The US Supreme Court agreed last week to hear the music industry's case to shut down file-sharing websites such as KaZaA, Grokster and eDonkey. These operations enable users to download millions of songs from each other for free instead of buying at least some of them in a music store. In aggregate, the traffic generated by music and video file-sharing can account for more than half of all internet traffic. Music companies, their revenues in decline, have been trying to suppress these "peer-to-peer" (P2P) practices in the courts, legislatures and by spreading deliberately defective copies of songs. They view P2P users as thieves who must be prosecuted. But traditional media companies should perhaps see P2P as to their long-term advantage because it helps create new markets and forms of distribution.

P2P is part of a large family of "grassroots" activities in the media. In the early years of broadcasting, radio amateurs congregated on the airwaves in the absence of commercial broadcasters. In the 1970s, personal computers were built by enthusiasts who successfully created the challenge for International Business Machines where giants such as RCA and the government-subsidised Bull had failed. Today, the open source movement has created Linux as an alternative computer operating system. The internet is perhaps the best example. It cannot be said that such voluntarist arrangements are more efficient than a market-based system. In theory at least, most of the arrangements listed above could be better created by companies with professional management, financing and marketing channels. Yet the frequency with which these grassroots movements emerge suggests some solid economic reasons behind them.

What all these activities have in common is that they are network operations. The more participants, the lower the activity's average cost and the higher its benefits to every participant. When the size of such networks is small, per-unit costs are high but benefits are low due to the small number of participants. Hence in many cases, the cost exceeds benefits and buyers will not show up, meaning insufficient "critical mass" for self-sustaining growth.

If an activity is ultimately desirable, one way ahead is for government to step in with subsidies, as in the early days of the internet. A second way would be for a company to underwrite the early deficit and then profit from subsequent growth. The problem is that competitors could access such a user base and share the benefits even though the early provider bore the costs of the original investment.

The third alternative is the community approach. For each member, belonging to a leading-edge group while beating the establishment becomes its own reward. Financially, the community activity lowers costs by contributing free labour to the common endeavour, such as skilled programmers' hours, and by sharing pirated content. Together, these efforts lower the number of participants required for critical mass. From there, the activity will often grow to a size sufficient for profitable commercial entry. Examples include commercial radio in the early 1920s; commercial internet providers in the 1990s; and, most recently, Apple with iTunes, its wildly successful music download service.

When such commercial entry takes place, private companies almost inevitably push aside the community that made it all possible. We can decry such evolution as a business takeover. Or we can celebrate it as part of the innovation process, in which community entrepreneurship plays an important but under-appreciated role.

Established media companies therefore should value the community efforts that create the user base for their own entry. While upholding the copyright principle, they should accept some early messiness in new applications in order to grow future markets. Twenty years ago, some of the same companies that are today challenging P2P also fought before the same Supreme Court against the video cassette recorder, citing the same piracy potential. They narrowly lost, but the VCR enabled widespread home video use that has proven immensely profitable to these companies.

It is not only about music. Today, with broadband internet emerging around the world, there are enormous secondary benefits to the economy and to innovation from rapid deployment of high-speed networks.

Entertainment uses are the "blockbusters" for broadband that will make it attractive to millions, thereby creating beneficial "network effects" that will enable other applications and future innovations. Suppressing P2P activities that prime the pump for subsequent commercial activity will only harm users, media companies and the digital economy as a whole.

The writer, professor of economics and finance at Columbia University and director of its Columbia Institute for Tele-Information, is a contributor to FT.com's New Economy Policy Forum, www.ft.com/techforum