Yeah, those “loss to piracy figures” are more or less made up.
Ars Technica is running a very informative story about the claims that various organizations and government agencies make about losses to piracy:
http://arstechnica.com/articles/culture/dodgy-digits-behind-the-war-on-piracy.ars
The ultimate conclusion is that the numbers are highly questionable, likely biased, and didn’t really deal with piracy in the first place.
The confusion is a result basically a 20 year “game of telephone” (yes… before P2P!), with all kinds of cross-citations obfuscating that the original numbers probably came from the publishers themselves.
Whether or not the selection of these particular numbers has been intentional (and selecting from the high end of a dubious estimate was likely intentional at least one time), this unexamined citation is certainly lending an undeserved air of authority and legitimacy to those numbers. For example, it sounds better if those numbers come “from the F.B.I.” or “from the U.S. Chamber of Commerce,” even if the monetary loss estimates were originally coming from self-reports of the industry itself. Furthermore, those numbers didn’t actually deal with “piracy.” For example, the estimates included revenue “lost” from countries with laws that weren’t as broad as ours! Those estimates, reportedly according to the initial author, “could admittedly be biased and self-serving.”
Of course, those aren’t the only numbers thrown around, but they are used quite a bit by a number of sources, including government agencies themselves and those responsible for passing the recent “PRO-IP” bill. (Meanwhile, the Orphan Works bill languishes…)
At any rate, we need better and more transparent research involving potential losses to infringement.