An Objective Analysis of Site Blocking: What the Research Shows
by Dr. Brett Danaher — 31 Oct 2017
I’ve just returned from my first visit to Taiwan, where I met with representatives from the ROC Government and the local creative industries, and participated in an international conference regarding the continued evolution of online governance. One of the issues discussed, which seems to be politically sensitive in Taiwan, concerned proposed amendments to Taiwan’s copyright law and specifically a provision for blocking for pirate websites. Apparently this has been considered in the past but has in each instance been met with strong opposition from citizens concerned about the impact it will have on use of the Internet – as a result, government has been reticent.
Arguments against piracy website blocking seem to mostly take one of two forms: there are arguments about it violating freedom speech or potentially “breaking the Internet” (to use a popular phrase), and there are arguments of the form “it won’t have any impact, you can still find ways to pirate any content you want, so why do it?” I’m not a legal scholar, so I don’t have any special expertise on freedom of speech. I’m also not an expert on the technical side of the Internet – although I do know that site blocking has been implemented in over 40 countries and in studying site blocking in several of those countries I’ve seen no evidence of a “broken Internet” or even a serious challenge to net neutrality. But I would leave the arguments about this to those whose expertise lies in the area of Internet policy regulation.
That said, I can offer my expertise on the argument that website blocking won’t have any impact because pirate content can always be found somewhere. I’ve been doing empirical research on this claim for the better part of a decade now; specifically, I’ve been looking at whether supply side antipiracy policies – enforcement actions that target piracy sites or protocols – can influence consumers to turn from illegal channels to legal ones. And I’ve been keeping an eye on the research of colleagues asking this same question. My findings are summarized in a peer-reviewed article published in Communications of the ACM.
Basically, there is no evidence of a supply side antipiracy action that completely removes popular content from the Internet – you can always find it somewhere. And there is evidence that antipiracy actions that do not make piracy sufficiently difficult have no meaningful impact on legal sales or total piracy, because consumers can either find other sites from which to pirate the content or circumvent the enforcement actions. These results are found by Aguiar et. al., by Poort et. al, and by Danaher et. al.
However, research also shows that taking actions against piracy sites can sometimes make piracy so inconvenient that a meaningful group of marginal consumers will turn their behavior from illegal sources to legal ones. For example, Mike Smith and I found that when Megaupload.com and Megavideo.com (the two largest piracy cyberlockers in the world) were both shutdown, many sites that linked to their content stopped working and many cyberlockers shifted their policies to be less tolerant toward copyright infringement. The result was a causal increase in revenues from digital movie sales and rentals of about 7.5%. Later, along with Rahul Telang, we found that even though the UK blocking the thepiratebay.org caused little decrease in total piracy and no increase in legal consumption, the UK blocking of 19 major video piracy sites in November 2013 caused a 12% increase in legal consumption and a large decrease in total piracy. Later, when 53 more piracy websites were blocked in November 2014, we found a similar effect.
The point is that supply side antipiracy actions fail when they only slightly raise the difficulty of pirating content, but seem to succeed in nudging consumption toward legal channels when they sufficiently raise the cost/inconvenience of pirating content.
Based on my trip, it appears that Taiwan’s creators want to explore piracy site blocking as a potential means of protecting content against piracy and increasing legitimate consumption. By all means, interested parties should debate what impact this could have on the Internet and/or whether it is legally desirable. But all of the research that I have undertaken indicates that one cannot argue against piracy website blocking by saying it won’t work – the data clearly indicate that when enough sites are blocked to make piracy sufficiently inconvenient, viewers migrate some of their consumption from illegal channels toward legal, revenue-generating channels.
My hope is that policy makers in Taiwan, and elsewhere, will not base their discussions and decisions solely on passionate rhetoric and political posturing, but will instead consider all sides of the argument – including empirical data – reasonably and objectively. Creators and Internet users in Taiwan and around the world deserve no less than that.
Dr. Brett Danaher is an empirical economist and data scientist who studies the effects of new, disruptive technologies on the media and entertainment industries with particular attention to how these technologies alter optimal firm strategy or government policy. He has served as an expert witness in several U.S. and European trials and is currently an Assistant Professor at the George L. Argyros School of Business and Economics at Chapman University.
An Objective Analysis of Site Blocking: What the Research Shows
by Dr. Brett Danaher — 31 Oct 2017
I’ve just returned from my first visit to Taiwan, where I met with representatives from the ROC Government and the local creative industries, and participated in an international conference regarding the continued evolution of online governance. One of the issues discussed, which seems to be politically sensitive in Taiwan, concerned proposed amendments to Taiwan’s copyright law and specifically a provision for blocking for pirate websites. Apparently this has been considered in the past but has in each instance been met with strong opposition from citizens concerned about the impact it will have on use of the Internet – as a result, government has been reticent.
Arguments against piracy website blocking seem to mostly take one of two forms: there are arguments about it violating freedom speech or potentially “breaking the Internet” (to use a popular phrase), and there are arguments of the form “it won’t have any impact, you can still find ways to pirate any content you want, so why do it?” I’m not a legal scholar, so I don’t have any special expertise on freedom of speech. I’m also not an expert on the technical side of the Internet – although I do know that site blocking has been implemented in over 40 countries and in studying site blocking in several of those countries I’ve seen no evidence of a “broken Internet” or even a serious challenge to net neutrality. But I would leave the arguments about this to those whose expertise lies in the area of Internet policy regulation.
That said, I can offer my expertise on the argument that website blocking won’t have any impact because pirate content can always be found somewhere. I’ve been doing empirical research on this claim for the better part of a decade now; specifically, I’ve been looking at whether supply side antipiracy policies – enforcement actions that target piracy sites or protocols – can influence consumers to turn from illegal channels to legal ones. And I’ve been keeping an eye on the research of colleagues asking this same question. My findings are summarized in a peer-reviewed article published in Communications of the ACM.
https://cacm.acm.org/magazines/2017/2/212432-copyright-enforcement-in-the-digital-age/fulltext
Basically, there is no evidence of a supply side antipiracy action that completely removes popular content from the Internet – you can always find it somewhere. And there is evidence that antipiracy actions that do not make piracy sufficiently difficult have no meaningful impact on legal sales or total piracy, because consumers can either find other sites from which to pirate the content or circumvent the enforcement actions. These results are found by Aguiar et. al., by Poort et. al, and by Danaher et. al.
However, research also shows that taking actions against piracy sites can sometimes make piracy so inconvenient that a meaningful group of marginal consumers will turn their behavior from illegal sources to legal ones. For example, Mike Smith and I found that when Megaupload.com and Megavideo.com (the two largest piracy cyberlockers in the world) were both shutdown, many sites that linked to their content stopped working and many cyberlockers shifted their policies to be less tolerant toward copyright infringement. The result was a causal increase in revenues from digital movie sales and rentals of about 7.5%. Later, along with Rahul Telang, we found that even though the UK blocking the thepiratebay.org caused little decrease in total piracy and no increase in legal consumption, the UK blocking of 19 major video piracy sites in November 2013 caused a 12% increase in legal consumption and a large decrease in total piracy. Later, when 53 more piracy websites were blocked in November 2014, we found a similar effect.
The point is that supply side antipiracy actions fail when they only slightly raise the difficulty of pirating content, but seem to succeed in nudging consumption toward legal channels when they sufficiently raise the cost/inconvenience of pirating content.
Based on my trip, it appears that Taiwan’s creators want to explore piracy site blocking as a potential means of protecting content against piracy and increasing legitimate consumption. By all means, interested parties should debate what impact this could have on the Internet and/or whether it is legally desirable. But all of the research that I have undertaken indicates that one cannot argue against piracy website blocking by saying it won’t work – the data clearly indicate that when enough sites are blocked to make piracy sufficiently inconvenient, viewers migrate some of their consumption from illegal channels toward legal, revenue-generating channels.
My hope is that policy makers in Taiwan, and elsewhere, will not base their discussions and decisions solely on passionate rhetoric and political posturing, but will instead consider all sides of the argument – including empirical data – reasonably and objectively. Creators and Internet users in Taiwan and around the world deserve no less than that.
Dr. Brett Danaher is an empirical economist and data scientist who studies the effects of new, disruptive technologies on the media and entertainment industries with particular attention to how these technologies alter optimal firm strategy or government policy. He has served as an expert witness in several U.S. and European trials and is currently an Assistant Professor at the George L. Argyros School of Business and Economics at Chapman University.
BACK TO ALL