M E D I A   L A W   

      I N T E R N A T I O N A L   ®


Copyright © Media Law International 2018. All Rights Reserved.

Specialist Guide to the

Global Leaders in Media Law Practice

1. Music Streaming as the New Frontier of Digital Music Distribution

Music streaming can yet be seen as an evolution of download, in that both require the availability of a hardware infrastructure upstream for the hosting of digital copies of musical works. The two diverge downstream, as streaming requires no permanent memorisation of the file on the user’s device.

Instead, packets of compressed digital data are received and temporarily stored (copied) in “buffers” (i.e. small portions of DRAM memory) of the recipient’s device, so that they are in almost real time decompressed, decoded and sent to a sound card, which emits sounds audible as music to the ear.

Therefore, differently from download, the playing of the file happens contextually, with no significant latency between the point in time the user clicks on a track and when the playing starts. Streaming is technically differentiated between live (or webcasting) and on-demand streaming. For the purpose of this paper, only the latter is considered.

The advent of streaming, made possible by the emergence of more effective storing and transmitting technologies, coupled with the increase of bandwidth, has dramatised the phenomenon of music dematerialisation, i.e. the complete separation between the ownership of a physical support and the enjoyment of music.

The streaming-led dematerialisation can also be viewed as a paradigm change, whereby users no longer buy music as a product but rather have access to a service. With streaming, the service component may be said to be as important as (if not to prevail over) the product component.

From a remuneration stand-point, the concept of streaming encompasses two main different models. The differentiation in this respect runs between the “premium” offering, whereby users subscribe for a periodic fee and have access to the full version of the service, and the “freemium” proposition, which is entirely adverting supported, so that users accept to expose themselves to advertising messages interrupting the flow of music, coupled with a usually impoverished version of the service.

Another taxonomical clarification is necessary with respect to the concept of “music streaming service providers”. Namely such a category should be taken to embrace not only pure musical content providers, like Spotify, Deezer and Tidal that control whatever is uploaded and streamed (henceforth referred to as “pure players”), but also music streaming services, such as first and foremost YouTube, that are intended as platforms for users to upload the content they wish, the so-called User Uploaded Content (“UUC”). UUC may consist firstly of entirely original content, in the form of the users’ own creations.

However a second form of UUC consists of either unauthorised wholesale reproductions of third parties’ content, such as songs, albums and music videos, or of derivative works, i.e. user generated content using parts of a third party’s protected work such as, for example, musical soundtrack for an otherwise original footage. It is this second category of UUC that is relevant for the purposes of this paper, because it represents for users an alternative to accessing music via pure players’ platforms.

Although YouTube has launched a paid monthly subscription service in a few countries, named YouTube Red, currently YouTube’s business model is essentially advertisement-based and does not even require users to register to it in order to access content.

2. The Economic Importance of Streaming

The 2017 Global Music Report of the International Federation of the Phonographic Industry (IFPI) shows that, over the past 15 years, the music industry’s revenues, generated in the aggregate from performance rights, synchronisation, physical and digital sales, have dropped by almost 40 per cent, with a steady declining curve since 2000.

However, the downward spiral has come to an end and the trend eventually returned to growth from 2015 and eventually in 2016 the global recorded music market experienced a revenue increase of 5.9 per cent compared to 2015 (see Figure 1 below), which the IFPI defines as “the fastest rate of growth since IFPI began tracking the market in 1997”.

Figure 1

The recorded growth is entirely attributable to digital revenues and, within that ambit, to streaming revenues in particular. When looking into the composition of the global music revenues by segment, the IFPI figures show that digital revenues (i.e. download plus streaming) in 2016 accounted for 50 per cent of the total market, with a growth of 17.7 per cent over 2015, which, significantly, was driven by a 60.4 per cent growth in streaming revenues.

Streaming of late has therefore been playing the lion’s share in the growth of the global recorded music industry, as it is signified by the observation that in the same period download revenues have indeed declined by 20.5 per cent and therefore streaming now represents the largest portion (59 per cent) of digital revenues. Streaming revenues, according to the report, may be attributed for the largest part to more than 100 million users of paid subscriptions, a number that has more than doubled, hitting 212 million, when considering the entire audience of both paid and ad-supported subscriptions.

These numbers provide the economic context to this paper and attest the centrality within the industry that streaming services have gained. They also show that the music industry’s economic crisis is far from being solved, as yet over 1/3 of its pre-crisis value is missing.

It is, however, all the most uncertain that streaming services are the remedy to the problem or whether instead, taking into account the music streaming providers’ prevailing business models, they would not only work as the sanctioning of an irreversible impoverishment of the industry at large or, to put it differently, of a stable different wealth allocation, to the detriment of artists and music labels.

In paragraph 3 below I will linger on the problems caused by the pivotal role now assumed by streaming services within the music distribution industry and I will, in particular, focus on the so-called Value Gap issue. In paragraph 4 I will then map out the legal solutions that, at EU level, are being envisaged to tackle the Value Gap issue. In paragraph 5 some conclusions will be sketched.

3. Setting the Problem: The Value Gaps

The growing popularity attained by music streaming services has attracted the scrutiny of the industry community, which have been increasingly vocal in manifesting its discontent for the mismatch between the skyrocketing consumption of music worldwide through streaming platforms and the yet proportionally very small revenues that streaming service providers return to right holders.

This is what is recognised as the Value Gap issue. This term is exclusively used within the music industry to stigmatise UUC streaming service providers, whose business model, according to critics, results in the most unbalanced distribution of resources for record labels and artists. Still the IFPI 2017 Global Music Market report dedicates a specific chapter to the so qualified Value Gap.

In it the issue is reported as one pertaining solely to UUC services, where YouTube is expressly mentioned as the player that extracts big revenues by selling advertising space in association with streams of music related UUC, while returning very little to right holders. The IFPI report does not shy away from pointing the finger at this Value Gap as the “biggest threat to the future sustainability of the music industry”.

This stance rests on a comparison between the revenues generated by subscription services and the revenues deriving from UUC streams, in both cases in proportion to the number of users. As is shown in Figure 2 below, numbers appear staggering, as total revenues that UUC streaming service providers reverse to the music community with almost 1 billion users are only 1/6 of those paid by subscription services with a user base as small as 1/4 of the former.

Figure 2

The music community holds that the unfair distribution of revenues generated by UUC streaming services is due to a flaw in the legal framework, both at EU and US level, allowing UUC streaming service providers, differently from pure players, to keep doing their business, without any need to negotiate and enter into license terms with right holders.

In Europe, the legal hole in question has been identified in the Safe Harbour (hereinafter the SH) that the E-Commerce Directive 2000/31/EC (henceforth the ECD) has erected for, inter alia, ISSPs (Information Society Service Providers) engaged solely in the hosting of information stored at the request of the recipient pursuant to Article 14 of the ECD.

These ISSPs shall be shielded from liability in the event that illegal content is uploaded, when certain conditions are met and notably that the ISSP does not have reasons to believe that illegal content is stored on its service and further that the provider, when notified that such an activity is being conducted, expeditiously removes or disables access to the relevant content.

The second condition is ordinarily referred to as the notice and take down procedure, by analogy with the (although much more detailed and specific) procedure under Section 512 of the US copyright Act.

Under the EU regime, the case law has further added that, in order for ISSPs to benefit from the SH under Article 14 ECD, it is necessary that they are neutral vis-à-vis the illegal content they store, i.e. that the host should not play “an active role allowing it to have knowledge or control of the data stored” (EUCJ ruling in the L’Orèal v. eBaycase C-324/09). In this connection, it has been observed that YouTube’s neutral role is questionable, to the extent that, inter alia, 80 per cent of watch time on the service is on content that the platform itself suggests.

Along with this, the lose terms in which the EU notice and take-down procedure is framed have been accused of not helping to define with certainty the scope for the host service providers’ exclusion of liability.

The music world has hence complained that the SH mechanism was fit during an era in which information society services were in their infancy, whereas in a mature digital market it has come of age and lends itself to be abused of by tech giants that leverage on the same SH to make money at the expense of the music community. On this basis, the community has overtly pressed the EU and US authorities for legislative reforms aimed at closing the gap.

The factual picture must be completed with a description of the Content ID tool that YouTube has put in place, as it helps grasping the nature of the Value Gap problem denounced by the music industry. Content ID is presented by YouTube as a system that copyright owners can use to easily identify and manage their content on the platform.

It is based on a database of files fed by content owners themselves, against which UUC is scanned by YouTube each time new content is uploaded. When content in a video matches a work listed in the database, right owners are informed and can decide whether to have the illegal content removed or otherwise monetize it, by having the advertising revenues directed to them. Content ID currently represents the larger source of revenues for right holders on YouTube.

In fact, according to some sources (https://thetrichordist.com), YouTube Content ID generates approximately 60 per cent of all revenues paid out by YouTube. Significantly however when the average per stream amount deriving from Content ID (USD 0.00030) is compared to the average per stream amount paid by pure players (USD 0.005), YouTube’s figures look even more like breadcrumbs, as it pays out only a fraction (7 per cent to 10 per cent) of the pure players’ bill.

Content ID requires no license royalty bearing agreements with the right owners and it is a benign concession made by YouTube, which otherwise could pay nothing at all and simply stick with the SH contemplated under the law. According to the aggrieved music community, the relationship with YouTube is therefore utterly unbalanced and unfair.

The Value Gap thus far does not provide the full picture of the debate around the fairness of the current distribution of the revenues generated from streaming services. It is also necessary to account for yet another manifestation of the Value Gap, this time denoting a fracture between artists (authors and performers), on the one side, and record labels, on the other side.

This different characterisation of the gap stems from the very meagre slice of the total revenues paid out by streaming service providers that end up into the artists’ pockets. The news abounds with artists’ complaints about the inconceivably low money they receive compared to the number of streams.

The root cause for the inequality in the revenue distribution perceived by artists has been spotted, by artists’ representative organisations, in the lack of transparency in the tripartite relationship to which artists are the weak party.

The making available of music through streaming platforms is in fact ordinarily governed by dealings, whereby record labels license-in the rights from artists; they then license-out the rights to distribute music to digital music services, which in their turn pay the agreed upon per stream revenues to the labels; the labels finally pay a cut of their profit to the artists.

Artists have no visibility of the actual terms of the financial relationship, which is intermediated by record labels, and also have difficult or no access to reliable data. Artists’ associations have put under the spotlight the opaqueness of the upstream relationship between labels and digital music services, as ensuing from the record labels’ negotiation strategy, which is centred around package deals, whereby they are thought to exact substantial down-payments from providers, in the form of attributable minimum advances for the service to secure the availability of the labels’ entire catalogue.

These advances are most likely not shared with artists and in any event there are no means for them to have access to the information, because the agreements are said by streaming services to be sealed by NDAs. Likewise NDAs also prevent authors from knowing the rates achieved for their own works.

EU authorities have been prompted to cure the two manifestations of the Value Gap by way of normative intervention. These calls have found their way in the Proposal for a Directive on copyright in the Digital Single Market of September 2016 (the Draft Directive or DD), which is illustrated below.

4. The DD’s Solutions to the Value Gaps

4.1 The Provisions Concerning UUC Streaming Services

The DD tackles the UUC streaming services related Value Gap under Article 13. As it can be inferred from the policy considerations set out in the DD’s preparatory and supporting documents, the Commission holds that the evolution of digital technologies, with the advent of Web 2.0, has turned the internet into the absolutely dominant outlet for access to copyrightable content.

However, as shown in para. 3, to date the number of users accessing content through UUC streaming players platforms pales in comparison with those accessing content on paid pure players’ platforms.

The Commission has found that this imbalance may create distortions, by producing a sort of negative umbrella effect. As a matter of fact UUC platforms, by giving access to plenty of music at no cost, detract from the value of music that users perceive and, as a consequence, pure players are compelled to keep service fees low, while the potential audience of willing paying subscribers remains comparatively limited.

The DD’s proposal is therefore based on a follow-the-money approach, as its goal is to divert part of the revenues filling the pockets of UUC platforms to the benefit of the music world. To this end, the proposed Article 13 seeks to tighten up the SH’s conditions provided for hosts under Article 14 of the ECD, as well as to clarify the right of communication to the public under the InfoSoc Directive (Directive 2001/29/EC),

with a view to ultimately deprive UCC service providers of the option of not concluding license agreements with right holders.

While referring to the DD’s full text available on https://ec.europa.eu/digital-single-market/en/news/proposal-directive-european-parliament-and-council-copyright-digital-single-market, the gist of the new provisions is summarised below.

The first innovation of Article 13 is that it creates a fourth category of ISSPs out of the three contemplated under the ECD, notably that of hosts which “store and provide to the public access to large amounts of works… uploaded by their users” (henceforth these ISSPs will be referred to as Large UUC ISSPs).

The DD provides no hint as to how the quantitative condition (large amounts of works) should be deemed fulfilled. This means that it will be an issue of fact, to be established on a case-by-case basis, and might indeed be the source of uncertainty on the actual scope of the rule.

The second important innovation of Article 13 is that Large UUC ISSPs will be required to act proactively and adopt “appropriate and proportionate measures”, such as “effective content recognition technologies” aimed at ensuring “the functioning of agreements concluded with right-holders for the use of their works” or in any event, even absent such an agreement “to prevent the availability on their services of works identified by right-holders through the cooperation with the service providers”.

Recital 38 to the DD is crucial in shedding light on Article 13. Notably the provision should be interpreted as redefining the conditions under which, under Article 14 of the ECD,hosts will be held to play an active role vis-à-vis the UUC, thereby losing the benefit of the SH.

This will be the case when ISSPs engage in “optimising the presentation of the uploaded works…or promoting them, irrespective of the nature of the means used therefor”. If the new SH conditions are not met, then ISSPs qualifying as Large UUC ISSPs, “thereby going beyond the mere provisions of physical facilities”, shall be obliged to conclude licensing agreements with right-holders.

In any event, if yet neutral hosts for the purposes of the SH, as redefined, Large UUC ISSPs shall still be subject to the obligation to take appropriate measures required under Article 13.

The newly created category of Large UUC ISSPs therefore shall always be obliged to put in place measures to prevent that protected works are illegally made available to the public. Also, the proposal seems to drastically imply that the provision by the ISSP of access to “large amounts” of UUC will per se qualify as an act of communication to the public relevant under Article 3 of the InfoSoc Directive, with the ensuing obligation for ISSPs to enter into license terms with the right holders, unless they genuinely qualify for the redefined SH under Article 14 ECD.

The proposal is prone to scepticism as to its compatibility with the existing liability regime for ISSPs under the ECD and the InfoSoc Directive.

The DD has notably chosen to intervene, through Recital 38, on the notion of “neutrality” of hosts which, under Article 14 ECD, would exempt them from liability. The importance of the Recital has been played down by supporters of the proposal as amounting to nothing more than codification of the existing CJEU’s case law on the notion of host “neutrality”.

However, this stance does not seem accurate. In particular, the Recital’s sentence “irrespective of the means used therefor” may turn out to rule out UUC providers’ exemption of liability whenever they adopt any means of content organisation, leaving the application of the SH to residual instances.

This, however, does not seem aligned with the case law, notably in the L’Oréalv eBay case (C-324/09), where the EUCJ did not divorce the notion of hosts’ responsibility from that of “knowledge and control” in Recital 42 ECD. The EUCJ did mention as examples of ISSPs’ non-neutral activity “the presentation of the content or its promotion”.

However,what seems the more law-compliant reading of the judgment is that only support tailored to individual users would qualify as a non-neutral, while generic automated assistance made available to the generality of users would not, as the latter would not imply – in accordance with Recital 42 - any knowledge on the ISSP’s part of the illegality of the content stored. Therefore, under the ECD and contrary to the statement made in Recital 38 of the DD, it is precisely “the nature of the means used” by the host that makes a difference to decide whether a service is neutral or not.

According to the DD, once neutrality is ruled out vis-à-vis the deployment of “any means” capable of optimising the content presentation, then the Large UUC ISSPs concerned will be exposed to direct liability under the existing copyright framework.

As a matter of fact, per Recital 38, they might by definition be held to go beyond the mere provision of physical facilities (which is as such considered neutral under the InfoSoc Directive) and therefore to perform an act of communication to the public under Article 3 of the InfoSoc Directive.

The above reported readings of Article 13 show that the DD might lead to a de facto surreptitious change of EU copyright law and of the ECD, hidden in a Recital, short of any consideration of the possible systemic repercussions and of any supporting case law’s interpretation legitimising such a view.

Criticism has also been voiced to the Large UUC ISSPs’ obligation to implement in any event preventive measures, as inconsistent with the provisions of Article 15 of the ECD (no general monitoring obligations) and even in contrast with the Charter of EU Fundamental Rights, because imposing a disproportionate burden on the freedom to conduct one’s business (Article 16 of the Charter), as well as end-users’ right to the protection of their personal data (Article 9 of the Charter) and to their freedom of expression (Article 11 of the Charter).

4.2 The Provisions Concerning the Artists’ Value Gap

The DD in articles 14 to 16, supported by Recitals 40 to 43, addresses what in this paper has been defined as the second facet of the Value Gap, concerned with the fairness of the remuneration that from streaming services trickles down to artists.

As discussed, this Value Gap essentially stems from two intertwined causes: the weak bargaining positions of artists, compared to the negotiating power of record labels and streaming service providers, as well as the information asymmetry for want of adequate information for artists on the exploitation of their works. The current tripartite configuration of the relationship sees artists relegated to the role of a nuisance to take care of only after the main agreement between record labels and service providers has been sealed.

The DD remedial proposal provides in the first place (Art. 14) for a transparency enhancing tool, under which artists should have access to “timely, adequate and sufficient information on the exploitation of their works and performancesnotably as regards modes of exploitation, revenues generated and remuneration due”.

Recital 41 clarifies that the implementation of this transparency obligation by Member States must take into account the specificities of each sector and to this end stakeholders should be consulted, in order to craft sector-specific requirements. To this end, the Recital encourages recourse to collective bargaining.

The second paragraph of Article 14 mandates the adoption of measures to secure effectiveness and proportionality of the transparency obligation, while, at the same time, contemplating as a de-minim is the case in which “the administrative burden resulting from the obligation would be disproportionate in view of the revenues generated…”. In this latter case Member States may adjust the transparency obligation.

The adjustment here can arguably be taken to allow for a substantial relaxation of the burden, probably to the benefit of smaller players.  

In a not dissimilar vein, Article 14 provides as an exception that the transparency obligation shall not apply when the contribution made by the relevant artists is not significant, having regard to the overall work or performance. The scope for application of this exception is not clear.

One could argue that it is intended to cater for situations in which a copyrightable asset qualifies as a work of joint authorship, for which however some of the contributions do not pass the materiality threshold for the relevant contributors to be considered as “authors” or “performers” for the purposes of the applicable copyright law.

If so, the specification would seem redundant, as existing principles of copyright law would suffice to bar those immaterial contributors from any right to remuneration for the exploitation of their works and, as a logical consequence, from the right to transparency.

Article 15 of the DD establishes a contract adjustment mechanism, which may be invoked when the remuneration originally agreed is disproportionately low compared to the revenues and benefits subsequently derived from the exploitation of the work.

This mechanism should enable artists to request additional appropriate remuneration. Recital 42 of the DD clarifies the provision, explaining that contracts for the exploitation of authors’ rights are normally of long duration, offering few possibilities for authors and performers to renegotiate them.

The same Recital implies that the contract adjustment mechanism in question, in the absence of an agreement between the parties, should consist in the recourse to a court of law or other competent authority that the Member State may identify.

Finally article 16 of the DD tackles the “fear factor”, i.e. the artists’ reluctance to enforce in court their rights against their contractual partners. The provision calls Member States to provide for an alternative dispute resolution procedure to adjudicate disputes around the transparency or the contract adjustment mechanisms. The DD does not provide any hints as to the type of ADR that should work to make artists’ light-heartedly take action against their strong counter parties.

The effectiveness of Article 16 is doubtful, as experience from other sectors shows that the weak party in a contractual relationship will simply not be willing to confront the opponent in any venue, unless and until mechanisms effectively shielding them are put in place, in the form of - for example - collective litigation mechanisms or delegation of representation to associations.

Unlike those concerning UUC streaming services, the provisions of the DD relating to the artists’ Value Gap have not attracted much debate or controversy. This should per se ring a bell, because if the strong parties in the contractual relationship had felt their privileges at risk of actually being dented they would certainly voice their concern.

Thus, the transparency obligations, if properly implemented, are a step in the right direction, because they should provide artists with means to at least make better informed decisions and also indirectly stimulate horizontal competition between record labels for the offering of better conditions

to artists.

Likewise the contract adjustment mechanism is, in abstract, a good solution that resembles well known instruments of copyright law purporting to secure artists their fair share of any increase in value of any subsequent sale of their works (the so-called droit de suite or the resale right of Directive 2001/84/EC). Yet, as mentioned, the prospects for effective enforcement of these rights remains uncertain.

From a systematic perspective, the norms of Articles 14 to 16 of the DD do not interfere with any other corpus of EU law and qualify instead as a limited harmonisation of contract law for a very specific sector, in the absence of any otherwise general harmonisation of this area of the law.

5. Conclusion

As I write these conclusive remarks I am on Spotify, listening to a Bach’s cantata of astounding beauty and once again I tell myself that streaming services are a blessing for music lovers. Yet I am part of a minority, the paying subscribers, and therefore one of the good-fellows in the eyes of the music community. The bad guys for the community are UUC platforms and their users.

The causal link between the success of UUC platforms and the impoverishment of the music industry is, however, at best uncertain, to the extent that the radical transformation that the copyrighted content industries are undergoing appears out of sight, bearing in mind the crucial role of UUC platforms for the promotion and discovery of music, most notably for independent labels, with monetisation coming from live performances, merchandising and other less traditional channels.

The EU Commission has nevertheless sympathised with the music industry’s view and veered towards the quite radical solution of Article 13 of the DD. The purely quantitative element – the “storing and providing to the public access to large amounts of works” – the solution this is based upon does not appear sophisticated enough to justify the unravelling of established tenets of the ECD and of the InfoSoc Directive.

Also, it could ironically entrench the position of the incumbent (YouTube) to the detriment of smaller players. In this connection, criticism has been voiced against the obligation in Article 13 DD for all Large UUC ISSPs to provide for content recognition technologies, regardless of whether or not they qualify for the ECD’s SH, as raising costs for small streaming providers.

On the other hand, in spite of the doubtful efficacy of the envisaged enforcement mechanisms, the DD’s Articles 14 to 16 appear to be spot-on, as they seek to adjust bargaining power imbalances under contract law, hence beyond the straits of the doctrine of dominance under competition law, in line with an established tendency in Europe.



Legal Issues in Music Streaming: The Value Gaps

Written by Massimo Maggiore,

Maschietto Maggiore Besseghini

Massimo Maggiore

Streaming is designed to enable web-based access, unconstrained by time and place, to virtually infinite music libraries, at the users’ demand. Users can listen to as much music as they wish, for as long as they wish on their devices, which work as pure music players, with their storage function relegated to a relic of the past. This basic description of how streaming works marks the main difference between streaming services and its immediate predecessor in the realm of digital distribution of music, i.e. music download.


 Middle East and North Africa

 North America


 Western Europe

 Central and Eastern Europe


 Featured Articles

 Firm Profiles

 Firm Rankings

 Order a Copy

 Enhance Your Profile


After practicing intellectual property and competition law for ten years for leading national and international law firms, Massimo Maggiore co-founded Maschietto Maggiore studio legale with Eva Maschietto in 2007. Since 2006, he has steadily collaborated with Bocconi University as an adjunct professor of marketing law. In 2014, he entered the rankings of Chambers & Partners Europe and the Legal 500 named him among the leading Italian Intellectual Property lawyers.He is the author of a number of publications and a frequent speaker at conferences regarding competition, intellectual property and IT law, in particular as it relates to copyright, open source software, trademark, retail sector, cloud computing and enforcement of competition law. Massimo is a member of the Milan Bar and is licensed to advocate before the higher courts.