A Free Market Model for P2P Content Distribution with Simulation

A free-market model is proposed which follows the current file-sharing trend to its logical conclusion. This model both preserves profits for the content publishing industry and, paradoxically, serves to reinforce the standard business model. This is accomplished by dispensing entirely with the notion of intellectual property rights and allowing the price of content to be determined naturally by market dynamics. All users of content have equal rights to distribute the content, and to profit from distribution. It is shown that this situation creates healthy profit for the content creator, with asymptotically diminishing returns for distributors on progressively lower rungs of the distribution network. At the bottom of the distribution network are the end users, who necessarily pay a fair price for content. This process is illustrated by way of a computer simulation.

Definitions

Content will refer to any form of data that can be transmitted on a network. We chose to use this term instead of domain-specific language (mp3, media, file, data, etc.) in order to reflect the fact that the proposed model applies equally to many diverse forms of content. Content might mean music, film, or other video, but it could just as well be artwork, software, designs, chemical formulae, books, press articles, financial analysis, scientific research results, and other kinds of data.

Similarly, intellectual property rights refers to legal rights involving various forms of intellectual property, including copyright and patent.

A distributor is an agent who purchases and distributes content with the aim of turning a profit.

The distribution network is a decentralized aggregate of connected distributors. For the purposes of this paper, the distribution network may be assumed to be a peer-to-peer (P2P) file-sharing system.

The consumer is an end-user whose interest lies primarily in viewing and using the content. The consumer places some subjective value on the content, which will vary from consumer to consumer.

Motivation

The elimination of intellectual property rights, it is often argued, would inevitably destroy any incentive for creativity. If content could be reproduced and distributed free of charge and without restriction, the argument goes, there would no longer be any incentive to create content, and the marketplace would descend into anarchy. This vision, however, is based on the faulty assumption that, barring the artificial intervention of the law, content would be passed from user to user at no charge. We contend that the law is in fact directly responsible for the illicit distribution of free content.

Laws forbidding unauthorized content redistribution have proven very difficult to enforce. The law additionally forbids the redistribution of content for profit without the proper licensing agreement. Since accepting payment links the seller directly to the transaction, this latter stricture is much more easily enforceable, and penalties for violation tend to be much more severe. This unfortunate situation leads to an artificial distortion of the marketplace whereby content is widely distributed at zero cost. This constraint effectively sets the market price for content to zero dollars, since zero is, by law, the lowest asking price.

Many industries have attempted to solve the problem of illicit free distribution of content by employing a variety of tactics, including various forms of digital rights management (DRM) as well as prosecution of consumers who share content. We contend that industry, government, and consumers would all be better served if the market price for content were allowed to assume its natural value. The model described in the following section attempt to accomplish this goal.

Model

The model that will be described here is the answer to the question: What if the artificial concept of intellectual property rights was eliminated, and all content owners were allowed to receive payment for its distribution? If content is no longer protected from redistribution, then the distribution rights are sold along with the content. The key to understanding this solution is that the buyer has no interest in distributing content for free, as he hopes to recover his investment. Allowing consumers to act as distributors and to be paid for their participation in the distribution network establishes a market rate for the content, thereby restoring a natural equilibrium to the market. A distribution network naturally emerges in which the producer of the content sells to a first-tier of distributors, who then sell to lower-tier distributors, and so on, until the bottom tier is reached. This bottom tier is composed of end consumers, who necessarily pay a fair price for the content since nobody has any incentive to give it away free of charge. Furthermore, when the demand becomes saturated (after some natural period of time), traders have no one left to whom they can sell the content, and it passes naturally into the public domain.

What if a philanthropist buys the content before it passes into the public domain and then chooses to distribute it freely, thereby undermining the value of the content for other interested parties? Although this is a remote possibility, it is unlikely that such a person would intervene at a level where the cost of the content is still very high. Furthermore, the impact of a single user on the network would be minimal. In order to have a significant impact on the marketplace, the user would need to purchase large amounts of upstream bandwidth, which is itself an expense. The question is parallel to asking what happens when a single stockholder chooses to unload all his stock for nothing. This simply does not happen in the real marketplace.

This model rests on the same liberal principles that provide the conceptual foundation of market economics in free economies.

Implementation

This system could easily be incorporated into popular P2P file-sharing protocols. All that is needed is the addition of an "ask" and a "bid" price for every file being made available or requested, respectively, by the user. Creators and publishers simply make their content available with an asking price, presumably one that is high enough to justify the expense of creating the content. Users who accept to pay the lowest asking price available on the network at any given moment initiate data transfer. Each file part is charged as an equivalent percentage of the agreed price for the whole file. File sharing software need only be linked to a micropayment system in order to process the multitude of resulting transactions. A full elaboration of the technical details of this implementation is beyond the scope of this paper.

Alternatively, users may sign up for flat-rate services, which in turn use these fees to purchase content on the open market. This business model is familiar to a number of industries, but with two important differences: users are free to record and redistribute whatever content they buy, and the service providers may operate independently from the control of the publisher of the content.

Speculation

Like any open market, this model allows for the possibility of speculation. A distributor may buy "long shot" content at a low price, believing that it has hidden value. If this value is later realized, the investment will pay off when demand increases. The model therefore rewards good predictive insight. The owner of low-value content also has every incentive to create higher demand through his own marketing and sales efforts.

As an example, consider the case of a distributor who spots a rising musical star. Because the creator is not yet a well known artist, he may be forced to distribute his work at a low price due to the market’s underestimation of the consumer demand. Recognizing the value in the content produced by this artist, the distributor buys a copy of the artist’s music. When the demand rises unexpectedly, our insightful distributor is able to sell many copies, both to other distributors and to consumers. Since the market has now recognized the demand, the price will rise temporarily, until the market begins to become saturated. Being one of a small number of content-owners, our distributor is well placed to earn a healthy reward for his insight. Also, note that the distributor’s interests are aligned with the artist’s, as his job also involves creating and increasing the demand for the content he has purchased.

Just as in the equities market, a good professional distributor makes it his job to seek out undervalued content and to buy it before it becomes popular. Distributors own large portfolios of content, with varying degrees of risk and potential for reward involved in each. Some content will do well, and others will lose money. On balance, a distributor will be able to profit to the degree of his predictive prowess and marketing skill.

Marketing

If the content creator shows or transmits his open content to a user, that user may then distribute it. If content is revealed for the purposes of advertisement, then no sale has taken place, and the content provider has lost the value of his content to the marketplace. How is it possible to advertise content that cannot be shown?

Various industries have already provided a number of innovative answers to this question. The music industry distributes a certain number tracks on a new album via radio and television without disclosing the entire contents of the album. Presumably, this is done to increase album sales. Online retailers commonly distribute parts of songs or other media free of charge in order to promote the sale of the product. Similarly, music artists often choose to disseminate some of their work freely in order to gain publicity, which they hope to translate into album sales. The software industry releases demo or trial versions of software with functionality limited in time or scope in the hope that users will find the software useful enough to buy it. The film industry launches movie trailers, effectively informing consumers about the films it hopes to sell without disclosing the entire film. Furthermore, consumers see films because of a variety of factors, none of which require them to see the films in their entirety before buying them. The cast, the director, and simple word-of-mouth are among these selling points.

Also, as in the flat-rate service example sited above in Implementation, content may be "pushed" to consumers without any need for specific marketing at all. In this case, the consumer trusts this agent to find quality content. If consumers feel that the service provider is not providing content of sufficient quality, they are free to stop paying the service fee and shop for a different service provider.

Examples from other domains abound. In fact, most traditional forms of marketing do not need to reveal the content in all its detail. Only enough information is given to pique the consumer's interest. One outstanding counterexample to this is the music industry, where only after music has been distributed free of charge to the consumer on radio and television networks are consumers then asked to pay to own the music. In contrast to other forms of content, it seems that music becomes valuable because it is popular, in a reversal of the natural situation in which content is popular because of its intrinsic value. However, as mentioned above, free music distribution is often, and should be, used to drive sales of paid content.

Finally, it should be noted that the distributors in the network have a vested interest in selling as many copies of the content as possible, and as quickly as possible, since the market price tends to decrease with time and increase with rising demand. It stands to reason that the distributors in this network benefit from a highly effective marketing campaign. The model therefore encourages the creation of highly efficient distribution channels, thereby increasing total sales.

Example

Our example will involve a content creator, a distribution network, and a number of end consumers. The content creator incurs expenses of $1000 in creating an original work. The creator then makes the work available on a P2P distribution network with an asking price of $800. Note that this price is less than the $1000 cost of the investment. The creator is banking on the possibility of selling more than one copy of the work to distributors and consumers.

Market research determines that consumers are willing to pay between $1 and $2 for the work. This is what we call the subjective "intrinsic value" of the work, as assessed by each consumer. Research further indicates that there may be about 2000 consumers interested in owning a copy. The total amount of profit available is therefore estimated to be about $3000. Understanding this, one distributor accepts to buy the work for the creator's asking price of $800. There are now two copies available on the network. The number of existing copies of the work available on the network is always public information. It is the responsibility of each distributor to estimate the potential for profit from his copy based on the number of copies on the network, the potential number of end-users who have not acquired a copy (but want one), the various amounts of money that consumers are willing to pay to own (and distribute) the work, and other data gathered from market research.

We now have a situation where two interested parties must recoup their investments. The content creator still needs to recover at least $200 (the $1000 initial cost minus the $800 gained from the first sale), and the distributor needs to recover his $800 investment. However, the market value of the content will begin falling rapidly as more copies are distributed. When the third copy is sold, the total potential for profit must be divided among three distributors instead of two.

This process continues, with each distributor (and the creator) continuing to sell copies at the ever-decreasing market rate. At any given time during this process, would-be consumers who do not yet own a copy have two options: buy the content at the going rate, or wait for the price to fall. Those consumers who are most eager to view the content will be willing to pay a higher price to see it sooner rather than later. (This situation is analogous to the model currently used by the film industry.) Furthermore, the consumer who buys the content sooner than other consumers, say for $5, becomes a part of the distribution network, and he may hope to sell a number of copies for $1 to future consumers. As with the distributors, the reward lies in understanding the future of marketplace: buying an obscure work which later becomes popular can prove very lucrative.

Eventually, every consumer who wanted to view the content owns a copy. This means that the market is saturated, and there are no further sales to be made. At this point, the value of the content naturally drops to zero. If new consumers appear, the demand has increased, and so the market value will rise again. For more discussion of market fluctuations, see the section entitled Speculation.

Simulation

In order to illustrate and validate the model, a simple simulation was written. A number of simplifying assumptions were made:

- The concept of bandwidth usage was not incorporated. Content is instantly copied from any peer to any other peer without any limitation.

- The number of consumers is fixed at 1000. This avoids the need for distributors to predict the demand for content.

- All distributors have complete knowledge of the market at all times, including the number of consumers who own the content, the number who do not, and each consumer's assessment of the intrinsic value of the content.

- The asking price is not incorporated into this simulation. We assume that the asking price is equal to the bid when a transaction occurs.

- Each distributor's assessment of the content's intrinsic value is assigned to a flatly distributed random value between $1 and $2, which remains constant (per distributor) for the duration of the simulation.

The simulation runs all distributors together asynchronously. In the beginning, only one distributor owns the content. This is the content's creator. Each content owner continuously looks for the highest bid among the distributors who do not yet own the content. The content is transferred to the highest bidder, and payment is transferred to the content owner. The bidder is now a new content owner, and will attempt to sell the content to other bidders. This process continues until all 1000 distributors/consumers own the content.

Much relies on the logic that each distributor implements to determine an appropriate bid price. This logic constitutes each distributor's intelligence. For the purposes of the simulation, each distributor determines the bid price as follows:

numOwners := 0

expectedReturn := 0

// try to estimate the expected return on investment

for all distributors:

if distributor owns the content

numOwners := numOwners + 1

else if distributor is not self

expectedReturn := expectedReturn + (distributor's assessment of

the content’s value);

// divide the total potential profit by the number of owners

// (plus one, because we would need to count self as an owner)

expectedReturn := expectedReturn / (numOwners + 1)

// the bid is the "fair price", i.e., the expect return plus the content's intrinsic value

bid := expectedReturn + (assessment of the content’s value)

Needless to say, this logic is naive, and any number of variations on this theme is imaginable. In one such variation, a certain fraction of distributors are considered to be only end consumers. These consumers have a "fixed bid" price because they are not interested in speculation. In other words, because they do not expect to earn money from sale of the content, they will only pay an amount equal to their assessment of the content's value.

Different methods of computing the distributors’ bid prices give different results. The results of one of these modified algorithms are shown below. In this example, 30% of the 1000 distributors are designated strictly as “consumers” who are unwilling to pay more than their own assessment of the content’s intrinsic value. In other words, these consumers do not expect to sell the content at all. For the remaining distributors, the square root of expected return is used to compute the bid price.

The results of the simulation show transaction prices falling sharply with time, as illustrated in Figure 1 below.

Figure 1

The maximum total amount of profit retained by any distributor at the end of the simulation was $95.67 for the original owner of the content (i.e., the content creator). A number of distributors lost a small amount of money, with the maximal loss being $2.42. Overall, 78.8% of the 1000 distributors were “winners” of the game. These distributors ended the simulation with a profit greater than their assessment of the content’s intrinsic value. For example, an distributor who assessed a value of $1.50 to the content and ended the simulation with a loss of only $1.00, having acquired a copy of the content, is considered a winner.

In Figure 2 below, the return on investment is calculated as follows:

ROI = (profit at end of simulation + content value) / (price paid for content)

Figure 2

Source code for the simulation is available on request.

Peaceful Coexistence

The free-market content distribution model has the seemingly paradoxical effect of reinforcing the standard business model. Because it exploits the natural tendency for people to prefer earning money to earning no money, illicitly distributed free content will become progressively less available on the network as more upstream bandwidth is consumed by paid offerings. Distribution of unlicensed copies is therefore discouraged in favor of distribution of open content. Content tagged as “open” would be tradable on the network, whereas the unauthorized selling of traditional content would incur the same penalties as usual. The standard business model forbidding unlicensed duplication of content not only continues to operate, but is protected by the saturation of free distribution channels with open content.

Comparison

There is no shortage of ideas for improving standard business practices in industries which exploit intellectual property for profit. A complete review of these ideas is beyond the scope of this paper, but links to a few of them can be found in the References section below. Many of these propositions employ highly domain-specific solutions that are not easily generalized. Others rely on the generosity and good will of consumers to pay for what they can receive for free. Some rely on a partial solution in which some content is provided free of charge while other content is paying. Merchandising and various services have been put forward as substitute revenue streams for content providers. Various taxation models have also been proposed. Most importantly, all of these models include some notion of intellectual property rights - rights which must be enforced across international boundaries at great expense to both government and industry.

Ethics

Industry leaders would have consumers believe that the sharing of content is equivalent to theft. This argument rests on the notion that, even once purchased by the consumer, the publisher still retains ownership of the content. Redistribution can therefore be considered “theft” because the content is not the consumer’s property to distribute. The consumer, despite having purchased the content, is not its rightful owner.

Because people believe in the ethical principle that forbids theft, the industry has leveraged this argument to equate file sharing with shoplifting. However, people have an intuitive understanding that copying a music file is somehow different from shoplifting, since there is no material loss to the creator of the media. Moreover, there can be no theft where both parties have agreed to exchange goods. Violation of a licensing agreement is not theft in any traditional sense of the word. It is the agreement itself that is in question, not ownership of the content.

One can imagine an alternative ethical principle in which owners of property are free to conduct transactions as they please with the property they own. This implies that, once acquired through legitimate means, property, intellectual or otherwise, belongs to its new owner unreservedly. People also have an intuitive understanding of this principle and are often outraged at attempts to restrict this freedom. Continuing the analogy to physical property, most people feel uncomfortable with restrictions imposed by an industry on their freedom to sell what belongs to them. Intellectual property should be treated no differently.

Conclusion

The model described here applies equally well to music as it would with any other media, or indeed any type of content. In this model, there is no need for expensive and damaging litigation, domain-specific distribution networks, or difficult and ethically questionable law enforcement. The content producer may choose to save the costs of marketing and distribution, allowing the other distributors in the network to perform these functions instead. Furthermore, the risks of speculation on unproven content can be offloaded, when desired, to the marketplace.

Additionally, this model encourages innovation to the same extent that the current model discourages it. Because content can be freely used by anyone who owns it, there is no longer any issue regarding improvements made to existing content. For example, an enterprising user may find that he has sufficient incentive to improve on a published technical design, thereby increasing the salability and value of his copy.

We believe that this model provides an elegantly simple, forward-looking, highly efficient, and profitable alternative to the current system. As Internet technologies continue to evolve and available bandwidth continues to grow, controlling the free flow of data will prove to be an increasingly intractable problem. The model encourages content producers to give up their belligerent attitudes towards networked distribution, and instead to view the network as a friend and partner.

References

A Business Model Involving Free File Sharing

http://www.techdirt.com/fotr/20030912/1032238_F.shtml

Describes a model for the recording industry based on merchandising and concert sales.

RIAA v. The People: Two Years Later

http://www.eff.org/IP/P2P/RIAAatTWO_FINAL.pdf

A brief history of the recording industries attempts to stem file sharing by using law suits, followed by a proposal for a "collective licensing regime".

Optimal Peer Selection in a Free-Market Peer-Resource Economy

http://www.eecs.harvard.edu/p2pecon/confman/papers/s6p3.pdf

This white paper describes a method for selecting a set of peers who can provide a file at lowest cost.

A New Way: Business Model for File Sharing

http://blogcritics.org/archives/2004/02/03/085028.php

Discusses various taxation-based models for funding media creation.

Presentation at the Federal Trade Commission's Public Workshop -- Peer-to-Peer File-Sharing Technology: Consumer Protection and Competition Issues

http://www.ftc.gov/bcp/workshops/filesharing/presentations/lincoff.pdf

The FTC agrees that the recording industry is taking the wrong approach to digital rights management. This paper suggests the creation of a "single, unified digital transmission right," at least for music.

Solving the Payment Problem for Open Source and P2P File Sharing

http://business.newsforge.com/article.pl?sid=03/08/26/143247&tid=33&tid=132&tid=3&tid=31

Makes the astute observation that, "Neither [the software industry not the entertainment industry] has an intellectual property problem. Instead, both have a payment problem.” "Utility computing" is proposed as a solution for the software industry, and a similar model is proposed to ensure payment for entertainment.

P2P Manifesto

http://rothko.hmdnsgroup.com/~montemag/p2pmanifesto_en.pdf

Advanced Peer-Based Technology Business Models: A New Economic Framework for the Digital Distribution of Music, Film, and Other Intellectual Property Works

http://shumans.com/p2p-business-models.pdf

Provides a thorough analysis of the state of the recording industry in 2002 and proposes an “open clearinghouse network” distribution system.

The Electronic Frontier Foundation White Papers

http://www.eff.org/wp/