In a study published earlier this month, professors Brett Danaher (of Wellesley College’s Department of Economics) and Michael D. Smith (from Carnegie Mellon University’s H. John Heinz III School of Public Policy and Management) conclude that the January 2012 shutdown of Megaupload positively affected the digital sales of movies for the two major film studios that provided data to the study.
Given the difficulty of collecting data in support of that conclusion — and the fact that the study was reportedly partially underwritten by the MPAA — the study is bound to be met with skepticism from copyright reform advocates and piracy sympathizers, but my reading of the study found it to be reasonable in its methodology, and quite measured in its claims, all things considered.
To cut to the chase, the study’s bottom line conclusion is that “in the 18 weeks following the (Megaupload) shutdown, digital revenues for these two studio’s movies were 6-10% higher than they would have been if not for the shutdown.”
The means by which the professors arrived at that conclusion is quite complicated, and beyond the scope of this post. Rather than focus on their methodology, I’d like to highlight some of the caveats the professors offer about their study and its conclusions — caveats that strike me as very reasonable, forthcoming and intellectually honest, but that will likely be summarily ignored by critics of the study, who are already focused on the extent of the relationship between the researchers and the studios that supplied them with data.
Early on in the study’s abstract, the authors note that anti-piracy measures are really only worth undertaking if they have a positive return for rights holders. Or, as the professors put it, “a necessary condition for supply-side anti-piracy polices to be worthwhile is that we must see a causal gain in media sales and revenues resulting from the reduction in piracy.”
The study then acknowledges that “[i]t is not clear that such a gain will be realized.”
“First, when a major filesharing site or protocol is eliminated, other alternatives exist,” the study notes. “Policy interventions against particular sites or protocols may simply transfer filesharing from one platform to another with no net effect on total piracy or sales…. Second, even if total filesharing activity decreases, if the discouraged pirates consist largely of consumers whose reservation prices for the content are lower than the market price, no sales gain will be realized. On the other hand, if the elimination of some of the largest and most convenient filesharing platforms leads some consumers to turn from piracy to purchases on legal channels, then revenues to content providers can increase. For these reasons, the question of whether shutting down major piracy platforms will causally impact sales to consumers is theoretically ambiguous and must be answered empirically.”
Even with confidence in the source data and their analysis thereof, the authors concede that their work has really just scratched the surface in terms of providing a sound basis for the development of public policy.
The following section is worth quoting at length, because it establishes (to me, at least) that the researchers are genuinely interested in producing solid data and a workable basis for policy, rather than just spouting a preconceived position in furtherance of an agenda.
“Finally, we note that while we believe our results strongly suggest that the shutdown of the popular Megaupload and Megavideo sites is causally related to an increase in digital motion picture sales, that there are several limitations associated with our study. First, Megaupload was a very well-known cyberlocker and its shutdown was highly publicized. As such, the shutdown of Megaupload influenced the policies of several other cyberlockers focused on piracy, and our results necessarily measure the “net impact” of the Megaupload shutdown across the cyberlocker industry, as opposed to just measuring the impact of Megaupload. In addition, our results only analyze the impact of Megaupload on digital motion picture sales. We are not able to measure the effect of this shutdown on other motion picture channels (e.g., DVD sales, theatrical sales) or on other product categories (e.g., music, books). Because we only observe 18 weeks following the shutdown, we also do not know whether the sales increase will persist or if these consumers will eventually find their way back to alternative piracy channels (in spite of the fact that we see no clear indication of such a reversion in the 18 weeks in our data). Finally, we note that our study only measures specific benefits of this regulation – it does not measure either tangible or intangible costs of this sort of intervention, and such costs should be considered carefully as part of any policy decisions.”
The full study can be downloaded here.