legitimate by the Board. As a conclusion, it was
found that the infrastructures of Turkcell and Telsim were deemed “essential facility”
during the phase of entry of GSM operators into the market.
In parallel with the application of the Doctrine, whether there is abuse of joint
dominant position was examined in the decision, as well. Upon holding that a (joint)
dominant position was possessed by Turkcell and Telsim in GSM
telecommunications infrastructure market, the said undertakings were found to have
abused their joint dominance by refusing to make a roaming agreement with Aria. It
was therefore established that Article 6 (a) of the Competition Act reading as “To
prevent, directly or indirectly, other enterprises in its area of commercial activities or
practices which aim to impede the activities of the competitors in the market” was
infringed.
Subsequent to such a conclusion, taking the necessary remedies in order to restore

the distorted competition, namely determination of technical and economic
conditions to be applied in the case was given to Telecommunications Authority.
Repeatedly (in the same manner as in TTA decision), depending on the invalidity of
the rule lex posterior derogate priori in the context of the relations between
Telecommunications Authority and Competition Authority,401 the decision conferred
that the Competition Authority was entirely competent in terms of investigation and
401
See supra, p. 144-145.
146
imposing a fine as regards the case. As was stated in TTA decision, the conviction
that each Authority are authorised on different legal basis and accordingly
empowered with different tools was asserted in the decision. What’s more, the
decision conferred a strict delineation between the powers and responsibilities of
each Authority, in which, power of sanction of Telecommunications Authority is
dedicated only to the licensing breaches and breaches of Telecommunications
Legislation not including a anti-competitive effect.
3.4 Assessment of the Turkish Case-Law in light of the Community Approach

on EFD
Notably saying, Turkish Competition Board regarded EFD as a Doctrine relying on a

stand-alone basis. At first glance, this seems to be in line with the EC case-law,
particularly with the framework drawn with the second line of judgments, i.e. T.
Ladbroke, E. Night Services and O. Bronner decisions. If the Doctrine is evaluated
within the framework of these judgments, we face highly strict rules and a limited
application. However, an over-zealous application of EFD is remarkable in the
context of the Turkish case-law.
That is to say, pertinent infrastructure and/or services have easily been deemed
‘essential facility’ throughout Turkish case-law, so far. When the ‘essentiality’
criterion is considered as the core of EFD, one should characterise a facility within
an ‘essential’ nature, in a cautious manner. This is so, because a Doctrine standing on
a relatively uncertain term ‘essential’ needs to be delineated with clear-cut rules. In
this context, nature and boundaries of the Doctrine as broadly construed in Turkish
case-law recalls the early phases of EC case-law, but not the second phase of the
Doctrine, in particular not the strict interpretation of the Oscar Bronner judgment.
In liberalisation periods of networked industries such as railways, airports,

telecommunications, etc. the Community Institutions, particularly European
Commission followed a comprehensive policy that aim at creating fully liberalised
markets. Not only competition law measures but also other socio-economic remedies
through the Common Community Policies were all directed towards this end in
147
1990s, and subsequent to these policies trans-national borders were started to be
removed in the way of realisation of a European Single Market. EFD has been used
widely in this process, and telecommunications sector became one of the foremost
areas influenced by application of EFD. On the other hand, as telecommunications
markets have been liberalised progressively, the need for applying EFD decreased
relatively, at least an unfettered application was left behind in time. At the last mile
of European case-law in Oscar Bronner, EFD has entered into a turning point where
demonstration that the facility concerned is ‘essential’ depends on some difficult
tests, i.e. indispensability, non-substitutability, etc. From this point of view, Turkey
appears to be going forward through a quasi-European transformation, not having
reached to the so-called last-mile, yet.
When the transformation of Turkish telecommunications sector is taken into account,

the commentators and practitioners might respect approach of Turkish Competition
Authority which usually incorporated an over-zealous application of EFD. However,
a number of issues pointed out in Board decisions especially in Roaming decision are
challengeable. As a matter of fact, the Roaming decision is a distinct decision in
terms of EFD for revealing a unique implementation as well as having far-reaching
results in the sector.
3.5 Assessment of EFD Decisions in Field of Telecommunications
Primarily saying, case-law regarding EFD in field of telecommunications is far less

established under the Turkish Competition Law. Simply, neither clear-cut rules
relating to ‘essential facilities’ nor the specific boundaries of the Doctrine are hardly
seen in the Competition Board decisions. However, in order for EFD to be
maintained and applied within the objective principles, precise and objective
delineation of relevant conceptions rather than reference(s) to the EC case-law is
strictly needed. This is why each antitrust case has its respectively distinct character
and entails peculiar remedies, by nature.
In almost every essential facilities case under Turkish Competition Law, two specific
(downstream and upstream) markets have been defined in accordance with the
148
Community rules. With regard to market definitions, European Commission took
care to comply with the EC Competition Rules and urged not to sacrifice the
competition law principles on ‘market definition’ to the rules of EFD. Confirming
this attitude, not only in its decisions but also through the relevant legislation, the
Commission called attention for market definition within the essential facilities
cases. In this regard, distinction between the telecommunications markets under the
downstream/upstream pattern was adopted by the Commission in the Access
Notice.402 Consonantly in Turkish telecommunications sector, through the Roaming
decision, markets for provision of ‘GSM services’ and ‘GSM telecommunications
infrastructure’ were specified as the two separate markets respectively deemed
‘downstream’ and ‘upstream’ markets.403 Similarly in TTA case, seven product
markets were identified, among which some of which were including downstream
products (generally regarding internet access services) some others were consisted in
upstream products (generally regarding infrastructure needed for access to end-
users);404 and abusive practise had taken place through vertical integration between
the so-called markets via restricting competition in the downstream (internet access)
market.
In contrast with the arguments used in market definition, competition law tools used
in order for reaching abusive practices appear quite subjective. Particularly in
Roaming decision, the discussions surrounding refusal to grant access ‘roaming’
reveal a somehow subjective way of thinking. For instance, in arguing existence of
abuse of joint dominant position in the case, oligopolistic market structure and
(consequently emerged) market distortions are demonstrated as indicators for joint
dominance. Such a demonstration seems to be built upon general evaluations rather
than specific instances, in that, inflexible demand for GSM infrastructure, sunk costs
incurred by new entrants and strong network externalities were shown as the basis of
joint dominance. Afterwards, notion of essential facilities and abuse of joint
402
Telecommunications Sector (1998) O.J. C265/2 (1998) 5 C.M.L.R. 821, at 45. See also supra note in
305.
403
Competition Board’s Decision dated 09/06/2003 and numbered 03-40/432-186, p. 13-17.
404
Competition Board’s Decision dated 06/11/2002 and numbered 02-68/821-333, p. 21.
149
dominant position were interestingly merged in order for the ultimate decision.
However, in proving that the GSM networks are essential to carry out GSM
communications services, the ‘indispensability’ and ‘non-substitutability’ tests were
not referred, depicting a contradiction to the strictly defined Community rules on
EFD.405 Given the Oscar Bronner criteria that apply to EFD; the economic and
technical deficiencies, i.e. disadvantageous conditions regarding base stations,
troubles related to huge costs of building infrastructure and difficulties arising out of
delays, which were pointed out to have existed in the decision do not match with the
strict requirements of EFD quoted above.
Besides, abuse of joint dominance accompanying application of EFD is hard to

justify against the ECJ’s attitude towards joint dominance. In France v. Commission
case, the ECJ found that, while collective dominant positions could be taken into
account (in the course of assessing a concentration under the MCR - within the
referred case), the Commission had to show that the oligopolists “in particular
because of correlative factors which exist between them, are able to adopt a common
policy in the market and act to a considerable extent independently of their
competitors, their customers, and also of consumers.”406 Regarding the Roaming
case, in which the instances developed out of the falling negotiations between GSM
900 and 1800 operators, it is actually hard to verify the existence of an abuse of joint
dominance, since it is not absolute that there was a common policy between the
respondents (Turkcell and Telsim) in the case.
On the other hand, if GSM networks had been deemed essential facility, mandating
access to such networks would have been justifiable under the sole application of
EFD or through sector-specific regulation alone, but not through a combination of
405
Revealing the so-called strictness, Oscar Bronner judgment established that in order for a refusal to
grant access to be unlawful the following conditions must be met: (i) The refusal must be likely to
eliminate all competition in the relevant market on the requesting party, (ii) The refusal must be
incapable to be objectively justified, (iii) The facility in question must be indispensable in order for
business of the requesting person to be carried on (inasmuch as there is “no actual or potential
substitute in existence.”).
406
150
EFD and Article 6.407 From this point of view, reaching an abuse of joint dominance
by resorting to EFD means neither applying EFD nor invoking the arguments related
to abuse of joint dominance is considered adequate on a stand-alone basis in the
context of concluding a competition breach. Then, a unique but vague conception in
the form of a ‘joint refusal to grant access to essential facility as an abusive practise’
is preferably used in the decision.
Another conflicting issue under EFD decisions in field of telecommunications (both

in TTA and Roaming decisions) relates to the overlapping powers of
Telecommunications Authority and Competition Authority. The decisions seem to
have solved this problematic issue by regarding the position of Competition
Authority as equipped with all relevant powers and responsibilities in case of a
competition breach. TTA and Roaming decisions brought an exception to this
general establishment in the event of arising of law breaches regarding
Telecommunications Legislation, not resulting in an anti-competitive effect,
including licensing breaches. According to the so-called decisions, other than the
referred infringements, Competition Authority is deemed solely competent in
investigating and concluding (with necessary remedies) the antitrust cases.408
407
It should be noted that Telecommunications Authority is also involved in the Roaming case, by
playing a role as both a mediator and an arbitrator between the parties, independently of the
investigation carried out by Competition Authority. In this regard, Telecommunications Authority
ordered Turkcell and Telsim to allow Aria to make roaming through their own networks in accordance
with Article 10/5 of the Law No. 406 which reads as follows: “Within the content of this Article,
mobile telecommunications, data operators or operators of other services and infrastructure as
determined by the Authority are also required to meet reasonable, economically proportionate and
technically feasible roaming requests of other operators working in the same field for permitting the
use of the customer equipment of the requesting operator on their telecommunication system.”
However, Turkcell and Telsim brought the Authority order to administrative and judicial courts as
well as international arbitration courts, and the order was rendered ineffective through the ligitations
that prolonged so much. After the said ligitations have concluded, in December 2003 and March 2004,
the Authority respectively imposed fines to Telsim and Turkcell for the reason that they infringed the
said Article. Though having risks of delaying that could arise from litigation,etc. the mandatory
roaming under sector-specific might be considered more justifiable and effective in comparison to
competition law measures. This is why the Article 10/5 of the Law No. 406 requires operators to meet
reasonable, economically proportionate and technically feasible roaming requests in a clear and
precise way.
408
See Competition Board’s Decision dated 09/06/2003 and numbered 03-40/432-186 (Roaming
Decision), p. 27-33 and Competition Board’s Decision dated 06/11/2002 and numbered 02-68/821-
333 (TTA Decision), p. 49-53.
151
However, as the Law No. 406 imposes the (ex ante) duty to make a roaming
agreement under specified conditions (if the request in question is reasonable,
economically proportionate and technically feasible), the ex post imposition of such a
duty is to be deemed ‘supplementary’ in nature. In order to refrain from a duplication
of sanctions that are imposable by the two Authorities, a concrete and an efficient co-
ordination must be held between the authorities. Then, one of the authorities must
take action ahead in a case co-operating with the other thoroughly, namely take the
responsibility of investigating the EFD case as the forerunner authority.409 Given the
fact that, competition authorities are deprived of adequate information regarding the
relevant sector and the market players, sector-specific regulators must in general be
empowered to act with the necessary tools and remedies, including imposition of fine
in a case. However, in some cases, the situation might be in opposite, where
competition authorities would be more equipped with the relevant data and more
capable to solve the case. In such cases, competition authorities must take the
initiative first and coordinate with the sector-specific authorities, substantively.
With regard to solving ‘overlap’ problems and securing collaboration between the
authorities, there exist a ‘Coordination Protocol’ between Turkish Competition
Authority and Telecommunications Authority which was signed on September 23,
2002 by each party. In the Protocol, procedures and principles were determined for
ensuring cooperation and coordination in transactions conducted by both Authorities
in the telecommunications sector, such as investigations, mergers/acquisitions,
exemptions/negative clearances and secondary regulations.410 In the working of the
Protocol, whereas the main concern should have been to discuss and determine what
should be done and who should take action in a case, this does not seem to have been
conducted in Roaming case. While the Protocol would be expected to have been
applied effectively, duplication of procedures and sanctions shows a failure in terms
of co-ordination in Roaming case.
409
Unfortunately, such a roadmap has not been followed in the Roaming decision and overlapping
powers of each Authority have crashed with each other and the same amount of fines were imposed to
the respondents (Turkcell and Telsim) by both Competition Authority and Telecommunications
Authority.
410
http://www.rekabet.gov.tr/word/rapor2002/dort.doc
152
In order for such administrative failures not to be faced, a governmental body (a
public authority equipped with reconciliatory as well as administrative powers)411
would settle the co-ordination problems. In such a settlement, the so-called public
authority must conclude the conflict with an applicable and effective solution
incorporating the determination of executive powers of each authority, relevant tasks
in imposing fine and designating requirements for exchange of information, in each
case. Briefly, how to initiate and coordinate the investigation must be clarified in the
course of a case within the settlement procedure. In order to clarify the nature of the
case and the boundaries of (overlapping) powers of each authority, the so-called
authority would rather be invested with adequate, effective and objectively justified
powers so as to make ultimate decision on the issue regardless of political concerns.
Importantly saying, though having different legal basis, methods (ex ante-ex post)
and procedures; the ultimate objectives of the two types of remedies are the same,
actually. That is to say, both competition law remedies and sector-specific rules aim
at securing effective competition in a relevant market. While the said authorities
impose on undertakings ‘duty to grant access to third parties’ via different tools and
powers, the content and consequences of such duties in many occasions do not differ,
by nature.
For instance, in Roaming case, the duty (being either in ex ante or ex post nature) to
grant access to GSM network on part of the dominant undertaking(s) means the same
burden in actual terms. Hence, it is not acceptable that the powers of each authority
conferred by Turkish Competition Act and Telecommunications Legislation (Law
No. 406 and 2813) are different from each other in terms of scope, purpose and
implementation because of their respective legal basis. Rather, in order to
complement each other and to reach the most effective decision, the two authorities
411
In order for coordination problems especially those incorporating overlapping powers and
responsibilities to be prevented, a state unit such as a State Planning Department or a similar authority
drawing and maintaining public strategies regarding competition policies would be empowered to
make the ultimate decision in case of conflict. In addition, irrespective of any conflict, some
coordinative steps would be taken under the leading role of the so-called public authority, inter alia
concrete lines would be drawn between the legal powers, some legislative measures can be taken in
order to eliminate forum shopping.
153
must act in a co-operative manner to solve the competition breaches, especially in
EFD cases. The leading role must be left one of the authorities according to the
features of the case, by taking some criteria, i.e. collection of data, close relation to
the case, etc. into account and in case of a conflict regarding overlapping powers, a
high-level body, namely a re-conciliator must engage in the case so as to solve the
conflicting issue.
As mentioned above, Roaming case was the most debatable essential facilities case
among others in Turkish case-law. Other than some points which are common with
TTA case such as ‘overlap’ problems between the authorities, the main challenge
towards Roaming decision was its evaluation of the ‘essentiality’ on part of the GSM
networks. Here is worth noting another related fact that, the ÇEA decision which
concerned an abusive practice in the field of electricity (power) transmission reminds
a crucial point in EFD analysis. As stated above, investigating the ‘essential’
character of a facility constitutes the core of EFD analysis and necessitates
cautiousness; however while such a cautious analysis is confronted in ÇEA
decision, an over-zealous and seemingly unstable application of EFD does exist in
Roaming decision. For instance, it must be noted that whereas the indicating factor
would have been the examination of technically and economically feasible
alternatives, such an examination had not been conducted in a detailed manner, in
Roaming case. Accordingly, Roaming decision incorporated sceptical and vulnerable
points in the light of both EC and Turkish competition law principles.
As to ÇEA case, the situation was different due to a wide range of constraints, inter
alia lack of concession for power transmission, inability to generate power without
transmission, geographical restraints, etc. which made access to and interconnection
with the transmission infrastructure actually indispensable, on part of the requesting
parties. That is to say, whereas - in ÇEA case - without access to transmission
infrastructure any power supplier could not carry out its activities and the duplication
of such an infrastructure is indeed impossible, there does not appear a similar
indispensability and non-duplicability in the context of GSM networks.
154
Briefly saying, though having recognised EFD under the Turkish Competition Law,
Turkish Competition Board decisions established far-reaching results and many
challengeable points which are still on the agenda of the competition law
practitioners and commentators and need to be re-evaluated in respect of competition
law principles.
3.6 Implications of Applying EFD for Turkish Telecommunications Sector
In recent years, major structural changes and significant regulatory reforms,

including a considerable progress in opening the telecommunications services to
competition has taken place in Turkish telecommunications sector. Turkey has for
the last few years accelerated liberalisation process in its telecommunications
markets, through which one of the major driving factors was the influence of EU
candidateship. A brief history of the Turkish telecommunications sector would
constitute a beneficial basis for further explanations.
Until 1994, Turkey’s telecommunications networks and services were developed and
offered through a state-run monopoly (under the name of ‘PTT’). With the enactment
of the Law No. 4000 in June 1994, the first step in the way of liberalisation of the
sector was taken. In accordance with this Law, PTT was divested into two parts and
telecommunications services started to be carried out by a separate company, Turk
Telekom.412
At the same time, the mobile communications market was opened to competition
progressively, with two operators (Turkcell and Telsim) performing under revenue-
sharing agreements with Türk Telekom in 1994. Internet service providers also
started to appear under service contracts with Turk Telekom. Such developments led
to amendments of the Law No. 4000 to liberalise some parts of the
telecommunications services. Under the new regime, two GSM 900 operators were
granted 25-year licences in 1998.413
412
With the Law No.4000, it was made possible to privatise 49% of Turk Telekom, as well.
413
With the amendments to the Law No.4000 and after being awarded the two GSM 900 licenses,
Turkcell and Telsim began to perform their activities under a separate license, being released from the
revenue-sharing agreement with Turk Telekom.
155
In January 2000, a new Law (Law No. 4502) separated policy and regulatory
functions in the sector by establishing an independent regulatory body, the
Telecommunications Authority. Through this legislation, Turk Telekom was released
from the state control (under the situation of K T - a quasi public company) and was
rendered a different status as a ‘private company’.414 One of the major steps taken via
this Law is the envisaging of the date of full liberalisation. According to the Law, the
monopoly (of Turk Telekom) in fixed voice telephony and establishment and
operation of infrastructure would be expired by 31st December 2003.415
Legal monopoly in fixed telephony services and telecommunications infrastructure

was removed by 31st December 2003, and at that time liberalisation process started
up. Correspondingly, besides GSM operators and ISPs (Internet Service Providers) a
number of companies named ‘long distance carriers’ were granted licenses. In spite
of commenced liberalisation with recently licensed operators, it is possible to say that
de facto monopoly of Turk Telekom is being perpetuated for the time being. This is
why in many relevant fields of telecommunications is still continuing the lack of
alternative networks, which makes Türk Telekom’s infrastructure to a significant
degree, non-substitutable. Besides the lack of efficient alternatives to the current
access networks, Türk Telekom’s established consumer inertia as well as its
vertically integrated companies acting in different markets makes difficult to reap the
benefits of liberalisation, in general.
It is known that awarding a number of licenses regarding cable TV, infrastructure,

and fixed wireless access is under the agenda of Telecommunications Authority for
the near future. However, in order to remove the concerns regarding effective
competition in Turkish telecommunications sector, a number of steps are remaining
to be taken by Telecommunications Authority. In particular, during liberalisation
process, it must be ensured that effective entries by new operators be accomplished
414
However, all shares of Turk Telekom has remained to be owned by Treasure.
415
The Law No. 4502 also stipulates that the monopoly of Turk Telekom will not be last more when
more than 50% of Turk Telekom shares are privatised even before the end of 2003.
156
in telecommunications markets. Thereby, Telecommunications Authority’s main task
seems to be facilitating new entries under reasonable, transparent and fair conditions.
In fact, any perfect licensing is not adequate on its own, and it is always – even in
competitive markets – needed to take remedies for granting access to third parties on
reasonable terms and conditions. In this perspective, one can say that at the centre of
Turkish telecommunications sector lie the problems with regard to access and
interconnection.
Against this picture, not only Turkish Competition Authority but also
Telecommunications Authority must be ready for any kind of competition breaches,
potential bottleneck cases, etc. As implied above, the existing access problems will
be able to increase during the process of liberalisation, and such an increase would
raise the necessity of a more sophisticated competition law that is to be in harmony
with sector-specific regulation. Such a harmonised and co-operative approach
between the relevant authorities is indeed necessary for preventing duplication of
procedures, imposition of excessive burdens, etc.
In a liberalised environment, new companies, innovative services and strategic

alliances could restrain the incumbent operators from monopolistic behaviours.
However without necessary measures, liberalisation would not bring out the desired
outcomes and in bottleneck situations there would be competitive failures due to the
lack of an established access regime. In order to reap the benefits of the liberalisation
on a country-wide basis, an efficiently harmonised access regime must be set out and
accompanied with the relevant competition policy tools, inter alia application of
EFD drawn with clear-cut rules. From this point of view, EFD would represent the
competition law measures as one of the effective tools for opening ‘essential’
network elements to third parties, in the liberalisation process. During this process,
Turkey proved its determination in demonstrating that the dominant undertakings
would be required to share its assets with the third parties, where necessary. In
particular, Turkish Competition Authority reached important decisions as regards
application of EFD in field of telecommunications, most of which served as mile-
stones in realm of competition policies.
157
Harmonising its legal framework with the EU Directives on the one hand, and coping
with the access problems on the other hand, Turkey is undergoing a period that is
difficult but quite similar to the transitional period of EU on the way of full
liberalisation. In this situation, the issue of access and interconnection in
telecommunications sector would constitute a major test for the application of EFD
in Turkey as had been the same in the EU context. In order to overcome the
bottleneck problems surrounding access and interconnection scheme, EFD would
thus be seen as an important tool not only by Turkish Competition Authority but also
Telecommunications Authority.
Considering that within the transition period from monopoly to fully completed
liberalisation, sector-specific regulation has a potential to play a key role in ensuring
competitive markets, Telecommunications Authority, as the sector-specific regulator
might resort to EFD as an alternative tool, under its access regime. However, such a
resorting would probably be in the form of implicit regulations rather than explicit
formulations. That is to say, the philosophy of EFD would favourably be embedded
into the sector-specific regulations such in regulations related to access and
interconnection, unbundling the local loops, etc. where the ‘essential’ character must
be conferred as the foremost criterion. In addition, the harmful effects of any denying
access to an essential network facility must be cited, and in lack of any legitimate
justification network access must be mandated on reasonable and fair conditions.
Briefly, the criteria that apply to EFD must be handled as a basis for sector-specific
obligations in reasonable terms.416
Summarising, after the removal of legal monopoly in Turkish telecommunications

sector by the beginning of 2004, as experience of other countries have shown, the
market is expected to show dynamic growth in terms of new market entry,
416
A provision of the German Telecommunications Act can be demonstrated as an example for using
EFD in telecommunications regulation. Under Article 33 of the German Telecommunications Act, any
provider of telecommunications services considered to possess a dominant position is required to
grant its competitors on the relevant market non-discriminatory access to essential services it uses
internally or sells to the market “to the extent that they are essential”.
158
investment and development of services.417 Acceleration of new entries and services
are of potential to generate more competition breaches and problems related to
access and interconnection. Essential Facilities Doctrine is capable to cope with such
(bottleneck) situations, especially in the environment of converging markets and
services. Given the fact that liberalisation speeds up convergence and innovation,
EFD is convenient to be placed in a more central position in Turkish
telecommunications policies.
417
OECD, “Regulatory Reform in the Telecommunications Industry-2002”, p.34. Available at
http://www.oecd.org/dataoecd/40/13/1840797.pdf
159
CHAPTER 4
4. CONCLUSION
Essential Facilities Doctrine has been one of the note-worthy developments in EC

Competition Law in the 1990s, which is usually deemed a notion originated from the
‘refusal to deal’ cases. Whereas some commentators state that the Doctrine is vital
for liberalisation aims, some others underpin the perception that the Doctrine is only
a useful label, being a specific form of ‘refusal to deal’ cases. However in EC
Competition Law, it is widely accepted that the Doctrine is regarded as an
unspecified type of the application of Article 82 of the Treaty, aiming at removing
the danger that could arise out of refusal of dominant undertakings to grant essential
resources to third parties. Hence, in EC case-law EFD is consciously invoked in
order to neutralise the abusive control of dominant firms over facilities which are
called ‘essential’ or ‘indispensable’.
When the initial stages of the Doctrine-based case-law are taken into account, there
did not appear even an implicit formulation of EFD and the first essential facilities
cases are generally acknowledged as a refusal to deal case. In these cases, it became
clear that the refusal to supply an already existing customer/competitor a facility
which is an input for sale of retail products, amounts an abuse of dominant position.
For reaching such a conclusion, a number of pre-conditions including ‘lack of
objective justification’ and ‘elimination of competition on the part of the competing
firm’ were to be met. In these cases, there was not a reliance on the notion
‘essentiality’, which is the core of EFD.
Following this line of case-law, the Magill case came into agenda of Community
Institutions, which concerned a dispute involving a refusal to supply a new customer
and prevention of emergence of a new product in the relevant market. In that case,
the ECJ rendered a decision regarding anti-competitiveness of a ‘refusal to grant a
license’, which attracted so much attention. The ECJ found three exceptional
160
circumstances that would support a finding that the refusal to license an intellectual
property right is abusive. Briefly, the rationale of the judgment was based on the
notion of reservation of a secondary market by denying access to a facility which is
an indispensable raw material for production of a new service.
The notion of ‘indispensability’ called upon in Telemarketing and Magill judgments

took its appropriate form to EFD in B&I Line and Sea Containers cases. In this
second line case-law, ‘essential facility’ is explicitly cited by the Commission,
whereas such an articulation has not been put forward by the European Court of
Justice. The referred Commission decisions emphasised the superior position of
harbour owners and heightened their obligations for their dual role (as acting both in
the downstream and upstream markets).
EFD was used in a different guise in the context of a number of decisions which
concerned railways. The Commission characterised rail services ‘essential’ in HOV
SVZ/MCN case, and referred them as being ‘necessary’ for rail transport operators in
European Night Services case. Finally in the Eurotunnel case, the Commission found
that the Channel Tunnel was an essential facility, and each half of the Tunnel was to
be used separately by the parties to the case. The remedies of the Commission in the
lastly cited two cases resemble those of a sector-specific authority, revealing the
Commission’s ambition for applying EFD in a wide spectrum. Notably, such an
ambition is manifested explicitly in the Commission decisions during the
liberalisation period.
In the Atlas judgment was established an explicit link between legal monopolies and
essential facilities. In spite of the existence of ONP (Harmonisation) Directives
which include sector-specific rules, it seems that the Commission was prone to use
EFD in order to impose specific obligations, i.e. non-discrimination, access
requirement, where necessary. The conditions imposed by the Commission under
EFD cases exhibited a more structural nature, when the relevant markets consisted of
network industries i.e. telecommunications, railways alike the situation in European
Night Services case.
161
The attitude of the Commission in such cases gave the way for using EFD as a
competitive measure for mandating third party access. One can see this rationale
behind the issue of the Commission’s Access Notice. After the enactment of Full
Competition Directive (Directive 96/19/EC), Access Notice was issued as an
important legal instrument for triggering access obligations, in order for enabling the
liberalisation to be more effective. The main concern of the Commission was
prevention of emergence of de facto monopolies after the removal of de jure
antecedents. In fact, during the transition period from monopoly to full liberalisation
of European telecommunications sector, the Commission gave a critical role to EFD
so as to apply the Doctrine in a wide-spread manner. In this context, the Access
Notice might be regarded as an explicit formulation of post-liberalisation policies,
which included an over-zealous application of EFD. Through the application of EFD
in an over-zealous manner, the balance to be drawn between the rights of the firms
seeking access and obligations of incumbent firms was shifted in favour of the
parties seeking access, namely towards service-based competition.
However, in the Oscar Bronner judgment, ECJ set a higher threshold to be met in an
EFD analysis, and facility-based competition is appropriately given way by this
decision. The ECJ provided a three-part test under which refusal to grant access
would constitute an abuse in the event that (i) the refusal to provide access to the
facility would be likely to eliminate all competition on the part of the undertaking
requesting access; (ii) access to the facility should be indispensable for the
competitor to carry on its business in that there is no actual or potential substitute to
the facility in existence; (iii) the refusal is incapable of being objectively justified.
In the Oscar Bronner judgment, the ECJ filled up the term ‘indispensability’ with the
‘non-substitutability’ test, going beyond the preceding decisions. The Court went
further and held that for an access to be capable for being regarded as indispensable,
it would be necessary at the very least to establish that it is not economically viable
to create for an ‘objective competitor’ comparable in size to the holder of the alleged
essential facility to replicate or duplicate the facility in question.
162
After the Oscar Bronner judgment, although one can regard EFD as having a stand-
alone basis under EC Competition Law, it became clear that demonstrating that a
product is indispensable or essential is not easy within the meaning of the Doctrine.
More precisely, the Oscar Bronner judgment proved that invoking EFD in a
bottleneck situation is not always acceptable; that is to say, the notion of ‘essential
facilities’ must be assessed within a stronger, analytical and economics-oriented
framework. Under such a detailed analysis, the short-term and long-term benefits of
opening an essential facility to third party access must be thoroughly investigated. In
this context, if access to production, distribution or sale of a facility is permitted too
easily, there would be no incentive for a competitor to develop competing facilities
and innovations. In aftermath of such a decision made after a superficial analysis,
competition is able to be increased in the short term, whilst it would probably be
reduced in the long term. Therefore, access to essential facilities should be mandated
upon a detailed analysis taking the short & long term conflicts and the balance
between service & facility-based competition into account.
Without a detailed analysis, access to allegedly essential facilities could be ordered

under some findings, depending solely on theoretical and comparative analysis which
is not reliable from a whole perspective. Such findings could be counterproductive
against sector-specific regulation where detailed conditions are developed for
mandating new entrants for access to incumbent’s facilities. The strict conditions
specified by sector-specific regulation may contravene with the conditions
determined through application of competition law by referring to EFD. Therefore,
competition authorities and sector-specific authorities must always be in coordination
with regard to anti-competitive practices which have elements that are suitable to be
assessed under EFD. In this coordination, detailed intervention of sector-specific
authorities should be mitigated with the manoeuvrability of competition authorities.
With respect to the co-operation between competition authorities and (sector-

specific) regulatory authorities, not only relevant data, but also powers of the
authorities must be exchanged, when necessary. In case of a bottleneck situation; one
163
of the referred authorities must take initiative as the forerunner body according to the
distinct characteristics of the each case. In this regard, the ‘Coordination Protocol’
signed between Turkish Competition Authority and Telecommunications Authority
could be given as an unsuccessful example in terms of distribution of tasks,
prevention of duplication of sanctions, etc. In the so-called Protocol, there exist only
provisions regarding exchange of relevant information, but not detailed guidelines
for implementing overlapping measures.
Ideally, each authority must take into account the ‘essential’ character of the resource
concerned, feasibility of providing facility, etc. when making decision as regards
liability for third party access. In an essential facilities case, focus should be directed
on special characteristics of the case, according to which each case would be handled
with respective remedies. If the case entails more telecom-specific analysis, sector-
specific regulator must lead the case, in otherwise situation(s) competition authorities
must take the initiative. In case of conflict, a high-level administration (a public
authority invested with extensive powers) must make the ultimate decision regarding
the handling of the case.
In context of handling essential facilities cases, sector-specific authorities would

seriously conduct EFD analysis in their decisions as well as take necessary steps to
adopt EFD in their respective regulations, at least in an implicit manner. These
propositions might be called onerous, however they must be taken into account in
order to remove any inconsistency between the procedures and the decisions of the
competent authorities.
Within the evolution of EFD from the Commercial Solvents judgment to Oscar
Bronner judgment, it can be concluded that EFD has potential to bring some added
value by providing an analytical framework to extend the range of Article 82 EC
beyond its traditional boundaries. The expansive as well as the adjustable nature of
EFD is able to be seen along the transition from the classical (refusal to deal) cases to
bottleneck (essential facilities) cases. Whereas, in the former cases could be seen two
easily identifiable markets, in the latter cases market definition becomes more
164
difficult due to the geographic and regulatory constraints. That is to say, in essential
facilities cases does not exist trade of goods or services, rather does opening of
network facilities to third parties.
Considering multiple types of networks and convergence between services, one can
not distinguish markets from each other under the classical formulation of
competition law. Thus, a market for access to a facility could not be specified clearly
in each case, particularly in cases regarding networked industries. Some instances,
which are also subject to regulatory oversight, i.e. unbundled access to the local loop,
incorporate the difficulties of relying on classical tests of competition law. In such
difficult instances, EFD has the capability to resolve (access) problems and cope with
the bottleneck situations. One can also observe in such cases that dominance
becomes less meaningful, and is likely replaced by the notion of essentiality as the
key competitive concern.
The application of EFD within the telecommunications sectors must be examined in

the general context of the rapid evolution of the markets. Potentially anti-competitive
behaviours generated by these rapid changes pose new challenges for competition
law and policies. Such challenges could be eliminated neither through the sector-
specific regulation alone nor via application of the classical methods of competition
law. In this framework, EFD would possess a central importance and have the power
to respond to the needs of the converging telecommunications markets.
In time, the more situations of convergence among rapidly changing markets are
confronted, the more apparent the constraints of the sector-specific regulation
become. Considering that the convergence between changing markets entails the
convergence of legal remedies, EFD would be the leading rule in the converged
marketplace. Accordingly, future implications of the changing markets would
increase the importance of EFD against the strict application of sector-specific rules.
While under sector-specific regulation, local loop unbundling is seen as the single
way allowing new entrants to compete with incumbent operators at local access
165
markets, and is deemed an essential facility for new entrants either implicitly or
explicitly, the situation can be changed from the point of view of EFD. Given the
alternative access technologies recently emerged at local access level, (i.e. digital
satellites, cable, and wireless local loop) EFD which offers a wider perspective in
terms of global changes and ‘convergence’ phenomenon could deny the ‘essential’
character of the incumbent’s local loop. The superiority of EFD as a tool of
competition law is excellently spelled out within the words of H. Ungerer, which is
quoted below:
Without going into further detail, suffice it to say that competition

law - in the form of a developed essential facilities concept - can
adjust, in a flexible manner, to situations of convergence, by
adjusting the market definitions used and without changing either
the regulatory framework or its basic principles.418
In essence, in EU telecommunications sector, the need for EFD is becoming

increasingly important due to the increased convergence of industries such as fixed
and mobile telecom, CATV, Internet, satellite etc. Since liberalisation accelerates
convergence, EFD also becomes a necessary measure in telecommunications markets
which are undergoing liberalisation process, such in candidate countries for EU
membership.
Being one of the candidate countries, Turkey rapidly digitalised telecommunications

networks and started to liberalise the services that are carried out through the so-
called networks. In parallel with the liberalisation process, a higher degree of
competition in provision of telecommunications services prevailed in the sector.
However, an efficient access regime that is to be harmonised with the competition
law rules is clearly became needed in terms of ensuring effective competition and
prevention of de facto monopolies. Confirming this need, Turkish Competition
Authority resorted to EFD in a number of its decisions to date. One of them
418
H. Ungerer, “The arrival of competition in European telecommunications”, p. 19, 3rd European
Forum on the Law of Telecommunications, Information Technologies and Multimedia: Towards a
Common Framework, Luxembourg, June 19, 1998.
166
concerned charging of an incumbent operator (Türk Telekom) its corporate users
below the competitive rates in field of providing broadband internet access services.
The other was related to an abuse of joint dominance through refusal to grant access
to the GSM network(s) that were deemed ‘essential facility’ to carry out GSM
telecommunications services.
Particularly in the latter case, there appear a number of points detrimental to

diligence of the Doctrine. Among them, interpretation of the criteria
‘indispensability’ and ‘non-substitutability’ in the context of GSM networks
constitutes the core point which seems to have depended on an unstable basis. This is
why under the Community Rules, more strict and clear-cut rules had been developed
as regards application of EFD. In particular, when the criteria of the Oscar Bronner
judgment are taken into account, it is hard to mention about a coherent application of
EFD. Rather, an over-zealous application of EFD is remarkable in Roaming case.
When the referred decisions are considered with the ongoing period of liberalisation,
some positive evaluations regarding the attitude of Turkish Competition Authority
could come into mind. However, since the GSM markets are entirely liberalised and
the new entrants have other possibilities than access to the existing networks which
are called essential facility, the Roaming decision is quite debatable. Against another
decision of Turkish Competition Authority, ÇEA , it is indeed controversial that the
claimant could not carry out its services in absence of roaming. This is why the
facility in the latter (ÇEA ) decision was the transmission infrastructure without
access to which, any power supplier could not generate power by itself, whereas the
GSM operators could perform their activities - even under less advantageous
conditions - without access to the existing networks in the former (Roaming)
decision.
Against this picture, it is advisable that a more coherent but less ambiguous
application of EFD is favourably to be adopted under the Turkish Competition Law.
Not only Turkish Competition Authority but also Telecommunications Authority
must recall attention towards a cautious application of EFD. In EFD cases, both
167
authorities must act in a collaboration and make their decisions in aftermath of a
more detailed and economics-oriented analysis.
Considering that actual and potential access problems could constitute insuperable
barriers to new entries during the liberalisation period, EFD must be used as an
alternative tool in eliminating such problems, in an effective manner. Not in case-by-
case analysis but also in implementation of the relevant legislation (i.e. sector-
specific rules), EFD would simply be deemed applicable. This is why EFD might
highly contribute to opening non-duplicable networks to competition in field of
telecommunications, when applied under clear-cut rules.
168
BIBLIOGRAPHY
ARDIYOK, . (2002), “Yerel ebekeyi Eri ime Açma Yükümlülü ü”, Rekabet
Kurumu, Uzmanlık Tezi
AREEDA, P. (1990), “Essential Facilities: An Epithet in Need of Limiting

Principles”, Antitrust Law Journal, Vol. 58, No: 3, p. 841-853
ARNOLD PORTER, (2002), “Introduction to the New EU Regulatory Framework

for Electronic Communications”, European Telecommunications Practice Group
http://www.arnoldporter.com/pubs/files/newsframewrok.pdf
A ÇIO LU, G. Ö. (2000), Avrupa Birli i ve Türk Rekabet Hukuku Çerçevesinde

Hakim Durumun Kötüye Kullanılması, Rekabet Kurumu Yayınları.
BAUMOL, W. J. and SIDAK, J. G. (1994), “Towards Competition in Local

Telephony”, The MIT Press and The American Enterprise Institute for Public Policy
Research.
BERGMAN, M. A. (2000), “The Bronner Case-A Turning Point for The Essential
Facilities Doctrine ?”, European Competition Law Review, No: 2, p. 59-63.
BIJL, Paul de and PEITZ, M. (2002), “Regulation and Entry into

Telecommunications Markets”, Cambridge University Press.
BIJL, Paul de (2000), “Competition and Regulation in Telecommunications

Markets”, CPB Netherlands Bureau for Economic Policy Analysis, The Hague.
BLACK, S. K. (2002), “Telecommunications Law in the Internet Age”, The Morgan

Kaufmann Publishers.
BLUMENTHAL, W., (2000) “Three Vexing Issues Under the Essential Facilities
Doctrine: ATM Networks as Illustration”, Antitrust L. J. No: 58, p. 855.
BOURREAU M., DOGAN P., (2001), “Regulation and Innovation in the

Telecommunications Industry”, Telecommunications Policy, Vol. 25, Issue. 3, p.
167-184.
BOURREAU M., DOGAN P., (2001), “Service-based vs. Facility-based

Competition in Local Access Networks”, Telecommunications Policy, Vol. 25, Issue.
3.
BRADLEY, Stephen P. and HAUSMAN, Jerry A. (1989), “Future Competition in

Telecommunications”, Harvard Business School Press.
169
BROCK, Gerald W. (1994), “Telecommunications Policy for the Information Age
From Monopoly to Competition”, Harvard University Press.
BRANNAN, J. (1999) “Open Broadband: An Essential Facility Doctrine Analysis”,

www.ukans.edu/-cybermom/CLJ/Broadband.htm.
BUIGUES, P. A., (2001) “European Policy on Local Loop Unbundling: Competition

Law Landscape and Implementation Issues”, Communications & Strategies, No. 42,
p. 51-66.
CAPOBIANCO, A. (2001), “The essential facility doctrine: similarities and

differences between the American and the European approach” European Law
Review, 26, no. 6, p. 548-564.
CAVE, M. and MASON, R. (2000), “The Economics of the Internet: Infrastructure

and Regulation”, Oxford Review of Economic Policy, Vol. 17, No. 2.
CAVE M. and PROSPERETTI L. (2001), “European Telecommunications

Infrastructures”, Oxford University Press and the Oxford Review of Economic
Policy, Vol. 17, No. 3, p. 416-31.
CAVE, M., MAJUMDAR, S. K., VOGELSANG, I. (2002), “Handbook of

Telecommunications Economics”, Elsevier Science B. V., 2002.
CAVE, M. (22-23.05.2003), “Infrastructure Liberalisation – Speed Up or Step Back:

European Telecommunications Liberalisation – Objectives and Assumptions”, TU
Delft 6 Annual International Conference, Warwick Business School.
COOK, P. 2002, “Competition Policy, Market Power and Collusion in Developing

Countries”, MSc Thesis, University of Manchester (Centre on Regulation and
Competition)
COTTER, T. F. (1999), “Intellectual property and the essential facilities doctrine”,

The Antitrust Bulletine, Vol. XLIV, No: 1, p. 233.
COWEN, T. (1995), “The Essential Facilities Doctrine in EC Competition Law:

Towards A Matrix Infrastructure", International Antitrust Law & Policy (Editor: B.
Hawk), Fordham University Law Institute, Fordham University, School of Law. p.
521-547.
COWIE, C. and MARSDEN, C. T. (1998), “Convergence, Competition and

Regulation”, International Journal of Communications Law Policy,
http://www.ijclp.org
CROCIONI, P. (2001), “Should Telecoms Liberalisation stop at call termination”,

Telecommunications Policy, No: 25, 2001, p. 39-58.
170
DOGAN, P. and GILLE, L. (22-24.09.2002), “Interconnection LRIC Model - A
World Bank Seminar in cooperation with TRA”, Ankara, Turkey.
DOHERTY, B. (2001), “Just What are essential facilities?”, Common Market Law
Review, 38, No: 2, p. 397-436.
DOLMANS, M. (1999), “Essential Facilities After Oscar Bronner Case”, IBC’s

Fourth Annual Conference on Telecommunication and EC Competition Law,
Brussels.
DOYLE, C. (2000), “Local Loop Unbundling and Regulatory Risk”, Journal of

Network Industries, Vol. 1, p. 33-54 http://www.cdoyle.com/papers/llurisk.pdf
ECONOMIDES, N. and WHITE, L. J. (1995), “Access and Interconnection Pricing:

How Efficient is the Efficient Pricing Component Rule?”, Antitrust Bulletin, Vol.
XL, p. 557-579.
EC Treaty (Consolidated Version of the Treaty Establishing the European

Community)
http://www.europa.eu.int/eur-lex/treaties/dat/ec_cons_treaty_en.pdf
European Commission, Recommendation on Relevant Product and Service Markets

(C(2003)497, published on February 11th 2003).
European Council, Directive 2002/19/EC of the European Parliament and of the

Council of 12 July 2002 (Access Directive).
European Council, Directive 2002/21/EC of the European Parliament and of the

Council of 7 March 2002 (Framework Directive).
European Commission, Commission guidelines on market analysis and the

assessment of significant market power under the Community regulatory framework
for electronic communications networks and services (2002/C 165/03).
European Council, Regulation (EC) No 2887/2000 of the European Parliament and

of the Council of 18 December 2000 on unbundled access to the local loop,
published on May12th 2000, O.J. L 336/4.
European Commission, Proposal for a Regulation of the European Parliament and of

the Council on unbundled access to the local loop, adopted 12 July 2000
Com(2000)394.
http://www.europa.eu.int/information_society/topics/telecoms/regulatory/new_rf/ind
ex_en.htm#ull
“Commission Recommendation on Unbundled Access to the Local Loop”,

C(2000)1059, 26 April 2000.
http://www.europa.eu.int/information_society/topics/telecoms/regulatory/maindocs/c
omgreen/index_en.htm
171
European Commission, Towards a new Framework for Electronic Communications
Infrastructure and associated services”, COM(1999)539.
European Commission, Information Society Directorate general, Brussels, 17

September 1999, XIII/A1, Explanatory Note.
European Commission, Access Notice on Application of Competition Rules to

Access Agreements in the Telecommunications Sector (1998) O.J. C265/2 (1998)
European Council, Directive 98/10/EC of European Parliament and of the Council of

26 February 1998 on the application of open network provision to voice telephony
and universal service in a competitive environment, OJ L 101.
European Commission, Recommendation of 8 April 1998 on interconnection in a

liberalised telecommunications market (98/322/EC) and Commission
Recommendation of 8 January 1998 on interconnection in a liberalised
telecommunications market (98/195/EC).
European Council, Directive 97/33/EC of the Parliament and the Council of 30 June
1997 on interconnection in telecommunications with regard to ensuring universal
service ad interoperability through application of the principles of Open Network
Provision (ONP), OJ No L 199/32.
European Commission, Green Paper on the Convergence of the

Telecommunications, Media, and Information Technology Sectors, and the
Implications for Regulation Towards an Information Society, COM(97) 623 (3
December 1997)
http://europa.eu.int/ISPO/infosoc/telecompolicy/en/comm-en.htm
European Commission, Directive 96/19/EC of 13 March 1996 amending Directive

90/388/EEC with regard to the implementation of full competition in
telecommunications markets, OJ L 74.
European Council, Directive 95/62/EC of European Parliament and of the Council of

13 December 1995 on the application of open network provision to voice telephony,
OJ L 131/6.
European Council, Directive 92/44/EEC of 5 June 1992 on the application of open

network provision to leased lines, OJ L 165/27.
European Commission, Guideline on the application of EEC competition rules in the

telecommunications sector (91/C 233/02),02), OJ C 233, 6.9.1991.
European Council, Directive 90/387/EEC of 27 June 1990 on the establishment of

the internal market for telecommunications services through the implementation of
open network provision, OJ L 192/1.
172
European Commission, Directive 90/388/EEC of 28 June 1990 on competition in the
markets for telecommunications services, OJ 1990 L 192/10.
European Commission, Directive 88/301/EEC of 16 May 1988 on competition in the

markets in telecommunications terminal equipment, OJ 1988 L 131/73.
European Council, Resolution of 30 June 1988 on the development of the common

market for telecommunications services and equipment up to 1992 [1988] OJ C
257/1.
European Commission, 1987 Green Paper (Towards a Dynamic European Economy:

Green Paper on the development of a Common Market for Telecommunications
Services and Equipment), COM(87)290,
ERGAS H., (1998), “Access and Interconnection in Network Industries”

http://www.necg.com.au/papub/papers-ergas-access.pdf
FAULHABER, G. R. (2002), “Access Access1 + Access2”

http://rider.wharton.upenn.edu/~faulhabe/Access.pdf
FURSE, M. (1995), “The Essential Facilities Doctrine in Community Law”,

E.C.L.R. Vol. 16, No: 8, p. 469-473.
GLASL, D. (1994), “Essential Facilities Doctrine in EC Antitrust Law: A

Contribution to the Current Debate”, European Competition Law Review, No: 6, p.
306-314.
GABELMAN A. (2001), “Regulating European Telecommunications Markets:

Unbundled Access to the Local Loop Outside Urban Areas”, Telecommunications
Policy, Vol.25, Issue. 10-11, p. 729-741.
GERADIN, Damien and SIDAK, J. Gregory, (22 January 2004), Seminar on

“European and American Approaches to Antitrust Remedies and the Institutional
Design of Regulation in Telecommunications”, Maastricht, The European Institute of
Public Administration (EIPA).
GILLETT, S. E. and VOGELSANG, I. (1999), “Competition, Regulation and

Convergence: Current Trends in Telecommunications Policy Research”.
HANCHER, L. (1999), “Case-law: Court of Justice (Oscar Bronner v. Mediaprint)”,

Common Market Law Review., p. 1289-1307.
HARZ, M. H. (1997), “Dominance and Duty in The European Union: A Look

Through Microsoft Windows at The Essential Facilities Doctrine”,
www.law.emory.edu
173
HAUSMAN, J. A. and SIDAK J. G. (1999), “A Consumer-Welfare Approach to the
Mandatory Unbundling of Telecommunications Networks”, The Yale Law Journal,
Vol. 109: 417, p. 417-505.
HELBERGER, N., “Access to Technical Bottleneck facilities: The New European

Approach”, Institute for Information Law, Faculty of Law, University of Amsterdam,
IViR Publications, http://www.ivir.nl/publications/helberger
HIRSH, M., RICHEIMER, G. A., “The essential Facilities Doctrine: Keeping the
Word ‘Epithet’ from Becoming One”.
http://www.rdblaw.com/News/PressReleases/Essentialfacilities.pdf
HITCHING, P. (1998), “Access to International Telecommunication Facilities”,

European Competition Law Review, No.2, p.85-98.
HOLM, J. (2000), “Regulating Network Access Prices under Uncertainty and

Increasing Competition: The Case of Telecommunications and Local Loop
Unbundling in the EU”, University of Copenghagen, Institute of Economics (MSc
Thesis).
HOLZNAGEL, B. (Winter 1998/99), “New Challenges: Convergence of Markets,

Divergence of The Laws?: Questions Regarding The Future Communications
Regulation”, International Journal of Communications Law and Policy, Issue 2.
IBC Global Conferences, (25 - 26th June 2002), Conference on “EU

Telecommunications Regulations & Law”, Brussels.
INTVEN, H., OLIVER J., SEPULVEDA E., (2000), “Telecommunications

Regulation Handbook”, Published by McCarthy Tetraultz. http://www.infodev.org/-
projects/314regulationhandbook/
ITU (International Telecommunications Union), (2001), “Trends in

Telecommunication Reform 2002-2001: Interconnection Regulation”, 3rd Edition.
ITU (International Telecommunications Union), (2002), “Trends in

Telecommunications Reform: Effective Regulation”.
ITU (International Telecommunications Union), (20 - 22 November 2002),

(International Telecommunications Union), Competition Policy in
Telecommunications, Document: CPT/04, Geneva.
ITU (International Telecommunications Union), (2003), “TREG Interconnection

self-training modules: Introducing Interconnection”.
http://www.itu.int/ITU-D/treg/selftraining/module1.asp
KV, (2002), Avrupa Birli i’nin Telekomünikasyon ve Bili im Teknolojileri

Politikası ve Türkiye Uyumu, stanbul.
174
JAUK, W. (Winter 1999/2000), “The Application of EC Competition Rules to
Telecommunications – Selected Aspects: the Case of Interconnection”, International
Journal of Communications Law and Policy, Issue 4.
KALLIALA J. (2000), Market Definition Under the EC Competition Law in the

Field of Voice Telephony, PILC Student Paper, Brussels.
KAUPER, T. E. (1989), “Whether Article 86? Observations on Excessive Prices and

Refusals to Deal”, B. Hawk (ed.), Fordham Corporate Law Institute International
Antitrust Law and Policy, Kluwer Law International, The Hague, The Netherlands.
p. 651-686.
KERF, M. and GERADIN, D. (1999), “Controlling Market Power in

Telecommunications: Antitrust vs. Sector-Specific Regulation”, Berkeley Tehnology
Law Journal, Vol. 14.
http://www.law.berkeley.edu/journals/btlj/articles/article_top.htm
KEZSBOM, A. and GOLDMAN, A.V., (1996), “No Shortcut to Antitrust Analysis:

The Twisted Journey of the Essential Facilities Doctrine”, Columbia Business Law
Review. Vol:1 No:1, s.602 http://www.ffhsj.com/firmpage/cmemos/0112041.html
KOENIG C., BARTOSCH A. and BRAUN J. (2002), “EC Competition and

Telecommunications Law”, International Competition Law Series, Vol. 6, Kluwer
Law International.
KOOPMAN, G., “Competition Policies and Telecommunications Regimes”

http://www.hwwa.de/Projekte/Forsch_schwerpunkte/FS/Hande/Publikationen/Koop
man%20edit.pdf
KORAH, V. (1998), “The Ladbroke Saga”, European Competition Law Review,

No:3, p.169-176.
LAFONT, J. J. and TIROLE, J. (1999), Competition in Telecommunications, The

MIT Press, London.
LANG, J. T. (1979), “The Monopolization and the Definition of Abuse of a

Dominant Position under Article 86 EEC Treaty”, Common Market Law Review,
Vol:16 p. 345-364.
LANG, J. T. (1994), “Defining Legitimate Competition: Companies’ Duties to

Supply Competitors and Access to Essential Facilities”, ”, B. Hawk (ed.), Fordham
Corporate Law Institute International Antitrust Law and Policy, Kluwer Law
International, The Hague, The Netherlands. p. 437-524.
LANG, J. T. (17.10.1996), “Conference on European Community Antitrust Law-

Innovation Markets and High Technology Industries”, Fordham Corporate Law
Institute, New York.
175
http://europa.eu.int/comm/competition/speeches/text/-sp1996_054_en.html
LANG, J. T. (September 2000), “The Principle of Essential Facilities in European

Community Competition Law – The Position since Bronner -” Notes for a lecture,
Copenhagen.
LANGLOIS, R. N. (December 1998), “Technological Standards, Innovations, and

Essential Facilities: Towards a Schumpeterian Post-Chicago Approach”, paper for
the George Mason University conference on Dynamic Competition and Antitrust,
Washington DC.
LAROUCHE, P. (2000), “Competition Law and Regulation in European

Telecommunications”, Hart Publishing.
LARRIERA, D. (2002), University of Strathclyde, “Pricing Telecommunications

Services in the UK”, MSc Thesis.
LARSON, A.C. and MUDD, D. R. (1999), “The Telecommunications Act of 1996

and Competition Policy: An Economic Hindsight”, Virginia Journal of Law and
Technology, Spring 1999, 4Va. J.L. & Tech. 1. http://vjolt.student.virginia.edu
LIE, E. (2002), “Competition Policy in Telecommunications”, Background paper,

ITU (International Telecommunications Union), Strategy and Policy Unit.
http://www.itu.int/competition
LIXIA, Li. (December 2002), “Local Loop Unbundling: International Experiences

and Implications for China”, MSc Thesis, University of Strathclyde.
LIPSKY, A. B., Jr. and SIDAK J. Gregory, (May 1999), “Essential Facilities”,
Stanford Law Review, Vol. 51, No: 1187.
LONG, D. C., LIEDEKERKE D. V. and RYAN, M. (1995), “Competition Aspects

of Interconnection Agreements in The Telecommunications Sector” (Final Report to
the European Commission-DGIV), Coudert Brothers.
LONGSTAFF, P. H. 2000, “Networked Industries: Patterns in Development,

Operation, and Regulation”, Program on Information Resources Policy, Center for
Information Policy Research, Harvard University, http://www.pirp.harvard.edu
MANNER, J. A. (2002) “Global Telecommunications Market Access”, Artech

House.
MELODY, W. H. (2001), “Telecom Reform: Principles, Policies and Regulatory

Practices”, http://www.itu.int/industry overview/
MUELLER, M. L., Jr. (1997), “Universal Service: Competition, Interconnection, and

Monopoly in the Making of the American Telephone System”, The MIT and AEI
Press.
176
NAFTEL, M. (1999), “Does The European Commission’s Telecommunication
Access Notice Send the Correct Economic Signals to the Market?”, Phoenix Center
Policy Paper No: 5.
NICOLINAKOS, N. (1999), “Access Agreement in The Telecommunication Sector-

Refusal to Supply and The Essential Facilities Doctrine under EC Competition Law”,
European Competition Law Review, No: 8, p. 399-411.
NICOLINAKOS, N. (2000), “The New Legal Framework for Digital Gateways-The

Complementary Nature of Competition Law and Sector-Specific Regulation”,
European Competition Law Review, No: 9, p. 408-414.
NICOLINAKOS, N. (2001), “Promoting Competition in Local Access Network:

Local Loop Unbundling”, European Competition Law Review, Issue: 7.
NIHOUL P. and RODFORD P. (2004), EU Electronic Communications Law,

Oxford University Press.
NIHOUL P., (Winter 1998-99), “Convergence in European Telecommunications: A

case study on the relationship between regulation and competition (law)”
International Journal of Communications Law and Policy, Issue. 2, p. 24-28.
http://www.digital-law.net/IJCLP/index.html
OECD, “Access Pricing in Telecommunications”, June 2004.
OECD (1996), “Essential Facilities Concept”, Series Roundtables on Competition

Policy No: 5, Vol. 4, No: 61.
OECD, “Regulatory Reform in the Telecommunications Industry-2002”,

http://www.oecd.org/dataoecd/40/13/1840797.pdf
OVERD, A. and BISHOP B. (1998), “Essential Facilities: The Rising Tide”,

European Competition Law Review. No: 4, p.183-185.
PITOFSKY, R., PATTERSON, D., and HOOKS, J. (2002), “The Essential Facilities
Doctrine Under U.S. Antitrust Law”, Antitrust Law Journal, Vol. 70, p. 45.
PITOFSKY, R. “The Essential Facilities Doctrine Under United States Antitrust

Law”, p. 23.
http://www.ftc.gov/os/comments/intelpropertycomment/pitofskyrobert.pdf
PIENPENBROCK H.-J. and SCHUSTER, Dr. F. 2002, “German

Telecommunications Law and the New European Regulatory Framework”.
ROSENTHAL M. (Winter 2002/2003), Brussels, “Open Access from the EU

Perspective”, International Journal of Communications Law and Policy, Issue 7.
177
SAUTER, W., “EU Regulation for the Convergence of Media, Telecommunications
and Information Technology: Arguments for a Constitutional Approach”,
ZENTRUM Für Europäische Rechtspolitik, ZERP-Diskussionspapier, 1/98.
SANFILIPPO, L. (1995), “Abuse of Freedom of Conduct: Neighbouring Market and

Application of Article 86”, European Business Law Review, Vol. 6, No: 3, p.71-75.
SAYGILI, . M. (2003), “Competition in Telecommunications Sector with Special

Concern to Internet Access Market”, Marmara Üniversitesi Avrupa Toplulu u
Enstitüsü, AB Hukuku A. B. D., Yüksek Lisans Tezi.
SHEPPERD, W.G. (1997) “Dim Prospect: Effective Competition in

Telecommunications, Railroads and Electricity”, The Antitrust Bulletin, Vol. XLII,
No: 1 p. 151-175.
SOMA J. T., FORKNER, D. A. and JUMPS, B. P., (1998), “The Essential Facilities
Doctrine in the Deregulated Telecommunication Industry” 13, Berkeley Tech. L.J.
p.565-590.
STOTHERS, C. (2001), “Refusal to Supply as Abuse of a Dominant position:

Essential Facilities in the European Union”, European Competition Law Review, 22,
no. 7.
SUBIOTTO, R. (1992), “The Right to Deal with Whom One Pleases under EEC
Competition Law: A Small Contribution to a Necessary Debate”, European
Competition Law Review, No: 6 p. 234-244.
TEKDEM R, Y. (2003), “AT Rekabet Hukuku’nda Anla ma Yapmayı Reddetme

Sorunu ve Zorunlu Unsur Doktrini”, Rekabet Kurumu Uzmanlık Tezleri Serisi No:2.
UNGERER, H. (2001), “Use of EC Competition Rules in the Liberalisation of the

European Union’s Telecommunications Sector: Assessment of Past Experience and
Some Conclusions”, Competition Policy Newsletter, No.2.
UNGERER, H. (July 2000), “Access Issues under EU Regulation and Anti-Trust

Law - The Case of the Telecommunications and Internet Markets”, Research Paper,
WCFIA Fellows Program 1999/2000, Harvard University (Weatherhead Center for
International Affairs)
UNGERER, H. (19.06.1998), “The Arrival of Competition in European

telecommunications”, 3rd European Forum on the Law of Telecommunications,
Information Technologies and Multimedia: Towards a Common Framework,
Luxembourg.
UNGERER, H., (1998), Ensuring Efficient Access to Bottleneck Facilities: The Case
of Telecommunication in the European Union.
http://europa.eu.int/comm/competition/speeches/text/sp1998_056_en.pdf
178
UNGERER, H. (10.06.1997), “Competition in Telecommunications - the Regulators’
Challenge”, Asia Telecom 97 Forum, Singapore.
VOGELSANG, B. and MITCHELL, M. (1997), “Telecommunications Competition:

The Last Ten Miles”, The MIT Press and the AEI Press, Washington, D. C.
WAL, G. van der W. (1994), “Article 86 EC: The Limits of Compulsory Licensing”,
European Competition Law Review, no. 4, p.230-235.
WEBB M. and TAYLOR M. (Winter 1998/99), “Light-Handed Regulation of

Telecommunications in New Zealand: Is General Competition Law Sufficient?”,
International Journal of Communications Law and Policy, Issue 2.
WHISH, R. (2001), Competition Law, 4. Ed., Butterworths, London.
WHISH, R. (2001), “Recent Developments in Community Competition Law”,

European Law Review, p. 234-235.
ZINHARA, C. (December 2002), “Dynamic Capabilities for Innovation: Incumbent

Mobile Operator’s Response to Competition”, MSc Thesis, Communications
Management, Strathclyde Business School.
Cases of European Court of Justice (ECJ)419
- Case C-7/97, Oscar Bronner GmBH & Co KG and Others v. Mediaprint

Zeitungs-und Zeischiftverlag GmbH & Co KG and Others [1998] ECR I-7791,
[1999] 4 CMLR 112.
- Atlas and Phonenix/GlobalOne [1996] OJ L 239/23.
- Case No. M. 1795 (2000) Vodafone/Mannesmann, IP/00/373 of 12 April 2000.
- Joined Cases 6,7/73, Commercial Solvents v. Commission [1974] ECR 223.
- Case 27/76, United Brands v. Commission, [1978] ECR 207.
- Joined Cases C-241 & 242/91P, RTE and ITP v. Commission, [1995] ECR I-743.
Cases of Court of First Instance (CFI)
- Case T-504/93, Tiercé Ladbroke SA v. Commission [1997] ECR II-923.
419
For the Community Court (ECJ and CFI) decisions that are mentioned in this study, refer to
http://www.europa.eu.int/jurisp/cgi-bin/form.pl?lang=en
179
- Joined Cases T-374, 375, 384 & 388/94, European Night Services v.
Commission, [1998] ECR II-3141.
- Case 311/84, Cenbtre belge d’études du marché, [1985] ECR 3261.
Cases of European Commission420
- Decision 88/589 of 4 November 1988, London European/Sabena [1988] OJ L

317/47
- Decision 92/213 of 26 February 1992, British Midland/Aer Lingus [1992] OJ L
96/34
- Decision of 11 June 1992, B&I Line plc/Sealink Harbours Ltd. [1992] 5 CMLR
255.
- Decision 94/19 of 21 December 1993, Sea Containers/Stena Sealink [1994] OJ L
15/8.
- Decision 94/663 of 21 September 1994, European Night Services [1994] OJ L
15/8.
Cases of United States Courts421
- Aspen Skiing Co. v. Aspen Highlands Skiing Corp 472 US 585 (1985)
- Alaska Airlines, Inc. v. United Airlines , Inc., 948 F 2d 536 (9th Circ 1991)
- Otter Tail Power Co. v. United States 410 US 366, 35 L. Ed. 2d 359, 93 S. Ct.
1022 (1973)
- MCI Communications v. AT&T, 708 F.2d 1081, 1132-1133 (7th Cir. 1983)
- U.S. v. Terminal Railroad Association 224 US 383 (1912).
- Associated Press v. United States 326 US 1 (1945)
- The Otter Tail Power Co v. U.S. 410 US 366 (1973)
- Intergraph Corp. v. Intel Corp., 195 F.3d 13346, 1358 (Fed. Cir. 1999)
420
For the Commission decisions that are mentioned in this study, refer to
http://europa.eu.int/comm/competition/antitrust/cases/
421
For the decisions of US Courts that are mentioned in this study, refer to
http://www.stolaf.edu/people/becker/antitrust/
180
- Apartment Source of Philadelphia v. Philadelphia Newspapers, Civ. A. No. 98-
5472, 1999 WL 191649
- Twin Labs v. Weider Health & Fitness, 900 F.2d 569-570 (2d Cir 1990)
- Hecht v. Pro Football, Inc. 570 F.2d 982, 992 (D.C. Cir. 1977), cert. denied, 436
U.S. 956 (197).
- City of Anaheim, 955 F.2d at 1381
- Case No. IV/M.1439, Telia/Telenor (1999) OJ [2001] L 40/1, [2001] 4 CMLR
1226.
Cases of Turkish Competition Board422
- Competition Board’s Decision dated 21/12/2000 and numbered 00-50/533-295

422
For Competition Board’s Decisions that are mentioned in this study, refer to
http://www.rekabet.gov.tr/ihlal.asp
181

Essential Facilities Doctrine in EC and Telecommunications

Uploaded by

Copyright:

Available Formats

Essential Facilities Doctrine in EC and Telecommunications

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Essential Facilities Doctrine in EC and Telecommunications

Uploaded by

Copyright:

Available Formats

ESSENTIAL FACILITIES DOCTRINE UNDER

A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF SOCIAL SCIENCES

MEHMET B LAL ÜNVER

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS

Examining Committee Members

Assoc. Prof. Dr. Erol TAYMAZ ___________________

Assist. Prof. Dr. Gamze A CIO LU ÖZ ___________________

Assist. Prof. Dr. Gül EFEM ___________________

Name, Last name : Mehmet Bilal ÜNVER

ESSENTIAL FACILITIES DOCTRINE

Ünver, Mehmet Bilal

M. Sc., Department of European Studies

Supervisor: Assist. Prof. Dr. Gamze A ÇIO LU ÖZ

September 2004, 181 pages

Keywords: Essential Facilities, Refusal to Deal, Indispensability, Competition

AT REKABET HUKUKU’NDA ZORUNLU UNSUR DOKTR N

Ünver, Mehmet Bilal

Yüksek Lisans, Avrupa Çalı maları

Tez Yöneticisi: Yrd. Doç. Dr. Gamze A ÇIO LU ÖZ

Eylül 2004, 181 sayfa

Anahtar Sözcükler: Zorunlu Unsur, Anla ma Yapmayı Reddetme, Kaçınılmazlık,

To my lovely wife who deserves forever love

In the second section of the thesis, the above-mentioned discussions will be

1. ESSENTIAL FACILITIES DOCTRINE IN COMPETITION

Competition Law serves many purposes, namely eliminating market imperfections

As a matter of fact, while aspiration of a peaceful and reliable supra-national system

As a corresponding matter, free market economy encompassing the ‘right to refuse to

Essential Facilities Doctrine (hereinafter usually called ‘EFD’) is an important

As could be inferred here, there is a number of pre-requisites for application of

Summarising, the obligations regarding ‘duty to deal’ or ‘duty to share essential

1.2 The US Case-Law

A monopolist’s refusal to deal under these circumstances is

1.2.1 Control of the Essential Facility by a Monopolist

Another discussion is related to the extent to what reliance on traditional dominance

1.2.2 A Competitor’s Inability Practically or Reasonably to Duplicate Essential

Given the lack of requirement of an equivalence between the market of essential

Regarding the characteristics of essential facilities, both the examination that

The second condition, that duplication be impractical or unreasonable, is a higher

1.2.4 The Feasibility of Providing the Facility

Determination of feasibility therefore must include an inquiry in order to verify any

In the context of ‘determination of feasibility’, several questions arise as to the

1.3 Assessment of the US Case-Law

As it can be inferred from the suggestions of P. Areeda,86 if the criterion ‘elimination

1.4 Essential Facilities Doctrine under EC Competition Law

In the context of EC Competition Law, the Essential Facilities Doctrine was

• dominant companies may not discriminate if the discrimination has significant

1.4.1 Main Characteristics of ‘Refusal to Deal’ Cases

However, pursuant to Article 82(b), an abuse of dominant position may be in the

(c) a withdrawal of access from an existing customer.

1.4.2 EC Case-Law on ‘refusal to deal’

The leading case on unilateral refusal to deal in EC Competition Law is Commercial

The anti-competitive aspect of the Commercial Solvents’ behaviour was particularly

In view of these conflicting arguments it is advisable to assert

It is noteworthy here that ECJ in United Brands, focused on the conduct of a

Telemarketing102 is another judgment in which the ECJ confirmed that an

Commercial Solvents also applies to the case of an undertaking

It must therefore be held in answer to the second question that an

Along with the aim of encouraging competition in many fields of Community-wide

The broadcasters’ refusal to provide basic information by relying

A dominant undertaking’s ‘reserving to itself the downstream market’ having been