Last month, I shared some thoughts on the subject of piracy as a tool for promotion, and I won’t repeat all that here.  Suffice to say, I’ve never understood why this particular argument has ever be taken seriously—other than the obvious reason that it offers a plausible sounding rationalization. But, even without thinking too deeply about the matter, just the anecdotal evidence hardly seems to support the premise at all.  Major works like blockbuster movies, which remain the most pirated content, don’t need piracy for promotion; if piracy only promotes to other “consumers” who also don’t pay, it’s a weak argument to make in the first place; and of course, there remains the inescapable logic that if the producers’ of the work didn’t ask pirates to promote for them, it’s not really a service, is it?

Still, the preposterousness rages on in the blogosphere, with any number of writers presuming to school creative producers in the orthodoxy of the future, including vague suggestions to “harness the power of piracy” rather than seek to eradicate it.  It’s a silly and often circular argument because, of course, nobody has ever really studied the comparison between direct loss vs. promotion attributable to piracy.  Until now.

Researchers at Carnegie Mellon University have, for the first time, compared the effect of cannibalization (i.e. box office losses) with the effect of promotion due to to piracy.  Conducted by Liye Ma, Alan L. Montgomery, and Michael D. Smith, their findings reveal that although piracy does have a promotional effect on actual sales, this positive is substantially outweighed by the negative effects of cannibalization.  Comparing both effects to the counterfactual of a market without any piracy at all, the study indicates that, without piracy, producers would have seen a 15% ($1.3bn) increase in box-office sales for the period from 2006-2008 and a 14% increase from 2011-2013.  Meanwhile, their findings indicate that piracy-related promotion contributes to 1.5% of current box-office sales.  So, piracy puts a little bit back, but nothing compared to what it takes away.

As with my post about the Singapore Study, I cannot comment with any authority on the methodologies use in this report.  My last math class was in the 11th grade, and I don’t know what this means:  X = ⎡⎣U C S+ S− I R ⎤⎦ʹ But the study’s conclusions do jibe with the common-sense assumption that black markets probably harm legitimate markets.