Regulating Collective Management Organisations by Competition: An Incomplete Answer to the Licensing Problem?

AuthorMorten Hviid - Simone Schroff - John Street
Pages256-270
2016
Morten Hviid, Simone Schroff and John Street
256
3
Regulating Collective Management
Organisations by Competition
An Incomplete Answer to the Licensing Problem?
by Morten Hviid, Simone Schroff and John Street*
© 2016 Morten Hviid, Simone S chroff, John Street
Everybody may disseminate this ar ticle by electronic m eans and make it available for downloa d under the terms and
conditions of the Digital P eer Publishing Licence (DPPL). A copy of the license text may be obtain ed at http://nbn-resolving.
de/urn:nbn:de:0009-dppl-v3-en8.
Recommended citation: Mor ten Hviid, Simone Schrof f, John Street, Regulating Colle ctive Management Organis ations by
Competition: An Incomplet e Answer to the Licensing Problem?, 7 (2016) JIPITEC 256 para 1.
Keywords: Collective Management Organisations; competition; licensing; reforms; EC; qualitative research
structures through reforms that increase the degree
of competition. This paper asks whether the reforms
have had the desired effect and shows, through qual-
itative research, that at least regarding the streaming
of music, competition has not delivered. Part of the
reason for this may be that the services required by
the now competing CMOs have changed.
Abstract: While the three functions of Col-
lective Management Organisations - to licence use,
monitor use, and to collect and distribute the revenue
- have traditionally been accepted as a progression
towards a natural (national) monopoly, digital exploi-
tation of music may no longer lead to such a fate. The
European Commission has challenged the traditional
A. Introduction
1
The licensing of copyright protected works has been
a feature of the music industry for decades, allowing
a large variety of users - bars, broadcasters, concert
venues etc. - to play music as part of the services they
offer. This system rests on central licensing agencies,
most commonly known as Collective Management
Organisations (CMOs). These administer the rights
of copyright holders from a central point, offering
licenses to the users. While the system has been in
place for a long time and has worked reasonably
well (although not perfectly) for analogue uses, the
rise of the internet and digital technology has been
a game-changer. By expanding the possibility of,
and demand for, cross-border uses, the traditional
system has come under considerable strain, and is
now reaching breaking point as new types of services
such as streaming emerge. These services need to
license musical works on an EU or global basis to
use the technology’s full potential. This is difcult
because, until now, the CMOs in the EU have been
nationally-based monopolies. To obtain a license that
covers Europe, 28 different licenses are, in principle
required. Such an arrangement clashes directly with
the EU’s ambition to create a Digital Single Market
(DSM).1 As a result, the current regulatory regime,
particularly as it relates to CMOs, has become a
1 For a discussion of the Digital Single Market, see European
Commission, Digital Single Market- Bringing Down Barriers
to Unlock Online Opportunities 2015 (available at
ec.europa.eu/priorities/digital-single-market/>, last
accessed 17/12/15).
Regulating Collective Management Organisations by Competition
2016
257
3
prime concern of the EU.
2 The functioning of CMOs in the digital domain is of
key importance for the DSM.2 The single market is,
after all, intended to allow for the free movement of
goods and services across borders, giving EU citizens
access to what they most prefer. The inability of
the current copyright system to issue cross-border
licenses to all users that require them means that
CMOs can provide services only on a member state
by member state basis, due to the threat of copyright
infringement (and therefore high costs) that any
unlicensed cross-border use would entail.3 The
result is geo-blocking: individual users are not able
to access services once they enter another member
state, even if they have paid for those services.
Rather than having a single market, online music
continues to operate through multiple separate
markets.
3 The EU’s response to this situation has been to issue
a Directive,
4
which is due to be implemented in 2016.
This Directive formalises competition between CMOs5
2 See, for example, Intellectual Property Ofce, Collective
rights management in the digital single market
available02015 (at:
uploads/system/uploads/attachment_data/le/401225/
collective_rights.pdf>, last accessed 15/4/16).
3 There have been agreements in the past that offered
MTLs, however, they have either not been renewed (see
Santiago Agreement), are limited to specic user groups
(Simulcasting Agreement), or have been found to be contrary
to competition rules (such as the CISAC Model Agreement).
For a detailed description of these agreements, please see:
Guibault and Van Gompel, Collective Management in the
European Union, in Gervais (ed.), Collective Management of
Copyright and Related Rights (Alphen Aan Den Rijn: Wolters
Kluwer, 2015, 3rd Ed.), 139- 174.
4 Directive on collective management of copyright and
related rights and multi-territorial licensing of rights
in musical works for online use in the internal market
(Directive 2014/26/EU) (CMO Directive).
5 Competition between CMOs for the management of rights
of rights has been introduced in the case law before. CMOs
have to be seen as dominant undertakings and are therefore
subject to competition rules (GVL v. Commission (Case
7/82, [1983] ECR 483, [1983] CMLR 645); RT v. SABAM (Case
127/73, Belgische Radio en Televisie v. SV SABAM and NV
Fonior, [1974] ECR 313, [1974] 2 CMLR 238). The results are
a number of restrictions on CMOs towards their members.
For the purpose of this article, the most important are
the following. It was held that CMOs cannot refuse the
management of rights by foreigners, even if they are not
resident in a country. This is especially the case if the CMO
has a dominant position and quasi monopoly (Case 7/82,
Gesellschaft zur Verwertung von Leistungsschutzrechten
mbH (GVL) v. Commission, [1983] ECR 483; [1983] 3 CMLR
645). In addition, right holders cannot be required to
assign all of their rights. (Case 127/73, Belgische Radio en
Televisie v. SV SABAM and NV Fonior, [1974] ECR 313, [1974]
2 CMLR 238)), in particular their online rights (Daftpunk)
(Case C2/37.219, Banghalter et Homem Christo v. SACEM, 6
August 2002). This is limited by economic viability though.
A CMO cannot be forced to accept only those rights which
are expensive to administer ( see for example, Case 127/73,
and places obligations upon them to serve better the
interests of users6 and right holders. In this paper,
we assess the possible outcome of this initiative, and
argue that there might be more effective ways to
address the problem posed by creating a DSM. We
conne our attention to the music market, which
is feeling the effects of the digital revolution most
acutely, at least among the creative industries. We
also focus our attention to the streaming of music.
Similar issues arise regarding the sales of digital
music, for example through electronic stores.7
B. Background: The role of the
CMO before digitalisation
4
As Handke and Towse point out, the licensing market
for musical works in an analogue world was (and
remains) highly complex.8 A large number of creators
and products (typically, artists and songs) have to be
matched with a similarly large number of diverse
users. Asymmetry of information in such a situation
creates prohibitive transaction costs for individual
licensing between a copyright owner and a user. In
other words, individual licensing represents a case
of market failure in which copyright owners and
Belgische Radio en Televisie v. SV SABAM and NV Fonior,
[1974] ECR 313, [1974] 2 CMLR 238, para. 10, 11 and 15).
6 User interests are only indirectly addressed, for example in
the transparency rules which are meant to give users the
information they need to choose licenses (see in particular
(Directive 2014/26/EU) (CMO Directive), art. 19- 22.).
7 Some of these issues have been addressed, either through
coalitions of old structures to create broader organisations
which can offer bundled clearing of copyrights; or
through new structures such as Merlin- a right clearance
organisation which can also offer MTLs. However, not all
Copyright Management Societies in the EU are members
of such structures. Thus transaction costs are incurred to
provide pan-EU availability of digital music. As Gómez and
Martens note: “We nd that in August 2013 there was still
substantial variation in availability in the iTunes country
stores across the EU DSM. Less than half of all song tracks
and music albums are available in all EU27 country stores.
Overall, music availability in the EU DSM is somewhere
between 73 and 82 per cent of what it could be in a fully
open DSM where all song tracks and albums would be
available in all EU27 countries.” (Gómez and Martens,
Language, Copyright And Geographic Segmentation in
the EU Digital Single Market for Music and Film, JRC/IPTS
Digital Economy Working Paper 2015, (available at
ssrn.Com/Abstract=2603144>, last accessed 15/4/16), 3-4.
However, Gómez and Martens do acknowledge that matters
are improving. It should also be noted that some right
holders’ business strategies relies on fragmented markets to
maximise prots, as for example in the audio-visual sector.
These attitudes may oppose the aims of the Commission but
nonetheless also affect the availability of pan- European
licenses.
8 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15).
2016
Morten Hviid, Simone Schroff and John Street
258
3
the users would both lose out. The copyright owner
would not generate the income they seek while the
user is not able to legally play the music they want.
The solution to this problem has involved several
intermediaries, including CMOs streamlining the
nancial transaction between the creator and user.
5 CMOs act to reduce the market failure.9 In general,
they have three functions: 1) to license works for
specic uses; 2) to monitor the use of works and
collect the revenue; and 3) to distribute the revenue
to its members.10 The CMOs collect the revenue for
low-value, high volume secondary uses; that is, uses
where the individual licensing fee is small but the
number of licenses which need to be issued add
up to a substantial revenue stream. CMOs manage
the rights of its members collectively, providing
blanket licenses to users. By managing the rights
collectively, they are able to lower the transaction
costs as well as provide a stable licensing framework.
In economic terms, they enable the market to
function by ensuring copyright effectiveness in
circumstances where copyright owners cannot
contract directly. A blanket license gives users –
especially broadcasters – the right to use any music
within the CMO’s repertoire. The blanket licenses
reduce the transaction costs because they do not
require negotiations on the price or the exact size of
the rights bundle for each individual transaction.11
6
CMOs have been a core feature of the licensing
market within the EU (and beyond) for more than a
century. Based on a system of reciprocal agreements
between CMOs, they have been able to license a
world-wide repertoire. A user can therefore use any
song they want and only pay their local CMO. The
transfer of funds across borders is carried out by the
CMOs themselves and is of no concern to the user. As
a result, CMOs have established a system of national
9 Haunss, The Changing Role of Collecting Societies on the
Internet, Internet Policy Review 2013, 1-8, Handke and
Towse, Economics of Copyright Collecting Societies, SSRN
2007 (available at:
cfm?abstract_id=1159085>, last accessed 14/12/15).
10 Andersen, Kozul- Wright, Z. and Kozul- Wright, R,
Copyrights, Competition and Development: The Case of the
Music Industry, United Nations Conference on Trade and
Development. Geneva: United Nations Conference on Trade
And Development 2000 (available at:
en/docs/dp_145.en.pdf>, last accessed 15/4/16), 21.
11 There is an extensive economics literature on what is often
termed “buffet” pricing inspired by the “all-you-can-eat
buffets”. Much of this literature has focused on behavioural
aspects, in particular those which lead to obesity, which
does not appear to be particularly relevant in our context.
The behavioural literature is summarised in Lambrecht
and Skiera, Paying Too Much and Being Happy About It:
Existence, Causes, and Consequences of Tariff-Choice
Biases, Journal of Marketing Research 2006, 212–223 as
well as Just and Wansink, The at-rate pricing paradox:
conicting effects of “all-you-can-eat” buffet pricing, The
Review of Economics and Statistics 2011, 193-200.
monopolies, which do not compete with each other,
but instead operate under a set of agreements which
determine the cost of licenses. While this broad
coverage in works, the economies of scale, and the
resulting monopoly status contribute to efcient
licensing in practice, it is also the source of the
European Commission’s main concern. While the
licences are “blanket”, their price may well differ
according to the type of organisation that requests
the blanket licence. No stakeholder is able to judge
the price charged and the lack of a viable alternative
has meant that a copyright holder has no incentive
to defect to a rival CMO, no matter how dissatised
they are.12 The CMO’s monopoly status has given rise
to typical concerns often attributed to monopolies
namely the potential abuse of a dominant position.13
Market prices cannot be established; neither for the
users in terms of how much they should pay for their
license, nor for the copyright owners, in relation to
the cost of administration that the system entails.14
7
EU case law has established that the CMOs are
undertakings which hold a dominant position,
meaning that they are subject to the full force of
competition law, including both article 101 TFEU
relating to concerted practices and article 102
TFEU relating to the abuse of a dominant position.15
This required restrictions on how they operated.
However, the CJEU also ruled that CMOs serve the
public interest, and that, therefore, competition
law was not to be applied rigidly.16 In other words,
while the Court found the monopoly status and
reciprocal agreements justiable in the broader
12 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15).
13 Assessing such abuses is made complicated by the two-sided
nature of the market, where the intermediary can decide
from which side of the market, copyright holders or users,
to extract rent, either in terms of funds or a “quiet life”.
14 Kretschmer, Access and Reward in the Information Society:
Regulating the Collective Management of Copyright (Poole:
Centre for Intellectual Property Policy & Management,
2005), 7.
15 Graber, Collective Rights Management, Competition Policy
and Cultural Diversity: EU Law Making at a Crossroads,
I-Call Working Paper 2012 (available at:
com/sol3/papers.cfm?abstract_id=2161763>, last accessed
15/4/16), 6.
16 See for example: Case C-395/87 Ministère Public v Tournier,
[1989] ECR-2521, para. 24; Lucazeau v SACEM ECR 2811; GVL
v. Commission (Case 7/82, [1983] ECR 483, [1983] CMLR 645;
Case 127/73, Belgische Radio en Televisie v. SV SABAM
ECR 51 313. All of these cases involved the application of
competition law and recognised that there are legitimate
interests that can limit its application in practice. Also
discussed in: Graber, Collective Rights Management,
Competition Policy and Cultural Diversity: EU Law Making
at a Crossroads, I-Call Working Paper 2012 (available
at:
id=2161763>, last accessed 15/4/16), 6.
Regulating Collective Management Organisations by Competition
2016
259
3
public interest, it also recognised the negative
impact the system could have on users and right
holders. For this reason, CMOs are required to offer
users reasonable licensing terms, while at the same
time giving their members as much freedom to
administer their rights as independently as possible
(as long as this is consistent with the functioning of
the CMO as a whole). Copyright owners should be
able to administer their rights individually insofar
as this does not impose undue costs on the CMO. For
example, while withdrawing all one’s works or the
online rights attached to those works is acceptable,
withdrawing the online rights for works A, B and
C, but not D, and G, is not, because keeping track
would be too expensive for the CMO.17 In essence,
the regulations have attempted to balance the
threat of monopolisation against effective rights
administration.18 However, given that there was no
viable alternative to the CMO system the Commission
tolerated it. The rise of the internet has changed the
rules of the game.
8
It should be noted that some authors have questioned
the treatment accorded to CMOs in the analogue
world. Katz in particular challenges the claim that in
the analogue world the CMO is a natural monopoly.
19
He observes that more than one CMO may operate
in a single territory. Unlike most other countries,
where the CMO is a monopolist, the US has three
CMOs managing musical works (ASCAP, BMI and
SESAC) of which one (SESAC) is rather smaller than
the others (less than 5% in 2000)20 and has coexisted
with ASCAP since 1931 and all three have been in
the market since 1941. The traditional argument in
favour of natural monopoly economies of scale
is not compatible with the persistent existence of
such a small rm.
9
Katz reminds us that, while the CMOs charge for
a blanket licence, they do not charge all users the
same price.21 Thus they use their monopoly power
17 Kretschmer, Access and Reward in the Information Society:
Regulating the Collective Management of Copyright (Poole:
Centre for Intellectual Property Policy & Management,
2005), 5.
18 Dietz, Legal Regulation of Collective Management of
Copyright (Collecting Societies Law) in Western and Eastern
Europe, Journal of the Copyright Society of the USA 2002,
908.
19 Katz, The potential demise of another natural monopoly:
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015,
541-593.
20 Katz, The potential demise of another natural monopoly:
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015, 554.
21 Katz, The potential demise of another natural monopoly:
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015,
541-593.
to engage in third degree price discrimination22,
charging different prices to different types of
businesses, a practice which has an ambiguous
effect on both total and consumer surplus. Katz also
points out that the existence of different licences
for performance of the work adds an extra tool to
the CMO to practice successful price discrimination
because it enables the CMO to identify the nature
of each user.23
10
In a supplementary article, Katz explores how his
argument would apply in the digital world. Given
his conclusion for the analogue world, it is hardly
a surprise that he is sceptical about the monopoly
argument.
24
However, given when it was written,
his paper has to engage in speculation. While it
undoubtedly was ahead of its time in 2006, and
many of the speculations have come to pass, it adds
little to the current debate. However, it does help us
understand why the Commission viewed the digital
world differently when it comes to competition.
C. The Digital Challenge
11
As digital technology, and especially the internet,
rose in importance, the needs of users changed
dramatically. A new breed of services came to the
fore, most notably, the streaming platforms (Spotify,
Deezer, Amazon Music, etc.). They differ from
analogue users in the kind of licenses they require.
Analogue users only require territorial licenses; their
services do not cross national borders25 and therefore
they do not require licenses that extend further.
However, the internet (and digitalisation) creates
the possibility of easy access to music irrespective of
tariff barriers or broadcasting regulations. Any legal
service seeking to exploit these possibilities requires
multi-territorial licenses. To cater to this need, CMOs
reacted rst by offering Simulcasting agreements,
providing cross-border licenses to internet radio.
The Commission accepted this solution as a
22 Firms engaging in third degree price discrimination offer
different prices to different identiable groups of buyers – a
classic example is different prices for different age groups.
23 Katz, The potential demise of another natural monopoly:
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015, 550.
24 Katz, The potential demise of another natural monopoly:
New technologies and the administration of performing
rights. Journal of Competition Law and Economics 2006,
245-284.
25 While strictly speaking not true, this is the assumption
which has been made in the industry, motivated by a view
that Broadcasters are (supposed to) focus on their national
audience, not least because of language barriers. The
exception is broadcasting with the Simulcasting Agreement
which resolves the issue by treating broadcasters as
geographically limited users and therefore as essentially
the same as analogue users.
2016
Morten Hviid, Simone Schroff and John Street
260
3
permissible exception under article 101(3) TFEU
(which began in 2004).
26
However, it remained the
exception, even as the Commission came to realise
that digital technology was not only changing user
requirements but the system as a whole.
12
The driver of change was not just user demand,
but the very nature of licensing itself. Handke
and Towse have argued that primarily, digital
technology makes the gathering and processing
of information much easier. Secondly, they argue
that it enhances market signalling: on one hand,
the use of individual works can now be assessed
with more precision than before; on the other
hand, there is potential for price discrimination
and charging every user what they are willing
to pay. Finally, as a result of these factors, CMOs
are able to reduce their costs.
27
New technologies
such as Digital Rights Management (DRM)28, which
enable rights to be administered individually,29 can
enhance efciency. This of course undermines the
CMO’s justication for their monopoly status,30
as there are now real alternatives to them.
26 Guibault and Van Gompel, Collective Management in the
European Union, in Gervais (ed.), Collective Management of
Copyright and Related Rights (Alphen Aan Den Rijn: Wolters
Kluwer, 2015, 3rd Ed.), 160-161. Other agreements such as
the CISAC model contract (leading to the cases European
Commission, Commission Decision of 16/07/2008 relating
to a proceeding under Article 81 of the EC Treaty and Article
53 of the EEA Agreement (Case COMP/C2/38.698 – CISAC)
(available at
cases/dec_docs/38698/38698_4567_1.pdf>, last accessed
17/12/15) and the CJEU decision CISAC v. European
Commission (Case T-442/08)) and the Santiago Agreement
(Notication of cooperation agreements (Case COMP/
C2/38.126 – BUMA, GEMA, PRS, SACEM),O.J. C. 145/2 of
17.05.2001) as well as the Barcelona Agreement (Notication
of cooperation agreements (Case COMP/C-2/38.377 – BIEM
Barcelona Agree-ments), O.J. C. 132/18 of 4.06.2002) were
found anti-competitive.
27 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15), 13.
28 In this context, DRM is a tool to control the type of access
one has to digital music. It controls both access and usage.
For a discussion of the merits of DRM, see e.g. Doctorow,
What happens with digital rights management in the real
world?, Guardian, 5 February 2014 (available at:
www.theguardian.com/technology/blog/2014/feb/05/
digital-rights-management>, last accessed 15/4/16).
29 Kretschmer, Access and Reward in the Information Society:
Regulating the Collective Management of Copyright (Poole:
Centre for Intellectual Property Policy & Management,
2005), 17.
30 Ficsor, Collective Management of Copyright and Related
Rights, WIPO 2002 (available at:
edocs/pubdocs/en/copyright/855/wipo_pub_855.pdf>, last
accessed 15/4/16), 98.
D. A new regulatory regime
13
The change in the Commission’s attitude rst became
clear when it refused to accept the Santiago and
Barcelona Agreements, which aimed to extend the
analogue licensing system to the digital domain.31
The Commission’s attitude was made even clearer
when it rejected CISAC’s model contracts. CISAC,
the world-wide umbrella organisation for CMOs,
devised model contracts to allow its members to
offer multi-repertoire, multi-territorial licenses.
The contracts had three core features: a national
allocation clause, an exclusivity clause, and a non-
intervention clause. Combined, the latter two had
the effect of maintaining the national delineation
of CMOs, guaranteeing their monopolies. While
these clauses were not new, the Commission now
considered them unjustied digital technology
meant that a local presence was not required to
ensure efcient enforcement.32 The Commission
argued that digitalisation enabled CMOs to compete
with each other in the eld of digital exploitation,
meaning online use in practice. Overall, it found
the model contract contrary to competition rules
under article 101 TFEU,33 although this decision was
overturned by the General Court in 2013.34 Instead,
CMOs should, the Commission believed, compete
with each other to attract members and users. This
in turn should lead to increased efciency in the
rights administration, aiding the emergence of new
markets.
35
The Commission shifted from viewing the
CMO as a necessary evil for ensuring the effective
licensing of works, to seeing it an as unnecessary
anti-competitive undertaking which harmed both
right holders and users. This stance was to become
clear Commission policy.
14 While Katz’s analysis casts doubt on the survival of
the past monopolising elements of collective rights
management, the move to digital exploitation
could, at least in theory, give rise to a new
monopoly element.36 This has so far attracted
31 Frabboni, Collective Management of Copyright and Related
Rights: achievements and problems of institutional efforts
towards harmonization, in: Derclaye (ed.), Research
Handbook in the Future of EU Copyright (Cheltenham:
Edward Elgar, 2009), 373-400.
32 Guibault and Van Gompel, Collective Management in the
European Union, in Gervais (ed.), Collective Management of
Copyright and Related Rights (Alphen Aan Den Rijn: Wolters
Kluwer, 2015, 3rd Ed.), 162.
33 European Commission, Commission Decision of 16/07/2008
relating to a proceeding under Article 81 of the EC Treaty
and Article 53 of the EEA Agreement (Case COMP/C2/38.698
– CISAC) (available at
antitrust/cases/dec_docs/38698/38698_4567_1.pdf>, last
accessed 17/12/15), 220-223.
34 CISAC v. European Commission (Case T-442/08).
35 Sparrow, Music Distribution and the Internet: A Legal Guide
for the Music Business (Aldershot: Gower, 2006), 1.
36 Katz, The potential demise of another natural monopoly:
Regulating Collective Management Organisations by Competition
2016
261
3
little commentary. With more data available
electronically, a comprehensive database of all
right holders and associated material would not
only be essential, but also display increasing return
to scale both in its creation and maintenance. For
full functionality it is important that the database
is comprehensive. Given the cost of establishment
and maintenance, it would be inefcient to have
two parallel fully comprehensive databases. By
contrast, the other elements - such as monitoring
and collecting money - seem to have less of a claim to
monopoly status once services become digital. Given
the international nature of such a database, there
is a serious issue as to who regulates the terms of
access and how the database is to be funded. Building
on existing databases held by CMOs, one possibility
would be for these to set up an institution to hold,
transform and maintain these databases. This has to
some extent already happened. Most CMO databases
(and all of the ones examined here) are part of CIS-
Net, the most comprehensive database for musical
works and their corresponding rights. It is owned by
FastTrack, which, in turn, is owned by the CMOs. The
question is whether competition among the CMOs
(in the EU) is sufcient to generate a comprehensive
database, whilst at the same time engendering a
meaningful and valuable choice.
15
In 2005, the Commission reported on the lack of
cross-border licenses for users in the online market.
It proposed that rights holders should be free to
choose their CMO, the rights that they assign to it and
their associated territorial reach.37 The underlying
rationale is a typical competition remedy: by giving
the individual the choice over the provider, they
can choose the service that most closely matches
their preferences. In other words, by allowing
right holders to vote with their feet, CMOs would
be bound to become more efcient in an effort to
not lose members. Furthermore, CMOs would issue
pan-European licenses, and by choosing their CMO
carefully, rights holders would be able to ensure that
each CMO would be able to offer coherent bundles.38
The Commission’s recommendation rejected
the analogue services’ use of reciprocal licence
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015,
541-593, Katz, The potential demise of another natural
monopoly: New technologies and the administration
of performing rights. Journal of Competition Law and
Economics 2006, 245-284.
37 European Commission, Commission Recommendation of
18 May 2005 on Collective Cross-Border Management of
Copyright and Related Rights for Legitimate Online Music
Services (2005/737/EC)” (available at
europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:3200
5H0737&from=EN>, last accessed 17/12/15).
38 Note that rights holders would have an incentive to seek out
CMOs who “managed” material similar to their own to give
that CMO more bargaining power vis-à-vis the users.
agreements and full harmonisation.39 Their approach
became law in the 2014 CMO Directive 2014/26/EU.
The Directive focused on more competition rather
than on harmonisation or an extension of the
traditional system of reciprocal agreements. The
Directive aims at providing an environment in which
competition can be fully effective. It sets minimum
standards for the transparency and supervision of
CMOs by their members and therefore the right
holders.40 Both of these are typical competition
remedies, which have been applied to areas such as
the energy market. In the case of the music industry,
EU policy is based on the distinction between the
analogue and the digital licensing market for musical
works, and the need to alter the role played by CMOs
in the latter. However, the major CMOs are already
meeting the Directive’s demands,41 so the question
is whether the legislative intervention will have its
intended effect. After all, if the database existed
and access was regulated/mandated, then the right
holder would genuinely have choice based on the
quality of service.42 To answer our question, we
investigated the problems that actually affect users
in the digital realm.
E. Methodology
16 To understand the current state of licensing in the
EU, we compared the experience of an analogue
user with that of one who seeks a license for online
exploitation. We simulated the path a potential
broadcaster or web-streaming service would
follow in acquiring a license, starting with the rst
search to identify CMOs all the way to the nal
license. It is assumed that the broadcaster seeks a
multi-repertoire, single-territory license because
they want to be able to use all kinds of music in
their programming which, by the nature of the
broadcasting sector, is assumed to reach a national
audience. By contrast, a web-streaming service would
also want to offer all kinds of music but on a multi-
territorial basis, making its programmes accessible
around the world, or at least within Europe to fully
exploit the potential of the Single European Market.
39 Kretschmer, Access and Reward in the Information Society:
Regulating the Collective Management of Copyright (Poole:
Centre for Intellectual Property Policy & Management,
2005), 13-14.
40 Directive on collective management of copyright and
related rights and multi-territorial licensing of rights
in musical works for online use in the internal market
(Directive 2014/26/EU) (CMO Directive), Part III.
41 Schroff and Street, The politics of the digital single
market: the case of copyright, competition and collective
management organisations (forthcoming).
42 For music, there may be an unnecessary stumbling block as
rights can only be assigned on an exclusive basis to a CMO
and therefore not to more than one collecting agent at the
same time.
2016
Morten Hviid, Simone Schroff and John Street
262
3
We used these two licensing scenarios to explore the
practical issues raised by providing multi-repertoire,
cross-national content.
17 The empirical research was designed in such a way
as to give a realistic picture of the situation and
challenges faced by practitioners. For this reason, it
was carried out by a research associate who has legal
training but is not working in the eld of intellectual
property or licensing copyright material. In our
view, this mimics the experience of those individuals
who have to acquire licenses for commercial
services. The researcher was asked to keep track of
how she identied relevant organisations, noting
down the challenges that she encountered. We chose
a representative set of European case studies: the
UK, France, Germany and Sweden. The ndings are
striking: while the analogue user nds a system in
place to satisfy their licensing needs, the same is not
true for those who want to run streaming services.
F. Findings: the problems
for online users
18 The main nding is that CMOs are either unable or
unwilling to satisfy the demand of online-services.
When a broadcaster seeks a license, all of our case
studies were able to provide them with a multi-
repertoire license for the rights in musical works.
This was because of the reciprocal agreements
that CMOs have with each other. In this sense, the
broadcaster has in this sense access to a one-stop-
shop. Table 1 below summarises the steps taken as
well as the key difculties in obtaining the right
to make copyrighted content available across
borders in the case of broadcasting. It is clear from
Table 1 that there are only limited difculties in
obtaining a licence for traditional broadcasting.
19 Table 1: Broadcasting
France Germany Sweden UK
Licenses
Required
SACEM (covers
other CMOs for
musical works)
SCPP/ SPFF
GEMA
GVL
STIM
SAMI
IFPI
PRS/ MCPS
PPL
Information
on Coverage
Yes Yes Limited Yes
Broadcasting
Tariff
available
online
Yes Yes No Yes
Indemnity for
Licensees
(coverage
of non-
members)
No Limited
(presumption
of
management)
No Limited
(presumption of
management in
some cases)
Information
available in
English
Partial (does
not include
substantive
licensing
information)
Partial (does
not include
substantive
licensing
information)
Yes Yes
20
In contrast, Table 2 below demonstrates the
considerably greater difculties encountered in
obtaining licences for web-streaming. The licenses
for online uses are a lot more complicated, not least
because the descriptions used by the CMOs are very
vague. Although some multi-territorial licenses exist,
it is not clear which works are covered by them. For
example, in the UK it is apparent that PRS, the CMO
for songwriters, composers and publishers, is able
to license the Anglo-American repertoire of certain
publishers on a multi-national basis. However, there
is no way to check what is actually included in this
description. They are not blanket licenses like the
ones available to broadcasters in the analogue
system. This means in practice that more than one
license is necessary to cover the same category of
works, increasing the cost for the user.
21
Secondly, just because the license is described as
multi-territorial, it does not follow that this involves
EU-wide coverage.43 For example, the French CMO
SACEM is only able to license France, Luxembourg
and Monaco the three countries in which it is
the main CMO anyway. It would have always been
able to license these even without a change in EU
policy.
44
Similarly, the other CMOs only offer licenses
that cover a very limited number of countries, but
none of them provided clear information as to what
countries were included. As a result, not only is
there no comparable one-stop shop, the territorial
gaps in the license are also unclear, giving rise to
major concerns regarding what is allowed. In sum,
while a broadcaster is provided with a one-stop-
43 A similar problem was observed for the sale of digital
music, see Gómez and Martens, Language, Copyright And
Geographic Segmentation in the EU Digital Single Market for
Music and Film, JRC/IPTS Digital Economy Working Paper
2015, (available at ,
last accessed 15/4/16).
44 The CMO Directive has been an issue at EU level for a
signicant amount of time before the Directive was nalised.
It is therefore possible that stakeholders anticipate the
changes early on knowing that they will have to comply at
some point. Having said this, SACEM has been able to issue
licenses for these three territories for years. It is therefore
unlikely that changes in EU policy had an effect on SACEM
in this case.
Regulating Collective Management Organisations by Competition
2016
263
3
shop to satisfy their licensing needs, the same
route is not available to online services wishing
to operate across borders. In fact, they struggle
with a more fundamental problem a lack of
information about the coverage of the license.
22 Table 2: Web-streaming
France Germany Sweden UK
Relevant
CMOs
SACEM but
does not cover
phonograms or
performances
GEMA but
does not cover
phonograms
or
performances
STIM but
does not
cover
phonograms
or
performances
PRS but does
not cover
phonograms
or
performances
MTL Licenses Not available.
SACEM offers a
license limited
to specic
territories but
not truly MTL.
No information
available from
other CMOs
Not available.
GEMA offers
an online
tariff covering
Germany and
some limited
multi-national
licenses
but not
Europe- wide.
No online
tariff by GVL
Not available.
STIM offers
some
limited MTL,
especially for
Scandinavia.
No
information
available
from other
CMOs
Not available.
PRS/ MCPS
offers a
license limited
to specic
territories but
not truly MTL.
No
information
available for
PPL
Tariff
available
online
Only SACEM
for limited
territories
Only GEMA
for limited
territories
Only for
STIM for
limited
territories
Only for
PRS/ MCPS
for limited
territories
Indemnity for
Licensees
(extent to
which non-
members are
covered)
No Limited
(presumption
of
management)
No Limited
(presumption
of
management
in some cases)
Information
available in
English
Partial (does
not include
substantive
licensing
information)
Partial (does
not include
substantive
licensing
information)
Yes Yes
23
A second notable insight from the two tables
is the similarities across the countries and
hence the relevant CMOs. If competition
was driving new or better licensing services,
one would expect to see more variation.
G. The Directive and the
limitations of competition
24
It would appear from our research that the Directive
not only offers no solution, but in fact worsens the
problem in some areas. The reasons can be found
in its inadequate conceptualisation of copyright,
especially its dynamics and the interests involved.
In fact, in its current form, it is likely to make the
situation more difcult for the majority of authors
and users, only really benetting a small group of
large right holders.
I. The User Lost in the Labyrinth
25 The real losers of the changes are the users in the
online environment. Having blanket licenses, as
broadcasters do, means that most of the identication
costs associated with licensing is carried by the CMO.
They have to identify the relevant right holders, and
they have to transfer revenues to sister CMOs for the
repertoire that is used. In the digital environment,
the cost is shifted entirely onto the user. Online users
however, have to identify the relevant right holders
because CMOs are not able to offer blanket licenses.
Instead, the user needs to contact a large number of
CMOs, hubs aggregating the repertoire of different
CMOs, and even individual right holders.
26 Where the local CMO cannot provide licenses with
multi-territorial cover, the user has to contact the
CMOs in all member states as well as those right
holders that have withdrawn their rights.45 This
poses major problems for all aspects of the licensing
process. First, there is the problem of identifying
the repertoire which requires an additional license
and the right holders associated with it.46 As we
have seen, statements about the scope of the
repertoire and rights managed by the CMO can
be very vague, rendering it difcult to tell what is
and is not included. Databases, such as CIS-Net, are
45 An alternative would be to “boycott” songs which were not
obviously covered by readily available licences. This may
lead to either pressure from the rights holders of those
songs to have them included or to migration of those rights
holders to another CMO.
46 How big the problem is depends on the activism of the
rights holders. If they are very active users of the services,
they will identify which CMO offers the best home in terms
of repertoire and shift their licences to that CMO. More
generally it is important not to treat the rights holders as
passive actors.
2016
Morten Hviid, Simone Schroff and John Street
264
3
not publicly accessible.47 As a result, the only way48
a user could guarantee a multi-repertoire, multi-
territorial blanket style license would be to contact
all CMOs or the necessary combination of hubs and
CMOs. This difculty is exacerbated by the absence
of authoritative and complete lists to identify
CMOs, especially across borders. Our research
revealed (especially the second row of Table 2) that
it is difcult to identify all relevant CMOs without
resorting to the academic literature a resource
which is not easily accessible to the general public. In
fact, determining which CMOs need to be contacted
has proved to be yet more complex because the
information provided is vague. A potential user has
to read around the topic, relying on blogs and similar
searches. While this may work in practice, the lack
of veriable information is a source of concern.
Furthermore, while CMOs provide signicant
amounts of information on their homepages, it
tends to be in their national language. In cases where
sections have been translated into English, they are
often signicantly smaller. In particular, translations
of licensing forms are not available. By comparison
(Table 1), broadcasting tariffs are clearly accessible
and explained by all the relevant organisations on
their websites.
27
Given the complexity of the task, users can never
be sure if they have actually covered everything
and potentially expensive infringement claims
remain a possibility. Given the problem of securing
the necessary complete clearance, one might
reasonably wonder whether it would be better not
just for the user, but possibly also overall, if it was
accepted that there might be occasional copyright
violations but that these would be resolved through
court settlements.
49
The key issues here are: what
the costs and nes are in cases of infringement;
whether the ne is proportional to the loss suffered
by a rights holder; and whether the latter ought to
have a duty to make it clear which CMO or other
vehicle is used for revenue gathering. The extent
of the damages depends too on the type of business
requiring the licence. In the case of YouTube-style
ones, they will be told to take something down. If
they do not comply reasonably fast, then they will
be held liable. If the service is a Spotify-style one
47 An interesting question is whether an exclusion could
be challenged on competition grounds as an abuse of
dominance. The databases may be seen as essential facilities
to which some users might be able to force access in return
for a reasonable fee.
48 Given the current set-up. As Katz points out, there are
alternative solutions if the rights holders and publishers
are ready to embrace them. Inspiration for this could
potentially be drawn from the e-book market. (Katz, The
potential demise of another natural monopoly: Rethinking
the collective administration of performing rights, Journal
of Competition Law and Economics 2015, 541-593.)
49 See Viacom International, Inc. v. YouTube, Inc., No. 07 Civ.
2103.
(i.e. no user uploads), then the service would be
liable straight away as licenses have to be sought
before the service is made available. One question
is whether the strategic re-assigning of rights can
be used to deceive or trick users in order to cash in
later, essentially by acting in a manner equivalent
to patent trolls.
28
There is very little information available on how
much licensees have to pay as a result of licensing
disputes. Most disputes are settled out of court and
the details are kept condential, even if they involve
a large number of plaintiffs complaining against
a licensing fee.
50
One of the few exceptions is the
example of NSM Music which was ordered to pay
£85,000 plus interest and legal costs after it lost a
licensing dispute with PRS for Music.
51
However, the
claims involved in these cases are substantial. In the
long-running dispute between the German GEMA
and YouTube, the demands reached €1.6 million for
the infringement of 1,000 songs that were uploaded
by users without consent.
52
In fact, it demanded 0.37
euro cent for each time a song is played.53 It is easy
to see how this could lead to very high costs once
the provider is found liable.54
II. Everyone Loses Out: the Income
29 The complexity of the current system is also likely
to lead to an overall lower licensing income, simply
because users cannot manoeuvre the system
efciently. As a result, they either do not offer a
service on the scale they would prefer, or they do
not pay all rights holders as they should.
30
Looking at the system in practice claries this.
Today, those CMOs that are able to offer truly multi-
territorial licenses are managing Hubs. Hubs refer
to the separate legal entities founded by a (large-
50 ITV et al v PRS and MCPS, Consent Order by the Copyright
Tribunal (cases CT 117,188, 199) (available at
bailii.org/uk/cases/UKIntelP/2012/o011911.pdf>, last
accessed 15/4/16).
51 PRS for Music, NSM Music ordered to pay PRS for
Music license fees 2011 (
com/aboutus/press/latestpressreleases/pages/
nsmmusicorderedtopayprsformusiclicencefees.aspx>, last
accessed 15/4/16).
52 LG München I: Keine Haftung des Plattformbetreibers für
Urheberrechtsverletzungen, MMR 2015, 831. The revision
was also turned down: OLG München: YouTube schuldet
GEMA keinen Schadenersatz, MMR-Aktuell 2016, 375539.
53 Schadensersatzprozess: Gericht weist erneut Gema-Klage
gegen YouTube ab, Der Spiegel 28 Jan 2016 (
spiegel.de/netzwelt/netzpolitik/youtube-gema-verliert-
vor-olg-muenchen-a-1074418.html>, last accessed 15/4/16).
54 It should also be noted that YouTube was not found liable in
this case due to secondary liability issues. The fee demand
itself was not determined as unreasonable.
Regulating Collective Management Organisations by Competition
2016
265
3
scale) right holder for the purpose of licensing.
Most of them cooperate very closely or are even
managed by one or more CMOs. As a result, these
CMOs are able to license this repertoire in addition to
their own repertoire. However, the CMO which had
originally held these works will not be able to issue
a license anymore.55 Major publishers have bundled
their rights in these Hubs but the repertoire is not
universal. Instead, repertoire coverage is divided
by publisher or even by sections of a publisher’s
repertoire (for example, Latin-American or Anglo-
American music). The management of Hubs overlaps
so that specic CMOs are able to license rights of
more than one repertoire. For example, PRS for
Music in the UK is involved in “Peer Music Publishing
Anglo-American repertoire, Imagem Anglo-American
repertoire, IMPEL Anglo-American repertoire, CELAS and
SOLAR56 (EMI and Sony/ ATV Anglo- American repertoire)
and Warner Chappell Music Publishing repertoire as a
PEDL partner”.
57
However, depending on what type
of repertoire the user requires, it is likely that they
will have to contact more than one CMO to cover all
the required rights. The multi-repertoire license has
been sacriced for multi-territorial coverage as the
licenses do not combine both.
31
In addition, these Hubs cover only musical works. All
other types of works for which licenses are required
cannot be cleared this way. Record labels which
own the rights in the performance and phonogram
usually manage their rights individually, but on
a multi-territorial basis. The exception is Merlin
which licenses for a range of Independent labels on
a multi-territorial basis.58 There is also some limited
55 Exploitation contracts that CMOs have with the right holder
do usually require the exclusive assignment of rights. The
Directive does not actually prohibit that. As a result, when
the rights are withdrawn, they are usually withdrawn
entirely, meaning that the original CMO does not manage
them anymore. The right holder is still able to license non-
commercial uses directly (article 5(3)) but these are not
relevant for this study. It should be noted though that if the
MTLs are based on the passport system, meaning that CMO
1 has mandated CMO 2 to manage the online licensing under
article 29 of the CMO Directive, then these agreements are
not exclusive. However, this is not the case for the HUBs
discussed here as these have the rights entrusted to them
directly.
56 SOLAR combines the Hubs from PAECOL (GEMA) and CELAS.
GEMA, Sony/ ATV Launches Joint Venture with PRS for
Music and GEMA (
sonyatv_launches_joint_venture_with_prs_for_music_
and_gema-1/>, last accessed 14/9/15).
57 PRS for Music. 2015. “Multi-Territorial Licensing” (
www.prsformusic.com/users/broadcastandonline/
onlinemobile/multiterritorylicensing/Pages/default.aspx>,
last accessed 10/9/15).
58 Merlin is a rights clearance organisation that manages the
rights on behalf of independent labels. In difference to other
organisations in this area, its licenses cover more than one
territory. In other words, the user can license the rights held
by many different independent labels in a one-stop-shop
by contacting Merlin. They do not have to go back to the
cooperation for cross-border licensing among the
CMOs in this area. For example, GVL, the German
CMO for performances and phonograms, offers
multi-territory licenses, but these cover only 20
member states
59
and is therefore not sufcient for
EU-wide clearance, which for example Europeana60
requires. Europeana only accepts works which will
be accessible in all EU member states.61 As a result,
it would be necessary in most cases to contact the
record label in order to clear the rights in the records
and performances; contacting the CMOs alone would
not be sufcient as they cannot provide adequate
MTL coverage. Finally, there is still no authoritative
list of Hubs and CMOs and of which works and rights
are covered, making the process more laborious.
62
As a result, the MTL licensing of musical works is
entirely divorced from other related rights, even
when they are intrinsically linked - such as musical
works and performances.
32 In practice, nding Hubs takes a signicant amount
of effort in practice. They are not prominently
featured or promoted by the CMOs. There is also no
database or similar facility to help users determine if
there is a Hub able to provide them with the license
they seek. Furthermore, even these projects are
very limited in scope. In fact, most focus is on the
Anglo- American repertoire. These Hubs also do not
have separate homepages with licensing facilities
that can be contacted directly; they are managed by
the CMOs. Thus, the number of potential actors has
increased, rather than decreased - another step away
from the one-stop-shop that broadcasters enjoy.63 If
musical works cannot be licensed, incomes cannot
be generated and therefore the incentivising effect
of copyright is itself weakened.
labels. It is an issue though that the actual membership is
not known and therefore may not represent a specic Indie
label in question. (, last
accessed 10/9/15).
59 GVL, Länderliste Web_radio (available at
de/rechtenutzer/webradio/laenderliste-webradio>, last
accessed 10/9/15).
60 Europeana is the common gateway where users can access
materials digitised and hosted by European cultural
heritage institutions. It can be accessed here:
europeana.eu/portal/>. The example is used here because it
has been actively promoted at EU level.
61 GVL, Länderliste Webradio (available at
de/rechtenutzer/webradio/laenderliste-webradio>, last
accessed 10/9/15).
62 The CMO Directive does envisage such a list and requires
the Commission to make it public. However, this has not
happened yet. (Directive 2014/26/EU) (CMO Directive), art.
39.
63 This highlights the fundamental trade-off between the
greater convenience of dealing with a single rm, a
monopoly, and that with a monopoly where there is no
competition. A similar dilemma has in the past arisen in
the case of “yellow pages”, where both advertisers and
consumers would prefer a single provider so long as that
provider did not abuse its monopoly position.
2016
Morten Hviid, Simone Schroff and John Street
266
3
33 This is made even worse in practice as the Directive
omits a key part of the licensing process. Licensing
music is a direct result of copyright law, especially the
right to control the public performances of musical
works and sound recordings. The Directive only sets
licensing standards for the multi-territorial licensing
of musical works. However, from a copyright
point of view, performing a work in public, such
as streaming or broadcasting it, requires a license
covering the performance and the recording of the
work. These are considered neighbouring rights and
administered by a distinct and separate set of CMOs.
34
The clearest indication of this is the lack of
streaming tariffs via CMOs for neighbouring rights.
For example, in Germany a broadcaster needs a
license from GEMA for the musical work and from
GVL for the performance and the sound recording.
For streaming the situation is more complicated and
more fragmented. The GVL, for example, does not
offer a streaming tariff on its homepage and in fact
also does not mention how to acquire the license
in practice. This means that a user has to contact
the right holder directly - a very onerous process
in practice, given the large number of record labels
and other right holders involved. The situation is not
any different in the other member states; in all our
cases the access to neighbouring rights for online
exploitation is limited in comparison to analogue
uses (such as broadcasting). In this respect, it is
unrealistic to expect users to acquire the correct
license in a system that is vague, highly complex and
unable to meet the demand. The failure of licensing
practices to change quickly enough could actually
harm the aim of copyright as a whole.
III. The Freedom of the Right Holder
35
For the user, the fragmentation of the rights is
the root of the problem. However, the Directive
explicitly allows copyright holders to split their
rights into bundles, based on the type of right and the
territorial scope. This results in a worsening of the
situation: the administration of rights has become
increasingly fragmented.64 In particular publishers
and record labels can now administer their rights
themselves, having withdrawn them from the
CMO system.65 However, they have not withdrawn
64 Cooke, Dissecting the Digital Dollar, Part One: How
streaming services are licensed and the challenges artists
now face (London: Music Managers Forum., 2015)
65 Arezzo , Competition and Intellectual Property Protection in
The Market for the Provision of Multi-Territorial Licensing
of Online Rights in Musical Works – Lights And Shadows
of The New European Directive 2014/26/EU, International
Review Of Intellectual Property And Competition Law
2015, 545, Directive on collective management of copyright
and related rights and multi-territorial licensing of rights
the works as a whole, but instead only the online
rights. In other words, while the CMO may be able
to license the work for broadcasting, it cannot do so
for online exploitation. This fragmentation places
a substantial burden on CMOs and right holders to
keep track of who holds what right to which work.
This task should not be underestimated.66 Some
CMOs themselves struggle to identify the specic
works and rights that they administer.67
36
Secondly, by allowing not only CMOs but also
independent rights management organisations
(which focus on licensing without the collective
component)68 to administer rights, the Directive
has effectively endorsed the licensing Hubs. Given
the demand for multi-territorial licenses, CMOs
have had to cooperate with each other and with
major publishers to offer multi-territorial licenses.
While these Hubs are managed by the CMOs, they
are distinct from them. This means that rather than
competing with each other to offer multi-territorial
licenses, CMOs are being hired by right holders to do
this via a clearing house system. It also means that
the usual social and collective features of the CMO,
a key element in the justication of their existence
is being marginalised.
37
In sum, the remedy which was supposed to bring
about CMOs to provide multi-territorial licenses
has instead cemented a fragmented system where
it is not clear what a license covers. Right holders
have got the power to choose where to register their
rights. Their decision will be determined both by
the nature and offerings of those who are willing
to have the rights registered to them and the users
of the services of those organisations, such as the
streaming services themselves and their consumers.
While it is unhelpful to look at this market through
the lens of the theory of two-sided markets, it is
important to keep in mind that to achieve the best
in musical works for online use in the internal market
(Directive 2014/26/EU) (CMO Directive), Part III.
66 In this respect some may argue that the CMOs are simply
left with the wrong “technology”, i.e. databases, which may
make the intervention by the Commission look harsh. An
unresolved question is whether the CMOs have been less
innovative than on other sectors because they were part of
a set of cosy monopolists or because there are some issues
which make it fundamentally harder to bring music into the
21st century.
67 Ranaivoson, Iglesias, and Vondracek, The Costs of Licensing
for Online Music Services: An Exploratory Analysis for
European States. Michigan State International Law Review
2013, 674.
68 CMOs license works and use some of their income for
services to the membership as whole, including social
insurances, pensions and cross-subsidising of genres. Rights
management organisations license works and distribute the
income to the right holders, without providing broader
services like CMOs do. As a result, the cross-subsidising from
successful to less successful right holders is signicantly
more limited.
Regulating Collective Management Organisations by Competition
2016
267
3
nancial outcome for the rights holders, creating
appropriate bundles of music is clearly valuable.
69
In other words, there is a natural tendency to have
CMOs that cover all works and represent all rights.
IV. Who Benefits?
38
In addition to the problems of rights fragmentation,
there are questions concerning the benet to
be derived for the majority of authors. The clear
winners of the changes are successful artists and
large right holders, such as publishers and labels.
They have the resources to administer their rights
on their own.
70
This trend has been most recently
conrmed by Arezzo who sees the publishers as
exploiting the new options.
71
Withdrawal of rights
in order to ensure efcient administration is not a
realistic option for most right holders, a problem
that is compounded by the fact that CMOs are not
required to use a common language.
39 In addition to the practical and technical issues not
addressed by the Directive, the Commission has
69 Arezzo , Competition and Intellectual Property Protection in
The Market for the Provision of Multi-Territorial Licensing
of Online Rights in Musical Works – Lights And Shadows
of The New European Directive 2014/26/EU, International
Review Of Intellectual Property And Competition Law 2015,
534-564.
70 Ficsor, Collective Management of Copyright and Related
Rights, WIPO 2002 (available at:
edocs/pubdocs/en/copyright/855/wipo_pub_855.pdf>,
last accessed 15/4/16), 97,Handke and Towse, Economics
of Copyright Collecting Societies, SSRN 2007 (available
at:
id=1159085>, last accessed 14/12/15), 10. They also have
more lobbying power and it is important to be alert to
the dangers that such lobbying power leads inappropriate
regulation and potentially slower convergence.
71 Arezzo , Competition and Intellectual Property Protection in
The Market for the Provision of Multi-Territorial Licensing of
Online Rights in Musical Works – Lights And Shadows of The
New European Directive 2014/26/EU, International Review
Of Intellectual Property And Competition Law 2015, 534-564.
For early predictions of this phenomenon, see Kretschmer
et al, The Changing Location of Intellectual Property Rights
in Music: A Study of Music Publishers, Collecting Societies
and Media Conglomerates, Prometheus, 1999, 163- 186; the
issue raised during the public consultation: Max Planck
Institute for Intellectual Property and Competition Law,
Comments on the Proposal for a Directive of the European
Parliament and of the Council on collective management of
copyright and related rights and multi-territorial licensing
of rights in musical works for online uses in the internal
market (available at:
cfm?abstract_id=2208971&download=yes>, last accessed
13/10/16), especially para. 17; and evidence that this is
already happening: 1709 Blog, Is Universal Publishing’s
exit from collective licensing a step backwards for music
industry ‘one stop’ aspirations? (available at:
the1709blog.blogspot.nl/2013/02/is-universal-publishings-
exit-from.html>, last accessed 13/1/16).
a highly simplistic view of author preferences. It
does not allow for how interests within the right
holder group may differ. Larger right holders have
an interest in leaving because, for them, economic
performance is key.72 Successful artists and
commercial copyright holders have an interest in
generating revenue compared to a less successful
artist who may rely on a wider distribution of their
works in order to generate a fan base.73 In terms of
rights administration, this translates into the larger
owners preferring efciency above other services
that CMOs provide (for example, social insurance).
40
Following the Commission’s logic, relying
on increased competition protected through
competition law can make CMOs focus both on
generating faster, more accurate practices, as
well as lowering overheads. However, there is no
accepted measure for CMO performance74 and
therefore neither for “efciency”. The one gure
indicating the cost of rights administration for
the copyright owner is the administration rate. It
measures the percentage of royalties that are used
for administration and indicates its relative cost.
This is the only directly comparable gure which the
CMO Directive requires to be published.
75
Therefore,
for copyright holders focusing on economic value a
lower administration rate is more attractive.
41
However, the reliance on administration rates
has two major drawbacks. First, in a world where
there is a choice between CMOs, this would seem
an inadequate measure of performance. Having a
measure which focuses solely on the cost side is
rather limited, since an artist is interested in the
absolute amount of money they receive. To be
satised with the current measure would mean
72 This is a well-known problem for cooperatives – and at least
for some aspects of the business model, one can equate a
CMO with a marketing cooperative. When cooperatives
have members with very diverse interests and aims, the
cooperative tends to malfunction and the more powerful
members tend to leave as they can do better on their own.
See e.g. Henriksen, Ingrid, Morten Hviid and Paul Sharp,
2012, Law and peace: Contracts and the success of the
Danish dairy cooperatives. The Journal of Economic History
72, 197-224.
73 Kretschmer, Digital Copyright: the End of an Era. European
Intellectual property Review 2003, 333-341.
74 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15), 6.
75 Directive on collective management of copyright and
related rights and multi-territorial licensing of rights
in musical works for online use in the internal market
(Directive 2014/26/EU) (CMO Directive), art 22 and Annex.
All of the other indicators which need to be published are
in absolute numbers, making them not directly comparable
across CMOs. For example, the collected revenue strongly
depends on the membership size, making the absolute value
in Euros a relative gure.
2016
Morten Hviid, Simone Schroff and John Street
268
3
always preferring a CMO which had low costs, but
which generated very little revenue, to one with
high costs but also high revenue.
42
Secondly, if CMOs choose to compete, as the
Commission intends, it would be on the basis of
the administration rate as an indicator of economic
efciency. This would attract the right holders with
the most valuable repertoire. The administration fee
is currently the same, irrespective of the actual cost
of collection. However, as more successful works are
easier to administer in practice, larger right holders
are cross-subsiding less successful ones.76 They
therefore have an incentive to leave and as a result,
the cross-subsidy is likely to unravel.
77
CMOs seeking
to prevent this are more prone to the inuence of
these larger right holders. As their threat to exit is
also the most credible, it will enhance their inuence
within CMOs.78 As CMOs have in practice signicant
leeway in determining both the tariffs as well as the
distribution policies,
79
smaller right holders are more
likely to be losing out.
43
A possible casualty of a more economic/competition
approach in this market is the demise of the social
and cultural features of the old CMOs. These required
a cross subsidy between artists. With the focus on the
economic value of the organisation, the incentive
to provide these subsidies will decrease. CMOs with
a stronger social component would be left with
repertoire of a lower market value, raising the
costs per work even more.80 At the same time, it is
hard to see the justication for these services being
bundled with the other activities of a CMO and being
protected through competition law. Channelling the
funds from online exploitation and bypassing the
established CMO system is likely to work in the same
way.
44
This situation feeds back into one of the main
issues raised by the effect of copyright. Copyright
protection, and especially its strengthening, is
usually linked to the harm it does to creators, rather
76 Wallis, Kretschmer and Klimis, Contested Collective
Administration of Intellectual Property Rights in Music- The
Challenge to the Principles of Reciprocity and Solidarity,
European Journal of Communication 1999, 14-15.
77 Competition typically leads to an unravelling of cross
subsidies.
78 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15), 10.
79 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15), 6.
80 Handke and Towse, Economics of Copyright Collecting
Societies, SSRN 2007 (available at:
com/sol3/papers.cfm?abstract_id=1159085>, last accessed
14/12/15), 10.
than to the larger corporations which do not create
works, but exploit them. This concern derives from
the assumption of “the romantic author”: the lone
creator who works independently.81 This paradigm
is further reinforced by the language used to
describe unauthorised use, most notably the moral
condemnation of piracy.82 A similar argument has
been made in relation to the term “extension for
performers”. Famous artists, such as Sir Cliff Richard,
have actively lobbied on this basis.
83
However, as the
licensing regime moves away from income from
shared performance rights as CMOs guarantee,84
the benets to the creator are further undermined.
The fear is that the regime is increasingly serving
the interests of the large stakeholders, whether
corporate or individual.85
H. Conclusion
45
Our empirical investigation clearly shows that the
current system in place for online music licenses
is falling signicantly short of the Commission’s
aims.86 First, it is nearly impossible to determine
who can offer an online license, and which works
and territories it covers. The information asymmetry
faced by users has been made even more problematic
by Hubs with limited coverage because it increases
the number of relevant players. (This issue has been
81 For a more detailed description, see Rose, Authors and
Owners (London: Harvard University Publishing, 1993)
and Campbell, Authorship, incentives for Creation and
Copyright in the 21st Century, Proceedings of the American
Society for Information Science and Technology 2006.
82 Ricketson and Ginsburg, International Copyright and
Neighbouring Rights- The Berne Convention and Beyond
(Oxford: Oxford University Press, 2006), 21.
83 Atkinson, Sir Cliff Richard’s victory: an extra 20 years of
copyright protection for sound recordings is only weeks
away (available at:
co.uk/2013/08/sir-cliff-richards-victory-an-extra-20-
years-of-copyright-protection-for-sound-recordings-is-
only-we.html>, last accessed 17/12/15).Cliff Richard does
not write songs, he only performs them. This makes him a
performer but not an author under copyright law. However,
it shows how the notion of creativity has expanded over
time.
84 Most commonly, the income is divided 1:1:1 between the
composer, lyricist and publisher, with payments directly to
the right holder.
85 For a detailed empirical analysis of copyright reforms from
this angle, see Schroff, The evolution of copyright policies
(1880-2010): a comparison between Germany, the UK, the US
and the international level. Doctoral thesis at the University
of East Anglia 2014 (available at
ac.uk/49708/>, last accessed 15/4/16).
86 The same is true for the sales of digital music, see Gómez
and Martens, Language, Copyright And Geographic
Segmentation in the EU Digital Single Market for Music
and Film, JRC/IPTS Digital Economy Working Paper 2015,
(available at , last
accessed 15/4/16).
Regulating Collective Management Organisations by Competition
2016
269
3
known for at least a decade, yet no obvious solution
has emerged). Secondly, the price of licenses is also
unknown. While a system of tariffs is supposed
to reduce the transaction costs by addressing the
information asymmetry, this is not the case for
online licenses. Standard online licenses are not pan-
European. At the same time, there is virtually no
information available on the cost of pan-European
licenses as granted by Hubs. Thirdly, rather than
competing with other CMOs, they are hiring out
their administrative capabilities to large scale right
holders, in particular publishers. All of the major
Hubs are associated and run out of the ofces of a
major CMO, in particular PRS, GEMA, SACEM and
SGAE. Their changes are not aimed at the individual
creator but instead large intermediaries. As these
Hubs are separated from the CMOs, the revenue
they generate is separate too, and may therefore
not contribute to the social/cultural aspects of
the CMOs’ work. In other words, CMOs are helping
large right holders to channel income past the
established system. As the major CMOs are already
complying with the CMO Directive’s provisions on
multi-territorial licensing, we are left to ask: what
is wrong with the EU’s attempt to meet the demands
of digitalisation?
46
The current insistence on rights being entrusted
by the right holder to a single CMO exacerbates
the problems. Right holders are unable to create
competition through multi-homing. As Katz argues,
CMOs were not necessarily natural monopolies under
the analogue regime and are even less likely to be so
under the new digital regime.87 Some components,
such as the databases of works and right holders
may be, but the collection of revenue and the single
assignment of rights clearly need not be. Because
there is a strong commercial interest on the part of
all stakeholders to have a comprehensive CMO at
least within genres monopolies are likely to emerge
naturally. It is difcult to see how competition will
remain. Whether this will ultimately lead all to be
in the same organisation or bodies organised along
the lines of a particular repertoire is difcult to
predict. One thing which seems abundantly clear is
that national organisations are unlikely to survive.
By allowing right holders to assign their rights in
any way they want, but not permitting simultaneous
assignment,88 the result is likely to be a new system
87 Katz, The potential demise of another natural monopoly:
Rethinking the collective administration of performing
rights, Journal of Competition Law and Economics 2015,
541-593, Katz, The potential demise of another natural
monopoly: New technologies and the administration
of performing rights. Journal of Competition Law and
Economics 2006, 245-284.
88 Under article 31 CMO Directive, simultaneous assignment
is possible in the very limited circumstances that the CMO
which usually administers the online use of works does
not offer multi-territorial licenses and has not mandated
another CMO to do so under the passport system. However,
of monopolies or oligopolies. The only difference
will be the basis of the distinction, from national
monopolies to repertoire-based ones.
47
Our reading of the Directive and our case studies
suggest that:
By mis-conceptualising CMOs, the remedies to
ensure more competition have had unintended
effects – for instance, the creation of clearing
houses managed by CMOs rather than
competition between CMOs;
The Directive does not go far enough rights
are still assigned on an exclusive basis and
therefore cannot be assigned to several agents
at the same time;89
In the matter of non-exclusive rights assignment
several CMOs can license a work, so the user is
not detrimentally affected; at the same time,
right holders can exclude some badly managed
CMOs, while remaining within the licensing
regime.
48
Our research has also enabled us to identify a number
of further questions:
Given that licensing is intimately linked to
the copyright system, should the copyright
system be reformed to accommodate changes
in licensing - in particular, for the protection of
consumers and less successful authors?
Are performing rights and their licensing really
different from other works and rights (for
example, e-books)?
What problems should a reformed licensing
system address? Is streaming equivalent to
other disruptive technologies and/or initiatives
in other markets such as Uber and Airbnb?
49
Given the importance of licensing practice to the
even in this case the multi-territorial online use cannot
only be assigned to one other CMO. It is therefore still a
single CMO which can provide the license in practice.
89 It is not required by the Directive that the same right for
the same work is assignable to more than one CMO. Indeed,
exploitation contracts explicitly prevent this. See for
example: BUMA/ STEMRA, Exploitatiecontract A (auteur)
(available at:
uploads/2015/05/PV2.BUM_.512.0914.08-A3-SPEC-
Exploitatiecontract-A-auteur-def.-d.d.-03.10.2014.pdf>, last
accessed 13/10), art 2(3) or GEMA, Berechtigungsvertrag
(Fassung April 2016) (available at:
leadmin/user_upload/Gema/Berechtigungsvertrag.pdf>,
last accessed 13/10/16), art. 1 and 1a; PRS for Music, Articles
of Association (available at:
com/SiteCollectionDocuments/About%20MCPS-PRS/prs-
memorandum-articles.pdf>, last accessed 13/10/16), art. 7.
2016
Morten Hviid, Simone Schroff and John Street
270
3
legitimacy and effectiveness of copyright more
broadly, linking the two directly at EU level is a
potential avenue of fruitful reform. Although beyond
the scope of this paper, future research should
investigate how the effect of copyright is shaped
by the licensing process. Key to this is the current
absence of copyright contract law rules to cushion
the effect of changes in the licensing practices for
less successful artists. Furthermore, research should
investigate the option of resorting to harmonisation
(potentially in combination with a re-adjusted
competition approach), as was done in areas of
protection to standardise licensing practices and
the availability of licenses across borders. Examining
the effects of copyright in this context is especially
important, given the on-going EU copyright review.
* Morten Hviid is the Director of the Centre For Competition
Policy, a member of CREATe and UEA Law School,
University of East Anglia, Norwich NR7 4TJ, UK.
Simone Schroff is an associated researcher with
CREATe and a member of the Institute for
Information Law, University of Amsterdam,
Vendelstraat 7, 1012 XX Amsterdam, The Netherlands.
John Street is a member of CREATe, the Centre for
Competition Policy and the School of Politics,
Philosophy, Language and Communication Studies,
University of East Anglia, Norwich NR7 4TJ, UK.
We would also like to acknowledge RCUK Centre for
Copyright and New Business Models in the Creative
Economy (CREATe) for funding this project. Any remaining
errors are ours.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT