The average American listens to more than three hours of music each day. Next to television viewing (which itself is becoming increasingly digitized and subject to copyright disputes), it is the largest leisure activity of Americans. Almost everyone buys CDs, and tens of millions of listeners use file-sharing programs.

In the Cato report Policing Pirates in the Networked Age, I argued that peer-to-peer (P2P) file sharing, unlike prior copying technologies like the photocopier and VCR, could cause actual harm to copyright owners-in particular, the recording industry. Even so, shutting down Napster might have been a poor tactic since file sharing via Napster’s successors, which do not rely on central file servers, would likely be more difficult to litigate against. Moreover, evidence put forward in the Napster case documenting harm caused to the record industry was weak (essentially nonexistent) at the time.

Since then events have unfolded at a rapid speed. My preliminary examination of record sale data appeared to contradict the theoretical analysis in my earlier Cato article, but record sales have since dropped significantly-with MP3 downloads a likely culprit. That preliminary finding made me, for a time, a hero to the legions who favor unfettered MP3 downloads. Although it was wonderful to bask in the positive media attention typically granted to anti-corporate (and anti-capitalist) types, the cheers turned to boos after a more thorough examination caused me to revert back to my original position that harm was likely. As sales swoon, the record industry has become increasingly desperate to stop MP3 downloads. Since MP3 downloading is projected to cause the industry to lose 25% or more of its revenues, it is not surprising that the industry feels besieged. This is a painful loss for any industry to absorb.

After its victory in the Napster case, the industry hit a roadblock in its attempt to shut down Napster progeny Grokster, Streamcast, and Kazaa, when a federal judge ruled that those noncentralized P2P file- sharing systems were little different from VCRs and thus not liable for the infringing behavior of their customers. This legal obstacle has derailed the strategy of going after the conduits for file sharing, so now another strategy has come to the fore.

The record industry was recently given permission by the courts to force ISPs to reveal the names of users thought to be engaged in substantial MP3 uploading. In an attempt to leverage this ruling before it might be overturned, the industry is in the process of bringing thousands of lawsuits against individuals making copyrighted files available on P2P networks. In mid-July, reports indicated that the Recording Industry Association of America had generated 871 subpoenas against ISPs and universities, asking for the names of individuals thought to be providing MP3s on file-sharing systems.

This is a strategy fraught with dangers for the recording industry. For the strategy to work, tens of millions of MP3 downloaders must be convinced to stop. To convince such a large base of users to change their behavior will require massive publicity. Bringing thousands of individuals to court will generate front-page stories. By attracting so much attention, the industry runs the risk of alienating its customers and inviting Washington to step in. Furthermore, the privacy implications of on-demand subpoenas are enormous. For example, Peter Swire of Ohio State University regards such subpoenas as a violation of due process: “On the RIAA view, your sensitive personal information…would be available to anyone who can fill out a one-page form.” Congressional hearings, anyone?

To date, Congress has largely been supportive of the copyright industry, to the point of proposing allowing record companies leeway to “hijack” the computers of downloaders. Nevertheless, if thousands of ordinary individuals are brought into court, and if the media portrays individuals sympathetically enough, congressional support could quickly evaporate. There is already some evidence of congressional concern. If public opinion turns against the record industry, Congress will likely abandon its current principles for more popular ones.

Lurking just below the surface of this dilemma is the possibility that the government will turn the recording industry into another post office-like franchise. A claim that is being increasingly repeated (particularly in the academy), and one that we are certain to hear more of if these lawsuits proceed, is that the government should step in to fix the problem by instituting a compulsory license, which would legalize downloading but require a legally set fee on products such as blank CDs which would be used to compensate artists. Proponents of these systems envision an era of enhanced artistic flowering (especially if the greedy record companies are taken down a peg or two alongside). Seemingly well-intentioned proponents of compulsory licenses have fallen into the same trap that has caught so many others-they expect government control to lead to an earthly musical paradise.

Government intervention must be a last resort. Under a compulsory license system, private contracts are preempted. Some government entity will decide how much tax one pays for a blank CD (even if the CD is not used to record music) and whether Shania Twain should get paid more than Eminem (in other words, how the pot of money is to be split). Market pricing and allocation are extremely hard to replicate by fiat. These decisions will inevitably become political (if nothing more than in the choice of commissioners). Imperfect markets are superior to an even more imperfect Ministry of Music.

It is far too early to write off market-based solutions to the problem of online distribution. Digital rights management, a technology still in its infancy, holds out hope of protecting works from unwanted copying. Some politicians who are friendly to the copyright industry have sought to mandate a form of this technology to be included in playback devices. Governments should not choose and mandate copy-protection technology, however, but instead should allow market experimentation.

The direction we take with the issue of music-sharing will reverberate to other markets. The U.S. has been lucky to not have a major government ownership role in media, providing another important reason for us as a society to not to get digital distribution wrong: if government must regulate mere entertainment, surely hard news and information are next. (PBS and NPR are nothing compared to the influence of a BBC). The perception that the current concentration of private media ownership is high has drawn howls of protest, but as the citizens of dictatorships know, we should not allow governmental control of the media to enter through the back door. Government control of music and movie distribution and such likely offshoots as awarding artistic grants to sympathetic musicians and filmmakers should be avoided at all costs.