New Study Estimates that Copyright Piracy Accounts for Almost 25% of Global Internet Traffic

Posted · Add Comment

On Monday, the Information Technology & Innovation Foundation hosted a panel discussion that focused on several new online-piracy-related studies, including the recent MarkMonitor study of online counterfeiting and piracy, and an ambitious new study by the firm Envisional, commissioned by NBC/Universal,  entitled An Estimate of Infringing Use of the Internet (summary, full report).

The Envisional study tries to estimate the percentage of Internet bandwidth used for piracy of copyrighted content other than “pornography,” a term that the study uses in the broad sense of “sexually explicit content.”  It concludes that about 1 of every 4 packets traversing the Internet contains infringing, non-pornographic content.  In other words, copyright piracy accounts for about 25% of global internet use.

What is remarkable about this study is that its 25% estimate significantly understates the true volume of global Internet copyright piracy because it excludes “adult” content.

When asked about what the numbers might look like if Envisional had included porn, Dr. David Price reported that their rough estimate was that about 75%-80% of “adult” content appeared to be infringing, and that if infringing adult content were included in the Envisional estimate, that would raise the overall estimate of infringing Internet use to about 35% or higher—in other words, about 1 of every 3 packets.

In addition to providing the best existing estimate of the overall volume and scope of Internet piracy, the Envisional study also provides many other useful findings and implications.  Three seem particularly important.

First, the Envisional study provides more bad news for those who still insist that because peer-to-peer networking itself has legitimate, efficient uses, existing popular implementations of “p2p” file-sharing have “substantial non-infringing uses.”  They don’t—under any rational definition of the phrase.  Here are the percentages of various forms of peer-to-peer traffic that Envisional found non-infringing: BitTorrent: 0.01%; Gnutella: 6.2%; eDonkey: 1.2%.  The Envisional study—just like prior studies—thus reconfirms that now-popular filesharing programs are so inefficient and unfit for lawful data exchange that they are used almost exclusively for piracy.  It’s difficult to imagine forms of “network management” more reasonable than those that those that might disadvantage traffic that is both burdensomely inefficient and almost entirely unlawful.

What may be worth noting is that P2P itself has changed dramatically.  Tools for P2P sharing of infringed content are the same as ever, but the technology behind P2P has moved into the mainstream.  Edge Caching is now the norm, many large enterprise software packages use forms of P2P for managing data across global networks, and legitimate video streaming technologies incorporate some of the efficiency modeling that originated in P2P.  So the effort of the file-stealing community to hang its hat on the ‘good technology’ coat rack is a false one.  The technology is being incorporated into legitimate uses, but not by the purported leaders in P2P like BitTorrent.

Second, Envisional offered equally bad news to those operators of piracy-friendly “cyberlockers”: Envisional found infringing content at the end of 91.5% of links to non-pornographic cyberlocker-hosted content.  That tells us something useful about the dishonesty of cyberlocker operators who want us to believe that they are so incompetent—so incapable of learning from the past—that they unintentionally built into their systems the very same “security vulnerability” that caused the early “cyberlocker” site,, to be hammered with a huge statutory-damage award a decade ago.  It’s one thing for site operators to claim that they cannot always predict how truly new technologies will be used—but quite another to ask us to believe that they are honestly incapable of learning anything from a decade of past experience.

Third, the Envisional study also highlights the importance of rationally resolving the often histrionic debates about the Combating Online Infringement and Counterfeits Act (COICA).  While Envisional estimated the percentage of global bandwidth used for infringement of non-pornographic content at about 25%, it estimated the percentage of U.S. bandwidth used for infringement at about 18%.  In other words, as in the brick-and-mortar world, legal commerce, rule of law, and U.S. statutory damages that enable copyrights to function as exclusive rights have made internet piracy less prevalent here than elsewhere.  That difference should vastly increase once courts reverse the Grokster-rerun called Viacom v. YouTube, and U.S. ISPs  conclude that voluntary cooperation with rightsholders is a far better prospect than watching their customers get sued or inadvertently share their sensitive personal data with identity thieves or pedophiles, or a governmental or corporate employer’s data with Iranian or Chinese spies (indeed, so much sensitive data that file sharing networks are now being data-mined by Wikileaks). As that happens, the critical question will be the one that COICA seeks to answer: How do we keep foreign pirates and counterfeiters out of our domestic internet-based markets?  During the 20th Century, when international commerce flowed through our seaports, the government fulfilled its WTO obligations, by using administrative inspections and seizures of infringing goods to protect our consumers, creators, and law-abiding domestic distributors from foreign infringers.

But as Google boasts in a recent whitepaper, during the 21st Century international commerce will increasingly flow—not through the ports on our nation’s coasts—but through the virtual “ports” in our computers.  That raises the critical question that COICA seeks to answer: In the age of the internet, how can the U.S. government best perform its traditional, WTO-mandated function of keeping foreign infringers out of our domestic markets?  It seems highly unlikely that we want customs to inspect packets in the same way that it has traditionally inspected shipping containers.  But if so, then how can the government execute its WTO obligations and protect our borders and domestic markets as the internet becomes an increasingly important means of international commerce?  This is the question that COICA seeks to answer, and the Envisional study should reinforce the importance of answering this question thoughtfully and effectively.

Thomas Sydnor II is a Senior Fellow in Intellectual Property at ACT. He researches and writes on intellectual-property law, particularly on copyrights.