A Copyright Exception for Monetizing File-Sharing. A proposal for balancing user freedom and author remuneration in the Brazilian copyright law reform

Request for Comments Draft, 18.3.2010

Volker Ralf Grassmuck*

Licensed under the Creative Commons BY-SA BR 3.0 Licence1

File-sharing has become wide-spread media cultural practice since Napster was launched in 1999. The response by the copyright industry has been repression. This has neither resulted in a decrease in the growth of file-sharing, nor in revenues for authors. Also since the start of Napster, a different approach has been suggested: permit file-sharing subject to a collectively managed levy. Just like the private copying exception, this will ensure both the freedom of citizens to use copyright protected works in certain ways and the right of creators to an equitable remuneration. The debate on a file-sharing permission started among specialists and spread to main-stream media and political parties, leading to a number of pilot projects and bills for implementing it in copyright law. The current copyright law reform provides Brazil with the opportunity to advance such a solution, thereby taking a leadership role in the international debate on the future of creativity in the digital era.

A new social contract

Pedro Alexandre Sanches: “The dissemination of free music on radio and TV is regulated, they [industry] get money for it, but nobody in the media is claiming that music is being given free to the consumer.”

Pena Schmidt: “They refused to understand the Internet as a means of communication. They refused to sit at the table and propose, discuss and collect money through ECAD [the Brazilian music collecting society]. They refused the right of ECAD to collect money on the Internet. If there is any, it’s a tiny amount, when in fact they should have joined and forced ECAD to become a distributor.”2

Copyright law was the core element in the social contract between authors and audiences that was negotiated in the 18th and 19th century. It worked well in regulating the relations between commercial parties in the culture industry. The digital revolution has fundamentally changed the media-technological basis of the production, distribution and consumption of cultural goods. Private persons, whose actions have until recently been outside the scope of copyright law, can now be producers and global distributors of creative works. Therefore a new social contract concerning culture has to be negotiated.

The overall goals of this contract remain the same: to ensure the possibility of all citizens to have access to and participate in the knowledge society and their freedom of expression, to ensure the freedom of a diversity of authors and artists to create and their right to an equitable remuneration for the use of their works, and to ensure the freedom of technologists to innovate, in particular the freedom of the Internet that has brought us a wealth of novel ways to communicate, cooperate and do business.

Also the two core elements of this contract remain: the willingness of authors and artists to create works and the willingness of audiences to pay them for doing so. This payment will continue in a range of ways, from market transactions, donations, corporate sponsorship to forms of collective redistribution like public funding and collective rights management.

For a large number of uses by a large number of individuals of a large diversity of works the conventional response of copyright law is collective management. When in the 19th century composers were unable to individually collect a remuneration from each cafe house and bar that performed their music, they joined hands and formed the first collecting societies that since then collectively collect the money for the community of music authors. When in the 1950s audio tape recorders became available for private use, enabling people to make non-commercial reproductions in their homes, in response the private copying exception was invented in 1965. Neither could the technology or their use for making reproductions be prohibited, nor could authors or even their collectives go after each individual owner of a tape recorder to collect their fair remuneration. Therefore the German legislature decided to permit private copying and required the producers and importers of tape recorders to add a copyright levy to the price of their devices. The music collecting society collects this levy and redistributes it to its members. This private copying exception was quickly adopted throughout Europe and other droit d’ateur countries.3 The levy was later extended to other recording devices like photocopying machines and video recorders and to recordable media.

Collective rights management organisations (CMO) thus developed as crucial institutions in the social contract between authors and audiences, ensuring authors a fair remuneration for mass-scale secondary and tertiary uses of their works,4 and ensuring audiences the informational freedom of private copying. Because of their special status, CMOs are subject to legal regulation and public approval and oversight. Internally they are membership organisations with democratic decision making. Currently in Germany there are 13 CMOs for different work categories (music works, music recordings, text works, images, movies and other audio-visual works). Peer-to-peer (P2P) File-sharing is a mass-phenomenon comparable to private copying, practiced by about half the Internet population.5 The adequate response is again a collectively remunerated copyright exception.

Repression does not work

Today there is a discrepancy between copyright law and the wide-spread practice of file-sharing. So far attempts to resolve this discrepancy were directed at repressive measures in order to make cultural practices conform to the law: in the form of technology (Digital Restrictions Management (DRM)), deterrent campaigns à la “Pirates are Criminals”6 and mass-scale civil and criminal proceedings. None of these have had any measurable impact on file-sharing. But instead of recognizing the failure and changing the approach, the same logic is bringing forth evermore extremist forms of repression. Excluding infringing citizens from the Internet for up to one year was pioneered in France and is now being called for by culture industries in many countries, including Brazil. The secretly negotiated multi-lateral Anti-Counterfeiting Trade Agreement (ACTA) is intended to introduce this digital death sentence globally. Deep Packet Inspection (DPI)6a for filtering file-sharing out of the Internet is now being tested by UK Internet Service Provider (ISP) Virgin Media7 and others.

All this is done under the unproven assumption that repression will improve sales opportunities for copyright products and with the proclaimed but equally unproven intention to increase revenues of authors and artists. A much more likely and, indeed, observable result of repression is not a decrease of file-sharing but an increase of tracker-less P2P networks,8 encrypted and anonymous P2P file-sharing,9 of closed trackers,10, file-hosting, sharing on Usenet, offshore hosting11 and hard disk-copying. Repression predictably calls in the next round of the technological arms race. As US law scholar Lawrence Lessig explains, the criminalisation of the whole generation of our children “can’t stop these activities, it can only drive them underground.” He points out one dramatic effect it does have: the erosion of the trust in the legal system.12

If cultural reality can not be made to conform to copyright law, then copyright law has to be adapted to reality: by legalizing what can not be prevented anyway and at the same time ensuring an equitable remuneration to authors.

The File-Sharing Exception

The model has been discussed under different names: “alternative compensation system” (William Fisher13), “noncommercial use levy” (Neil Netanel14), “licence globale” (Alliance Public Artistes15), “culture flat-rate” (privatekopie.net and FairSharing.de), “contribution créative” (Philippe Aigrain16). Variations of its details are, of course, being discussed but the contours of the general model have emerged by now. Following the precedent of the private copying exception, the goal is:

a legal permission for private online sharing of published copyright protected works for non-commercial purposes subject to a collectively managed levy.

The permission refers to private, natural persons, thus excluding companies and other legal institutions. It refers to published works, ensuring the right of first publication to the author. Secondary and tertiary uses are already regularly collectively managed. It refers to non-commercial uses, i.e. anybody earning money from the use of another’s work will continue to be required to obtain a license. “Sharing” refers to both up- and downloading. While downloading is already covered by the private copying exception in some countries,17 permitting uploading requires an exception to the exclusive right of making available. “Online” refers to networks using the Internet protocol, both wire-based and wireless. “Levy” refers to a fixed sum to be paid by the beneficiaries of the permission and allocated by CMOs to authors based on the measured popularity of their works. Finally the permission should be implemented in copyright law in order to achieve legal certainty for authors, performers, exploiters and Internet users alike.

The Culture Flat-Rate in Copyright Law

Three models for implementing such a file-sharing permission in copyright law have been suggested.

Mandatory collective management of the exclusive making-available right
This was first implemented by Hungary. Silke von Lewinski from the Max Planck Institute for Intellectual Property in Munich analysed the Hungarian provision and found it in accordance with international and European copyright law.18 The French Alliance Public-Artistes commissioned a legal study from France’s most renowned copyright scholar André Lucas who also found mandatory collective management compliant with French, European and international law.19 This model was supported by French Members of Parliament from both the socialists and the conservative parties who passed it into law in December 2005. Alas, the decision was reversed soon after.

Extended collective licensing
This instrument is widely used in Nordic European countries since the early 1960s for broadcasting and cable re-transmission and has recently been applied to the reproduction of works for educational purposes and the digitisation of works in libraries, museums and archives.20 It extends a license concluded between a CMO and a group of users of certain rights to authors, performers and exploiters who are not a member of the CMO. These non-members usually have the right to opt-out of such an agreement. With respect to file-sharing this model has been discussed particularly in Italy,21 leading to two bills introduced in parliament in July 2007 and in April 2008.

Copyright exception
A third option is to model a file-sharing exception on the time-tested private copying exception. This has been tested by Alexander Roßnagel and his team at the Institute of European Media Law (EML) on commission from the German and European Parliament factions of the Green Party.22 Their study has shown that such an exception is feasible within the framework of existing German and European law, even though it requires changes in both. They conclude: “The introduction by law of a culture flat-rate therefore requires amendments to both national and European law, yet it remains nothing less than the logical consequence of the technological revolution ushered in by the internet.”

It seems that by subjecting the complete scope of the making-available right to mandatory collective management, the first model is going too far, while the second model is not going far enough. By permitting exceptions to the exception it would still require policing the boundary between licensed works and those that have been opted-out of the agreement. Therefore a clearly defined copyright exception is the best option to achieve legal certainty for all parties involved.

Which Works to Include?

Empirical research shows that nearly all categories of copyright protected works are being shared to varying degrees depending on the characteristics of the different P2P protocols.23 Therefore the permission should extend to all categories that are also covered by the private copying exception. Whether computer software that has been exempted from the private copying permission and thus also from receiving a share from the levy, and computer games should also be included is up for discussion with the respective industries.24

Who should Pay?

The beneficiaries of the permission, i.e. the individual private Internet users, owe the creators of the works they share. Internet Service Providers (ISPs) do not download music or movies, just as the makers and operators of photocopying machines, audio recorders and MP3 players do not copy. The argument of a “contributory liability” is not convincing: By the same token the providers of electricity, search engines, computer monitors, chairs etc. would contribute to file-sharing. Conversely, no Internet users would cancel her broadband subscription if P2P were to disappear.

However, for practical reasons one can not expect consumers who acquire these devices, media and services to pay the copyright levy in a separate transaction. Therefore legislature in many countries has tasked the producers and importers of devices and media with collecting the private copying levy. For reasons of transparency and fairness, German copyright law since its 2008 reform requires that end-consumer bills separately indicate the copyright levy included in the price for these products.25

Likewise, ISPs and mobile phone companies that provide Internet access to private homes are the logical parties to add the file-sharing levy to their monthly customer bills and transfer the money to the CMOs. LAN houses provide a large number of Brazilians who cannot afford broadband at home with access to the Internet. Assuming that their clients actually use file-sharing applications for up- and downloading copyright protected works (this needs to be empirically assessed), LAN houses, like ISPs, should add a copyright levy to the price they charge. The rate, however, cannot be so high as to exclude a significant portion of the population from Internet access altogether. Just as with the current plans for establishing a national broadband service, public policy has to balance the interest of society to include all citizen in the opportunities of the digital age and the interests of authors and publishers as well as those of ISPs and LAN houses.

Are they willing to pay?

The Swedish music collecting society STIM in a survey published in February 2009 found that 86.2% of responding Internet users are willing to pay a monthly copyright levy entitling them to file-sharing.26 This willingness has also been proven by bands like Nine Inch Nails and Radiohead who released albums for free download and received significant amounts of voluntary payments from their fans. The online indie label Magnatune.com releases all its albums under a Creative Commons license expressly permitting to file-share them, which effectively makes payment voluntary. Magnatune also offers their customers to pay a price of their own choice on a scale from 4-14 Euro. Rather than paying the lowest possible price, the average payment is between 8 and 9 Euro, clearly indicating that listeners are willing to pay creators a price they deem fair.

The same willingness was proven for computer games when in October 2009 the developer studio 2DBoy offered its game “World of Goo” on a pay-what-you-like basis. Remarkably, they found the average price paid higher for GNU/Linux users than for Windows users and, mapping average payment per country onto per capita GDP, they found the “generosity factor” to be exceptionally high in Brazil.27 Quite the opposite from industry claims that what is gratis is considered worthless, one can conclude that people acculturated in free and sharing culture are more aware that creators need to be remunerated and more willing to behave accordingly.

But I don’t file-share

The copyright levy should be mandatory for all Internet users. Just like permitting opt-out for individual works, making payment optional would require policing the boundary between those who pay and those who do not, which would largely defeat the public policy purpose of the file-sharing exception.

An objection often raised against a levy mandatory for all Internet users is: “I don’t file-share. Why should I pay?” This should be alleviated by differentiating the rate by access speed. E-mail-only dial-up access should be exempt. Given that half of all Internet users file-share already and nearly 90 percent are ready to pay for legalised P2P, one can expect the number of people to whom this objection applies to shrink even further, once a file-sharing exception is introduced. Also non-file-sharers benefit from decriminalisation and wider access by gaining a richer cultural sphere.

Cross-subsidising is already common in many cases. Taxes of citizens without children are used for funding schools. Someone who buys a detergent in a supermarket pays for the advertising-funded movie on “Free TV” that she is not watching. The private copying levy on a recordable DVD is due even if the buyer uses it for a back-up copy of her own data. Finally, if 86.2% of the Internet population are willing to pay for the right to file-share, may the other 13.8% stop it? If so, by the same logic we would not have public broadcasting, opera, health care, police or national defence.

How much?

Objectively, it is impossible to determine the positive impact due to its “discovery effect” versus the “damage” of file-sharing that a levy might compensate.28 Subjectively, the pay-what-you-like models give an indication as to how much certain works are worth to certain people. In actuality, rate-setting in collective management is a very difficult procedure. In case of the private copying levy, rates are negotiated between CMOs and the associations of device and media producers. For the file-sharing levy, negotiations would include not only ISPs but also artists and Internet users who are paying in the end. Also public mediation by the newly proposed Instituto Brasileiro de Direito Autoral (IBDA) would be helpful if not even essential for successfully concluding an agreement in the public interest.

Ever-since Fisher had calculated the fair amount to be roughly 5 US$ per month,29 five has been the magic number in the debate, varying between Dollars, Pounds, Euros, Reais etc. For a Brazilian household that can afford broadband Internet access, five Reais per month is not a prohibitive amount. Assuming 6.6 million households with broadband Internet access in Brazil,30 R$ 5 per month yields an annual amount of R$ 396 million.

Revenues for music CDs and DVDs reportedly decreased by 31.2% (or R$ 141,7 million) to R$ 312.5 million in 2007,33 while in 2009 ECAD (Escritório Central de Arrecadação e Distribuição34) was able to distribute 17.06% (or R$ 46.34 million) more revenues from collective management of music rights to its members than in the previous year.35 Movies generated revenues of R$ 966 million in 2008 at the box office which was an increase of a full 25% over the previous year.31 Movie DVDs reportedly saw a drop by 10.83% from 27.2 million units sold in 2007 to 24.7 million in 2008.31a Conservatively assuming a sales price of R$ 40,00 this amounts to a decrease by R$ 120 million. The Brazilian book market showed a slight annual increase (by 6.03%) as well, to R$ 2.286 billion in 2007.32

The increases in cinema and book revenues occured in spite of wide-spread file-sharing, and there are reasons to assume that also the decrease for recorded music and movie DVDs are unrelated to it.36 But even if one would assume that file-sharing is the single cause for the decline of the market for music and movie discs (by approximately R$ 261.7 million) and that the levy would have to compensate for it, the levy proceeds of 396 million Reais would be more than sufficient to do so.

A realistic approach, of course, cannot start from such a simplistic and flawed assumption. It will have to take into account the complex dynamics in each of the sectors and include the whole range of revenue channels for creative works like live performances, cinema screenings, commissioned works, merchandizing etc. that have been shown to be positively impacted by file-sharing.

For the music industry it has been shown that the revenues from 2000 to 2008 remained stable, with the decrease in recorded music made up for by increased revenues from live music and collective management.37 Recorded music sales is shifting rapidly from CDs to digital distribution. Commercial download services by ensuring quality, speed and freedom from malware will be able to compete with legalised P2P which, being open networks, will continue to suffer from these three issues. After all, iTunes was established at a time when P2P use was already widespread. In particular new business models based on fairness, benefit-sharing and transparency like Magnatune and those promoted by the Fair Music Initiative38 will increasingly attract both artists and paying audiences. According to the International Federation of the Phonographic Industry’s (IFPI) 2010 annual report, album downloads globally rose an estimated 20% in 2009, with Internet and mobile downloads and streams now accounting for more than a quarter of all recorded music industry revenues worldwide.39

An important public policy goal is cultural diversity. Since 2000, the worldwide annual release of new music albums has more than doubled.40 Thus the digital environment is clearly promoting diversity. The increase is due to the activities of independent labels and, since it occured during the time of the rise of file-sharing, one can conclude that file-sharing is helping rather than hindering cultural diversity.

Not the least, a realistic approach has to start from recognizing the fact that current levels of authors’ income are far below the average national income in all professions, that a few stars gain a disproportionate percentage of the revenues, that female authors earn significantly less than their male colleagues and that the typical author income since 2000 has been decreasing.41 These facts are unacceptable for a society that defines the culture and creative industries as its central dynamic. In fact, audiences are aware of this unbearable situation of authors and artists, leading to voluntary payments that are on average higher than the forced payments in services such as iTunes.42 Therefore, it is not unlikely that negotiations about fair rates for the file-sharing levy that primarily involve artists and audiences will result in higher rates than if collecting societies and industry associations were to conduct them alone.

Thus, rather than “compensation” of alleged “damages”, the system should strive to create “sustainable resources for creative activities in the digital era” that ensure that this creativity can flourish and grow.43 The seemingly simple question “How much?” actually leads to the core of the social contract between artists and audiences that is currently being negotiated.

Who should receive Payments?

The levy is due to those who create the works that are shared under the new exception, i.e. authors and performing artists as well as the “auxiliary service providers in the creative process” as German law professor Thomas Hoeren aptly called exploiters. Authors (composers and lyricists, literary authors, film makers, photographers etc.) and publishers as well as musicians and record labels are joined together in their respective collective rights management organisations (CMOs).

The pool of lump-sum payments by Internet users thus has to be distributed first to the CMOs for the different work categories (music, audio-visual works, text, images etc.) based on the measured proportion of these work categories in file-sharing networks, and then within the CMOs to the individual members based on the actual popularity of their works.

In Germany, the thirteen existing CMOs pay at least 50% and up to 100% of their proceeds to authors, while the remainder goes to the exploiters (publishers or record labels).44 In the case of ECAD the situation is more confusing because its members are not authors, musicians and exploiters directly but ten associations of these groups.45 This two-tiered structure creates additional obfuscation concerning money flows and leads to overall adminstrative costs of close to 30% that appear to be among the highest in the world.46 While it is hard to understand in the analogue world how authors and musicians are willing to accept a system in which nearly one third of the money due to them for the use of their works goes into the apparatus for collecting it, it is clear that in the digital online realm collective rights administration will be automated to a maximum degree, ensuring that the maximum amount possible reaches those who acutally create cultural goods.

However, the most recent figures about the percentages of the proceeds that ECAD and its member associations in the end actually distributed in 200547 indicate that also in Brazil more than two thirds of the money for author rights go to authors and more than half of the money for neigbouring rights goes to musicians.

In order to counter-act the star effect of very few artists receiving the largest share and to foster cultural diversity, the community of creators organized in the CMO might decide to make pay-out regressive, i.e. with increasing popularity the percentage per unit would decrease.48 In addition to the remuneration for creators, a part of the collective funds is used for cultural, educational and social purposes: for supporting young artists, the production of new works and the environment of creation and dissemination itself.49

Record labels and publishers are not opposed to flat-rate licensing per se, as we can see in the growing number of contracts with telecommunications companies like Nokia (“Comes with Music”), Internet Sevice Providers like Neuf Cegetel in France and TDC in Denmark and with service providers like Spotify in which complete music catalogues are licensed for a flat-rate payed by the user or embedded invisibly in the price for other products and services (like Nokia’s mobile phones or the products the advertising for which pays streaming services). These Business-to-Business flat-rates are expressly positioned as legal alternatives to P2P file-sharing. In fact, they have nothing to do with the cultural practice of sharing. If users can download songs at all these are usually wrapped in “terminator DRM”: if they leave their ISP, the license expires and they loose all the music they collected from the service.50 For authors and musicians these B2B flat-rates make remuneration nontransparent because the contracts are clouded in non-disclosure agreements.

Secrecy and control of authors and consumers — that is what industry likes about these B2B flat-rates, and this is why they dislike a copyright law-based file-sharing flat-rate that is collectively managed and publicly supervised and ensures authors at least 50% of the revenues.


In order to allocate the levy fairly to authors, performers and publishers, the number of downloads of their works needs to be measured as accurately as possible. CMOs are already distributing some of their proceeds based on actual usage data. Concert organisers and DJs, for example, have to produce play-lists so that the artists whose works are performed can get paid. But performance rights on radio and TV are often, including in the case of ECAD, subject to “indirect distribution” based on a sample of broadcast stations and a sample of the most frequently played songs. In the case of music played in commercial establishments such as restaurants and stores and in the case of the private copying levy, distribution is the most imprecise, based on conjectures from sales and airplay. The latter two methods lead to systematic distortions that favour the most popular artists to the disadvantage of independent label artists. In the Internet environment empirical measurements can be much more precise, encompassing and fair, avoiding the distortions of the so called black box money of the analogue age, ensuring payments to artists deep into the Long Tail and thereby supporting cultural diversity.

A number of methods have been suggested and tested for this purpose. P2P market researchers like Big Champagne51 and infringement investigators like Logistep52 monitor P2P from within, creating detailed reports for their clients. Big Champagne CEO Eric Garland leaves no doubt that the Internet is better empirically suited for close measurement than any other medium.53

Once P2P has been legalised, cooperation from file-sharers themselves could be recruited as well. They would install a module that plugs into Vuze, BitTorrent, Miro, Ares, Mozilla Firefox and other applications used for downloading. This module logs the metadata of each work54 the user downloads from the Internet and sends an anonymised monthly report to an administrative site that then calculates the total number of downloads in a given territory. Audioscrobbler is an example of such a voluntary reporting agent deployed in Last.fm.55 Noank, a prototype file-sharing flat-rate system developed by Fisher and field-tested in Hong Kong, also includes a detection and reporting module that plugs into any media-player of the user’s choice, including iTunes and Windows Media Player.56 Both of these log only music and they log the songs the user listens to rather than downloads. Whether plays or downloads should be the measure of popularity on which to base pay-outs is up to debate. However, how often a DVD is watched or a CD listened to in the private sphere is currently irrelevant to both copyright law and remuneration. A shift from number of reproductions to number of plays would need a solid justification.

A third method suggested, among others, by French economist Philippe Aigrain, also uses plug-ins, but is based not on open participation but on a representative sample of households who volunteer to have their media use monitored. Such a system is already in use by market research company Nielsen that claims to monitor and measure more than 90% of global Internet activity today.57

Aigrain argues that fraud attempts, that this system would surely attract, can be countered by using the data of only 5% of the panel members, chosen randomly. Large-scale collusion or automated generation of fraudulent data is fairly easy to detect and will be subject to deterrent sanctions. He then calculates a feasible sample size that, even if 95% of the data were discarded, would still detect works that are downloaded only several thousand times per year, thus ensuring a positive impact on cultural diversity.58

During the introduction of the file-sharing levy these and possibly other methods of measuring will be used concurrently in order to check the results against each other and optimize the methodology to ensure fraud prevention, remuneration of the greatest possible diversity of cultural expressions and cost-effectiveness.

Collective Rights Management

It is evident that the significance of collective rights management is greatly increasing in the digital age. When WIPO in 1999 newly established its Copyright Collective Management Division (CCMD), it explained its rationale thus: “The experience of recent years has increasingly confirmed that the individual exercise of rights is impractical; … Collective management is an essential tool for the efficient exercise of rights; collective management societies therefore play an important and very useful role, both for authors/creators and for users. This is definitely why they have experienced considerable development in parallel to the increased use of works made possible by new technology.”59 A recent report by a French government commission urges the expansion of existing and the creation of new collective management arrangements in order to simplify online licensing, including the proposal to subject the making-available right to mandatory collective management.60

The Brazilian Ministry of Culture (MinC) as well made it clear in its discussion paper for the current copyright law reform61 that it favours the extension of collective management. It proposes to expand the scope of the private copying exception and introduce a collectively managed levy on it. It also encourages the formation of a CMO for the public performance rights of audiovisual works, one for reprographic rights — in the hope “to finally resolve the conflict between the owners of literary works and the teachers and students of educational institutions” — and additional ones for other categories of works.

At the same time, criticism in particular of music CMOs by their members and users is also increasing. In Germany a public petition initiated by concert organiser Monika Bestle in May 2009 urging the Bundestag to review the conduct of the music CMO GEMA and start a comprehensive reform of that organisation received more than 100,000 signatures.62 It will lead to a public hearing in the first quarter of 2010.

Like many other music CMOs across the globe, ECAD is criticised for its lack of transparency, internal democracy and equity in distributing proceeds to its members. Its conduct gave rise to several parliamentary investigations. The most recent one by the Legislative Assembly of São Paulo in April 2009 led by congressman Bruno Covas concluded that the collective management of music rights is in a “state of institutional anarchy.” The final report states that “this anarchy allows ECAD to exceed its financial, legal and statutory obligations, giving rise to irregularities and evidence of crimes such as forgery, tax evasion, embezzlement, illicit enrichment, conspiracy, formation of a cartel and abuse of economic power.”63

To address this issue, the MinC proposes to create a new regulatory agency, the Instituto Brasileiro de Direito Autoral (IBDA) under the Ministry of Culture. It will supervise, regulate and promote the collective administration of rights, provide administrative dispute mediation and organise the registration of works. The latter would greatly aid in the identification of works in the measurement of downloads and in allocating the levy share to its beneficiaries.

If collective management is crucial for the future of creativity in the digital age, then an institutional framework ensuring internal democracy, fair representation, transparency in the allocation of funds and public oversight is crucial for the future of collective management. The IBDA promises to be a key instrument on the way towards this goal.


The MinC’s proposals for reforming the Brazilian copyright law constitute an important framework for the emerging new social contract between authors and audiences. However, it has shied away from addressing the pressing issue of mass-scale user-distribution of copyright protected works.

This paper is intended to encourage the MinC and others involved in the copyright law debate to consider a collectively managed file-sharing exception as a model for striking a balance between authors’ rights and users’ rights with respect to this important digital challenge.

The debate on this model was initiated ten years ago by copyright law scholars and soon joined by members of the music community and other creative sectors, members of CMOs, consumer and Internet user organisations, economists, technologists and eventually by political parties. Today large-scale open debates take place on a new social contract over creativity: in the broad alliance of artists, consumers and the Internet community “Création Public Internet” in France,64 in the ongoing effort to negotiate an agreement between creative communities and the public that seeks to improve access to and income for knowledge goods under the name “The Paris Accord”65 and in the forum “Artists-to-fans-to-artists” initiated by musician Billy Bragg.66 Members of Parliament in countries like France and Italy are tabling bills to implement a file-sharing permission in copyright law. The Isle of Man is about to start a trial on it.67

In Brazil the process of reforming copyright law exceptions and the collective management system is well underway. The country should seize the opportunity, adopt a model whose time has come and lead the way into an equitable future that combines the freedom to create and the freedom to share.

The decisive question is how we as citizens of the knowledge society want see ourselves: Do we prefer to see ourselves as consumers with the choices of products and services the market offers, and as objects of market research, advertising, surveillance, technological restrictions, deterrent campaigns and juridical repression? Or do we see ourselves as partners in an arrangement where we all provide creatives whose works we enjoy and share with each other with decent working and living conditions to create them?


* VRG is a media-sociologist focussing on the digital revolution and the ensuing changes in cultural practices. He has conducted research on the knowledge order of digital media, on copyright and the knowledge commons at Free University Berlin, at Tokyo University and at Humboldt University Berlin and is currently a visiting professor in the Grupo de Pesquisa em Políticas Públicas para o Acesso à Informação (GPOPAI) at the Escola de Artes, Ciências e Humanidades (EACH) of the Universidade de São Paulo (USP). He can be reached at <ralf.grassmuck at usp.br>. He wishes to thank the members of GPOPAI for invaluable insights and great discussions, in particular Maria Carlotto, Pablo Ortellado, Felipe Sentelhas and Jorge Machado. Very special thanks go to Bráulio Araújo who translated the text into Portuguese.

1 <a rel=”license” href=”http://creativecommons.org/licenses/by-sa/3.0/br/”><img alt=”Creative Commons License” style=”border-width:0″ src=”http://i.creativecommons.org/l/by-sa/3.0/br/88×31.png” /></a><br /><span xmlns:dc=”http://purl.org/dc/elements/1.1/” href=”http://purl.org/dc/dcmitype/Text” property=”dc:title” rel=”dc:type”>A Copyright Exception for Monetizing File-Sharing. A proposal for balancing user freedom and author remuneration in the Brazilian copyright law reform</span> von <span xmlns:cc=”http://creativecommons.org/ns#” property=”cc:attributionName”>Volker Ralf Grassmuck</span> steht unter einer <a rel=”license” href=”http://creativecommons.org/licenses/by-sa/3.0/br/”>Creative Commons Attribution – Share Alike 3.0 Brazil Licence</a>.

2 é pau, é pedra, é pena. Entrevista com Pena Schmidt, 20 March 2009, http://pedroalexandresanches.blogspot.com/2009/03/e-pau-e-pedra-e-pena.html

3 See Bernt Hugenholtz, Lucie Guibault & Sjoerd van Geffen, The Future of Levies in a Digital Environment, IvIR, University Amsterdam, March 2003, p. 11 ff., http://www.ivir.nl/publications/other/DRM&levies-report.pdf

4 The primary use is the sale of a work by its author to a publisher or the sale of a recording by a band to a record label. A secondary use is the broadcasting of a music record by a radio station or the private copying of a record onto audio tape. A tertiary use is the recording of a radio broadcast of a record onto audio tape.

5 E.g. in a UK study from spring 2008 commissioned by industry organisation British Music Rights (BMR) and conducted by the University of Hertfordshire (“Music Experience and Behaviour in Young People”, http://www.ukmusic.org/cms/uploads/files/UoH%20Reseach%202008.pdf) 63% of respondents admitted to downloading music from P2P file-sharing networks. In Brazil, according to Nielsen (IBOPE) NetRating in March 2008, 41.4% of all residential Internet users in the country were using P2P and file-hosters to download music, films and TV series (http://idgnow.uol.com.br/internet/2008/05/05/p2p-9-4-mi-de-brasileiros-baixam-conteudo-pela-internet-diz-ibope/). According to Ipoque’s measurement, South America ranks third among the regions studied behind Eastern Europe and Southern Africa, with 65% of its Internet traffic generated by P2P (Ipoque, Internet Study 2008/2009, http://www.ipoque.de/resources/internet-studies/internet-study-2008_2009, p. 5.)

6 “Raubkopierer sind Verbrecher” is an ongoing campaign by the marketing company of the German movie industry: http://www.hartabergerecht.de/

6a In contrast to shallow packet inspection which only looks at header information, DPI devices inspect the actual content of Internet data packets and allow to report, redirect, rate limit and block any content that meets pre-defined criteria. DPI is used by ISPs, enterprises and governments for a variety of purposes, including the detection of spam, viruses and other forms of attacks, criminal investigations, prioritizing certain applications like Voice-over-IP while slowing down or blocking peer-to-peer traffic, injection of personalized advertising, surveillance, censorship and copyright enforcement.

7 Virgin Media to trial filesharing monitoring system, The Register, 26.11.2009, http://www.theregister.co.uk/2009/11/26/virgin_media_detica/

8 A technique known as distributed hash tables (DHT) allows to locate files by querying other peers in the BitTorrent swarm, removing the need for centralized trackers (s. Nate Anderson, Pirate Bay moves to decentralized DHT protocol, kills tracker, Ars Technica, 17 November 2009, http://arstechnica.com/tech-policy/news/2009/11/pirate-bay-kills-its-own-bittorrent-tracker.ars). The Pirate Bay has moved to this next generation architecture. To a user, the index of The Pirate Bay (http://thepiratebay.org/) feels like before, but now it does not host torrent files any more, but only “magnet links.” TPM is thus nothing more than a search engine for .torrent files, like Google (http://www.google.com/search?q=filetype%3Atorrent).

9 For example, OneSwarm (http://oneswarm.cs.washington.edu/) is a privacy preserving friend-to-friend network (F2F) developed at the Computer Science Department of Washinton University with a grant from the US National Science Foundation (NSF). OneSwarm uses public-private keys to encrypt participant’s IP addresses, manages them in a distributed hash table (DHT) and encrypts the file-exchanges with SSL.

10 Adam Frucci, The Secret World of Private BitTorrent Trackers, Gizmodo, 19.02.2010, http://gizmodo.com/5475006/the-secret-world-of-private-bittorrent-trackers

11 Internet pirates find ‘bulletproof’ havens for illegal file sharing, The Guardian, 05.01.2010, http://www.guardian.co.uk/technology/2010/jan/05/internet-piracy-bulletproof

12 E.g. in his talk at re:publica 09, quoted in: VRG, The World Is Going Flat(-Rate), IP Watch, 11 May 2009, http://www.ip-watch.org/weblog/2009/05/11/the-world-is-going-flat-rate/

13 Promises to Keep, Stanford University Press 2004, http://www.tfisher.org/PTK.htm

14 Impose a Noncommercial Use Levy to Allow Free P2P File-Swapping and Remixing, November 2002, http://www.utdallas.edu/~liebowit/knowledge_goods/netanal%20levy.pdf

15 http://www.lalliance.org/

16 Philippe Aigrain, Internet & Création, Editions InLibroVeritas, October 2008, http://www.ilv-edition.com/pdf_ebook_gratuit/internet_et_creation.pdf; see also Réponse de l’UFC Que Choisir á Mission Zelnik, 28 September 2009, http://www.creationpublicinternet.fr/blog/public/mission_C_I_reponse_UFC-Que_Choisir.pdf

17 Like Switzerland and the Netherlands. S. Allard Ringnalda, Mirjam Elferink en Madeleine de Cock Buning, Auteursrechtinbreuk door P2P filesharing, Regelgeving in Duitsland, Frankrijk en Engeland nader onderzocht, WODC, July 2009. http://www2.law.uu.nl/priv/cier/Documenten/PDFned/RAPPORT%20Filesharing%20WODC%20DEF%2090902.pdf, p. 20

18 Silke von Lewinski, Mandatory Collective Administration of Exclusive Rights – A Case Study on its Compatibility with International and EC Copyright Law, UNESCO e.Copyright Bulletin, No. 1, January – March 2004, http://portal.unesco.org/culture/en/files/19552/11515904771svl_e.pdf/svl_e.pdf

19 Carine Bernault & Audrey Lebois under the supervision of Professor André Lucas, Peer-to-peer et propriété littéraire et artistique. Etude de faisabilité sur un système de compensation pour l’échange des œuvres sur internet, Institut de Recherche en Droit Privé de l’Université de Nantes, June 2005, http://alliance.bugiweb.com/usr/Documents/RapportUniversiteNantes-juin2005.pdf; English translation, March 2006: http://privatkopie.net/files/Feasibility-Study-p2p-acs_Nantes.pdf

20 http://www.kopinor.org/layout/set/print/content/view/full/2090

21 Promoted by the researchers at NEXA, Center for Internet and Society at Politecnico di Torino: http://nexa.polito.it/licenzecollettive

22 EML, The Admissibility of a Culture Flat-Rate under National and European Law, 13.03.2009, http://www2.malte-spitz.de/uploads/emr_study_culture_flat_rate.pdf; the German original version is available here: http://www.gruene-bundestag.de/cms/netzpolitik/dokbin/278/278059.kurzgutachten_zur_kulturflatrate.pdf

23 On eDonkey the largest category is music. BitTorrent supports larger files, so the largest categories are audio-visual and software works, including games. See e.g. Ipoque, Internet Study 2008/2009, op. cit.

24 The German game developer association GAME supports both the file-sharing exception and the inclusion of games (Malte Behrmann, executive director of GAME, in an interview with VRG on 19.6.2009)

25 For the same reason of transparency, some Brazilian movie theaters indicate on the movie ticket the copyright levy paid to ECAD for the music performed in the screening.

26 STIM, Pirates, file-sharers and music users. A survey of the conditions for new music services on the Internet, February 2009, http://www.stim.se/stim/prod/stimv4eng.nsf/Productions/B5CA55F631B0F152C125759E0030BD74/$File/pirates_filesharers_music_users.pdf

27 http://2dboy.com/2009/10/26/pay-what-you-want-birthday-sale-wrap-up/

28 Studies have shown consistently that those who downloaded from P2P networks are significantly more likely to buy a CD or a game, go to a concert and go to a movie than non-file-sharing internet users (S. e.g. Dutch Ministry for Economic Affairs, Ups and Downs. Economic and cultural effects of file sharing on music, film and games, January 2009, http://www.ivir.nl/publicaties/vaneijk/Ups_And_Downs_authorised_translation.pdf).

29 Promises to Keep, Chapter 6: An Alternative Compensation System, p. 21, http://cyber.law.harvard.edu/people/tfisher/PTKChapter6.pdf.

30 Comitê Gestor da Internet no Brasil, Pesquisa sobre o Uso das Tecnologias da Informação e da Comunicação no Brasil TIC Domicílios e TIC Empresas 2008, São Paulo 2009, http://www.cetic.br/tic/2008/index.htm.

31 Público de cinema cresce mais de 25% no País em 2009, A Tarde, 13.1.2010, http://www.atarde.com.br/cultura/noticia.jsf?id=1337032

32a (Source: Ancine and União Brasileira de Vídeo.) The majority of this drop is attributed to the concentration process in the rental sector by Lojas Americanas acquiring Blockbuster in 2007, leading to a sharp decrease in the number of video rental shops and a corresponding drop in sales of DVDs to these companies. End-consumer sales dropped from an all-time high of 21.5 million units in 2007 to 20.1 million in 2008. The end of the novelty of the DVD and of the sales bubble it created is another factor in the complex market dynamics, leading to a drop of the retail price. Competition from Internet, computer games and other forms of home entertainment taking up ever more of consumers’ time is yet another element. “Piracy” is only a marginal factor concludes Sousa in her analysis (Ana Paula Sousa, Efeito dominó – Como a queda do mercado de DVDs no Brasil pode afetar o cinema, Revista Filme B, May 2009, http://www.filmeb.com.br/portal/html/materia10.php).

32 Câmara Brasileira do Livro Sindicato Nacional de Editores de Livros, Produção e Vendas do Setor Editorial Brasileiro, Relatório, São Paulo, August 2008, p. 7.

33 Associação Brasileira de Produtores de Discos (ABPD), http://www.abpd.org.br/estatisticas_mercado_brasil.asp

34 http://www.ecad.org.br/

35 Números do Ecad apontam recorde. Valor distribuído aos artistas em 2009 foi 17,06% maior do que no ano anterior, Estadao, 06.03.2010, http://www.estadao.com.br/estadaodehoje/20100306/not_imp520212,0.php

36 See Felix Oberholzer-Gee & Koleman Strumpf, File-Sharing and Copyright, Harvard Business School Working Paper 09-132, 2009, http://www.hbs.edu/research/pdf/09-132.pdf

37 See e.g. Daniel Johansson and Markus Larsson, The Swedish Music Industry in Graphs. Economic Development Report 2000 – 2008, Trend Maze and Royal Institute of Technology, December 2009, p. 6. The dynamics in other countries are comparable to those in Sweden, as the upcoming research report of GPOPAI will show for the Brazilian music industry.

38 http://fairmusic.net/manifest/lang-pref/en/

39 IFPI, Digital Music Report 2010, http://www.ifpi.org/content/section_resources/dmr2010.html

40 Felix Oberholzer-Gee & Koleman Strumpf, File-Sharing and Copyright, Harvard Business School Working Paper 09-132, 2009, http://www.hbs.edu/research/pdf/09-132.pdf, p.1.

41 Martin Kretschmer and Philip Hardwick, ALCS Study: Authors’ Earnings from Copyright and Non-Copyright Sources: A Survey of 25,000 British and German Writers, 13 July 2007, http://www.cippm.org.uk/alcs_study.html. The study is the largest of its kind, and one of the first that systematically sets authors’ income into a context of earnings data available for other professions, and in other countries. It is also the first that was able to control the results against collecting society payments, as well as tax, insurance and labour force data held by government statistical offices.

42 Leah Belsky, Byron Kahr, Max Berkelhammer and Yochai Benkler, Everything in its right place: Social Cooperation and Artist Compensation, Berkman Center Working Paper, upcoming, March 2010.

43 Philippe Aigrain & Suzanne Aigrain, Sharing and the Creative Economy: Culture in the Internet Age, p. 23 ff., draft, January 2010, http://paigrain.debatpublic.net/?page_id=171&lp_lang_view=en

44 GEMA, the German CMO for music works, pays roughly two thirds to authors, i.e. composers and lyricists, and one third to publishers (see GEMA-Jahrbuch 2009/2010, http://www.gema.de/presse/publikationen/jahrbuch/ausgabe070800/, p. 279). GVL, the German CMO for music recordings pays 50% to performers and 50% to record labels for broadcast and rental, and 64:36 in favor of musicians for public performance (See GVL Verteilungspläne 2008, https://www.gvl.de/pdf/verteilungsplaene-2008.pdf, p. 2) VG Wort, the German CMO for text works, pays 70% to authors and 30% to publishers. Revenues from clipping services go to 100% to authors (see Verteilungsplan der Verwertungsgesellschaft Wort, 2009, http://www.vgwort.de/files/verteilungsplan_vgwort_2009.pdf, p. 1).

45 The six “effective” and four “administrative” member associations are listed here: http://www.ecad.org.br/ViewController/Publico/conteudo.aspx?codigo=21. The two that receive the by far largest share of the distribution by ECAD are União Brasileira de Compositores (UBC) and Associação Brasileira de Música e Artes (ABRAMUS).

46 Neither ECAD nor its member associations publish annual reports. ECAD states on its website: “Of the total proceeds, 17% are intended for ECAD and 7.5% for its member associations for their administration expenses. The remaining 75.5% are paid out to the affiliated rights holders.” (http://www.ecad.org.br/ViewController/publico/conteudo.aspx?codigo=25). This yields a total administrative overhead of 24.5%. Calculating the differences between annual arrecadação and distribuição from 2004 to 2008 as published by ECAD (http://www.ecad.org.br/ViewController/publico/conteudo.aspx?codigo=52) yields actual administrative costs withheld by ECAD varying from 23.26% in 2006 to 16.44% in 2005. The total amount distributed by member associations in the years 2003 to 2005 published by Abramus (http://www.abramus.org.br/arquivos/2003.pdf, …2004.pdf and …2005.pdf) shows that the administrative overhead of the member associations has grown from 7,83% in 2003 to 12,75% in 2005, leading to total administrative costs of between 31,07% in 2003 and 26,29% in 2004.

47 Available at the site of ABRAMUS: http://www.abramus.org.br/arquivos/2005.pdf

48 This is how GVL, the German CMO for music recordings, pays to its artist members. Based on the reported income from the primary exploitation of their works, musicians and vocalists who earn up to € 50,000 per year get 100%, income between € 150,000 and 200,000 receives only 25% and beyond that only 10% (see GVL Verteilungspläne 2008, https://www.gvl.de/pdf/verteilungsplaene-2008.pdf, p. 1)

49 Required by law to do so, German CMOs currently set aside 10 to 25% of the common funds for such purposes. Aigrain (2008) proposes to use 50%.

50 This creates a strong lock-in effect. When a customer is dissatisfied with her current ISP or gets a better cost/performance ratio from a competitor, she will think twice of moving when the price is losing large parts of her music collection. These terminator DRM policies call for an examination for anti-competitive activities.

51 http://bigchampagne.com/

52 http://www.logistepag.com/

53 See interview by Music Ally, January 2010: http://musically.com/blog/2010/01/12/metrics-eric-garland-bigchampagne/

54 E.g. based on audio- and video fingerprints, i.e. digital summaries of works that allow their identification without the need to embed identifiers like digital watermarks into files. See Audible Magic (http://www.audiblemagic.com) for a commercial service and MusicBrainz (http://musicbrainz.org/doc/AudioFingerprint) for a free software implementation.

55 http://en.wikipedia.org/wiki/Last.fm#Audioscrobbler_plugin

56 http://www.noankmedia.com/howitworks.html

57 http://en-us.nielsen.com/tab/product_families/nielsen_netratings

58 Aigrain, 2008, op. cit., p. 93 ff.

59 World Intellectual Property Organization, Permanent Committee on Cooperation for Development Related to Intellectual Property, Promotion and Development of Collective Management of Copyright and Related Rights (PCIPD/1/7), 25 May 1999, http://www.wipo.int/edocs/mdocs/mdocs/en/pcipd_1/pcipd_1_7.pdf, p. 3.

60 Mission Zelnik, Creation et Internet, Rapport au Ministre de la Culture et de la Communication, January 2010, http://www.culture.gouv.fr/mcc/Espace-Presse/Dossiers-de-presse/Rapport-Creation-et-Internet

61 Diagnóstico das discussões do Fórum Nacional de Direito Autoral e subsídios para o debate. Documento apresentado no III Congresso sobre Direito de Autor e Interesse, 10.11.2009, Públicohttp://www.cultura.gov.br/site/2009/11/10/diagnostico-das-discussoes-do-forum-nacional-de-direito-autoral-e-subsidios-para-o-debate/

62 https://epetitionen.bundestag.de/index.php?action=petition;sa=details;petition=4517

63 http://brunocovas.com.br/CPI-do-Ecad-chega-ao-fim/

64 http://creationpublicinternet.fr/

65 http://www.tacd-ip.org/blog/the-paris-accord/

66 http://a2f2a.com. S.a. Billy Bragg, Canadian songwriters push for new approach to downloading, By Jennifer Ditchburn (CP), 20.11.2009, http://www.google.com/hostednews/canadianpress/article/ALeqM5h7DWd6762i_yzihYrpbI9yHK-bBQ

67 The plan was first announced at MIDEM 2009 (See: Inside the Isle of Man’s £1/month unlimited music plan, 25.2.09, http://arstechnica.com/media/news/2009/02/inside-the-isle-of-mans-1month-unlimited-music-plan.ars) and has greatly advanced since then.


  1. In response to this text, Richard Stallman reminded me of his 1992 proposal “The Right Way to Tax DAT”. He did not, of course, refer to P2P file-sharing which was invented only seven years later. He rather responded to a bill dealing with digital audio recording that was passed into US copyright law later that year as the Audio Home Recording Act (AHRA)”.

    Richard’s proposal, I think, is interesting for two reasons. First, the AHRA is, to my knowledge, the only private copying exception subject to a levy in Anglo-American copyright law. The US equivalent to the Continental European private copying exceptions is the Fair Use doctrine. It does not require payment of a levy but it is not a proper permission either, but rather a defence against allegations of infringement subject to a four-factor test to be applied by a court in each individual case. Therefore, it is interesting that a common law jurisdiction for accommodating digital audio recording found a solution very similar to the private copying exception in civil law. What is even more interesting is that the father of Free Software in principal endorsed this balancing of the freedom to make private copies and the freedom of technology from restrictions with a collective payment to the authors whose works are being copied.

    Secondly, while Richard agrees that a levy should be collected on digital audio recorders and media, he proposes a number of modifications of the model that was discussed in 1992 which are also relevant to the file-sharing levy proposed here. He argues that the distribution of the levy pool should be based on measuring — by means of a survey — the actual songs being copied. This is an essential part of the file-sharing levy as well.

    Then he writes that the money should not distributed in strict proportionality to the number of times an artist’s work is copied, because “then a large share will go to make a few superstars even richer than they are now.” Instead, he suggests that as the number of copies of a work increases its share of the levy decreases. This amounts to a re-distribution from superstars to non-stars but, like Richard, I think it is justified by the priority interest of society to promote cultural diversity. It also has a precedent in progressive income taxation and in the way at least one collecting society currently distributes its proceeds to its artist members (see footnote 48 in the text above).

    Finally, he suggests that the entire levy should go to music authors and musicians and none of it to publishers and record labels. He argues that “the musicians and composers are the ones who truly create the music. In principle, we could do without record companies entirely.” While I agree that artists should be the primary beneficiaries of the system, I’m not sure about excluding publishers and labels. There are not only the four global major labels but thousands of independent labels that are responsible for 80% of the world’s album releases. If artists wouldn’t get anything out of contracting with them and they could do everything on their own, labels wouldn’t exist any more. Thus assuming that publishers and labels provide a valuable service to artists and audiences, how can we justify excluding them from the financial returns from file-sharing the works they helped create? I do think that it is a structural problem that authors and publishers, musicians and labels make up the same collecting societies. It prevents authors and artists from speaking up for their own interests which are not the same as those of the exploiters of their works. Ideally, re-booting the collecting society system would involve setting up separate organisations for these two groups. But even so, I’m not sure publishers and labels should not get a fair share from the file-sharing levy. What would a fair share be, 50%, 25%? How could one establish this fairness? If what publishers and labels are doing is really dispensable, can we expect that artists will stop working with them shortly? What do you think?

  2. Fantastic contribution. Thanks for posting that. I was not really knowledgeable of your website, however will return back again much more often now. Adding you to my personal favorites.

  3. Pingback: PPC Manager

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.