p2pnet.net News Feature:- The digital 21st century has presented the members of the record label cartel and and major studios with challenges which, so far, they’ve individually and collectively failed to meet.

In Big Music’s corner are: Warner Music Group (US), Universal Music Group), France, Sony BMG Music Entertainment, Japan and Germany; and the EMI Group, UK. The Big Seven studios are: Disney, Warner Bros, MGM, Universal, Fox, Paramount; and, Sony.

Under the Digital Media Project, the Berkman Center for Internet & Society at Harvard Law School have just released an important paper which states, "The online environment and new digital technologies threaten the viability of the music and film industries’ traditional business models."

Below, in full, is the core paper.

Go here for more detailed information about each of the four business models available in the appendices, or download the complete document here.

The appendices are: 

• Digital Media Store
• P2P Stores
• Collective Blanket Licensing
• Ancillary Products and Services

Read on >>>>>>>>>>>>>>>>>>>>>>>>

CONTENT AND CONTROL
Assessing the Impact of Policy Choices on Potential Online Business Models in the Music and Film Industries

Introduction
New digital technologies and the online environment pose significant challenges to the traditional business models of the music and film industries. The digital era threatens current revenue models by changing the environment in which copyright operates. To prevent unauthorized copying of their works, copyright holders have traditionally relied on practical barriers as well as their legal exclusive rights to control reproduction and distribution. The new technologies vitiating those practical barriers?peer-to-peer (P2P) services, digital compression technologies, and others?are demonstrating just how empty those legal rights may be and how poorly matched they may be with cultural norms and practice. Consumers are exploiting the exciting potential for greater interactivity and involvement with content, but also the opportunity to acquire content illicitly, and are thus finding themselves in conflict with many of those who make content possible.

The industries have responded in two ways. First, incumbents, particularly the major record labels and studios and their respective trade associations, have focused their initial energies on protecting traditional roles and revenue streams through the legal system. Three areas of concern (which we will refer to as "policy nodes") have seen the greatest industry activity and ensuing public debate: technical interference with and potential liability of P2P services; the civil and criminal liability of infringing consumer distributors; and the implementation and legal reinforcement of digital rights management technologies (DRM). The industries have brought numerous civil suits against companies and consumers across these nodes, while lobbying vociferously for enhanced forms of government intervention and enforcement.

The second response consists of attempts by new industry players - and, more recently, some incumbents ? to develop new business models designed to capitalize on novel features of the online environment. These new models range from slight variations of offline models to more radical re-conceptualizations of the roles of and relationships among content creators,¹ intermediaries, and consumers.

Government intervention to privilege any particular business model at this point would have significant effects beyond achieving private benefits for those following the model in question. Different business models offer disparate economic surplus and social costs and benefits, in terms of the quantity and diversity of content created and distributed; the public’s rights in copyright, such as fair use and first sale; and technological innovation, among other factors. Legislation designed to support particular models or incumbents could suppress alternatives that would otherwise have proven better for society as a whole. Although government action may be necessary at some point, it should respond to empirically demonstrated problems - not predicted crises - and must be carefully designed to balance competing interests and concerns.

This paper seeks to identify potential implications of the actions sought from policymakers across the three policy nodes on the future of the music and film industry.² To do so, it describes the organizing principles and structures of four exemplar business models and how policymaker choices across the nodes would help or hinder their adoption and viability. The models represent a disparate range of responses to the industries’ current challenges and are presented according to the degree of change they represent from the traditional business model.

The first model, the Digital Media Store, is essentially the offline, traditional retail model moved online. The P2P Store model embraces distribution technologies and authorized consumer distribution while continuing to seek per-copy revenues. The Collective Blanket Licensing model acknowledges the difficulties of controlling consumer access and use and offers flat-fee, blanket licenses for non-commercial distribution and copying. Finally, the Ancillary Products and Services model lets consumers distribute copies for free, leveraging file sharing to drive other revenue streams.

Organization and methodology
To contextualize our examination of the interaction of the policy nodes and the business models, we first consider why and how digital technologies challenge the traditional business models of the film and music industries. We then describe industry activity across the policy nodes to date and the social costs and benefits of their actions and proposals at each. These discussions provide a foundation for the analysis of each model and the impact of policy actions on it in turn. The discussion of each business model in the body of the paper is similarly structured to allow for ease of comparison. For each, there is also an appendix offering a more in-depth, standalone evaluation.

While evaluating each model on a range of dimensions, our analysis is limited in the following ways. The four models are not exhaustive of all potential outcomes and variations. We generally analyze the models as if adopted by the whole industry, but, just as some of the models co-exist today, it is possible, indeed likely, that different industry segments would pursue different models in the future. The paper also does not attempt to identify which model represents a better outcome or is a more likely one. Finally, this paper does not attempt to address the market for physical media of films and music except as it relates to the emergence of the online market.

The impact of digital technologies 3
Increasingly efficient distribution and audio compression technologies threaten traditional business models by enabling consumers to distribute content easily and cheaply. The capabilities for copying and distributing content formerly reserved to well-capitalized entities with expensive equipment and expertise are now readily available to the general public. As obtaining high-quality copies for free becomes trivially easy, the ability to charge for access to copies becomes tenuous. Even though evidence that file sharing has caused losses to the music industry is controversial4 and film industry revenue is currently on the rise,5 online infringements reasonably can be expected to reduce revenues in the long run.6

Digital technologies can also reshape the roles of certain intermediaries within the industries and their relationships to content creators. A number of functions and entities currently stand between creator and consumer, with record labels and film studios being the most important. Within the existing dominant business structure, record labels and film studios provide the substantial sums necessary to create, distribute, and market content. For undertaking large capital investments and ensuing financial risks, the labels and studios often receive certain exclusive rights7 from creators and share in the works’ returns. In the music industry, labels recoup all costs before artists see profits and, indeed, few artists ever do.8 A small minority of artists, heavily invested in and promoted by the labels, generates the vast majority of record sales and royalties. Similarly, major studios invest substantially in a few productions as well as star actors and directors; these productions make up large percentages of industry revenue, and creators’ income is rather concentrated, though less so than in the music industry.9

The digital environment can reduce costs and alter or eliminate certain functions. Traditionally, the high costs of creation were accompanied by high costs for physically manufacturing the media, marketing the work, and distributing it, from packaging to shipping to retailing. In the digital environment, new production technologies have dramatically lowered the threshold cost for creating music content, and, to a substantially lesser extent, in films.10 The cost of manufacturing each additional digital copy is practically zero. The online environment provides myriad, potentially cheaper and better-targeted opportunities for marketing.11 Finally, the Internet enables low cost distribution among content creators, retailers, and consumers, perhaps liminating the need for standalone retailers altogether.12

These changes may enable greater independence for content creators as well as increased content diversity. Recording artists and film creators might improve their negotiating position with labels and studios if the upfront investment needed was less substantial. In some cases, it may enable content creators to act altogether independently of the traditional intermediaries. Moreover, the class of content creators who can achieve financial viability might also be larger given the reduced revenues needed to see a profit. In particular, the Internet may help make creators in niche genres more sustainable. The online environment can help content creators and consumers find one another and execute transactions that in the past would have been infeasible. The degree to which these possibilities may be realized is different for the film and music industries, of course, due to the generally significantly greater costs of creating films.

THE THREE POLICY NODES
Distribution technologies like P2P, which enable efficient consumer-to-consumer distribution, are the subject of the first node. Second and closely related is the treatment of consumers who illicitly distribute content. The use and enforcement of DRM is the final node.

P2P and other distribution mechanisms
P2P file sharing, among other online distribution mechanisms, enables individuals to find, acquire, and share content with great efficiency. Although the technology is equally advantageous for licit content, P2P is overwhelmingly used for infringement today. As a result, the industries’ primary response to date has been to attempt to interfere with P2P in various ways. The industries try to reduce the ease and convenience of P2P through "spoofing"13 - flooding P2P systems with bogus versions of files. Rights holders have also sued P2P service providers under theories of secondary copyright infringement.

The efficacy of these approaches is questionable. Technologies to counter spoofing have already been developed.14 Though the original Napster and other similar services were forced to shut down,15 courts have so far rejected suits against fully decentralized P2P providers, which have no control over their users.16 Even if these services were held liable (under current or newly legislated doctrines), they cannot be shut down, and suppressing the software’s distribution, potentially by offshore companies or hobbyist programmers, may prove impossible.17

In response to these difficulties, the industries are in some cases working with new P2P services to try to capture revenues from users of the technology rather than simply trying to fight it.18 Of the four models considered here, only the Digital Media Store would benefit from entirely precluding P2P systems. In distinct ways, the others embrace P2P networks for the marketing opportunities, decreased distribution costs, and large online consumer base they offer.

For policymakers, the options related to P2P must be assessed on a number of dimensions. The role of P2P in supporting or hindering the viability of certain online business models is important. P2P technology is essentially a tool for searching for and transmitting files which is indifferent to whether it is used for licit and illicit purposes. For this reason, along with threatening the legitimate uses of P2P, attaching liability to use of the technology will almost inevitably impact myriad related technologies. Fearing liability, technology creators will constrain their innovation and resultant technologies, potentially limiting novel legitimate uses and tools.19

Illicit consumer distributors of content
Beyond the distribution systems are the individual distributors themselves. Over 60 million Americans have participated in P2P file sharing.20 Some consumer distribution is entirely legal, authorized by the copyright holder or otherwise permitted by law. Only the industries’ reaction to infringing distribution and downloading is of concern here.21

Copyright law creates substantial civil and criminal penalties for unauthorized downloading and distribution of copyrighted content. The music and film industries have sought legal redress against thousands of allegedly infringing distributors ("uploaders").22 Their primary intent is to increase the true cost of "free" content, deterring consumers from file sharing. While criminal penalties could pose a more drastic threat to infringers, current standards under US law render criminal prosecutions difficult to pursue in most P2P file sharing cases.23

The Digital Media Store, P2P Store, and Collective Blanket Licensing models continue to derive revenues from copies of artistic works and could benefit from deterrence through lawsuits. However, the latter two enable and embrace consumer distribution only in particular authorized ways. In the Ancillary Products and Services model, all non-commercial distribution is freely allowed.

For policymakers, evaluating the treatment of unauthorized consumer distributors is a complicated calculus. Infringing uploading and downloading violates the law24 and thus should be subject to sanction. Lawsuits and prosecution could be beneficial not only in enabling business models, but also in reconciling norms with the law.

However, the legal regime is most successful when it is consistent with prevailing norms and practices, and the ability of lawsuits and prosecutions to reduce infringement significantly is controversial at best, futile at worst. Some surveys suggest that the initial barrage of lawsuits reduced P2P use, but studies have shown that P2P traffic has continued to increase,25 and P2P systems are now adding anonymizing features that may impede the industries’ efforts.26

In turn, enforcement may do more harm than good. As millions violate the law as part of their daily lives, "respect for the law" in general may decline.27 To some, it also seems excessive and unfair to levy heavy sanctions nearly at random on individual infringers.28

Altering copyright law or copyright holders voluntarily licensing consumers as distributors could create the same social benefits of reconciling norms with law without the costs of aggressive enforcement. Indeed, enabling legitimate distribution may also create social benefits by transforming consumers into active participants in the shaping of culture, rather than passive recipients of artifacts created by a few institutions. Lateral rather than top-down distribution would be accompanied by an increase in the influence of peers’ interests and tastes.

Digital rights management technologies (DRM)
Copyright holders can employ DRM to control how their intellectual products can be used, including restrictions on playing, copying, and distributing content. Its utility to the industries is potentially manifold: for example, it can theoretically limit infringing distribution and enable rights holders to price discriminate, exacting payment for various consumers’ uses.

Current law and industry practice reinforce DRM in several ways. The Digital Millennium Copyright Act (DMCA) prohibits the manufacture and distribution of devices that circumvent DRM controlling access to or copying of a copyrighted work; it also prohibits the act of circumventing access controls. Though the DMCA has a limited set of exceptions, typical defenses to infringement, like fair use, do not apply.29 Rights holders are pervasively implementing DRM in the offline and online markets and have successfully sued circumvention device creators under the DMCA.30

The FCC has also recently mandated that all devices capable of receiving digital television programming manufactured in July 2005 and later must recognize a "broadcast flag" - information that specifies how received content can be copied. Devices may only output protected content in a degraded form or to other broadcast flag compatible devices and must be robust against user access or modifications that might permit access to the unmediated digital stream. Technology creators must receive permission from the FCC to distribute such digital television devices. 31

Many believe that DRM is an illusory barrier to piracy. Even if DRM were able to preclude most people from distributing a given work, even one unencrypted copy can quickly propagate through a P2P system. No DRM is uncrackable, and, even without circumventing, files can be re-encoded into an unencrypted format once burned to CD or as they are outputted in analog form. 32

If DRM cannot prevent piracy, its usage restrictions may only diminish the value of acquired content and thus decrease consumer demand. Some in the content industries are seeking new business models that would not benefit from DRM. DRM may be of appeal to the Digital Media Store and P2P Store, but will be of no use to the other Collective Blanket Licensing and Ancillary Products and Services models, which allow and encourage unrestricted use of copyrighted materials.

Policymakers must weigh the potential of DRM to enable owners to protect their artistic works with its unintended consequences for consumer fair use and first sale rights.33 DRM can effectively deny consumers their rights under fair use and first sale, limiting various personal and transformative uses. In this way, DRM may undermine the potential of the digital environment’s enabling consumers to interact with and manipulate content.34 Policymakers must also assess the impact on innovation. Rights holders and DRM standards creators may use DRM to "lock out" disfavored digital media device and software creators by refusing them licenses.35 DRM mandates also limit technology creators’ flexibility to design tools with novel uses.36

THE BUSINESS MODELS

As described above, the four business models were developed to illustrate a spectrum of potential outcomes, not necessarily mutually exclusive but offering alternative responses to the challenges and opportunities of digital technologies and the online environment. The models, Digital Media Stores, P2P Stores, Collective Blanket Licensing, and Ancillary Products and Services, range from highly similar to vastly different from the traditional offline business model and from seeking a high degree of control over content and consumer behavior to none. Thus, as illustrated below, action across the policy nodes to enforce or diminish such control would have different impact on each.

This paper assesses each of these models across a range of dimensions, evaluating them comparatively here in the body of the paper and then in more individual depth in the appendix for each model. To preserve equal treatment, we describe and evaluate them conceptually, although we do cite existing attempts to implement them in our discussion of supporting evidence. First, we consider how the model may alter the roles of players within the industry as well as affect the diversity of content produced. Second, we consider the model’s potential for sustaining industry viability, describing reasons for and challenges to success. Third, we examine each model’s interaction with the three policy nodes, noting the model’s broader social costs and benefits in this regard. Finally, we suggest options across the three nodes that policymakers could employ to support the model, and then analyze the potential consequences of doing so.

DIGITAL MEDIA STORE 37

Overview
As a translation of the traditional brick-and-mortar retail model for physical media into the online market for digital media, the Digital Media Store (DMS) model would offer the least disruption to the traditional practices and structures of the media industries. In the model, retailers would offer licensed content for sale to consumers at websites or through software applications. By ensuring consumers have a reason to purchase content at online Stores, the industries would benefit from the suppression of piracy via technical and legal means.

Model structure
In this model, rights holders, whether content creators or intermediaries like record labels, would license their works to Stores. Consumers would go to centralized locations online to obtain works. Stores would offer consumers access to works under pay-per-download (PPD), all-you-can-eat subscription services, or hybrid fee structures (e.g., subscriptions allowing a fixed amount of downloads per month). To increase their market share and sales, Stores might promote particular works or creators through advertising or recommendation systems just as brick-and-mortar stores do today. DRM might limit various uses of works, including portability or use of works after a limited time period (thus enabling rental models).

Potential impact on industry actors
This model has the potential to bring content creators and retailers closer together than they currently are, changing the roles and revenues of existing intermediaries. Although many content creators will continue to rely upon intermediaries for their specialized expertise, this model would contribute to the declining cost structure by enabling inexpensive distribution. As such, this model might reduce the need for well-capitalized intermediaries. Moreover, it could support a broader spectrum of creators than does the traditional physical system. Because digital storage and presentation are relatively inexpensive, Stores can host a practically infinite number of creators.38

Requisites
To succeed, the DMS model must successfully "compete with free," or, in other words, convince consumers to pay for media content from the Stores rather than downloading it for free on P2P networks or other sources. To do so, the industries must increase the value of what they offer such that consumers find it worth paying for, or to increase the true cost of obtaining media content from other sources without paying for it. For example, the Stores may seek to be easier to use, more reliable and more convenient than P2P networks. The media industries may also continue to increase the true cost of obtaining unauthorized content using P2P and other vehicles through interference with P2P networks and infringing users and by employing DRM to inhibit infringing distribution.

Evaluation

The DMS model currently dominates online efforts to obtain revenues from films and music. Thus, the challenges and supporting evidence regarding it are the most specific and empirically well developed of any of the models. While at least a segment of online consumers appear willing to purchase legitimately through Stores, it is as yet unclear whether that segment will be sufficient to support the model and the industries. Further, as discussed in the Introduction, current evidence suggests that present efforts to fight piracy may be ineffective.

Challenges
Over the last year, numerous Music Stores have debuted with lower prices, larger catalogs, greater purchasing flexibility and lesser usage restrictions, allowing CD burning and portability. However, they continue to struggle to offer a product that can truly compete with what is available via P2P. For example, the iTunes Music Store (iTMS) has a one million-song catalog, the largest of any service, but that equals merely 20% of total released recordings and an even smaller portion of all recordings given the number of rare and live recordings one can find on P2P.39 Purchasing options are also restricted in a variety of ways. Some record labels force Stores to sell only complete albums despite data showing that consumers prefer singles.40

Similarly, some songs are licensed for PPD purchasing but not for streaming through subscription services and vice-versa.41

Stores are also burdening content with DRM to control consumer behavior. Among other uses, DRM can be used to tie media to specific hardware or software; for instance, Apple’s iTunes Music Store’s (iTMS) songs can only be moved to an Apple iPod. Also, rental subscription services currently do not offer any portability. Research suggests that such restrictions dampen consumer interest.42 DRM is also likely to be used to enable the same product to be priced differently according to the uses a consumer can then put it to; for example, Napster 2.0 may use it to enforce a higher price for a portable version of a song in a forthcoming subscription service.43 The potential benefits of such pricing or reducing piracy through DRM must be weighed against the extent DRM will drive consumers towards free, unencrypted content on P2P.

Finally, the Stores’ pricing may still be too high. Based on higher purchasing volumes during Stores’ temporary price cuts, some claim that the labels would be better off licensing at lower rates, allowing Stores to lower their prices.44 While all-you-can-eat subscription models may provide a better choice for highvolume users, their prices are above the average consumer’s spending on music.45

The Film Stores face similar challenges in competing with free and are currently in a particularly primordial state. Of the two main film per-copy download services, Movielink and CinemaNow, only CinemaNow offers permanent downloads, albeit from an extremely limited portion of its already small catalog. Both services’ prices are often above what one would pay at a brick-and-mortar Blockbuster. Recently, RealNetworks and Starz debuted a subscription service, but it too has a limited catalog. All services have DRM restrictions and are only beginning to offer limited portability.

Supporting evidence
Despite the Stores’ limited success to date, survey data and the Music Stores’ growing market share suggest reason for optimism. The launch of iTMS and other Music Stores has helped lead to a marked increase in the size of the legitimate market for music downloads. In 2003, it blossomed to $36 million and $47 million for US PPD and subscriptions revenues, respectively, and the market is expected to grow to $271 million in 2004, according to Jupiter Research.46 The significance of this growth should not be overstated, as the online market still represented merely 0.8% of gross music industry revenue in 2003; however, it does demonstrate that at least some consumers will pay rather than pirate given a compelling offering.

Further, in polls taken recently and during the original Napster’s existence, P2P-using consumers indicated that depth of catalog, flexibility of purchasing and use, and convenience are often just as important as price.47 Thus, if the Stores can improve these features they may be able to increase their market share further. In 2003, Jupiter Research found that half of surveyed consumers reported they would pay if offered prices and DRM flexibility similar to those now offered by PPD services.48 Moreover, competition in this emerging market is likely to produce improvements. Already, Music Stores are beginning to compete on price, exclusive content, and flexibility of content use.

The film market is much smaller, but that is to be expected in view of the facts that broadband connections are needed to download films quickly and that relatively few consumers as yet want to watch films on their computers. Technical improvements to Movielink and Cinemanow49 and the Starz subscription service’s debut may help bolster interest, just as the increasing variety of Music Stores did for music. Policy implications

While not inherently necessary to the model, the DMS would benefit from policies across the three nodes to help inhibit piracy. The potential benefits of successfully exerting such control over content must be weighed with the unavoidable social costs of seeking that control. Impact from actions across the three nodes

The industries would benefit from preventing piracy through DRM, interference with P2P, and lawsuits against direct infringers. While enabling a viable model for the industries would be socially beneficial, as described in the Introduction, these tactics might not meaningfully decrease piracy and have negative social consequences in regard to the fairness of and respect for law; inhibition of innovation; and restrictions on legitimate consumer uses.

Policy options to support this model
Policymakers in favor of this model would support industry proposals to increase their ability to control content and increase the sanctions facing those who evade it. Total eradication of piracy would likely be the industry goal, but policy might also be aimed at the more modest goal of control over content for a short period after its initial release.50

First, policy makers could increase the civil and criminal penalties facing consumer infringers and, to increase the likelihood of facing them, could increase Justice Department resources for criminal prosecutions and decrease the standard of proof required for a finding of guilt (proposed in the 2004 IP Protection Act51). Second, legislation could make it easier to sue P2P services (proposed in the 2004 INDUCE Act52) and legalize interference with uploaders via "interdiction"53 (proposed in the 2002 Berman Bill54). Finally, policy makers could (proposed in the 2002 Hollings Bill and recent requests to the FCC)55 pass DRM mandates that required all new analog and digital media devices to be DRM-enabled, perhaps even limiting the use of unencrypted content.56

The efficacy of adding new technological tools or legal reinforcement to the industries’ arsenal against piracy is far from given. Each attempt so far to impede piracy has resulted in technological counterattacks or workarounds and a continued arms race seems nearly inevitable.57 However, to the extent such policy actions would actually increase the control the music and film industries were able to exert over copyrighted content, certain social benefits would accrue. Complete content lockdown and substantial reduction of piracy would allow the industries to deploy the Stores largely without having to compete with free, whereas a protected initial window would maintain control during the portion of a work’s life when it is likely to garner a significant portion of its returns. In turn, the industries’ outlook for the online future would be significantly improved. Such control would also enable much greater price discrimination; without competition from free and greater difficulty in evading DRM, consumers could be offered, for instance, one price to listen to a song once, with additional fees for extra listens, moving it to a portable player, backing it up on CD, and so on. Such differential pricing can be socially beneficial by increasing total consumer access to artistic works and by increasing rights holders’ revenues. 58

The efforts to exert control would have significant social costs. The more severe penalties for file sharing as well as the heightened strictures on innovators and users of content would enhance the social costs across the three nodes. Tactics like interdiction may also interfere with legitimate activities.59 DRM that tracks and exacts payment for each use may raise privacy concerns and restrict lawful consumer uses.60 Such social costs increase as policies seek closer to total control of content and eradicating piracy. Furthermore, consumers’ evasion of the legal and technical strictures may eventually require more severe measures, generating increased collateral damages.

Further, by seeking to advantage this business model over others, these policies may reinforce the traditional industry structure and at the expense of certain content creators. For example, if the industries rely on high sales during the limited initial period of impeded piracy, this approach might intensify the industries’ current tendency to invest in a few immensely profitable hits while marginalizing the vast majority of content creators. Generating high initial sales will require large upfront costs for marketing and the technological impediments to piracy, and, in this way, this approach may make risk-spreading intermediaries like the record labels and film studios more valuable. In this way, the benefits of the dropping cost structure and greater independence of content creators might diminish.

THE P2P STORE 61

Overview
The P2P Store model would maintain the industries’ reliance on selling copies and, in that sense, is closely related to and compatible with the DMS. (For that reason, this discussion in this section is more abbreviated, drawing substantially on the preceding analysis.) However, rather than resist P2P networks and other consumer-to-consumer distribution channels like the DMS, this model would accept that they will persist and attempts to embrace them. To do so, the model would harness consumers as lawful second-order distributors.

Model structure
The P2P Stores’ primary role would be transactional, paying rights holders for works and accepting payment from consumers. Rights holders would individually negotiate licenses of their works to P2P Stores. Content hosting and distribution, while potentially P2P Store activities, would not be core to them.

At least two variations of the model can be imagined. In "superdistribution," consumers would host and distribute licensed content through any P2P network or other distribution systems. Consumers would then download content from each other; downloaded files might allow a certain number of free plays and then force the consumer to pay to retain access. To encourage consumers to distribute licensed content, they might receive a cut of the revenue or other rewards if those who receive the works "downstream" from them then purchase access. P2P Stores and rights holders could seed P2P networks to catalyze this scheme. In "closed networks," consumers would join a P2P network in which only licensed, authorized files can be shared with and downloaded from others. The P2P Stores would create and maintain this P2P environment. In both cases, consumers could be provided with PPD, subscription, or other payment options, and works could be encumbered by DRM.

Potential impact on industry actors
Widespread adoption of this model might affect the diversity of artists and the way money is allocated within the industry. This model empowers artists to provide their works directly to their consumers even more than does the DMS. By co-opting consumers to share distribution efforts and act as viral marketers, this model can reduce the financial investment and infrastructure required to achieve content sales, and in doing so reduces creators’ need for large multifaceted intermediaries. Consumers may be exposed to a broader spectrum of creators and artistic works, as lower costs might enable creators to achieve financial viability without selling as many copies of their works as is currently required.

Requisites
As was the case with respect to the DMS model, the P2P Store model can succeed only if the opportunity to purchase content legally is valued by consumers more highly than the opportunity to access content illegally without payment. Because the P2P Stores model also relies on widespread consumer-driven distribution, consumers must be willing to share files that are not accessible without payment.62 Further, successful interference with P2P and other decentralized distribution networks would be counterproductive, though reducing piracy through direct infringement lawsuits and DRM would be beneficial.

Evaluation

The viability of this model depends on its ability to compete with free. Although the reliance on selling copies is familiar to the industries, decentralized distribution is a new concept. Existing P2P Stores that do exist are currently less established than DMS. P2P Stores such as the closed network Wippit and superdistribution services Weedshare and Altnet have obtained licenses from numerous independent artists,63 and, while the music industry reportedly will license new closed network ventures,64 the extent to which they will embrace this model’s strategy is unclear.65 This model offers the industries the opportunity to work with new technologies and consumer preferences, but it also requires the industries to give up more control.

Challenges
Price, DRM restrictions (or lack thereof), size of catalog, consistency and convenience of products, and flexibility of purchasing options all factor into the P2P Store model’s potential for success. It faces similar challenges to the DMS model in these regards. The industries may also be hesitant to license the P2P Stores because they may further diminish popular attitudes that unauthorized uses of P2P are wrong or unlawful. Moreover, to embrace the P2P Store, the industries would have to forfeit whatever benefits the DMS derives from legal and technical interference with P2P systems. The industries certainly could abandon such interference and employ the two models jointly, but the industries may not feel ready to take this step.66

Supporting evidence
On the other hand, the P2P Stores may be a more attractive model for the industries and consumers. For the industries, this model offers reduced costs of distribution and content hosting, consumer participation in marketing, and the opportunity to work with their consumers and new technologies. For consumers, this shift to encouraging behavior they already engage in and perhaps prefer is a primary benefit. Consumers may enjoy the sense of community and opportunity to find content, recommendations, and individuals of similar tastes that file sharing enables. While consumers can obtain this benefit from unlawful sharing, payment or other rewards for lawful sharing may make P2P Stores more attractive. In a recent Gartner | G2 poll of online music service users, 25% indicated that a sharing capability increases a service’s appeal.67 Although some Digital Media Stores have incorporated limited sharing functionality,68 some consumers may prefer to stay in P2P environments rather than participating through a centralized commercial site.

Policy implications

The P2P Stores benefit from lawsuits against illicit file sharers and DRM to the extent they prevent piracy. However, the model relies on P2P systems and other distribution technologies being preserved. Policymakers may be asked to aid the P2P Stores in these regards.

Impact from actions across the three nodes
As noted in regards to the DMS, DRM and direct infringement lawsuits may be entirely inefficacious, and, despite the potential social benefit from enabling the model’s viability, may cause net social harms. Unlike the DMS, this model may be socially beneficial by allowing consumers to actively participate in distribution and continue peer sharing, albeit in a for-fee form. Furthermore, by embracing distribution technologies like P2P, this model would enable innovation.

Policy options to support this model
Policymakers favoring this model should reject efforts to burden P2P networks through the imposition of secondary liability and the encouragement or specific legalization of technical tactics such as spoofing. However, strengthening penalties for illicit file sharing, DRM mandates and anti-circumvention laws, and other technical impediments to infringing distributors could support this model if they successfully diminish piracy. As noted in the DMS section, these greater strictures would enhance potential social harm.

COLLECTIVE BLANKET LICENSING 69

Overview
This model avoids fighting illicit consumer behavior as a primary strategy in favor of enabling consumers legitimately to continue their currently illicit activities. Accepting that consumers prefer and will continue to download unlimited, unencrypted content from decentralized distribution services like P2P, this model provides a mechanism by which rights holders can be compensated. Rights holders would form a collective blanket licensing (CBL) organization, and, rather than attempting to collect per-use fees for each specific item, would offer consumers a flat-fee license to access and use all works, without restrictions over copying or further distribution. The CBL organization would count the uses and remit payment to rights holders accordingly. Although CBL has functioned well for music performing rights organizations (PROs) such as ASCAP and BMI, moving to this model for the industries’ primary revenue stream could have dramatic implications for industry structure and revenue distribution.70

Model structure
The model’s functioning would be roughly similar to the PROs’ blanket licensing and Professor William Fisher’s proposal for an "Entertainment Co-op."71 Rights holders would register their works with the CBL organization, permitting it to grant to consumers the right non-commercially to copy and distribute the works online. The organization would offer a license to these rights that covers the entire catalog of registered works in exchange for a flat fee, perhaps in the form of a monthly subscription. Individual consumers could purchase the licenses directly, or services such as P2P software distributors or ISPs could pay the fee for each of its users, enabling them to take advantage of the license and passing on the cost in some fashion. Licensed consumers might only be allowed to distribute files in systems that limit access to licensees;72 however, anyone would be allowed to offer such legitimate distribution systems, licensed consumers would be able to download from anyone, and no DRM or similar encumbrances would be employed. To determine payments to rights holders, the CBL organization would track usage of licensed content. Among other possible methods, it could measure what is being shared and downloaded on P2P networks; in addition, it could recruit users to allow software to track all usage, similar to the Nielson television surveys. Relative popularity based on this counting would determine remuneration for individual rights holders.

Potential impact on industry actors
By managing initial distribution and licensing, a CBL system could help engender financial returns for more content creators without need for approval or support from anyone other than fans. Without aid from other intermediaries, creators could produce their own recordings, register them, and receive compensation through CBL. Certainly, many creators might still need intermediaries to help produce and market their works, but the lower costs of the digital environment could mean that lower revenues could be achieved through the CBL while still achieving meaningful profits for creators. The CBL system and its lower cost structure could thus mean that a greater diversity of creators and works would be available to consumers. Further, consumers’ ability to obtain all content for a set price could lead them to sample, and thus fund, a broader variety of creators, supporting greater diversity.

Requisites
For this model to prove viable for the media industries, the cost of licensing must be worth the value of becoming lawful. Consumers must prefer to pay for the license rather than illegally obtain the same unrestricted content for free. Even though this model is in part born out of the futility of suppressing illicit use altogether, the industries might still hold out the threat of infringement lawsuits to encourage consumers to become licensed. Finally, this model requires that the industries (or an independent organization that obtains licenses from the industries) are able technologically and administratively to set and collect fees and accurately to estimate consumption patterns.

Evaluation

This model would represent a radical change for the industry, and, for that reason alone, might incur significant resistance. Aggregating enough licensed content and effectively implementing the CBL’s many functions are significant challenges. On the other hand, the model may be more attractive to consumers because they can continue to find and use music as many already do today, through P2P networks or other online means and unencumbered by DRM. While typical market mechanisms may be ideal, CBL might be the best practical option in the face of severe copyright enforcement difficulties.

Challenges
Rights holders, particularly industry incumbents like the labels and studios, may worry that the CBL will not maximize their revenue or that of the industry as a whole. By lumping all goods together at one preset price, the CBL may decrease value in the system by removing the market signals that allow for efficient pricing. The CBL would preclude rights holders’ ability to set their own prices according to what the market would bear, their individual cost of creation, and other factors.73 In this way, joining the CBL may forfeit an individual rights holder’s possible competitive advantages in favor of overall industry viability and cooperation. To the extent antitrust oversight would be necessary, as it is with the PROs, a CBL might further reduce copyright holders’ control.74 Incumbent labels and studios may also be hesitant to the extent enabling this model would reduce their importance.

In addition, the accuracy of the usage counting is a significant concern. Measuring downloads as well as Nielson-style sampling may provide sufficient estimates; however, users might attempt to "game" the system, using a creators’ works simply to send money to them, and over- and under-compensation may occur due to counting inaccuracies. Under-compensation would be of significant concern to less popular creators, for sampling would be most likely to leave them out.75

Consumer adoption will depend heavily on the price point. The flat fee must make a sufficient number of consumers feel that they are receiving more value than in a per-copy system and from illicitly downloading for free. The CBL organization must also have a broad and deep catalog to be compelling to consumers. Of course, rights holders may be hesitant to join the organization before evidence exists that consumers are willing to subscribe.76

Supporting evidence
Even though the CBL has its imperfections, it could provide a relatively preferable option for the industries if it can capture substantial revenue from the many consumers who would otherwise download illegally. ASCAP was formed to enable joint copyright enforcement in the face of widespread infringement; however, it evolved to offer efficient blanket licensing for a pool of copyrights, a development that finally made obtaining licenses viable and compelling to radio stations and other performance outlets.77 Even a low fee of $5 per month would generate substantial revenues if collected from the current 60 million American file sharers.78 Successfully charging such a low fee across a large subscriber base could also be more profitable because it is comparatively cheaper than charging a higher price with greater need for enforcement on nonsubscribing infringers.79

At an appropriately low price point, consumers may find compelling the opportunity to avoid the legal risk of file sharing. Moreover, all-you-can-eat access would empower consumers to try out more content, as each marginal consumption would be experienced at zero cost. Finally, many market mechanisms, like bundling fees into ISPs and other services, may help bring consumers into the fold.

Policy implications

This model is not unconcerned with control over content through DRM or interference with P2P, though it might still benefit from wielding the threat of infringement lawsuits. Rather than increasing restrictions across the nodes, policymakers eager to enable this model would limit copyright holders’ control over content, compelling the model’s adoption and licensing of file sharers.

Impact from actions across the three nodes
By embracing distribution technologies and rejecting DRM, this model would enable technological innovation and sustain consumers’ ability to interact with and manipulate digital media. It would also enable the benefits of engaging consumers as active participants in distribution. The industries’ efforts to increase the real costs of infringement through lawsuits will likely continue in some form, and, although these may incur similar social costs of enforcement, they will likely be lower because the purpose of using this model is to accept consumer behavior and vitiate the need for aggressive enforcement. As noted, it may be cheaper to charge lower fees with low enforcement.

Policy options to support this model
The most aggressive policy option in support of this model is to mandate its adoption through government action. Along with securing the social benefits noted above, the value of this approach is that it would ensure a deep and broad catalog of works at a cost that provides profits to the industries but does not impose excessive monopoly rents on consumers. In one approach, the government would require that the industries form a CBL organization for non-exclusive blanket licenses. The grant of copyright rights would be conditioned on registering with the organization. The government would then directly set licensing rates or provide oversight to ensure the organization provides fair rates, perhaps by using rate court proceedings

similar to those involved in ASCAP’s or BMI’s antitrust consent decrees.80 Alternatively, the government could create a CBL-like organization and, rather than collect license fees, use taxes or levies on related digital media goods and services to fund distribution of payment for content usage to licensors. In so doing, the government would remove all prohibitions on non-commercial copying and distribution. Unlike the industry-run approach, the tax-based approach would eliminate infringement lawsuits for the covered rights and would guarantee a particular level of compensation to rights holders in aggregate, forcing consumers to pay into the system.

This kind of active government involvement presents several potential problems. Continued reliance on property rules, contracts, and markets is generally preferable to expansive government involvement.81 A government-run system could suffer from price stagnation in the licensing fees, enormous negotiation costs to determine the fees, and distortions from industry rent seeking. The stakes are high for a change of the magnitude of the CBL, and, once instituted, a compulsory license may be difficult to abandon if it does not prove viable or optimal.82 The government-run approach presents particularly serious concerns. Levies on associated goods and services (such as CD burners or ISP subscriptions) or using income tax receipts to fund the CBL will distort the market and impose costs on consumers regardless of their consumption of copyrighted content.83 Finally, and perhaps most importantly, government control over the CBL would give the government a role in providing access to and compensation for intellectual content, a culturally uncomfortable if not dangerous structure.84

A less aggressive policy approach in support of this model is simply to refuse to skew the balance of consumer rights and industry control in favor of the industries’ piracy-fighting arsenal, thus adding to the industries’ incentive to support (or acquiesce in) the emergence of a CBL. For instance, the DMCA’s anticircumvention provisions could be repealed, and secondary liability for P2P could be eased. Should either or both industries choose to organize in this fashion, the government could play an enabling role at that time.85

ANCILLARY PRODUCTS AND SERVICES 86

Overview
Abandoning attempts to control copying and distribution as expensive and futile, this model would essentially transforms the economic role of digital recordings from revenue generators to marketing tools for alternative revenue streams. Content creators and rights holders would actually encourage the non-commercial digital redistribution of copies to promote ancillary products and services. Such revenue streams plausibly include but are not limited to: live performance and theatre exhibitions, merchandise sales, endorsements and other advertising roles, voluntary contributions, and fan club services. While some content creators might be able easily to adapt to this model, it would be a drastic change for most creators and current intermediaries. Industry-wide adoption faces significant obstacles. 87

Model structure
Along with creating and marketing their artistic works, content creators would have to provide and market revenue-generating ancillary products and services. Just as today, some creators would likely be comfortable doing these activities themselves, while others would outsource these activities to third-party intermediaries. To maximize the market size for their ancillary products and services, content creators and their intermediaries would allow and encourage their fans and consumers to redistribute works, thus provide marketing for the content creator. Websites that offered downloads would be only a starting point for distribution; among other ways, consumers might receive recordings attached to emails from friends, through P2P networks, or bundled with other digital or physical products.

Potential impact on industry actors
Under this model, the role of current labels and studios as intermediaries would change dramatically. Along with their already altered roles due to reduced production, distribution, and marketing costs, these intermediaries would no longer be able to derive any revenue from selling copies. Studios directly support various ancillary products’ and services’ production and promotion while capturing part of the revenue already, while labels have only begun to do so.88 Labels and studios would have to transform themselves to focus primarily on these functions in order to survive. On the other hand, new intermediaries might emerge to supplant these displaced players, or creators might perform these functions themselves.

This model may also differentially impact the viability of content creators. As discussed more below, not all creators are well suited to these alternative streams. Creators who are currently able to sell copies and generate revenues from ancillary products and services without broad marketing campaigns would have an easier transition than the heavily marketed superstar recording artists and blockbuster films. Indeed, these creators might benefit significantly if the public’s attention and thus dollars were less preoccupied by a small group of superstars.89

Requisites
To match the industries’ current net revenue levels, the total dollar value of the market for ancillary products and services would have to increase, either through increased prices or purchasing. The free access to copies must channel consumer purchasing into these other revenue streams. Especially in this model, offsetting the lack of revenue from selling copies with reduced costs of producing, distribution, and marketing may be critical to viability.

Evaluation
The necessity of significant growth in the existing market for ancillary products and services to compensate for the loss of revenues from copies and to cover the continuing cost of creating and marketing copies presents a significant challenge. Though it is not clear whether this model could support the industries at their current size, current trends and some anecdotal evidence suggest that this model does present a provocative opportunity for at least some creators and segments of the media industries, especially the music industry due to its smaller cost structure.

Challenges
That recording artists rarely see any profit from selling copies is oft cited in criticisms of the traditional business model. However, record labels rely on revenue from recording sales to produce, distribute, and market the artist’s music. For those artists who currently sell copies through record label contracts, even for those whose sales do not result in significant direct profits for the artist, the loss of the record label activities consequent to the loss of copy revenues could decrease the potential market for ancillary products and services.

The meaningful reduction in such costs that digital technologies can offer, as discussed in this paper’s Introduction, is thus an important factor for this model’s viability. If these costs were greatly reduced and the gross margins of the ancillary products and services were higher than those of copies of artistic works, the alternative products and services could offer profit levels even at reduced total revenue levels. In music, this factor creates the greatest problems for superstars; the labels heavily market these artists in order to drive sales and, likewise, generate substantial revenue from selling copies due to extensive marketing campaigns. If the total revenue pool and consequently marketing budgets were smaller, fewer artists and intermediaries would be able to invest in such marketing.

Similar problems exist for the blockbuster films. More generally, the magnitude of the cost structure for films may mean that even significant reductions would not reduce the total to a sum recoupable from ancillary products and services without copy revenues. The average production and marketing costs for a feature film in 2002 were $58.8 million and $30.62 million,90 respectively; in contrast, for a major label record album, costs in 2001 were around $125 thousand and $100 to 500 thousand, respectively.91 Moreover, unlike in music, digital technologies do not create significant reductions in production costs for films.92 Finally, some content creators may be ill suited to the alternative revenue streams. For instance, merchandizing by classical cellists and fan clubs for film directors seem awkward fits; also, many artists may wish to focus on their art rather than hawk the ancillary products and services. Meanwhile, creators who already have or can develop a substantial and avid fan base will be more able to profit from advertising participation, concert tours, and merchandise licensing. Although some creators already embrace and gain profits from the ancillary products and services, it is unclear how many of those who currently do not could begin to do so sufficiently to compensate for lost copy revenues.

Supporting evidence
Available evidence suggests that the alternative revenue streams’ ability to support the entire industry at present levels is not entirely likely. On the other hand, the evidence suggests that reliance on profits from ancillary products and services is quite conceivable for some creators and can still provide for the industries, albeit at lower profit levels. Markets exist for a number of ancillary products and services, which could be pursued further in light of the industries’ abandoning revenue from selling copies. For the above reasons, creators who currently sell few copies and do not rely on high production and marketing investment will have the highest chance of success due to their indifference to the loss of copy revenues and the low copy costs they will need to cover with alternative revenue streams.

The most lucrative alternative revenue stream for films and many musicians may be live performances. In music, The Grateful Dead encouraged free copying and distribution of their live performances, yet grossed a reported approximately $50 million per year from touring in 1990-1995.93 Indeed, for most creators, live performance revenue already represents their greatest direct revenue source.94 For films, even with perfect copies readily available, the theater experience may continue to be of desirably higher quality and to offer a communal and social experience difficult to replicate at one’s home. Films have sometimes been re-released in theaters to great success, despite their availability in cheaper formats such as VHS.95 The film industry may be able to retain, at least partially, a revenue stream that currently makes up about 25% of current revenues.96 Licensed and direct merchandise sales also already represent a significant source of revenues for certain feature films and musicians. Total gross sales of licensed merchandise in 2002 are estimated at $1.5 billion for the music industry and $14 billion for licensed merchandise from films, TV and animation.97 Although the subject matter and genres of some film and music works do not lend themselves to these revenue streams, niche markets have been developed for merchandise for even lesser known creators. The String Cheese Incident, for example, has had only modest album sales but generates $3.5 million annually from merchandise orders as well as extensive touring.98

Although total amounts for these industries are difficult to estimate, rights holders and creators have generated revenues from participating in advertising. Musicians and occasionally directors endorse products and services, appear in advertising campaigns, and have tours and concerts sponsored by advertisers.99

Product placement deals have provided significant revenues for feature films. The most recent James Bond film, for example, had product placement deals worth $70 million.100 BMW’s short films, all by well-known directors and all featuring BMWs as part of the action, are an example of the blending of artistic creation and commercial promotion.101

Creators are also generating non-sales revenues directly from fans. Membership in fan club sites can provide revenues for some creators. David Bowie, for example, charges a fee for access to his personal site, which offers information about him, chat rooms, his artwork for sale, and banking and ISP offerings from partner companies.102

Creators can also solicit voluntary contributions. Although music site Magnatune does charge for downloads, it resembles pure "tip jar" models in that all songs can be freely redistributed and can be listened to for free at Magnatune’s website. The average musician on Magnatune currently receives $1500 annually.103 Other anecdotal evidence also suggests that fans are willing to be patrons for creators, donating a "ransom" for the creation of new content. Marillion, a popular band now independent after 20 years with a major label, was able to produce and market two albums with funds provided in advance by fans.104

Policy implications

In this model, rights holders would give up control across the three nodes. Given that rights holders are free to license their rights to copying and distribution however they choose, policymakers would not need to take action to enable this model. However, they could take actions to prompt its acceptance.

Impact from actions across the three nodes
Because this model does not rely on the imposition of control across the three nodes, it could be very beneficial to the public. If this model could support the industries, it would sidestep the emerging conflict between copyright holders’ business incentives and the digital era. The model would enable technological innovation; sustain consumers’ ability to make personal and transformative uses of content; engage consumers as active distributors and allowing peer sharing; and avoid the potentially culturally damaging effects of the lawsuits.

Policy options to support this model
To encourage use of this model, policymakers could limit tools that aid rights holders in selling copies, like the DMCA and secondary liability rules that could affect P2P. In the extreme, policymakers could remove the prohibitions on the non-commercial reproduction and distribution of audio and video recordings. Such a policy would ensure the above societal benefits; however, given the alternative revenue streams’ uncertain ability to support the industries, locking all content creators into this model could have potentially devastating consequences.

CONCLUSION

By examining four potential business models and how policy actions would impact them, this paper has sought to provide lawmakers with an improved understanding of the context and potential implications of the policy choices facing them across three key nodes. Decisions to act at each of these nodes will have mportant consequences for the music and film industries, for citizens as consumers and as members of a shared culture, and for the future of technology innovation.

Our analysis does not identify which model will succeed or is preferable because we believe available evidence is insufficient to justify confident judgments on these questions. Digital technologies and broadband access are still relatively new, and the opportunities they offer both the consumers and the producers of digital recordings have yet to be fully explored. Those online business models that exist are still in their infancy, and their long-term viability and impacts remain unclear.

Some view this uncertainty as a crisis needing immediate resolution. The music and film industries, particularly the major labels and studios, have claimed that copyright is failing to fulfill its mission of rewarding creators in the online environment and that piracy will eventually cripple their traditional business models. Thus, they clamor for government action across the three nodes to protect the industries - or, more realistically, industry incumbents.

We believe that seeing only peril in threats to the traditional business models and incumbents may be premature. As demonstrated by the models described in this paper, the industries’ online future may be markedly different from their offline past, but that is hardly sufficient reason to seek to forestall it. Rather, the uncertainty presented by the future is reason for government caution, not preemptive action. Aggressive legislative action to privilege any business model above others at this point would be problematic.

Insulating the traditional business model would require re-imposing greater control on content and consumer behavior despite potentially liberating digital technologies, which, as discussed, would incur significant social costs. New business models would potentially be suppressed despite their potential to leverage technology to create even greater economic surplus and other net social benefits in terms of incentives for creativity, innovation, artistic diversity and fair use. At the same time, compelling adoption of another model because it requires less control (or for other reasons) would also distort the forces acting to optimize models. The industries are not collapsing today, and legislation in anticipation of a potential future collapse is likely to be at best poorly tailored given the nascent state of the online market. Observing the development of the models and their consequences would allow policymakers to act with superior information and better-targeted legislation, if legislation is necessary at all.

In the past, Congress has acknowledged these risks and chosen to forbear from immediate action to amend copyright and other doctrines when incumbent models have been threatened by new technologies.105 For example, copyright law did not initially grant composers any rights in these reproductions when piano rolls were invented. When Congress eventually did act, it created a compulsory license that has been arguably critical in allowing the recording industry to blossom. When the Betamax and VCR were developed, the film and television industries called on Congress for aid; yet, Congress chose not to act, and today, VHS and DVD sales and rentals are the largest industry revenue streams.

We do not object to aggressive government intervention on principle, nor do we argue that it will never be necessary to alter copyright or other aspects of the legal framework to ensure an appropriate balance among rights holders and the public in the digital era. However, we do argue that it is currently premature. As current and new players experiment with business models, whether those described here or others not yet imagined, policymakers will have an opportunity to assess their viability and impact across all interests before they act.

REFERENCES:

1 Throughout this paper we generally use the term "content creators" to refer to recording artists and the host of players involved in producing a sound recording or film. These may be the same or separate from "rights holders."

2 This paper’s analysis takes place in the context of a broader study, the Berkman Center’s Digital Media Project, of the transition from analog to digital media and its impact on copyright and associated legal doctrines, the content industries, artists, consumers, and shared culture. See http://cyber.law.harvard.edu/media.

3 Please see Appendix I for further discussion of the impact of digital technologies on industry cost structures and roles.

4 See Stan Liebowitz, Will MP3s Annihilate the Record Industry? The Evidence So Far, (June 2003), at http://wwwpub.utdallas.edu/~liebowit/knowledge_goods/records.pdf (arguing file sharing has caused decline); Koleman Strumpf and Felix Olberhozer, The Effect of File-sharing on Record Sales; An Empirical Analysis, (March 2004) at
http://www.unc.edu/~cigar/papers/FileSharing_March2004.pdf (reaching conclusion file sharing has had no statistically significant impact); Stan Liebowitz, Pitfalls in Measuring the Impact of File-sharing, available at http://www.utdallas.edu/~liebowit/intprop/pitfalls.pdf (criticizing Strumpf and Olberhozer); Rafael Rob and Joel Waldfogel, Piracy on the High C’s: Music Downloading, Sales Displacement, and Social Welfare in a Sample of College Students, National Bureau of Economic Research Working Paper 10874, October 2004, available at http://papers.nber.org/papers/W10874; Kai-Leung Hui and Ivan Png, Piracy and the Legitimate Demand for Recorded Music, 2 Contributions to Economic Analysis, article 11 (2003); Justin Hughes, "Spin and Rinse in the Rhetoric of Peer to Peer" (unpublished paper 2004).

5 "Deluge of DVD in a watershed year: spending on the disc rose 40% in 2003," Video Business 24(2): 1 (Jan. 12, 2004) (discussing recent film industry success).

6 See Liebowitz, Pitfalls in Measuring the Impact of File-sharing, supra note 4 (arguing that this should be a rebuttable presumption given economic theory).

7 For a detailed analysis of copyright in the digital era, see Copyright and Digital Media in a Post-Napster World, Gartner¦G2 and the Berkman Center for Internet & Society at Harvard Law School, available at http://cyber.law.harvard.edu/home/uploads/254/2003-05.pdf, update available January, 2005.

8 See William Fisher, Promises to Keep, (Stanford, CA: Stanford University Press, 2004), 54-58, 77-78 (hereinafter "Fisher, Promises").

9 See Fisher, Promises, at 78-79.

10 For music, see Fisher, Promises, at 22 (noting that "$1000 worth of ? hardware and software" may be sufficient); Raymond Shih Ray Ku, The Creative Destruction of Copyright: Napster and the New Economics of Digital Technology, 69 U. Chi. L. Rev. 263, 305-306 (2002). Harold Vogel notes that major releases have typically cost $125 thousand, or upwards of $300 thousand if substantialstudio time is needed. Harold Vogel, Entertainment Industry Economics, (New York, NY, USA: Cambridge University Press, 2001, 5th ed.), 162. See also M. William Krasilovsky & Sidney Shemel, This Business of Music (8th ed. 2000), 23, cited in Mark S. Nadel How Current Copyright Law Discourages Creative Output: The Overlooked Impact of Marketing, Berkeley Technology Law Journal, Vol. 19, 785-856, 798 (Spring 2004) (hereinafter "Nadel, Impact of Marketing"), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=489762 (placing production costs at between $80-150 thousand). For films, see the Motion Picture Association of America (MPAA) Worldwide Market Research, U.S. Entertainment Industry: 2002 MPA Market Statistics 19, 20 (2003) (claims that in 2002 the average cost of producing was over $58.8 million with another $30.62 million for marketing and advertisement); Peter Henderson, "Technology Lets Garage Studios Change Hollywood," Reuters via USA Today (Feb. 24, 2004), at http://www.usatoday.com/tech/news/techinnovations/2004-02-24-making-movies_x.htm (noting technological advances making it easier to create films, but noting that small budget films still cost in the tens of thousands).

11 See Fisher, Promises, at 18-24. For music, major labels marketing standard releases often expend $100,000 and over $500,000 on premier acts, though independent labels’ and artists’ expenses are generally much less. See Vogel, id., at 163; RIAA, Costs of a CD, http://www.riaa.com/news/marketingdata/cost.asp ("marketing and promotion costs [are] perhaps the most expensive part of the music business today."); Lynne Margolis, "Independent’s Day," Christian Science Monitor (Apr. 21, 2003) available at http://www.csmonitor.com/2003/0411/p13s02-almp.html (discussing lower cost structure of independents). For films, see MPAA Worldwide Market Research, id.

12 See Fisher, Promises, at 20.

13 See James Maguire, "Hitting P2P Where it Hurts," Wired News (Jan 13, 2003), at http://www.wired.com/news/digiwood/0,1412,57112,00.htm.

14 See Mike Linksvayer, "Morpheus with Bitzi Anti-spoofing," (Oct 7, 2004) at http://gondwanaland.com/mlog/2004/10/07/morpheuswith-bitzi-anti-spoofing/.

15 A&M Records v. Napster, 284 F.3d 1091 (9th Cir. 2002); Fisher, Promises, at 116-120.

16 MGM v. Grokster, 2004 U.S. App. LEXIS 17471 (9th Cir., 2004) (finding that the Grokster and Morpheus services, which have no control over their users, were not liable for the infringements of their users).

17 See Peter Biddle et al., The Darknet and the Future of Content Distribution, (October 15, 2002), available at
http://crypto.stanford.edu/DRM2002/darknet5.doc (predicting that highly efficient distribution networks "will remain a fact of life").

18 See, e.g., John Borland, "Music rebels seek to tame P2P," News.com (Nov. 16, 2004) available at http://news.com.com/Music+rebels+seek+to+tame+P2P/2100-1025_3-5453788.html.

19 See Mark Lemley and R. Anthony Reese, Stopping Digital Copyright Infringement Without Stopping Innovation, forthcoming
publication, available at http://intel.si.umich.edu/tprc/papers/2003/210/Stopping_Copyright_Infringement_Without_Stopping_Innovation.htm; "Testimony of Gary Shapiro on behalf of the Consumer Electronics Association" at the IICA hearing, available at http://ce.org/s2560test.pdf
(discussing threat to innovation in light of bill proposing tougher liability standards); Briefs for Defendants in appeal to MGM v.Grokster, available at http://www.eff.org/IP/P2P/MGM_v_Grokster.

20 Ipsos News Center, American Consumers Continue to Embrace Digital Music, (Dec. 4, 2002) available at http://www.ipsosna.
com/news/pressrelease.cfm?id=1685.

21 Estimates vary depending on system and estimator. See A&M Records v. Napster, 114 F.Supp.2d 896, 911 (N.D.Cal. 2000) ("as
much as eighty-seven percent of the files available on Napster may be copyrighted and more than seventy percent may be owned or administered by plaintiffs."); Amicus Brief of ACLU in support of Defendants at 13-14, appeal to MGM v. Grokster, 259 F.Supp.2d 1029, available at http://www.eff.org/IP/P2P/MGM_v_Grokster/20030926_aclu_amicus.pdf (placing the percentage of infringement on Grokster and Morpheus systems at around 75%, in contrast to plaintiff’s 90% estimation).

22 John Borland, "RIAA files new round of P2P lawsuits," News.com (Nov. 18, 2004) available at http://news.com.com/RIAA+files+new+round+of+P2P+lawsuits/2100-1027_3-5458594.html (music industry has sued 6,952); John
Borland, "MPAA touts lawsuits, new P2P-fighting software," News.com (Nov. 16, 2004) available at http://news.com.com/MPAA+touts+lawsuits%2C+new+P2P-fighting+software/2100-1025_3-5454939.html.

23 Infringement must be willful and "have a total retail value of more than $1000." 17 USC 506(a). See also Fisher, Promises, at 147 (discussing this issue in greater detail).

24 Of course, not all uploading may count as infringing distribution or reproduction. See Berkman Amicus Brief in Capitol Records et. al. v. Noor Alaujan, available at http://cyber.law.harvard.edu/media/capitol_amicus. Similarly, some downloading may be legal, though some substantive defenses were rejected in A&M Records v. Napster, 284 F.3d 1004, 1013-1020 (9th Circuit 2001).

25 See Pew Internet & American Life Project, Sharp decline in music file swappers: Data memo from PIP and comScore Media Metrix, (Jan. 4, 2004) at http://www.pewinternet.org/reports/toc.asp?Report=109; Cachelogic, The True Picture of P2P File-sharing (Fall 2004), at http://www.cachelogic.com/research/slide1.php; Thomas Karagiannis et. al., Is P2P dying or just hiding?, Presented at Globecom (November 2004) available at http://www.caida.org/outreach/papers/2004/p2p-dying/; Hughes, supra note 4.

26 See John Borland, "Covering Tracks: New privacy hope for P2P," News.com (Feb. 24, 2004) at http://news.com.com/2100-1027-
5164413.html.

27 See Lawrence Lessig, Free Culture, (Penguin Press: Mar. 25, 2004), 202, available at http//blogspace.com/freeculture/Harms; Fisher, Promises, at 243 (discussing "culturally unhealthy" impact of millions violating the law).

28 See Lemley and Reese, supra note 19.

29 17 USC 1201-1205. Exceptions in 1201(d)-(j). Regarding the lack of a fair use exception, see Universal v. Reimerdes, 111 F. Supp. 2d 294, 321-324 (S.D.N.Y 2000).

30 See, e.g., Universal v. Reimerdes, 111 F. Supp. 2d 294 (precluding interoperability of DVD players without a license);
RealNetworks v. Streambox, 2000 U.S. Dist. LEXIS 1889 (W.D. Wash. 2002) (prohibiting creating compatible program to access
streaming media files).

31 47 CFR 73.9002(b) et. seq. (Nov. 4, 2003) and see http://hraunfoss.fcc.gov/edocs_public/attachmatch/FCC-03-273A1.pdf; EFF
Digital Television Liberation Front, available at http://eff.org/broadcastflag/; DefCon12 slides, (July 31, 2004), available at http://www.eff.org/broadcastflag/defcon12/ (describing these requirements).

32 See Biddle et al., supra note 17 (arguing that DRM does not affect the viability of P2P and that businesses might be better off avoiding DRM entirely); Cory Doctorow, Microsoft Research DRM Talk, (June 17, 2004), available at
http://www.craphound.com/msftdrm.txt.

33 See Dan Burk and Julie Cohen, Fair Use Infrastructure for Rights Management Systems, 151 Harvard Journal of Law and
Technology 42, 43-54 (Fall 2001); R. Anthony Reese, The First Sale Doctrine in the Age of Digital Networks, 44 Boston College Law Review 577 (2003).

34 See Fisher, Promises, at 28-31 (discussing enhanced abilities to interact with content).

35 See Berkman Center Digital Media Project Team, iTunes: How Copyright, Contract, and Technology Shape the Business of Digital Media, (April 10, 2004), 44-48 (discussing lock-out in the current digital music market).

36 See Paul Horn, Eliot Maxwell, and Susan Crawford, Promoting Innovation and Economic Growth: The Special Problem of Digital Intellectual Property, report from Committee on Economic Development, 31-43 (March 3, 2003), available at
http://www.ced.org/docs/report/report_dcc.pdf.

37 For further discussion of the Digital Media Stores model, see Appendix II.

38 Music Stores are already finding that almost all of their content is streamed or purchased at least once. "Post-Conference Report on Midem 2004," Digital Media Wire, at http://www.digitalmusicforum.com/midemreport2004.htm; Chris Anderson, "The Long Tail," Wired (Oct 2004) at http://www.wired.com/wired/archive/12.10/tail.html.

39 See Leander Kahney, "Music Biz Buzzing Over iTunes," Wired News (May 2, 2003) at http://www.wired.com/news/print/0,1294,58706,00.html.

40 See Brian Garrity, "All Aboard The Digital Train?" Billboard (Sept. 20, 2003); "Online music sales could be sad song for big record companies," LA Times via Milwaukee Journal Sentinel (April 28, 2004) available at http://www.jsonline.com/bym/tech/news/apr04/225616.asp.

41 See Katie Dean, "New Napster Off to Solid Start," Wired (Nov 3, 2003) at http://www.wired.com/news/digiwood/0,1412,61023,00.html.

42 See Paul Gluckman, "Building Business on Legal Downloads Isn’t Easy, Panelists Say" Washington Internet Daily (Feb. 10, 2004) (citing complaints about lack of interoperability); Berkman Center and GartnerG2, Copyright and Digital Media in a Post-Napster World, (Aug. 2003) at 17, available at http://cyber.law.harvard.edu/home/uploads/254/2003-05.pdf (discussing portability as expected use). See also iTunes: How Copyright, Contract, and Technology Shape the Business of Digital, supra note 35, at 53 (citing GartnerG2 statistics regarding declining demand if consumers cannot resell content).

43 See "Napster Previews First Portable Digital Music Subscription Service," (Sept. 1, 2004) available at
http://www.napster.com/press_releases/pr_040901.html (pricing at $14.95, above current subscription rate of $9.99).

44 See Steven Levy, "Forecast: Song Costs May Fall like Rain," Newsweek (Sept 27, 2004) available at http://www.msnbc.msn.com/id/6037780/site/newsweek/; Amy Harmon, "What Price Music?" New York Times (Oct. 12, 2003) at
http://www.nytimes.com/2003/10/12/arts/music/12HARM.html?pagewanted=2&ei=5007&en=ccd1925d9ba7f1d2&ex=1381377600&
partner=USERLAND.

45 See Census Bureau, Statistical Abstract of the U.S., 2003: The National Data Book, (January 2003) (projecting consumer
spending at $58.81 and $57.27 for 2003 and 2004, respectively); "US per capita consumer spending on each of nine forms of
entertainment / media forecast for 2002 in dollars," Plunkett’s Entertainment & Media Industry Almanac (Jan. 7 2002) (placing consumer spending at $71/person in 2002). Most subscription services currently charge $9.99 per month or about $120 per year. See http://www.mp3.com/tech/services_index.php for pricing information.

46 Forrester Research, "For downloads, things are looking up", News.com (Feb. 4, 2004) at http://news.com.com/2030-1069-
5153314.html; May Wong, "The downfall of downloads?" AP via Kansas City Star (Oct 3, 2004) available at
http://www.kansascity.com/mld/kansascitystar/business/personal_finance/9803225.htm?1c.

47 See Phil Leigh, Online Music Starts Rocking, (December, 2003) at http://insidedigitalmedia.com/research/rocking.pdf (citing research by now defunct Webnoize that breadth of choice and convenience were more often cited than price); Horn, et. al., supra note 36, at 49 (citing research by now defunct Webnoize that breadth of choice and convenience were more often cited than price); Harris Interactive, "Reasons that Teens Downloaded Without Paying, 2003" (October 2003), reprinted by eMarketer in radio interview with Phil Leigh, at http://www.insidedigitalmedia.com (finding 59% wanted to get "one or two songs", 48% wanted music quickly, 46% found music too expensive, 44% wanted music for free, 40% wanted songs unavailable elsewhere).

48 See "The Apple Music Store," Jupiter Research (May 30, 2003) at http://weblogs.jupiterresearch.com/analysts/gartenberg/archives/Jupiter%20Apple.pdf.

49 See Paul Korzenioski, "Could It Finally Be Showtime For Online Movie Downloading Sites?" Investor’s Business Daily (Dec. 9, 2003).

50 The Digital Media Project is pursuing an on-going study of two scenarios related to these extreme and moderate policy solutions, the Technology Lockdown and Technology as Speed Bump Scenarios: http://cyber.law.harvard.edu/media.

51 See http://www.corante.com/copyfight/archives/COE04B52_LC.pdf.

52 See http://www.lessig.org/blog/archives/COE04694_LC.pdf.

53 Interdiction entails downloading a song from a P2P user such that no one else can download it at the same time. In this way, rights holders can effectively block transmission of the content. See Randy Saaf, Written Testimony for the Oversight Hearing on "Piracy of Intellectual Property on Peer-to-Peer Networks" submitted to the House Judiciary Committee (Sept 26 2002), at http://www.house.gov/judiciary/saaf092602.htm.

54 See H.R. 5211, available at http://thomas.loc.gov/cgi-bin/bdquery/z?d107:h.r.05211:.

55 S. 2048, "Computer Broadband and Digital Television Promotion Act", http://www.eff.org/IP/SSSCA_CBDTPA/20020321_s2048_cbdtpa_bill.pdf; MPAA, Content Protection Status Report (April 25, 2002)
at http://judiciary.senate.gov/special/content_protection.pdf.

56 The aim of such restrictions would be to render useless unencrypted content illicitly acquired via P2P or other distribution systems. Such restrictions were feared when the CBDTPA was proposed; see Electronic Frontier Foundation, "Alert: Congress Calls for Public Participation on Digital Media Technology Mandates," (Apr. 9, 2002), at http://www.eff.org/IP/SSSCA_CBDTPA/20020322_eff_cbdtpa_alert.html.

57 See Testimony of Edward Felten on "Piracy of Intellectual Property on Peer-to-Peer Networks," submitted to House Judiciary Committee (Sept, 29 2002) (cautioning against technological arms race).

58 See Fisher, Promises, at 167.

59 See Electronic Frontier Foundation, "Comments on the Berman P2P Bill," (Aug. 2, 2002), at http://www.eff.org/IP/P2P/20020802_eff_berman_p2p_bill.php.

60 See Julie Cohen, A Right to Read Anonymously: A Closer Look at Copyright Management in Cyberspace, 28 Conn. L. Rev. 981
(1996).

61 For further discussion of the P2P Stores model, see Appendix III.

62 While DRM might be useful in enforcing payment in some instances, this model in no way requires DRM as a barrier to infringing redistribution. That is, once a consumer purchases the song, he could automatically receive an unencrypted copy (see, e.g., the superdistribution service, http://www.faircopy.com/). The consumer could then, quite easily, distribute that unencrypted copy, and thus DRM is posing no barrier to infringing distribution, nor does it pose any barrier to legitimate uses of lawfully purchased copies.

63 See Amicus Brief of Altnet in support of Defendants, appeal to MGM v. Grokster, 259 F.Supp.2d 1029, available at
http://eff.org/IP/P2P/MGM_v_Grokster/20030929_altnet_amicus.pdf (discussing Altnet’s licensors); Weed, CD Baby members use
SML’s Weed to sell Music, (June 24, 2004), at http://weedshare.com/web/releases/06-24-04_WEED_RELEASE.html,
http://weedshare.com/web/releases/12-11-03_WEED_RELEASE.html (on Weed); Tony Smith, "Wippit adds 10,000 BMG tracks to
catalog," The Register, (Mar. 15, 2004), at http://www.theregister.co.uk/content/6/36254.html (on Wippit).

64 See Smith, id.; John Borland, "Music rebels seek to tame P2P," News.com (Nov 16, 2004) at
http://news.com.com/Music+rebels+seek+to+tame+P2P/2100-1025_3-5453788.html?tag=nefd.pop.

65 See Andy Sullivan "Labels blacklist song-swap companies" Reuters (July 15, 2004) available at
http://uk.news.yahoo.com/040715/80/ey4fj.html.

66 We also acknowledge that, with all channels that allow illegitimate distribution shuttered, closed networks and superdistribution could still exist in authorized channels. However, one cannot embrace distribution schemes like superdistribution today while still continuing to spoof those songs or sue the distribu