File-sharing Networks Return with Legitimate Ways to Share Music -- and Make MoneyPublished: August 06, 2008 in Knowledge@Wharton
After the U.S. Supreme Court declared in 2005 that Internet file-sharing sites Grokster and StreamCast had illegally aided their customers' efforts to share pirated copies of copyrighted music and video files, many commentators predicted the demise of businesses that depended on online file-sharing.
But the technology that Napster, the pioneer of music file-sharing, Grokster and StreamCast unleashed has returned, supported by a business plan that respects copyright laws. Three years after the high court's ruling, several start-ups say they have found ways to make peer-to-peer (often called P2P) file-sharing legal and perhaps profitable.
"Although the early use of P2P networks was for digital piracy, P2P networks are increasingly being used for legitimate content distribution, including music, video and software," writes Kartik Hosanagar, a Wharton professor of operations and information management, in a recent paper on P2P business models. "For example, Grooveshark, rVibe, We7 and iMesh are firms that use P2P networks to distribute music to users. The music is licensed from the music labels, the files are distributed from users' machines and the P2P firms provide software and billing. A number of technologies have also emerged to prevent piracy in P2P networks. As a result, distribution of digital products through P2P networks is likely to become more prevalent."
Still, Hosanagar says that these new commercial outfits need to tweak their business plans. He predicts that they will make more money if they are savvier about pricing content and about paying their customers to share it. In "Dynamic Referrals in Peer-to-Peer Media Distribution," Hosanagar and two co-authors at the University of Washington, Yong Tan and Peng Han, create a mathematical model of the ebb and flow of supply and demand on a peer-to-peer network. Their model suggests that file-sharing firms should often pay high fees to users who provide content to other users -- sometimes higher than the retail price of the file itself -- and that they should vary these referral fees and their prices according to the demand for particular files. (The three scholars have also recently completed a related paper on the topic titled, "Diffusion Models for Peer-to-Peer (P2P) Content Distribution: On the Impact of Decentralized, Constrained Supply.")
How P2P Networks Operate
To appreciate their arguments, it helps to understand how peer-to-peer file-sharing works.
Each computer in a peer-to-peer network acts as both a store and a customer. Users of these networks can provide content by making the files on their computers available to other users of the network. In return, they can use the network to reach into other customers' computers and copy the files that have been made available. Participants in a peer-to-peer network can share any kind of content, though music has proved most popular. The P2P networks provide software that enables their customers to organize their content, search the network and swap files with each other. These firms often charge a fee for each file acquired though the network, and use that revenue to pay royalties to content creators and referral fees to users who share their files.
An obstacle to the growth of these networks, the paper's authors write, is that many customers are eager to copy files but reticent to make their own files available to others, a practice called free riding. The free riders may have files that other users would like to have, but can't find. Vibe and Grooveshark try to induce free riders to share files by paying referral fees for making the files available to other customers. But Hosanagar says that they may not be paying enough. "rVibe pays 5 cents for a 99-cent track. Grooveshark has recently changed its policy, but I think it was originally 10 cents a song. Our conclusion is that you want to offer really high payments early on and that payments shouldn't be fixed. We found in many cases that the [referral] payment could be higher than the price. For example, you might initially pay $1.50 for a 99-cent file."
Why would a company pay more for a product than the eventual sale price? Is that not a recipe for bankruptcy? Not at all, Hosanagar says. It's a logical response to the law of supply and demand. "The initial high payment brings in a lot of distributors, and it gets them to share files when there's scarcity." As a result, he says, the firm ensures that its customers can get the files they want and don't shop elsewhere or resort to illegal file-sharing. By encouraging early distribution, the arrangement also feeds the buzz that any media company seeks: as more and more people hear a song or see a video, they may recommend it to friends who then may buy it, too. Those users, in turn, distribute the file, scarcity abates, and the firm can gradually reduce the referral fee.
Let Demand Determine Price
Retail prices for content on a peer-to-peer network should work similarly, Hosanagar says. That is, they should be flexible and reflect demand. Firms would then have two pricing options. If they want to create buzz, they may initially set a low price for files to encourage people to buy them. Or they may decide to price based on scarcity, charging more early on to capture maximum revenue from the zealots who'll snap up anything new from a favorite artist. Later, once that hardcore demand has been sated, they could charge less. "The optimal strategy seems to be to price the product low at first and pay the majority of the amount [collected] to the P2P distributor," Hosanagar says. "So initially your profits are lowest. Over time, you increase your price and reduce your referral payments." On the Internet, firms can easily implement flexible prices. "In the past, it was difficult to do customized pricing," notes Hosanagar. "The music industry has generally been one that was slow to adapt."
What's more, organizing a company as an enabler of peer-to-peer sharing potentially provides a wider menu of offerings for customers. "Apple is going to find it impractical to negotiate deals with a lot of independent, unknown artists, while with a Grooveshark or rVibe, the independent artist signs up for the service, creates an account and uploads his songs," says Hosanagar. That allows the artist's fans, no matter how small a group they are, to download songs and pay whatever price the artist and Grooveshark agree to (with Grooveshark taking a small percentage of each sale).
Likewise, that obscure musician will have a tough time arranging a distribution deal with iTunes. "Apple isn't going to want to negotiate contracts with a bunch of unknowns," Hosanagar points out.
The most ardent advocates of peer-to-peer networking argue that it will eventually displace Apple because consumers will tire of Apple's inflexible pricing, and content providers will rebel against its stranglehold on online distribution.
"Personally, I don't see the P2P model displacing iTunes," Hosanagar says. "But I do think it's here to stay. I recollect a statement that I read in early 2000 where somebody said P2P was a solution in search of a problem. That's moot now," with many people around the world sharing files on legal peer-to-peer networks.
Still, the question remains: Can sharing become a booming business like iTunes? Certainly many consumers want to swap files online, but it's not as clear that they're willing to pay to do so. The brief history of file-sharing is littered with failed ventures.
In the Beginning
For practical purposes, Napster, the brainchild of a Northeastern University freshman, created the niche. The service allowed anyone to swap -- or in the opinion of the major music labels, steal -- music files online. If you wanted a song by the White Stripes and someone on the network shared it, you could download and play it on your PC. Napster kept, on a central server, an index of all the music available on its network, which made searching easy. Music labels sued, arguing that Napster was abetting the pirating of copyrighted music files. A federal court agreed and Napster was eventually liquidated.
When Grokster and StreamCast entered the market, they provided file-sharing software but didn't create central indexes. Their software allowed users to search each other's computers, seeking files that they wanted. When the major labels sued them, they argued that they couldn't control what people did with their software and, without central indexes, didn't even know. The U.S. Supreme Court didn't buy their arguments and ruled that they, too, had abetted theft.
But file-sharing lived on. Solid estimates of the extent of illegal file sharing are scarce, but some range as high as one billion songs a year. rVibe and Grooveshark are trying to persuade at least some of these folks to purchase downloads by combining appeals to guilt and greed. Grooveshark's web site stresses that musicians go hungry if people don't pay for their music, and of course it pays those referral fees.
Rock musician Peter Gabriel has endorsed another approach. He's an investor in a peer-to-peer file-sharing firm called We7. Its customers can download free songs with short ads at the beginning. "The revenue generated from these advertisements goes to artists, labels and other rights owners," We7 explains on its web site. "You get music for free, and the artist gets fairly paid." The ad disappears after four weeks. Or customers can elect to purchase a file outright and skip the ad. Users can also share files.
Which of these models will triumph? Hosanagar isn't sure.
"I can think of three or four outcomes we might see. There might be free content that's used to stimulate demand for the other things, like concerts and T-shirts. There might be free ad-supported content. Or there might be a model where you buy the songs, but it will not be the rigid pricing model that we see today. Or lastly, it might be a model where payment is on a per play basis rather than a per-purchase basis."