Friday, June 18, 2010 - The Progress & Freedom Foundation Blog

Why Copyright Industry Costs-of-Piracy Studies Correctly Ignore the "Positive Economic Effects" of Criminal Racketeering

I just published a new paper called Punk'd: GAO Celebrates the "Positive Economic Effects" of Counterfeiting and Other Criminal Racketeering ("Punk'd"). It debunks efforts to claim that a recent report from the Government Accountability Office (GAO) reasonably concluded that counterfeiting and piracy have important "positive economic effects" and that all copyright-industry estimates of piracy's economic costs are "bogus," "baseless" and "false." A few of the many examples of such nonsense include Content Industry Piracy Claims Are Bogus and GAO Concludes Piracy Stats Are Usually Junk, File Sharing Can Help Sales.

Punk'd explains why these backfiring claims are the real "bogus junk." This post summarizes some of its arguments.

In April, GAO released a report entitled Observations on Efforts to Quantify the Economic Effects of Counterfeit and Pirated Goods (the "Observations Report"). Just recently, the copyright-hating blog Techdirt called this Observations Report "excellent."

Actually, it was inane. This Observations Report can fairly be said to have reached the following conclusion: "[I]t is difficult, if not impossible, to quantify the net effect of counterfeiting and piracy on the economy as a whole." Punk'd explains why this partly non-responsive and partly unauthorized restatement of the obvious and irrelevant told us nothing while empowering the worst defenders of corporate piracy to blame copyright and trademark owners because counterfeiters and pirates tend to conceal their crimes.

Granted, this sad truth must be taken in context. Over the years, GAO has done many thoughtful analyses of intellectual property rights (IPRs) and their importance to the U.S. economy. But humans are fallible, and in the Observations Report--many things went very wrong.

But that outcome was not ordained. For years, GAO had argued that other federal agencies had failed to show the "leadership" and "accountability" required when dealing with something as important to the present and future American economy as IPRs. Consequently, in the PRO-IP Act of 2008, Congress convinced the President to direct GAO--not some other agency--to "conduct a study to determine how the federal government could better protect manufacturers by quantification of the impacts of... counterfeit goods on... the overall economy of the United States." See Punk'd at 2-3.

This mandate was narrow: GAO was not asked to report on copyright piracy, nor was GAO itself required quantify the costs of counterfeiting--just to assess how the government generally could best help to quantify the economic costs of counterfeiting.

Nevertheless, a series of errors--some internal and some externally driven--needlessly complicated this narrow mandate until the result was an Observations Report that re-imagined the PRO-IP Act's reporting mandate to be nothing more than an absurd call to restate the long-obvious and celebrate the "positive economic effects" of criminal racketeering.

GAO was never asked to report on whether we can precisely quantify the "net economic effects" of counterfeiting, piracy or other forms of criminal racketeering.

Estimating the annual effects of counterfeiting on the U.S. economy is challenging. But the Observations Report repeatedly and needlessly complicated this task until it fairly reported that the task-as-complicated had become essentially impossible. Punk'd shows that three errors led to this result.

First, the Observations Report doubled the complexity of GAO's assigned task by assuming that enacted statutes do not mean what they actually say. GAO presumed that when the PRO-IP Act ordered GAO to report the economic costs of "counterfeit goods," it meant that GAO should on report on both counterfeiting and all forms of piracy.

This violated the first principle of statutory interpretation, and it was a needless insult to Congress. A glance at the U.S. Code--or the TRIPS Agreement--should have proven that Congress knows the difference between trademark counterfeiting and copyright piracy.

Next, the Observations Report quadrupled the complexity of its assigned task by believing anonymous "experts" who argued that when estimating the costs of counterfeiting and piracy, we must calculate their "net costs" by offsetting the costs that illegal and criminal acts inflict upon those whose federal civil rights are being violated by the "positive economic effects" that violating the federal civil rights of others confers upon counterfeiters, pirates, and their associates.

That is absurd. Laws outlawing counterfeiting and piracy require law-abiding Americans to discount to $0 the "positive economic effects" that they could expropriate via counterfeiting or piracy. Presumably, those legally bound by this discount-to-$0 convention include those employed by the Legislative Branch that outlawed counterfeiting and piracy. Nor is this conclusion debatable: the remedial provisions of the Copyright Act, the Lanham Act, and the Criminal Code--particularly the Racketeer-Influenced and Corrupt Organizations Act (RICO)--foreclose any claim that Congress wants anyone to consider the "positive economic effects" of counterfeiting, piracy or other forms of criminal racketeering. Calculating the "net costs" of criminal racketeering is thus absurd--that is why no other GAO report has required cost-of-crime data to quantify the "positive economic effects" that crimes confer upon criminals. See Punk'd, at 3 n.8.

Finally, the Observations Report interpreted the PRO-IP Act to have asked GAO to answer an absurd question: Could the government precisely quantify both the economic costs and the speculative "positive economic effects" of both counterfeiting and piracy?

"No" is the obvious answer to even a quarter of that question. Indeed, the Observations Report correctly notes--like many GAO reports before it--that the economic costs of "illicit activities" can never be precisely quantified; counterfeiting and piracy are no exceptions. But we already knew that: Consequently, the PRO-IP Act cannot be interpreted as a demand to restate the obvious. For example the actual-damages and statutory-damages provisions of the Copyright Act show that the first Congress to recognize that the costs of copyright piracy often cannot be precisely quantified was--literally--the First Congress.

Moreover, past GAO handling of cost-of-crime data would have led Congress to expect that GAO would know that any practical "quantification" of the economic costs of counterfeiting would necessarily be a rough-but-reasoned estimate that would have to be extrapolated from either the best available data or new data that might be reasonably and expeditiously created. For example in its 2007 report Public and Private Entities Face Challenges in Addressing Cyber Threats, GAO noted that the precise costs of computer-crimes were unknown, but then reported cost-estimates derived from an FBI survey collecting industry data. See Punk'd, at 4 n.9, 13, 15 nn.44-45.

Cumulatively, these three errors made the Observations Report entirely unresponsive to the statutory mandate imposed by the PRO-IP Act. The Act required an answer to the following question: What should the federal government do to "better protect manufacturers" by the assisting with the "quantification" of the economic costs of counterfeiting? No answer to that question is expressed or implied in the Observations Report.

Claims that the Observations Report found that industry cost-of-piracy studies are "baseless," "bogus" and "false" are themselves baseless, bogus, false--and backfiring.

For the usual defenders of deliberate, for-profit piracy--trade associations the Consumer Electronics Association and "public interest" groups like Public Knowledge and the Electronic Frontier Foundation (EFF) the badly flawed Observations Report was merely a tool to be mischaracterized for political advantage. They thus argued that the Observations Report had found that copyright-industry cost-of-piracy estimates were all "baseless," "bogus," and "false."

Their claims were nonsense. The Observations Report had embarked on a pointless quest to confirm that it would be "difficult or impossible" to precisely quantify the "net economic costs" (costs-minus-benefits) of counterfeiting and piracy. Consequently, when the Observations Report or its anonymized "experts" critiqued three copyright-industry cost-of-piracy studies, they were merely explaining why those studies had not precisely quantified the "net" costs of piracy--something that none even tried to do. Nevertheless, those generally anonymous critiques did not even arguably find that those industry studies were "baseless," "bogus" or "false."

And we know this for a simple reason: the three copyright-industry cost-of-piracy studies discussed in the Observations Report may have provided only reasoned estimates of the economic costs of piracy, but they are superior in detail, transparency, and methodology to almost all data underlying cost-of-crime estimates relied upon for decades by GAO, FBI, DHS, and many other federal and private entities. See Punk'd at nn.8,45,48.

Consequently, those who argued that the Observations Report proved that the three discussed costs-of-piracy studies were "baseless," "bogus," and "false," necessarily accused GAO, FBI, DHS, and many other agencies of misleading Congress for decades with even more "baseless," "bogus," and "false" cost-of-crime data. Worse yet, when trade associations representing the computer, communications, or consumer-electronics industries, hurled such accusations, they almost certainly accused their own member-companies of submitting equally "baseless," "bogus," and "false" cost-of-crime data to the government.

And then, there is EFF. For nearly a decade, EFF has been urging the architects of internet piracy to affect willful blindness to conceal the scope of the harm that they cause. For example, so called "decentralized" file-sharing programs like Morpheus, Grokster, and LimeWire are technologically inefficient absurdities explicable only as means to facilitate piracy. See, e.g., Tim Wu, When Code Isn't Law, 89 Va. L. Rev. 679, 717 (2003) ("P2P design shows that avoiding copyright requires important deviations from the optimal design for speed, control, and usability.").

Nevertheless, in I Am a Lawyer: Peer to Peer File Sharing After Napster, EFF urged developers of such these obviously piracy-adapted programs to "plausibly deny knowing what your end users are up to" by building "a level of 'plausible deniability' into your... business model." EFF thus advised developers to engage in potentially criminal conduct in order to avoid civil liability. See, e.g, In re Aimster Copyright Litig., 334 F.3d 643, 650 (7th Cir. 2003) (willful blindness can prove criminal intent). Indeed, a former LimeWire executive has now testified that an EFF attorney advised him "to prevent... the preservation of incriminating evidence." Arista Records v. Lime Group, 2010 U.S. DIST. LEXIS 46638 at *27 n.15 (S.D.N.Y. 2010), amended in other part by Order of May 25, 2010, Arista Records v. Lime Group, 06 CV 5936 (KMW), (S.D.N.Y.)(informing the Plaintiffs that the Court would "benefit from more briefing" on the crime-fraud exception to the attorney-client privilege); see also Brief for the United States as Amicus Curiae at 3 n.1, MGM Studios, Inc. v. Grokster, Ltd., 04-480 (Jan 24, 2005) (noting that EFF's client refused to let the U.S. Department of Justice review the full record because it feared "criminal investigation") .

Such conduct reveals EFF's hypocrisy: In its Politico advertisement, EFF sanctimoniously wailed to Congress about the need for a "fact-based... copyright policy"--after a decade of urging the architects of copyright piracy to engage in potentially criminal conduct in order to avoid knowing how much harm they were causing. Fortunately, this hypocrisy also proves why GAO and many other federal agencies were right to have relied for decades on cost-of-crime estimates derived from industry data via methodologies less detailed, transparent, and restrained than those used in copyright-industry costs-of-piracy studies. Two factors were at work:

In short, cost-of-crime studies and data derived from lawful industries may be imperfect. But in the context of copyright piracy, the better-than-most industry studies represent the best data available--at least until EFF convinces pirates to stop averting their eyes in order to generate better data. Imperfect-but-reasoned estimates of the costs of any illegal or criminal activity are vastly preferable to prudish ignorance that empowers criminals and punishes victims during an eternal wait for the Godot of quantitatively precise cost-of-crime data.

posted by Thomas Sydnor @ 3:54 PM | Copyright , Cyber-Security , IP , Innovation , Internet , Trademark