The Thing We Don’t Talk About in Piracy Estimates
Today the House Judiciary committee is holding a hearing on the impact of copyright on the economy [webcast here]; I have already weighed in on it here. One of the important elements of the hearing should be to address empirical deficiencies in what we know about the effect of infringement. A recent report by the National Academies summed it up: not much.
The problem with research in this area is that it is hard to conduct reasoned analysis when some subjects are off-limits. There are certain assumptions which simply cannot be questioned in economic loss analyses involving IP. It is not unlike the fact that the early study of geology suffered because for generations, it was socially unacceptable for geologists to even discuss evidence undermining the previously unchallenged notion that earth had sprung swiftly into being at the decent hour of 9 AM on a crisp Monday in October. When we cannot even acknowledge the ramifications of certain empirical issues — at least, not in polite company — academic progress is unlikely.
So what is The Issue of Which One May Not Speak? The fact that money not spent on pirated content is, in many cases, still spent.
The U.S. Government Accountability Office pointed this out in a widely discussed report in 2010, observing that “effects of piracy within the United States are mainly redistributions within the economy for other purposes and that they should not be considered as a loss to the overall economy.” Money does not “just vanish.” A Swiss Government commission made a similar observation the following year.
Nevertheless, critics excoriated the GAO report and others like it for simply observing that intra-economy transfers are often redistributive, instead of destructive. Polite people just don’t say things like that.
An old humorous “Adult Swim” commercial pokes fun at this.
Citing a then-recent estimate of the costs of film piracy to the economy, the ad observes that it is false to claim that this money is “lost” to the economy – rather, those are lost to the industry.* When infringement occurs, the infringed rightsholder loses a sale opportunity. Money saved by those infringing content is still spent on other things, however, such as baby formula. Thus, the commercial arrives at the tongue-in-cheek observation that instead of costing the economy, “film piracy feeds babies.” However, television piracy, it concludes with facetious gravitas, “is destroying this great nation of ours. Naughty pirates.”
Several years old now, the ad is nevertheless surprising because it acknowledges the taboo subject: some degree of infringement is not wealth destruction but rather wealth redistribution.
The fact that infringement may be redistributive instead of destructive does not make it acceptable, of course. A violation of a government-granted right is normatively undesirable, because it flouts an entitlement that – at least in theory – reflects the will of the public. This is bad. Even if infringement is “only” redistributive, we still make strong normative societal judgments against involuntary wealth redistribution. This happens regardless of whether it results from law (e.g., by tax policy), or contrary to law (e.g., infringement).
Normatively bad isn’t the same as an economically bad, however. Not all normative transgressions necessarily have macroeconomic consequences. And yet those two items are invariably linked when studies consider infringement. Infringement is bad, therefore we must assign an economic cost to its badness. Hence, study after study makes the repeatedly discredited assumption that every infringement is a lost sale, usually calculated at the highest retail price for which the good was offered, and every lost sale represents a commensurate economic loss.
None of this should diminish the opprobrium we assign to infringement: a reduced return means decreased incentive and may translate to reduced creative output. There must be some truth to this incentive theory, or centuries of IP law are based on a flawed premise. In this sense, redistribution is destruction if it deters future investment. That’s a slightly more complex model – one which has yet to be built.
Clearly, intellectual property is important to our economy — as is open competition, and the free exchange of ideas. These three forces are each valuable tools in the “innovation toolbox”, and allowing any one of them to be undermined – including intellectual property – may impair innovation, along with other important social goals. But as long as the empirical evidence around the policy conversation is so impoverished, we won’t be making well-informed decisions.
* This observation is only true with respect to the U.S. economy. Some amount of piracy occurs abroad, and the “savings” accruing to consumers are invested in foreign economies; this is certainly a loss to the U.S. economy.