In November 2007, the Songwriters Association of Canada (SAC) released a proposal for the monetization of the file sharing of music in Canada. This article attempts to determine whether or not Canada, given its international and bilateral treaty agreements, could ever adopt the SAC‘s proposal. The article approaches this analysis through the ― three-step test, which was adopted under the Berne Convention in 1971 and enshrined in the subsequent TRIPS Agreement and NAFTA; the article also analyzes whether or not the Proposal is compatible with Canada‘s obligations under the Rome Convention. The article concludes that, without amendments to the international treaties to copyright of which Canada is a part, a proposal like SAC’s could not be successfully enacted.
This article has also been published. See, Sookman, Barry. “The SAC Proposal for the Monetization of the File Sharing of Music in Canada: Does it Comply with Canada’s International Treaty Obligations Related To Copyright?.” Osgoode Hall Review of Law and Policy 1.2 (2014): 102-136. A PDF of the article is also available here.
On 16 November 2007 the Songwriters Association of Canada (“SAC”) issued a press release announcing a Proposal for the Monetization of the File Sharing of Music (the “Proposal”). The main aspects of the Proposal are the following:
- The Copyright Act (the “Act“) would be amended to establish a new right: “The Right to Equitable Remuneration for Music File Sharing”. The term “Music File Sharing” is defined in the Proposal “as the sharing of a copy of a copyrighted musical work without motive of financial gain”.
- The amendments to the Act would create exceptions or limitations (the “proposed file sharing exception”) which “would make it legal to share music between two or more parties, whether over Peer-to-Peer networks, wireless networks, email, CD, DVD, hard drives, etc.” This “new right would authorize the sharing of music with other individuals”.
- The proposed file sharing exception would apply “to the majority of the world‘s repertoire of music”.
- The proposed file sharing exception would not apply to “parties who receive compensation for file sharing”. Therefore, commercial entities such as iTunes and PureTracks would not benefit from the exception. Music sites like iTunes and PureTracks would continue to be licensed directly by creators and rights holders “and would continue to develop attractive ‘value added’ services and security features that keep them distinct from the file sharing activities”.
- Creators and rights holders would receive equitable remuneration in the amount of $5.00 per Internet subscription.
- The Proposal states that “the amount of income generated annually could adequately compensate the industry for years of declining sales and lost revenues, and would dramatically enhance current legal digital music income. Sales of physical product would continue to earn substantial amounts, albeit gradually decreasing.”
SAC stated that it sees “this model being adopted internationally” and is “working with Creators’ groups around the world to effect a global system of remuneration for the sharing of music files”. The Proposal attracted considerable debate and public comment. This article examines part of the debate: whether the Proposal could ever be implemented without violating Canada‘s international and bilateral treaties related to copyright. As explained below, the Proposal would violate the three-step test enshrined in the Berne Convention, the TRIPS Agreement, and NAFTA. It would also breach Canada‘s obligations under the Rome Convention. Accordingly, the Proposal, like other proposals to establish “alternative compensation systems”, “non-commercial use levies”, and “non-voluntary licenses” to authorize non-commercial P2P file sharing of music, could not be implemented in the manner proposed without abrogating international agreements related to copyright.
CANADA‘S APPLICABLE INTERNATIONAL TREATY OBLIGATIONS
Proposals for exceptions to copyright are evaluated for conformance with Canada’s international obligations related to copyright. For example, exceptions to create levies for private copying, which create substantially narrower exceptions to exclusive rights than the Proposal, are structured to ensure they do not abrogate international obligations related to copyright. A proposal that would permit file sharing of music would be subject to this scrutiny by the Canadian Government. Such scrutiny is hardly unusual: the German Ministry of Justice, for instance, declined to adopt a new exception to copyright with respect to non-commercial file sharing, expressly referring to the three-step test under the Berne Convention.
Whether the Proposal would comply with Canada’s obligations under the Berne Convention, NAFTA, the TRIPS Agreement and the Rome Convention is therefore of more than academic interest. Canada has ratified the Berne Convention, Rome Convention and TRIPS and is a party to NAFTA. There are 163 contracting parties to the Berne Convention, 86 contracting parties to the Rome Convention and 151 WTO Members have acceded to the TRIPS Agreement. Canada‘s most significant trading partner, the United States, along with Mexico, are parties with Canada to NAFTA. Canada cannot easily resile from or change these agreements. Further, both the TRIPS Agreement and NAFTA contain dispute resolution mechanisms which can result in sanctions for the violation of these treaties.
Under the Berne Convention, the TRIPS Agreement and NAFTA, Canada must provide copyright holders with the exclusive right to make and to authorize reproductions of musical works. Producers of sound recording must be given the exclusive right to make and to authorize reproductions of sound recordings pursuant to the Rome Convention, NAFTA and the TRIPS Agreement. Because sound recordings of popular songs intrinsically contain a musical work, Canada must comply with each of these international treaties, conventions and agreements with respect to both sound recordings and musical works. Further, Canada must provide copyright holders with the exclusive right to authorize the communication of musical works and performances of musical works to the public. Both the reproduction right and the right to communicate works to the public are relied upon by rights holders to commercially exploit music in Canada, including over digital networks like the Internet.
THE PROPOSAL‘S INCOMPATIBILITY WITH CANADA‘S THREE-STEP
Under the Berne Convention, the TRIPS Agreement, and NAFTA, Canada agreed to confine limitations or exceptions to the reproduction right for musical works to circumstances in which all three of following conditions (known as the “three-step test”) are met: (1) the limitation or exception is limited to “certain special cases”, (2) “the reproduction does not conflict with a normal exploitation” of the work, and (3) the limitation or exception “does not unreasonably prejudice the legitimate interests of the author”. Under NAFTA, Canada agreed to the same three-step test for exceptions or limitations to the exclusive reproduction right associated with sound recordings.
The three-step test has its origins in the Berne Convention (1971). Article 9(2) of the Berne Convention states the following:
It shall be a matter for legislation in the countries in the Union to permit the reproduction of such works in certain special cases, provided that such reproduction does not conflict with a normal exploitation of the work and does not unreasonably prejudice the legitimate interests of the author.
Article 13 of the TRIPS Agreement is modelled after Article 9(2) of the Berne Convention. It states the following:
Members shall confine limitations or exceptions to exclusive rights to certain special cases which do not conflict with a normal exploitation of the work and do not unreasonably prejudice the legitimate interests of the right holder.
Apart from the difference in the use of the terms “permit” and “confine” and “author” and “right holder”, the main difference between Article 9(2) of the Berne Convention and Article 13 of the TRIPS Agreement is that the former applies only to the reproduction right. The wording of Article 13 does not contain this limitation and applies to all copyright rights in works to be provided by the Berne Convention and the TRIPS Agreement. Accordingly, it applies to both the reproduction and communication to the public rights.
NAFTA goes further than either the Berne Convention or the TRIPS Agreement. Under NAFTA the three-step test applies to limitations or exceptions related to sound recordings as well as to works protected by copyright such as musical works.
Each of the three conditions of the three-step test is given a distinct meaning to avoid a reading that could reduce any of the conditions to redundancy or inutility. Further, the three conditions apply on a cumulative basis. Each is a separate and independent requirement that must be satisfied before an exception or limitation will comply with the three-step test. The failure to comply with any one of the three conditions results in the exception or limitation being disallowed. For example, an exception that is a special case may nonetheless conflict with the normal exploitation step and an exception that does not unreasonably conflict with the normal exploitation condition could nonetheless unreasonably prejudice the legitimate interests of the author.
The Proposal does not meet any of the three conditions of the three-step test. It therefore fails to comply with Canada‘s treaty obligations on three separate grounds. An analysis of the three-step test in relation to the Proposal is set out below.
FIRST STEP: THE PROPOSAL IS NOT A “CERTAIN SPECIAL CASE”
The first step, the “certain special case” limitation condition, imposes at least two, and according to some authorities three requirements. Limitations or exceptions are a “certain special case” only if they are (i) “clearly defined”, (ii) narrow in scope and reach, and (iii) can be justified on a sound policy rationale.
The word “certain” means “known and particularized, but not explicitly identified; determined, fixed, not variable: definitive, precise, exact”. The term “certain” means that under the first condition an exception or limitation in national legislation must be clearly defined. This guarantees a sufficient degree of legal certainty.
To satisfy the first condition more is required than that the exception or limitation be clearly defined. It must be also a “special” case. It must accordingly be narrow in its scope and reach. In the WTO Decision, a case which addressed whether S.110(5) of the United States Copyright Act complied with S.13 of the TRIPS Agreement, the WTO Panel defined the meaning of the word “special” as follows:
We also have to give full effect to the ordinary meaning of the second word of the first condition. The term “special” connotes “having an individual or limited application or purpose”, “containing details; precise, specific”, “exceptional in quality or degree; unusual; out of the ordinary” or “distinctive in some way”. This term means that more is needed than a clear definition in order to meet the standard of the first condition. In addition, an exception or limitation must be limited in its field of application or exceptional in its scope. In other words, an exception or limitation should be narrow in quantitative as well as a qualitative sense. This suggests a narrow scope as well as an exceptional or distinctive objective. To put this aspect of the first condition into the context of the second condition (“no conflict with a normal exploitation”), an exception or limitation should be the opposite of a non-special, i.e., a normal case.
There is debate whether the first step imposes a third, public policy requirement. Although several authorities have suggested that there is no such additional requirement, a number have concluded that a special case must be supported by some sound, social and legal-political reason to justify its application. For example, the WIPO Guide to the Berne Convention interprets the term “special case” as follows:
This means that the use covered must be specific – precisely and narrowly determined – and that no broadly-determined cases are acceptable; and also that, as regards its objective, it must be “special” in the sense that it must be justifiable by some clear public policy considerations.
Prof. Ficsor in the Law of Copyright and the Internet expresses a similar opinion:
First, the use in question must be for a quite specific purposes: a broad kind of exemption would not be justified. Secondly, there must be something “special” about this purpose, “special” here meaning that it is justified by some clear reason of public policy or some other exceptional circumstance.
Jörg Reinbothe and Silke von Lewinski in their treatise The WIPO Treaties agree that limitations and exceptions should be based on a “specific and sound policy objective”. They state that “policy areas of concern or relevance to limitations and exceptions may be public education, public security, freedom of expression, the needs of disabled persons, or the like”.
The interpretation of the term “special case” was canvassed in an opinion rendered by the Hungarian Council of Copyright Experts which addressed whether private copying from illegal sources such as P2P networks would comply with the three-step test set out in the Berne Convention. The Council held it would not. In reaching this conclusion, the Council interpreted the term “special case” as requiring a sound public policy reason:
First “step”: exceptions and limitations may only cover “special cases”. This condition is partly of a quantitative, and partly of a quantitative, normative nature. It is of a quantitative nature in the sense that exceptions and limitations may only be applied in a narrower scope, and it is of a qualitative, normative nature in the sense that there is a need for some sound, social and legal-political reasons to justify their application.
Based on the foregoing, to determine whether the Proposal complies with the first condition of the three-step test, it is necessary to determine at a minimum whether the Proposal is “clearly defined” and narrow in scope and reach, and, according to at least some authorities, whether it can be justified by a sound public policy purpose. The Proposal does not meet any of these conditions.
THE PROPOSAL IS NOT CERTAIN
As noted above, to be “certain” an exception must be clearly defined to provide a sufficient degree of legal certainty. The Proposal does not meet this test. The Proposal states that it targets online sharing of songs, including the sharing of music on P2P networks”. It also states that it targets illegal downloading over mobile phone networks. However, the Proposal defines Music File Sharing “as the sharing of a copy of a copyrighted musical work without motive of financial gain.” This definition could include all copying of musical works done without a profit motive. Examples given in the Proposal are sharing music over P2P networks, wireless networks, email, CD, DVDs and hard drives. On this view, any copying from any source and onto any media would be covered. It could also cover musical ringtones. There are real questions as to what is the intended scope of the exception. In any event, the Proposal does not comply with the requirement that an exception be a “certain” special case.
THE PROPOSAL IS NOT NARROW IN SCOPE AND REACH
More fundamentally, the proposed file sharing exception is also not narrow in scope or reach. In examining the scope of the proposed file sharing exception, it is appropriate to take into account, among other factors, the potential users who could be excepted from liability, the acts that would be excepted, and the impact of the proposed exception on other substitutable sources of music. The Proposal has the following scope:
- It would permit music file sharing by every Canadian citizen as long as it is without motive of financial gain.
- It would permit sharing of both musical works and sound recordings, since both rights subsist in music that is reproduced electronically.
- It would permit sharing illegally made (infringing) copies of music.
- The purpose of the sharing could be to obtain the only copy of the music ever acquired by the downloader. As such, the Proposal would permit not only copying for secondary purposes, e.g., to obtain a second copy for use on a different format, but to acquire the only copy that individuals may ever acquire. It would thus directly substitute (or replace) purchased copies.
- The file sharing technologies used to share music could be any file sharing service, including P2P services like the former Napster, Grokster, or Kazaa; Web and BitTorrent sites like QuebecTorrent, IsoHunt, and TorrentBox; and not-for-profit music sharing sites.
- The media onto which music could be copied would be unlimited and would include computers, PDAs, digital audio recording devices (DARs) including iPods, cellphones, CDs, and DVDs.
- The Proposal would permit unauthorized copying onto all sorts of other hardware, such as hardware equipped with memory chips, radio and television sets, digital cameras, digital video units, car stereos, automobile information systems, watches, kitchen appliances and so forth.
- The proposal would except from infringement both downloads and streams. It would, accordingly, also permit streaming through a variety of sites and services including com; linking sites, like youtvpc.com, www.addictivejunk.com and www.peekvid.com; and sites, like youtube.com, that streams music and other files.
- Copying of every type and genre of music would be excepted from liability. This could even include music embodied in other content, such as movies, games and software.
As the above demonstrates, virtually every Canadian with a mobile phone or an Internet connection, a personal computer, a CD or DVD burner, an iPod or other DAR would fall under the exception. Literally any imaginable copying would be included, unless it is done with a motive of financial gain. The Proposal in effect is an unlimited license to copy music. The Proposal, far from being a “special case” that is narrow in scope and reach, would be the opposite. It would become a normal case. The net result would be a near total levitation of the copyright system, where exclusive rights would effectively cease to exist with respect to digital copying. The proposed file sharing exception would accordingly not meet the first condition of being a certain “special” case.
That the Proposal would not comply with the first condition of the three-step test is supported by authorities that have considered this question. Bob Rietjens recently considered whether an alternative compensation system to permit P2P file sharing for noncommercial purposes would comply with the special case condition of the three-step test. He concluded that “P2P licences are not compliant with the first step of the three-step test. A P2P licence does not qualify as narrow in scope, both in regard to the number of potential users and in regard to the types of works covered.”
Prof. Jane Ginsberg also concluded that a broad exception to exclusive rights to permit digital copying over P2P networks cannot be characterized as a specific case:
“Because more and more works are marketed directly to end users, private copying should no longer be characterized ‘certain special cases’: it is becoming a leading mode of exploitation.”
The Hungarian Council of Copyright Experts considered this question in determining whether private copying from illegitimate P2P sources could comply with the first condition of the three-step test. The Council concluded that it could not, because it would make digital copying the normal case, rather than a special case, and because it could not be justified on sound public policy considerations:
If private copying were allowed also from illegal sources, it would conflict with the criteria of “special cases”. First, it would transform the “special case” foreseen for the application of the right to remuneration into a general form of reproduction. Second – and this in itself would be sufficient to exclude the applicability of any exception or limitation – it would not correspond to the condition that for the recognition of a “special case”, a sound and well-founded social, legal-political justification is needed. As regards private copying from illegal sources mentioned by the petitioner, there are two typical sources. The first one is that a work is made public, distributed or communicated to the public illegally, in particular by making it available to the public through the Internet (as it has happened recently in the case of the Hungarian film “Üvegtigris 2” (“Glass Tiger 2”)), and the other one is that the technological protection measure applied by the owner of rights is circumvented and the work thus distributed or communicated to the public serves as a source of private copying. To permit free uses or to limit the exclusive right of reproduction to a mere right to remuneration in such cases not only would not correspond to the quantitative, qualitative and legal-political criteria of “special cases”, but it would also be in clear conflict with the very raison d‘être, objectives and fundamental nature of copyright. The message delivered by it would be devastating: do not care that you copy from an illegal source; do not worry that you do so without the author‘s consent; copyright is an out-of-date institution; on the Internet, everything is free; just copy any work and use it in this beautiful new world of complete freedom! In fact, the illegal web-sites do “advertise” themselves exactly with this kind “revolutionary ideology”.
For the above reasons the Proposal fails the first step of the three-step test.
SECOND STEP: THE PROPOSAL WOULD CONFLICT WITH THE NORMAL
The second condition of the three-step test requires that a limitation or exception not conflict with “a normal exploitation” of the work or sound recording. The interpretation of this condition involves the construction of two terms: “exploitation” and “normal”.
The ordinary meaning of the term “exploit” connotes “making use of” or “utilizing for one‘s own ends”. The term refers to an activity by which copyright owners employ the exclusive rights conferred on them to extract economic value from their rights.
The term “normal” reflects two connotations. The first is an empirical standard of what is regular, usual, typical or ordinary in a factual sense. The second reflects a somewhat more normative, if not dynamic, approach that takes into account potential technological and market developments. An exception or limitation to an exclusive right will rise to the level of a conflict with a normal exploitation of the work if excepted uses enter into economic competition with the ways that the rights holders normally, or could potentially, extract economic value from that right in the work. All forms of exploiting a work which have, or are likely to acquire, considerable economic or practical importance must be reserved to rights holders. Accordingly, the phrase “normal exploitation” includes those forms of exploitation that currently generate significant or tangible revenues as well as those forms of exploitation that, with a certain degree of likelihood and plausibility, could acquire considerable economic or practical importance.
The WIPO Guide summarizes the elements of the second condition as follows:
Second step (which may only follow if the exception or limitation has not “failed”
at the first step; that is, it is a special case): an exception or limitation must not conflict with a normal exploitation of works. This means that all forms of exploiting a work (that is, extracting value of the exclusive right of reproduction in the work through exercising it) which ha[ve], or [are] likely to acquire considerable economic or practical importance, must be reserved to the owner of this right; and that exceptions or limitations must not enter into economic competition with the exercise of the right of reproduction by the rights owner (in the sense that it must not undermine the market for the work in any way whatsoever).
The WTO Decision summarized the second condition as follows:
We believe that an exception or limitation to an exclusive right in domestic legislation rises to the level of a conflict with a normal exploitation of the work (i.e., the copyright or rather the whole bundle of exclusive rights conferred by the ownership of the copyright), if uses, that in principle are covered by that right but exempted under the exception or limitation, enter into economic competition with the ways that right holders normally extract economic value from that right to the work (i.e., the copyright) and thereby deprive them of significant or tangible commercial gains.
There can be little doubt that the online music market is increasingly replacing physical sales of music as the primary marketplace for virtual sales. Internationally, the market is rapidly migrating from physical to digital formats and the digital distribution of music is expanding at a fast pace. According to Forrester half of all music sold in the United States will be digital in 2011 and sales of digitally downloaded music will surpass physical compact disc sales in 2012. This growing online distribution market uses a plethora of technologies to offer licensed file sharing including P2P systems and services. Technologies like BitTorrent DNA have the potential to effectively deliver licensed musical content to millions of potential buyers of musical content. In Canada, music is also being purchased over the Internet from legitimate distribution services. The market for paid downloads is growing and there is a growing migration from physical to digital formats.
Based on the foregoing, the Proposal does not meet the “normal exploitation” condition. The proposed file sharing exception would expropriate rights that hold considerable economic and practical importance to rights holders, an importance which is likely to continue to grow significantly. Further, the broad scope of the proposed file sharing exception would undoubtedly result in music sharing that would enter into economic competition with commercial sales channels.
As described above, the “normal exploitation” condition not only includes current exploitation methods. It also includes those modes of exploitation which, with a certain degree of likelihood and plausibility, could acquire considerable economic or practical importance. In this regard, digital rights management makes it increasingly possible to control unauthorized copying of works and to develop innovative new business models desired by consumers. The Canadian Government has stated that it intends to enact legislation to implement the WIPO Copyright Treaty (“WCT”) and the WIPO Performances and Phonograms Treaty (“WPPTI”). A core objective of these treaties (the “WIPO Treaties”) is to create a favourable legal environment, including by protecting TPMs, to bolster the market for distributing digital content to end users. The current, exploitation of music over digital networks already has brought this form of exploitation within the “normal exploitation”condition. Potential future uses of technological protection measures and digital rights management information makes the Proposal‘s broad exception for copying including copying over digital networks even more untenable.
It may be argued, as SAC tries to do, that an exception coupled with a levy system may be justified based on the obstacles being encountered by copyright owners in enforcing their rights in the digital environment. Levy systems have, in fact, been established in Canada and around the world as a means of compensating copyright holders against de minimis, or technically infringing copying, that is too expensive or complicated to prohibit. It does not follow, however, that exclusive rights should be effectively replaced by a right to receive remuneration because of difficulties in enforcement. Prof. Ficsor addresses this argument as follows:
The idea emerges, time and again, that, if the exclusive rights cannot be exercised in the traditional, individual way, they should be abolished or reduced to a mere right of remuneration. It is not, however, justified to claim that, if a right cannot be exercised in a way in which it has been traditionally exercised, it should be eliminated or considerably reduced. In such a situation, there is no reason for drawing the conclusion that a non-voluntary license system is needed. There is a much more appropriate option, namely the collective administration of exclusive rights.
Further, the Proposal does not contemplate a limited or de minimis exception and hence does not meet the “special case” first condition. In addition, even though there is substantial unauthorized copying, there is still a substantial marketplace for sales of music on CD formats and a growing legitimate online distribution market place. These marketplaces would be effectively undermined by an exception to permit unlimited file sharing of music. The difficulties in enforcing rights in the digital environment is not, in these circumstances, a sufficient basis to overcome the second condition of the three-step test.
Bob Rietjens addresses and rejects the argument that the inability of rights holders to enforce their rights against P2P file sharing is a sufficient basis to grant a broad exception from infringement:
The Panel [in the WTO Decision] took a much broader scope regarding the normative test: ‘We believe that an exception or limitation to an exclusive right in domestic legislation rises to the level of a conflict with a normal exploitation of the work . . . if uses, that . . . are covered by that right but exempted . . . enter into economic competition with the ways that right holders normally extract economic value from that right to the work…’
The proper question thus is: do the acts allowed under the exception enter into competition with the ways that right holders normally exploit the work? This broader test entails that it is not merely the financial interest in the part that is split off from the copyright owner‘s bundle of rights that is relevant. One must assess whether the exempted acts enter into competition with all the ways in which right holders normally extract economic value from their copyright. The ‘half a loaf of bread is better than none at all’ argument is, therefore, not compliant with the second step of the three-step test. How does this relate to the P2P licence? In short: the fact that the copyright industry cannot enforce certain acts is not relevant as a levy system would undermine the normal exploitation forms used by the copyright industry.
The argument made by SAC that the exception only applies to non-commercial file sharing also does not enable the Proposal to pass the “normal exploitation” condition. It is neither likely nor plausible to assert that a wholesale non-commercial exception for file sharing would not significantly, and potentially mortally, impair the marketplace for traditional CD sales as well as the online digital sales of music. These sales and services would be expected to compete with free file sharing. Consumers would believe that the $5 per month payment gives them the entitlement to obtain music for free. In such circumstances, it cannot reasonably be expected that consumers would believe they should pay twice to acquire music. Bob Rietjens agrees:
It should be considered whether the legalization of P2P networks has an effect on the sales of CDs, as sales of CDs are a normal form of exploitation. P2P has a negative effect on sales of CDs as ‘freely’ downloadable music seems to be a close substitute for off-line commercially purchased music. Thus, legalized P2P use conflicts with a normal exploitation of the work, in the sense that it competes with current forms of exploitation that generate revenue, ie the sale of CDs.
A similar argument goes for legal music download sites. These sites seem to boom at the moment. There are currently around 150 legal music download websites on the Internet of which 100 are in Europe. iTunes, MyCoke-Music.com, PureTracks, Sony‘s Connect, Napster, etc. are doing good business. At any European site there are at least 450,000 tracks available. Some four million Europeans have already paid for online music, but there would hardly seem to be an incentive to use these sites if you are already entitled to download music anyway (and are already indirectly paying for it) under a P2P licence. This view is supported by the Panel‘s view that substitution effects should be taken into account. With regard to the ‘business’ exception, the Panel stated that it was relevant that establishments might be induced to ‘switch from recorded or live music, which is subject to the payment of a fee, to music played on radio or television, which is free of charge’. Apparently, substitution effects are important for the second step of the three-step test.
Further, in determining whether the second condition has been met, a possible conflict with a normal exploitation of a particular exclusive right cannot be counterbalanced or justified by the mere fact of absence of a conflict with another mode of exploiting the work even if the other form of exploitation would generate income for rights holders. Accordingly, reserving to rights holders the commercial marketplace for the distribution of music including a potential to offer “value added” services like iTunes or PureTracks does not remove the conflict of the proposed file sharing exception with the normal exploitation condition.
The fact that rights holders would be compensated under the Proposal by the $5 levy on Internet subscriptions cannot justify the file sharing exception. A right to receive equitable remuneration is not a factor that can be taken into account in considering whether an exception or limitation would conflict with a normal exploitation of a work. Equitable remuneration can only be considered in the third step in determining whether an exception would cause unreasonable prejudice.
Several authors who have studied proposals for non-voluntary licenses to permit non-concluded that such proposals would fail to comply with the second condition of the three-step test. Bob Rietjens concluded that the second step of the three-step test cannot be met by a levy system designed to provide compensation for non-commercial licensed copying over P2P networks:
The P2P licence does not survive the scrutiny of the second condition of the three-step test. The acts exempted by a P2P licence (ie file sharing) compete with the normal exploitation of works by the copyright industry.
Christophe Geiger expressed the same opinion in considering whether the three-step test could be met by the creation of non-voluntary licenses to legalize P2P file sharing:
Yet it seems hard to reconcile such solutions with the second step, even by adopting a restrictive conception of the notion of “normal exploitation”. Such a solution would certainly encroach directly on the main market of online exploitation of works and would therefore violate the three-step test.
Alexander Peukert, in a comprehensive law review article on the subject, concluded that legalized P2P file sharing schemes are incompatible with even the most restrictive view of the second condition:
Applying these definitions to the proposals raises severe doubts regarding their compatibility with the three-step test. The reason is that the uses covered by the proposed non-voluntary license, i.e. copying and distributing content online by way of up- and downloads or streaming are a source of income today and will probably become even more important in the future. Right holders increasingly establish commercial platforms offering their content for download or streaming. Assuming that a complete shift to commercial online distribution by way of streaming is plausible at least for music and perhaps motion pictures…non-voluntary licenses covering non-commercial file sharing can hardly be considered to be in line with the second criterion of the three-step test.
The conclusion that the Proposal would not comply with the ―normal exploitation” condition of the three-step test is supported by leading authorities that have interpreted the three-step test. Martin Senftleben concluded:
If a specific privilege, like the exemption of time-shifting, does not pose substantial difficulties, a broad limitation generally privileging strictly private use in the digital environment certainly does… [I]t is inevitable to conclude that the broad privileges serving strictly personal use which are known from the analogue world are likely to conflict with a normal exploitation of copyrighted material in the digital environment. If the digital revolution really takes place and more and more works are directly marketed to end-users, this emerging “leading mode of exploitation” will be threatened by the general exemption of private copying. That the privilege would then erode the economic core of a wide variety of works can hardly be denied. It would encroach upon a typical major source of income.
Prof. Ficsor expressed the same opinion after canvassing statements made by prominent copyright experts at a WIPO worldwide symposium dealing with issues of the impact of digital technology on copyright and related rights:
As these views clearly indicate, private copying through the global information network is emerging as a major form of utilization of works. Accepting the idea that every user of the network should have the privilege to make a reproduction freely for private, personal purposes – considering the breathtaking growth of the Internet population – would mean accepting that normal exploitation of works would become impossible, and not only on the global network but, as a consequence of free availability of perfect copies, also in major ‘off-line’ markets. Such a practice which so obviously conflicts with basic forms of normal exploitation of works should not be allowed under Article 9 of the Berne Convention (and equally not under the TRIPS Agreement and the WCT which incorporates this provision by reference). And since we are faced here with serious conflicts with normal exploitation and not only with ‘simple’ prejudice to the owners of rights, it is not sufficient to handle the problem through the recognition of a mere right to remuneration, with all the inadequacies of such remuneration systems and with all the doubts about their compatibility with international norms concerning the right of reproduction and national treatment.
The Hungarian Decision also specifically addressed whether an exception to permit copying from unauthorized sources made available through P2P networks would comply with the normal exploitation step of the three-step test. After concluding that such an exception, even if enacted as part of a levy scheme, would not comply with the first step, the Council expressed the opinion that it would also fail the second step:
In the case of the second “step” this does not require too much effort, since on the basis of the findings outlined above concerning the first “step”, the answer seems obvious. The “on-line” making available of works in digital form, along with copy-protection technological measures, for private copying (downloading) has become a form of normal exploitation. In view of this, the permission of this kind of private from illegal sources would be in clear conflict with this form of normal exploitation of the works concerned.
Based on the foregoing, the Proposal would also fail the second step of the three-step test.
THIRD STEP: THE PROPOSAL WOULD UNREASONABLY PREJUDICE THE
LEGITIMATE INTERESTS OF RIGHTS HOLDERS
Given that the Proposal would not pass either the first or second steps of the three-step test, it is unnecessary to consider whether the third step would be met. Given the very serious impacts of the Proposal on right holders, however, even the third step of the three step test could not be met.
The analysis of the third condition of the three-step test requires two determinations. First, one must determine the “legitimate interests” of right holders at stake. Then, it is necessary to determine whether a prejudice reaches an “unreasonable” level.
The word “legitimate” is commonly defined as (a) conformable to, sanctioned or authorized by, law or principle: lawful; justifiable; proper; (b) normal, regular, conformable to a recognized standard type. The term “interests” may encompass a legal right or title to a property or to the use or benefit of a property including intellectual property. It may also refer to a concern about a potential detriment or advantage, and more generally to something that is of some importance to a person. The notion of “interest” is thus not necessarily limited to actual or potential economic advantage or detriment.
The term “legitimate interest” relates to lawfulness from a legal positivist perspective. However, it also has the connotation of legitimacy from a more normative perspective that calls for the protection of interests that are justifiable in the light of the objectives that underlie the protection of exclusive rights. The interests to be protected are those that are justifiable in the sense that they are supported by relevant public policies or other social norms.
A limitation or exception will “unreasonably prejudice the legitimate interests” of the rights holder if it unreasonably deprives the copyright owner of the right to enjoy and exercise the right of reproduction as fully as possible, or where it causes or could cause an unreasonable loss of income to the copyright owner. Remuneration paid under a compulsory licensing scheme can be considered in determining whether an exception causes unreasonable prejudice. But remuneration will avoid unreasonable prejudice only in justifiable cases.
The third step allows a balancing between the interests of rights holders and legitimate societal interests. The words “not unreasonably prejudice” allows the making of exceptions that may cause prejudice of a significant or substantial kind to the authors’ legitimate interests, provided that the exception otherwise satisfies the first and second conditions of the three-step test, and as long as it is proportionate or within the limits of reason; it must not be unreasonable. In determining what prejudice the author should reasonably be required to tolerate, both the quantity and quality of the potential prejudice must be assessed. Prof. Ricketson explains this with an example of the possibility of creating exceptions to address, respectively, the activities of individual photocopying, home taping, and file sharing:
[C]ompare the activities of individual photocopying, with individual home taping, and individual file sharing. The first might well pass all three steps, although maybe not from a cumulative point of view; the second looked at cumulatively, might pass step 2 but not step 3, leading inevitable to some kind of compulsory license. The third should not pass step 2, because of its open-ended character, and would not therefore arise for consideration under the third step. In this regard, national legislators should not fall into the trap of assuming that any unreasonable prejudice that might otherwise result to authors can always be assuaged through the imposition of a compulsory license: by definition, there will always be some “prejudices” that cannot be remedied in this way. While Article 9(2) is far from providing a “bright line” rule that can be readily applied, the individual and cumulative effect of each of the three steps, in particular the third, is to highlight the need for care, moderation, and constraint in constructing any compulsory licensing scheme under national law.
The Hungarian Decision also considered whether an exception to permit file sharing from unauthorized sources could be justified under the third step of the three-step test. The Council of Copyright Experts expressed the opinion that it could not. Their view, with which we agree, is that such an exception would be neither “legitimate” nor “reasonable”:
As pointed out above, in the case of the third “step”, the words “legitimate” and “unreasonably” have a normative, value-oriented meaning, and that, in judging whether a prejudice to the legitimate interests of authors is reasonable or unreasonable, also the possible legitimate interests of third parties and of the public at large should be taken into account. The panel believes that there is no need to elaborate on the reasons for which it would be nonsense to claim that, in addition to free private copying from legal sources, the permission of such copying from illegal sources could also be recognized as a legitimate interest of users, and that the prejudice caused by this to the owners of rights would not be unreasonable. The right to remuneration alone would not be suitable to reduce the prejudice thus caused to a reasonable level.
The Proposal would fail the third condition of the three-step test, because it would, essentially, create an expropriation of copyright holders’ rights to exploit music. Exclusive rights in music would effectively cease to exist and rights holders would become almost totally dependent on the equitable remuneration levy. The exception would likely compromise the credibility of the copyright system as many users would probably consider themselves legitimately free to use and abuse copyrighted works in any manner they see fit. The net result of the Proposal would be a near total “levitation” of the copyright system. The proposed file sharing exception would unreasonably prejudice the legitimate interests of rights holders.
THE PROPOSAL‘S INCOMPATIBILITY WITH THE ROME CONVENTION
Canada could also not implement the Proposal without violating Canada‘s international obligations under the Rome Convention. Under that convention producers of phonograms must be given the right “to authorize or prohibit the direct or indirect reproduction of their phonograms”. Under the convention contracting states can provide for exceptions or limitations for: (a) private use; (b) use of short excerpts in connection with the reporting of current events; (c) ephemeral fixation by a broadcasting organisation by means of its own facilities and for its own broadcasts; and (d) use solely for the purposes of teaching or scientific research.
None of the above exceptions or limitations would be broad enough to permit the Proposal to be implemented into law. The exception for “private use” is limited to a use that is neither public or for profit. It is principally relevant to copying of phonograms in the privacy of the recipient’s home. Exceptions for private use are generally confined to the making of single copies by users. Private use does not, however, extend to acts of commercial scale reproduction including acts of making copies for the purposes of distribution to others, or making copies by means of P2P file share services. “The Rome Convention thus limits the definition of “private” copying in a way similar to the private copying exception in the Canada Copyright Act; a copy made for the personal use of the maker, and not for the purpose of distribution, or for the purpose of trade or for communication to the public by telecommunication. The peer-to-peer “sharing” that is at the essence of the SAC Proposal is not permitted under the Rome Convention.”
SAC itself acknowledges that its Proposal goes further than an exception for private copying. The Proposal makes this clear in the following statement:
The new right would make it legal to share music between two or more parties, whether over Peer to Peer networks, wireless networks, email, CD, DVD, hard drives etc. Distinct from private copying, this new right would authorize the sharing of music with other individuals.
Article 15(2) of the convention also permits exceptions or limitations in the following situations:
Irrespective of paragraph 1 of this Article, any Contracting State may, in its domestic laws and regulations, provide for the same kinds of limitations with regard to the protection of performers, producers of phonograms and broadcasting organisations, as it provides for, in its domestic laws and regulations, in connection with the protection of copyright in literary and artistic works. However, compulsory licences may be provided for only to the extent to which they are compatible with this Convention.
This paragraph permits exceptions or limitations where they fall within the boundaries provided by domestic laws with respect to the protection of copyright in works. It allows for the same type of exceptions or limitations that are provided to works to apply to phonograms. It is also an indirect reference to the three-step test in the Berne Convention as the basis of exceptions and limitations in domestic copyright laws must, as a matter of that convention, conform to that test. Since the Proposal cannot be implemented with respect to musical works without violating the three-step test under the Berne Convention, Canada would be equally unable to enact legislation to implement the Proposal with respect to sound recordings under the Rome Convention.
Moreover, under Article 15(2) compulsory licences can only be enacted to the extent they are compatible with the convention. The Rome Convention allows for compulsory licenses only in limited circumstances, none of which would apply to permit the expropriation of producers’ reproduction rights in sound recordings in exchange for a mere right of remuneration as contemplated by the Proposal.
The failure of the Proposal to comply with the provisions of the Rome Convention provides a further basis for concluding that it would violate the TRIPS Agreement. Under Article 14.2 of TRIPS, producers of phonograms must enjoy the right to authorize or prohibit the direct or indirect reproduction of their phonograms. Article 14.6 of TRIPS permits exceptions or limitations to this right only to the extent permitted by the Rome Convention. Accordingly, since the Proposal could not be implemented without violating the Rome Convention, it could not be implemented into law without also abrogating Canada‘s obligations under TRIPS.
THE PROPOSAL COULD NOT BE IMPLEMENTED WITHOUT ABROGATING
OR AMENDING THE BERNE CONVENTION, TRIPS, NAFTA AND THE
For the reasons given above, the Proposal could not be implemented by Canada without violating Canada‘s international obligations under the Berne Convention, Rome Convention the TRIPS Agreement and NAFTA, unless each of those international instruments were amended to make the Proposal possible. Indeed, many of the proponents of such schemes have expressly acknowledged that their proposals could not be implemented without changes to international agreements related to copyright. Prof. Fisher, for example, notes that the Berne Convention and the TRIPS Agreement would need to be modified before his alternate compensation proposal could be implemented. Prof. Lessig expressly acknowledges that some of his proposals “would require changes to or the abrogation of some treaties”, identifying in particular the Berne Convention, the TRIPS Agreement and the WIPO Treaties as those that would need to be changed or breached. Other individuals who have studied the compliance of non-voluntary licenses with the three-step test concur. These proposals, like the Proposal, do not meet the internationally accepted test for exceptions and limitations to exclusive rights.
Prof. Litman, recognizing that a non-voluntary license regime would not comply with United States treaty obligations, proposed an opt-out mechanism, which would allow rights holders to choose not to participate in the scheme (the SAC proposal contains no such opt-out mechanism). However, it is clear that she also doubts whether her proposal for a statutory default rule, even with an opt-out procedure, would comply with United States treaty obligations. She states that such a process would “be deemed at least arguably compliant” with United States’ treaty obligations under the Berne Convention and the WIPO Copyright Treaty. However, she notes that her “proposal’s Berne-compatibility is optional” and she also apparently doubted whether an opt-out proposal would comply with Article 5(2) of the Berne Convention, which prohibits imposing any formalities on the enjoinment or exercise of rights.
The doubts expressed by Prof. Litman as to whether a non-voluntary alternate compensation system could be established based on an opt-out mechanism are well founded. Under Article 5(2) of the Berne Convention, the enjoinment and the exercise of the rights to be provided “shall not be subject to any formality; such enjoinment and such exercise shall be independent of the existence of the protection in the country of origin of work”. “Formalities” are any conditions or measures, independent from those related to the creation of a work, without the fulfilment of which the work is not protected or loses protection. Registration, deposit of the original or a copy, and the indication of a notice are typical examples. An opt-out model would be a condition or measure affecting the enjoinment and the exercise of rights holders‘ exclusive rights. Alexander Peukert, after considering the opt-out models suggested by Profs. Litman and Prof. Fisher, came to the same conclusion:
To sum up, the opt-out model as suggested by Jessica Litman and William Fisher is a state-required formality for the enjoyment of minimum exclusive rights. It is thus not in line with Art. 5(2) [of the Berne Convention]. The reason is again that international copyright law persists in the notion of exclusive rights, even in the digital network environment. These exclusive rights have to come into existence without further formalities as the statutory default. To provide for exclusivity only under the condition that the right holder opts out of a levy/tax system does not meet this requirement.
The SAC Proposal for the monetization of file sharing of music in Canada would, in return for a right of equitable remuneration, effectively legalize the copying and distribution of music over online networks and onto any other conceivable media and devices. The Proposal would, if implemented effectively, result in unlicensed file sharing becoming the norm in Canada. The Proposal would usurp markets currently being exploited by copyright owners, including markets that are growing in size and importance. The proposed music file sharing exception would, in effect, create a competitive marketplace in which rights holders would be expected to compete for business with free music sharing by individuals through the use of P2P and other file sharing systems. The overall effect would be the expropriation, or near expropriation, of copyrights in musical works and sound recordings. Copyright holders‘ only realistic form of compensation would be the proposed levy system.
For the reasons given above, the Proposal would not comply with any of the three conditions in the three-step test. Accordingly, it could not be implemented without abrogating Canada’s international obligations related to copyright or without amendments to the Berne Convention, Rome Convention, the TRIPS and NAFTA.
 “A Proposal for the Monetization of the File Sharing of Music From the Songwriters and Recording Artists of Canada” Songwriters Association of Canada, online: Songwriters Association of Canada http://www.songwriters.ca/studio/proposal.php.
 Berne Convention for the Protection of Literary and Artistic Works, Sept. 9, 1886; revised July 24, 1971 and amended 1979, 1 B.D.I.E.L. 715.
 Agreement on Trade-Related Aspects of Intellectual Property Rights, 15 April 1994, 33 I.L.M. 1197 [TRIPS].
 North American Free Trade Agreement Between the Government of Canada, the Government of Mexico and the Government of the United States, 17 December 1992, Can. T.S. 1994 No. 2, 32 I.L.M. 289 (entered into force 1 January 1994) [NAFTA].
 Rome Convention: International Convention for the Protection of Performers, Producers of Phonograms and Broadcasting Organisations 26 October 1961, 496 U.N.T.S. 44 [Rome Convention].
 See N. W. Netanel, “Impose a Non-Commercial Use Levy to All Free Peer-to-peer File Sharing”, Harvard J. of Law & Technology (2003) at 19 [“Netanel”]; W. W. Fischer, Promises to Keep: Technology, Law and the Future of Entertainment, (Stanford: Stanford University Press, 2004) at 199-251 [“Fischer”]; L. Lessig, The Future of Ideas, (New York: Random House, 2001) at 254 seq. [“Lessig”]; Raymond Ku, “The Creative Destruction of Copyright: Napster and the New Economics of Digital Technology” (2002) 69 U. Chi. L. Rev. at 263 [“Ku””]; Peter Eckersley, “Virtual Markets for Virtual Goods: The Mirror Image of Digital Copyright?” (2004) 18 Harvard J. L. & Tech. 85 [“Eckersley”]; Jessica Litman, “Sharing and Stealing” (2004) 24 Hastings Comm. & Ent. L. J. [“Litman”].
 See Bob Rietjens, “Copyright and the Three-Step Test: Are Broadband Levies Too Good to be True?” (2006) 20 (3) Int’l. Rev. L. & Computers 323 [“Rietjens”]; Alexander Peukert, “A Bipolar Copyright System for the Digital Network Environment”, (2005) 28:1 Hastings Comm. & Ent. L. J. [“Peukert”]. It has also been suggested that the federal government would not have the constitutional authority to enact a broad based levy on ISPs to provide compensation for unauthorised file sharing. See Jeremy F. de Beer, “The Role of Levies in Canada‘s Digital Music Market Place”, (2005) 4:3 Cdn. J. L. & Tech. 153.
 Wanda Noel et al “Free v. Fee”, (2006) 23 C.I.P.R. 1; See also Taking Forward the Gowers Review of Intellectual Property: Proposed Changes to Copyright Exceptions, U.K. Intellectual Property Office, at paras. 31-32 [Noel].
 Ricketson et al, International Copyright and Neighbouring Rights, Vol. I, 2d ed. (2006) [“Ricketson”], paras. 10.31-10.34; Mihály Ficsor, The Law of Copyright and the Internet (2002) [“Ficsor”], §3.16; Opinion of the Council of Copyright Experts, No. SzJSzT 17/06 of May 11, 2006 (Hungary) at paras. 8-10 [“Hungarian Opinion”]; Remuneration of Private Copying in Australia at 570-71, 582 [“Aust. Report”]
 Peukert, supra note 8 at 51.
 The WTO dispute resolution process has been used to address alleged failures of countries to honour their treaty obligations related to copyright. See United States – Section 110(5) of the U.S. Copyright Act, WTO Report of the Panel, WT/DS160/R, 15 June 2000 [the “WTO Decision”]. Canada – Patent Protection of Pharmaceutical Products, WT/DS114/R, 17 March, 2000 [“WTO Patent Decision”]. Also, on September 25, 2007, the WTO Dispute Settlement Body established a panel to review China‘s Protection and Enforcement Measures of Intellectual Property Rights, WT/DS362/7. Disputes under the Berne Convention and the Rome Convention can be referred to the International Court of Justice by a party to a dispute. Berne Convention, supra note 3 at Article 33; Rome Convention, supra note 6 at Article 30. Further, since members of TRIPs and NAFTA must comply with the applicable provisions of the Berne Convention, a breach of that convention by Canada could also subject Canada to sanctions under TRIPs and NAFTA.
 Berne Convention, supra note 3 at Article 9(1); TRIPS Agreement, supra note 4 at Part II Article 9.1 (requires compliance with the Berne Convention); NAFTA, supra note 5 at Article 1701 (requires compliance with the Berne Convention).
 Rome Convention, supra note 6 at Article 10; NAFTA, supra note 5 at Article 1706(1) (requires the right to authorize or prohibit the direct or indirect reproduction of sound recordings); TRIPS Agreement, supra note 4 at Part II Article 14.2 (requires the right to authorize or prohibit the direct or indirect reproduction of sound recordings). See also, Article 10 Convention for the Protection of Producers of Phonograms Against Unauthorized Duplication of Their Phonograms [“Phonograms Convention”] (1971) (each Contracting State must protect producers of phonograms against the making of duplicates and against the importation of such duplicates for the purpose of distribution to the public and against the distribution of duplicates to the public).
 Berne Convention, supra note 3 at Article 11; NAFTA, supra note 5 at Article 1705(c) (requires authors to have the right to authorize or prohibit the communication of works to the public); TRIPS, supra note 4 at Article 9.1 (requires compliance with the Berne Convention).
 Unauthorized copying of musical works and sound recordings can infringe the reproduction right in ss.3(1) and 18(1) of the Act. Copies are made each time a musical work or sound recording is uploaded or downloaded. Sookman: Computer, Internet and Electronic-Commerce Law (Thomson Carswell) at s. 3.7(a). The reproduction right is implicated when a work or sound recording is copied onto servers for the purposes of offering a streaming or download service. Statement of Royalties to be Collected by CMRRA/SODRAC Inc. for the reproduction of musical works in Canada by Ontario Music Services in 2005, 2006 and 2007, Copyright Board, March 16, 2007 [“CSI Online Decision”]. The rights to authorize communications to the public and to communicate works to the public are also implicated in the posting of a musical work on a website for the purpose of transmission to the public and the transmission of performances and downloads to the public. Society of Composers, Authors and Music Publishers of Canada v. Canadian Association of Internet Providers,  2 S.C.R. 427; Canadian Wireless Telecommunications Association v. Society of Composers, Authors and Music Publishers of Canada 2008 FCA 6; Tariff No. 22.A (Internet Online Music Services) 1996-2006, Copyright Board, October 18, 2007.
 Berne Convention, supra note 3 at Article 9(2); TRIPS supra note 4 at Part II, Article 13; NAFTA, supra note 5 at Article 1705(5); Ricketson, supra note 10 at paras. 13.11-13.25.
 NAFTA, supra note 5 at Article 1706(3).
 WTO Decision, supra note 12 at paras. 6.71-6.74.
 NAFTA, supra note 5 at Articles 1705.5, 1706.3.
 WTO Decision, supra note 12 at para. 6.97; WTO Patent Decision, supra note 12 at para. 7.21.
 WTO Decision, supra note 12 at para. 6.97; WTO Patent Decision, supra note 12 at para. 7; Ficsor, supra note 10 at 91-92; WTO Decision, supra note 12 at para. 6.74; M. Senftleben, Copyright, Limitations and the Three-Step Test (2004) [“Senftleben”] at para. 4.3.; Ricketson, supra note 10 at paras. 13.11-13.25.
 Ficsor, supra note 10 at paras. 5.55, 10.03; WIPO Guide to the Copyright and Related Rights Treaties Administered by WIPO (2003) [“IPO Guide”], at CT-10.2 at 213.
 WTO Decision, supra note 12 at para. 6.108; Ricketson, supra note 10 at 764.
 WTO Decision, supra note 12 at para. 6.109; Ficsor, supra note 10 at paras. 5.55, 10.03; WIPO Guide, supra note 23 at CT-10.2; Ricketson, supra note 10 at §13.11.
 WTO Decision, supra note 12 at para. 6.109.
 WTO Decision, supra note 12 at paras. 6.111-6.112; Ricketson, supra note 10 at para. 13.11.
 WIPO Guide, supra note 23 at CT-10.2.
 Ficsor, supra note 10 at para. 5.55.
 Jörg Reinbothe & Silke von Lewinski. The WIPO treaties 1996 : the WIPO Copyright Treaty and the WIPO Performances and Phonograms Treaty : commentary and legal analysis. (London : Butterworths, 2002) at 124 [Reinbothe].
 Hungarian Decision, supra note 10 at 12-13.
 See Noel, supra note 9 at 12 also adopting the more rigorous two-prong test in interpreting the term “special”.
 See paras. 3 and 4 of the Summary of Proposal.
 See para. 1 Summary of Proposal.
 See Proposal at para. 5.
 The Proposal is not even limited to digital copying, however. It could conceivably also include the reproduction of musical scores, musical compositions, and copying of sheet music
 WTO Decision, supra note 12 at paras. 6.113, 6.127, 6.131-6.133, 6.148.
 Brent Hugenholtz et al, “The Future of Levies in a Digital Environment, Final Report, Institute for Information Law”, March 2003 at 41 [IViR Final Report].
 Eckersley, supra note 7 at 155 argues that blanket licenses in respect of downloads can meet the “special case” condition because “downloads that are part of a specially organized public reward mechanism are peculiar and limited, and in the ordinary English usage of the expression, they are certainly a special case of infringement.” Eckersley‘s reliance on a “public reward mechanism” is an irrelevant factor under the first condition, as the existence of compensation is only relevant in considering the third condition. His conclusion that a blanket download license would be “peculiar and limited” is hard to reconcile with its actual scope and reach.
 Rietjens, supra note 8 at 332.
 Ginsberg et al, “Private Copying in the Digital Environment”, in Kabel et al, Intellectual Property and Information Law (Kluwer Law International 1998) at 149-151.
 Hungarian Decision at para. 36 (emphasis added). Alexander Peukert and Carine Bernault contend that the non-commercial reproduction of musical works through P2P networks would comply with the first condition of the three-step test. Their views assume that the first condition only requires some clear public policy consideration to support a P2P non-voluntary license. Peukert, supra note 8; Carine Bernault et al, “Peer-to-Peer File Sharing and Literary and Artistic Property, A Feasibility Study Regarding a System of Compensation for the Exchange of works via Internet”. June 2005. To the extent such condition exists, their views conflict with the persuasive opinion of the Hungarian Council of Copyright Experts set out above. In any event, any exception must also be narrow in scope and reach. Their opinion fails to take this part of the three-step test into account.
 WTO Decision, supra note 12 at para. 6.165; Ricketson at 768; WTO Patent Decision, supra note 12 at para. 7.54.
 WTO Decision, supra note 12 at para. 6.166; Ricketson, supra note 10 at para. 13.16; WTO Patent Decision, supra note 12 at para. 7.54.
 WTO Decision, supra note 12 at paras. 6.177-6.183; Ricketson, supra note 10 at paras. 13.18-13.19; Ficsor, supra note 10 at 284-285; Gowers, at para. 34; Peukert, supra note 8 at 33-34.
 WIPO Guide, supra note 23 at CT-10.2 (emphasis added).
 WTO Decision, supra note 12 at para. 6.183.
 In-Stat, The Online Music Market: Downloaded Music Will Outpace Physical Media Bought Online in 2007 (“Healthy growth in the online music market will continue for the rest of the decade, with worldwide sales growing from $1.5 billion in 2005 to $10.7 million in 2010”); In-Stat, Online Music and Video: New Distribution Channels Emerge (“Online sales of digital music represented 6% of the worldwide music market in 2006, up from 4% in 2005, the high-tech market research firm says. By 2011, online sales will represent 26% of all music purchased worldwide.”); PricewaterhouseCoopers, Global Overview—Global Entertainment and Media Outlook: 2007-2011 (New York: PWC, 2007) (“The composition of the market is rapidly migrating from physical to digital formats… We expect digital distribution of music to expand at a 26.8 percent compound annual rate, more than tripling to $23 billion in 2011 from $7 billion in 2006.”); International Federation of the Phonographic Industry, IFPI:07 Digital Music Report (London: IFPI, 2007) (“Digital music sales are estimated to have almost doubled in value worldwide in 2006, reaching an estimated trade value of around US$2 billion. Digital channels accounted for an estimated ten per cent of music sales for the full year 2006, up from 5.5 per cent in 2005.”).
 “Music Downloads to Surpass CD Sales by 2012”, Boston Business Journal, February 19, 2008 (“Digital music sales will grow at a compound annual gross rate of 23% over the next 5 years, reaching $4.8 billion in revenue by 2012. In contrast, by 2012, CD sales will be reduced to $3.8 billion.”).
 Organisation for Economic Co-operation and Development, Digital Broadband Content: Music, Report DSTI/ICCPI/IE(2004)12/FINAL (Paris: OECD) at 78-79; Michael Einhorn, “Gorillas in our Midst: Search for King Kong in the Music Jungle”, (2008) 55 J. Copyr. Soc‘y, at 145.
 Private Copying III Decision (“PC III”), at 13 (Copyright Board finding that music is being purchased over the Internet from “legitimate distribution services””); Private Copying 2005, 2006 and 2007 (May 11, 2007) (“PC IV”), at paras. 60-61 (Copyright Board finding that 6% of all private copies were lawfully purchased online); CSI Online Decision, at paras. 2, 6 (Copyright Board concluded that, “we are now in the midst of experiencing the next radical change: the authorized download over the Internet of digital files containing sound recordings of musical works.”); PricewaterhouseCoopers, Global Overview—Global Entertainment and Media Outlook: 2007-2011 (New York: PWC, 2007) (“The composition of the market is rapidly migrating from physical to digital formats. We expect digital distribution of music to expand at a 26.8 percent compound annual rate, more than tripling to $23 billion in 2011 from $7 billion in 2006.”); Tom Jurenka ―Internet Industry Overview, a Report Prepared for Gowling Lefleur Henderson LLP in connection with SOCAN Tariff 22″ (November 2006) at 5 (“Music is one of the great e-commerce success stories of the Internet. In 2005 Canadians ordered just over $7.9 billion worth of goods and services over the Internet, 16.4% of which was music of all categories”); “Music on the Internet: A Canadian Perspective”, Aaron Research, November 2006 (describing the Canadian sites and services involved in online music distribution).
 Gowers at para. 168 (“DRM‘s are a legitimate and acceptable tool used by rights holders and are increasingly being used to support new business models.”); IViR Final Report, supra note 38 at 1 (“In the digital environment, technical protection measures and digital rights management systems make it increasingly possible to control how individuals use copyrighted works. Rights holders and media distributors are now in a position to apply, and are increasingly using, such systems to identify content and authors, set forth permissible uses, establish prices according to the market valuation of a particular work, and grant licenses directly and automatically to individual users. Unlike levies, electronic copyright management systems make it possible to compensate right holders directly for the particular uses made of a work. Where such individual rights management is available there would appear to remain no need, and no justification, for mandatory levy systems.”); PC III at 11, 62 (Copyright Board finding that “TPMs” and “DRM systems” are increasingly being used by rights holders “to control the distribution and use of, and access to, music and other content.” The Board predicted that the “more widespread TPMs and DRM systems become, the more rightsholders are likely to make content available legitimately, and therefore, the more consumers can be expected to have otherwise paid for private copying rights.”).
 See Copyright Reform Process: A Framework for Copyright Reform, at 4 (June 2001); Supporting Culture and Innovation (Oct. 2002); Interim Report on Copyright Reform, at 2-5 (May 2004); Government Statement on Proposals for Copyright Reform (Mar. 2005); CIPO: Frequently Asked Questions (Apr. 2005); Counterfeiting and Piracy are Theft: Report of the Standing Committee on Industry, Science and Technology, at para. 5 (June 2007).
 Ficsor, supra note 10 at C10.34 (“While in the case of traditional off-line and off-air ‘home taping’, the exercise of such a right was impossible, on the interactive digital network – on the basis of technological measures and rights management information – it is very possible.”); Reinbothe, supra note 30 at 126 (“[L]evy systems have usually been based on the reasoning that…the exclusive right of reproduction cannot be enforced. This reasoning looses its strength in the digital environment, which may well result in the exclusive right of reproduction to be enforceable, particularly with respect to digital private copying.”); Eckersley, supra note 7 at 155-157 (Arguing that the “normal exploitation” step cannot be met if DRMs can be used to prevent widespread unauthorized sharing of files); Gregory Hagen et al “Canadian Copyright Reform: P2P Sharing, Making Available and the three-step Test” (2006) Ottawa Law and Technology Journal, 3:2 (arguing against amendments to the Copyright Act to implement the WIPO Treaties because such amendments would enable copyright holders to control the use of their works thus preventing the establishment of a levy system for copying in the digital environment because such a system would fail the three-step test).
 Eckersley, supra note 7 at 155-157. Eckersley argues that an exception could be justified if technological protection measures cannot prevent widespread unauthorized sharing of information goods.
 Ginsburg, at 149-150 (“private copying could be understood as non-infringing because it was de minimis…but too expensive and complicated to prohibit”); Ricketson, supra note 10 at para. 13.33 (“private use would appear to be confined to the making of single copies, and the basis for it a kind of de minimis argument”).
 M. Ficsor Collective Administration of Copyright and Neighbouring Rights (Geneva: WIPO, 1990) at 6, (emphasis added) quoted by Daniel Gervais, “Collective Management of Copyright and Neighbouring Rights in Canada: An International Perspective” (2001).
 Rietjens, supra note 8 at 331 (emphasis added).
 Rietjens, supra note 8 at 331 (emphasis added).
 WTO Decision, supra note 12 at para. 6.172; Reinbothe, supra note 30 at 125.
 Reinbothe, supra note 30 at 126-127; Ficsor, supra note 10 at paras. 5.58, 10.33¬10.34; Senftleben, supra note 22 at paras. 4.3.2, 4.3.3.
 Rietjens, supra note 8 at 332.
 Christophe Geiger, “The Role of the Three-Step Test in the Adaptation of Copyright Law to the Information Society, UNESCO, Copyright Bulletin, January-March 2007, at 8-9.
 Peukert, supra note 8 at 34 (emphasis added).
 Senftleben, supra note 22 at 204, 206.
 Ficsor, supra 10 at para. C 10.34.
 Hungarian Decision at para. 37.
 WTO Decision, supra note 12 at para. 6.222.
 WTO Decision, supra note 12 at para. 6.224; WTO Patent Decision supra note 12 at para. 7.68.
 WTO Decision, supra note 12 at para. 6.223.
 WTO Decision, supra note 12 at para. 6.224; WTO Patent Decision, supra note 12 at paras. 7.68-7.73; Ricketson, supra note 10 at paras. 13.23-13.25.
 Ficsor Copyright, at paras. 5.57, 10.03; WTO Decision, supra note 12 at paras. 6.220¬6.229.
 Senftleben, supra note 22 at para. 4.3.2.
 Gowers at para. 34.
 Ricketson, supra note 10 at para. 13.27; Ficsor, supra note 10 at paras. 5.58, 10.33¬10.34; J. Reinbothe, supra note 30 at 126-127; Senftleben, supra note 22 at paras. 4.3.2, 4.3.3.
 Reinbothe, supra note 30 at 127.
 Ricketson, supra note 10 at para. 13.27.
 IViR Report, supra note 38 at 41 (commenting on the implications of applying levies to multi-purpose digital machines such as PCs).
 Rome Convention, supra note 6 at Article 10. The term “phonogram” is defined to mean “any exclusively aural fixation of sounds of a performance or of other sounds”. The term “reproduction” is broadly defined to mean “the making of a copy or copies of a fixation”. Ibid. Article 3.
 Ibid. At Article 15(1).
 WIPO Guide to the Rome Convention and to the Phonograms Convention, (WIPO 1981), Commentary 15.2 to Article 15 to the Rome Convention. The WIPO Guide expressly notes, “As to phonogram producers, the ease with which recording of high quality can be made these days places the idea of private use in a new dimension.” Ibid.
 Ricketson, supra note 10 at para. 19.12
 Ibid. at para. 13.33 (“private use would appear to be confined to the making of single copies”).
 Ibid. at para. 19.12.
 Rome Convention, supra note 6 at Ss. 80(1); 80(2).
 Proposal at para. 5. (emphasis added)
 WIPO Guide at RC-15.3; Guide to the Rome Convention and to the Phonograms Convention (WIPO 1981) at para. 15-7-15.9.
 Ricketson, supra note 10 at para. 19.16.
 As detailed above, Canada would also be unable to enact the Proposal with respect to sound recordings because under NAFTA exceptions and limitations to the reproduction right in sound recordings must comply with the three-step test and the Proposal would not comply with that test.
 Ricketson, supra note 10 at para. 19.16 (pointing out that compulsory licenses are allowed under Art. 7(2)(2) (broadcasting of performances), Art. 12 (broadcasting of phonograms), and Art. 13(d) (communication to the public of certain broadcasts)).
 Fisher at 247-248. Prof. Fisher expressed the view that there could be some discretion with respect to recordings of musical compositions pursuant to Article 13 of the Berne Convention. However, Article 13(1) only permits the imposition of reservations and conditions “to authorize the sound recording” of a musical work. This provision addresses the fixation of a musical work onto a sound recording, which was likely intended to encompass the recording of a musical work by instruments capable of reproducing them mechanically. See WIPO Guide at BC-13.1 (explaining the history of the provision). Article 13 does not permit exceptions for otherwise reproducing musical works. Nor would it allow any exceptions to Canada’s obligations to provide rights holders with the exclusive right of authorizing reproductions of sound recordings under NAFTA.
 Lessig, supra note 7 at 330-331. Canada has not yet ratified the WIPO Treaties. Accordingly, this article does not address the Proposal’s compliance with those treaties.
 Rietjens, supra note 8 at 332 (“The interpretation by the Panel [in the WTO Decision] of the TRIPS three-step test indicates that only those exclusions that allow de minimus use are allowed, regardless of lack of effective or affordable means of enforcement or social or cultural policy reasons. P2P licensing is a clear breach of countries’ obligations under the TRIPS three-step test. In the EU/USA WTO dispute, the USA decided not to amend its copyright laws. Instead it decided to pay a yearly damage (‘buy out’) to the EU. In contrast, it is clear that countries cannot buy their way out of a P2P licence as the amount of damages would simply be too high. P2P licensing might be academically challenging, but it seems of little practical importance, as P2P licensing is not compliant with the TRIPS three-step test.”); Geiger at 8 (“Such a solution would certainly encroach directly on the main market of on-line exploitation of works and would therefore violate the three-step test.”); Peukert, supra note 8 at 78-79 (“This paper has shown, however, that none of the currently discussed models is in accordance with obligations contained in international copyright law. The BC, TRIPS and WCT rest upon the notion of legal and technological exclusivity enjoyed by the copyright owner. They are opposed to the implementation of statutory, non-voluntary licenses covering non-commercial file sharing. Only if the right holder is free to decide whether she wants her work to be subject to a levy/tax system is exclusivity as the fundamental requirement of international copyright law accomplished. What is more, exclusive rights and individual licensing have to be the legal default. Therefore, only an opt-in model according to which the right holder has to register the work for the levy/tax system can be implemented in national law without the need to amend international copyright treaties or terminate membership.”). See also OECD Report at 92 (“New compulsory licenses for P2P could also be found to interfere with obligations under the major international agreements dealing with copyright such as the Berne Convention and the WIPO Copyright Treaty.”); Ficsor, supra note 10 at para. C10.33-10.34.
 Litman, supra note 7 at 43-44 and footnotes 166-167.
 WIPO Guide, supra note 23 at § BC-5.7.
 Peukert, supra note 8 at 65. See also Daniel Gervais, “Towards a New Core International Copyright Norm: The Reverse Three-Step Test” bepress Legal Series, paper 214, (March 2004) at 71-73. (“Because the Berne Convention, the substantive obligations of which were incorporated in the TRIPS Agreement, severely limits the availability of compulsory licensing, at least for the rights of composers, any licensing system should thus be voluntary. I believe this is also in line with traditional copyright policy: if a rights holder does not want his work licensed, then he should have that right”). Prof. Gervais has expressed the opinion that an opt-out mechanism could be implemented as part of an extended collective licensing regime. An extended collective license is not a non-voluntary license or a remuneration regime for private copying. Daniel Gervais, Application of an Extended Collective Licensing Regime in Canada: Principles and Issues Related to Implementation, study prepared for the Department of Canadian Heritage, June 2003 at 17, 26-27. (Prof. Gervais also noted in discussing extended collective licensing that a compulsory license with no opt-out rights can only exist “where international conventions allow the implementation of a compulsory license, as is the case with cable retransmission”. Ibid.