¶ 1 Leave a comment on paragraph 1 0 One of the primary reasons that the traditional music industry views P2P as such a threat is because it disrupts conventional power relations. During the 20th Century, the major labels built their empires on the basis of a distribution cartel; only the “big six” (as things stood in 1999, before recent waves of consolidation) had the economic might and the political heft to saturate retailers and airwaves alike with their music. Independent artists and labels were either left out in the cold, or charged an exorbitant rate to participate in the marketplace. And music fans were essentially left with a choice between musical Coke and Pepsi; unless they were fortunate enough to live near an indie music retailer or a free-form radio broadcaster, their options were limited to whatever the major labels were promoting at that moment in time.
¶ 2 Leave a comment on paragraph 2 0 P2P changed this dynamic profoundly, by leveling the playing field and lowering the barriers for music distribution. While this didn’t erase the strategic benefits accruing to large industrial organizations (marketing still costs a fortune), it did undermine one of the core mechanisms by which they accrued and retained market power. Yet when the music industry critiques P2P and decries the “piracy” that takes place on file sharing networks, it rarely does so by complaining that its distribution cartel has been compromised. Instead, it makes a more direct economic argument, arguing that consumers use file sharing networks as an alternative to paying for music, and therefore that every download on a P2P network can be viewed as a “lost sale.”
¶ 3 Leave a comment on paragraph 3 0 Clearly, this notion is absurd if taken literally; so much music is downloaded freely from the internet that if each downloaded song were sold at market value then the total amount of money spent would outstrip the music industry’s revenues, even in the best of years, by orders of magnitude. District Judge James P. Jones, adjudicating a case brought against Daniel Dove, a member of a BitTorrent tracker site called EliteTorrents, has pointed this out, as well. In his decision, he faults the music industry’s logic, observing that the “RIAA’s request problematically assumes that every illegal download resulted in a lost sale,” and pointing out that “[i]t is a basic principle of economics that as price increases, demand decreases. Customers who download music and movies for free would not necessarily spend money to acquire the same product.” Yet logic would dictate that if not every download represents a lost sale, at least some of them must. And doesn’t this subset of downloads directly hurt the music industry’s bottom line?
¶ 4 Leave a comment on paragraph 4 0 Researchers have been grappling with this question – attempting to assess and quantify the impact of P2P on music sales revenues – at least since Napster’s debut. I was among the first to publish findings on this subject, as an analyst for Jupiter Research in 2000. At the time, my clients included the RIAA and all of the major labels, so my purpose was to help the industry assess whether a genuine threat existed, and to develop market strategies that would mitigate or accommodate these new technologies. We employed a very robust methodology, fielding a survey of over 2,000 US “online music fans.” At one point in the survey, we asked whether respondents’ purchasing habits had increased, decreased, or remained consistent since they first started visiting music sites. At another point in the survey, we asked whether they had ever used Napster. Given the range and order of questions on the survey, there was no way respondents could know that we were looking for a statistical relationship between these two factors. Our results were surprising, even to us:
¶ 5 Leave a comment on paragraph 5 0 No segment of respondents was more likely as a whole to have increased its music purchasing than the segment of Napster users was. . . . Napster users were 45 percent more likely to have increased their music purchasing habits than online music fans who don’t use the software were. This trend holds true regardless of factors such as age, income, online tenure (the number of years that an individual has been using the Web), and overall music purchasing level.
¶ 6 Leave a comment on paragraph 6 0 In other words, we found that, among online adults who liked music, Napster was actually helping music sales. Though some were no doubt using the service as a replacement for traditional music retail, others were using it as a vehicle to discover and sample new music, increasing their enthusiasm about music products and driving them to purchase more. In 2002, I published follow-up research, based on a newer Jupiter survey, showing that file sharing continued to have a mixed effect on music purchasing habits, with a net positive effect overall. This time, we found that file sharers were 75 percent more likely than the average online music fan to have increased their music purchasing habits since they started visiting online music sites.
¶ 7 Leave a comment on paragraph 7 0 In the decade since then, dozens of researchers from around the world have published scores of studies on this subject in both academic and commercial publications, and the results have run the gamut from positive to neutral to negative for the industry. While a thorough review of all the relevant literature is beyond the scope of this chapter, there are several recently published meta-analyses that attempt to summarize and/or integrate this literature. Business professors Felix Oberholzer-Gee and Koleman Strumpf have been among the most active in this area, publishing a number of highly-cited articles on P2P and media economics since 2004. As they argue in their most recent work on the subject, “[b]ecause the theoretical results are inconclusive, the effect of file sharing on industry profitability is largely an empirical question.” Yet, reviewing the empirical literature, they find that “the results are decidedly mixed.” While “the majority of studies find that file sharing reduces sales,” there are several that “document a positive effect,” and “an important group of papers reports that file sharing does not hurt sales at all.” In short, there is no research consensus on the subject, either theoretically or empirically. Similarly, technology journalist Drew Wilson has recently published an extensive series on P2P news website ZeroPaid.com analyzing 20 published research reports related to P2P’s economic effects. He has found that a great deal of the research undermines the RIAA’s claims, and that some of the corroborating research uses spurious logic or questionable methodologies.
¶ 8 Leave a comment on paragraph 8 0 If we can conclude anything at all from the research in this field, it’s that the relationship between P2P and music economics is anything but simple. Studies have produced variant findings in part because different groups of people share music in different ways, at different times, under different circumstances, for different reasons. Similarly, the music industry has undergone significant changes in recent years due to a variety of factors, many of which are so closely related to P2P that it’s hard to control for one and measure the impact of the other independently (I’ll discuss these other factors in my next chapter). Even framing the question introduces difficulties. If we look only at music “sales,” are we ignoring revenues that accrue from non-retail sources such as licensing and subscriptions? If we look at “industry profitability,” which firms count as “industry” and which don’t, and whose numbers are we going to use to assess profit and loss? If we are interested in the total “economic impact” of P2P, do we take into account second-order effects like sales of concert tickets and merchandise, or word-of-mouth marketing? To my knowledge, nobody has yet addressed these questions definitively, and it’s entirely likely that a definitive answer is downright impossible. Far from being an unmitigated threat to the bottom line for artists, composers, labels and other stakeholders in the music economy, P2P is more of a digital Rorschach test; any assessment of it is far more likely to reflect the viewer’s biases and preconceptions than to represent an objective measure of its total impact on the marketplace.
¶ 11 Leave a comment on paragraph 11 0  We used regression analysis, typically considered a statistical indicator of causality, and a stronger indicator of meaningful relationship than correlation.