Meta-analysis of studies on file-sharing

A post by Slashdot user Dangerous_Minds summarizes a series ZeroPaid's Drew Wilson, who has been examining 20 file-sharing studies from the decade-plus-long filesharing wars. Time and again, the studies show that the effect on markets is marginal, and that the big entertainment companies are opposed to file-sharing a means of suppressing competition and innovation:

While most writers would simply criticize the study and move on, Wilson took it a step further and looked in to what file-sharing studies have really been saying throughout the years. What he found was an impressive 19 of 20 studies not getting any coverage. He launched a large series detailing what these studies have to say on file-sharing. The first study suggests that file-sharing litigation was a failure. The second study said that p2p has no effect on music sales. The third study found that the RIAA suppresses innovation. The fourth study says that the MPAA has simply been trying to preserve its oligopoly. The fifth study says that even when one uses the methodology of one download means one lost sale, the losses amount to less than $2 per album. The studies, so far, are being posted on a daily basis and are certainly worth the read."

What Various Studies Really Reveal About File-Sharing


  1. Does how niche the market is make a difference? Or the medium? (Books vs. music vs. TV vs. Hollywood vs. Bollywood…) I don’t think the impact of textbook piracy would be negligible for that market, since most customers are only buying out of necessity.

  2. I’m not sure it’s fair to call Wilson’s work a meta-analysis. His writing and “analysis” demonstrates his lack of exposure to economics research. 

    The papers he has discussed thus far, and my own quick survey of the literature, have fundamental flaws. Yet Wilson’s criticism of the Phoenix report (which does rely on the faulty premise that p2p downloads represent lost sales) skips the math and criticizes the authors for not citing others who have shown what they are trying to prove (using a pretty standard model).
    From what I skimmed of his other work, he never questions the methodology of any of the papers that have a pro-file sharing conclusion, despite their universal reliance on what I believe to be a terrible instrumental variable, questionable data sets, and (largely) data from the early 2000s when file sharing behavior was much different than it is now (napster anyone?).I applaud the effort, and I agree with his politics on the issue, but this is at best a sloppy analysis that I can’t imagine convert any readers (or policymakers) to our side.

    1. Totally agree here, and it’s interesting that the commenter above seems to have approached this from an economics angle. It would be great if we could consistently use the term “meta-analysis” (even though I know jargon differs between disciplines), which is a quantitative analysis of the literature, usually done, at least in biomedicine, in a type of literature review known as a systematic review. Here, they don’t seem to have combined results of individual studies quantitatively, and I couldn’t find anything about how the articles they analyzed (or reviewed) were selected (i.e., no documented, replicable search strategy for the literature on the topic). 

Comments are closed.