A reply to ACTA critics

Last week was another busy week for developments in ACTA. There were meetings in Mexico by representatives from Canada and its key trading partners to further flesh out how to address the worldwide problems with counterfeiting and piracy. Meanwhile, anti-copyright critics were busy filling the blogosphere and press attacking the proposed treaty.

Of course, the actual text of the treaty is not know by the public. So commentators have relied on their interpretations of leaked documents to try and derail it.

When the criticisms are examined, it will be readily apparent that certain ACTA critics misapprehend what has been disclosed and make assertions or reach conclusions that cannot be justified based on the leaked documents.

Prof. Geist’s Assertions

For example, Prof. Geist published several recent blogs dealing with ACTA.  In one blog, Estimating The Cost of a Three-Strikes and You’re Out System, he attempts to argue against any agreement that would include a graduated response system. He starts his blog with the statement that the discussions in Mexico

“is likely to turn to the prospect of supporting three-strikes and you’re out systems that could result in thousands of people losing access to the Internet based on three allegations of copyright infringement. Leaked ACTA documents indicate that encouraging the adoption of three-strikes – often euphemistically described as ‘graduated response’ for the way Internet providers gradually send increasingly threatening warnings to subscribers – has been proposed for possible inclusion in the treaty.”

He refers to countries such as New Zealand, UK, and France as countries that either have or are implementing such systems.

However, as I pointed out previously in response to similar statements by Prof. Geist, from what is known from the leaked documents, what is actually being proposed is a requirement that ISP safe harbors be tied to eligibility conditions that ISPs “put in place policies to deter unauthorized storage and transmission of IP infringing content”. This self-regulatory approach currently exists in s.512(i) of the DMCA and in comparable legislation in Australia and in other countries.

Both the leaked EU Memo and the EU Analysis of the US proposal makes this clear.

Section 3 of the EU Memo states in part:

“On the limitations from 3rd party liability: to benefit from safe-harbours, ISPs need to put in place policies to deter unauthorized storage and transmission of IP infringing content (ex: clauses in customers’ contracts allowing, inter alia, a graduated response).  From what we understood, the US will not propose that authorities need to create such systems. Instead they require some self-regulation by ISPs.”

The EU Analysis document states:

“The aim of paragraph 3(b) is to establish a system that can be considered to make the exemptions from liability subject to specific conditions…

The proposed paragraph 3(b)(i) adds an important prerequisite for the limitations on liability to apply: the intermediary must adopt and reasonably implement a policy ‘to address the unauthorized storage or transmission of materials protected by copyright or related rights’…”

The leaked documents, therefore, do not support the conclusion that a state sanctioned graduated response system is being proposed as part of ACTA.

Prof. Geist argues that the system being proposed “is likely to turn to the prospect of supporting three-strikes and you’re out systems that could result in thousands of people losing access to the Internet based on three allegations of copyright infringement.”  However, regardless of whether Prof. Geist is referring to a state sanctioned graduated response system or to a self-regulatory eligibility condition, his statements show he continues to misapprehend or inaccurately describe how these systems work.

State sanctioned graduated response systems have the following characteristics: (1) rights holders monitor P2P networks for illegal downloading activities; (2) rights holders provide ISPs with convincing proof of infringements being committed by an individual at a given IP address; (3) educational notices are sent through an ISP to the account holder informing him or her of the infringements and of the consequences of continued infringement and informing the user that content can be lawfully acquired online; and (4) if the account holder repeatedly ignores the notices, a tribunal may take deterrent action, with the most severe sanctions reserved for a court.

Prof. Geist says that this is three strikes “euphemistically described as “graduated response”. In fact, graduated response is “euphemistically” described by Prof. Geist and others as three strikes in order to demonize it.  As, I have previously described , the graduated response systems such as what has been enacted in France and what is being implemented in the UK or New Zealand have built in protections to avoid precisely the consequences Prof. Geists say they have. Prof. Geist continues to slag ACTA by misdescribing how these graduated response systems work.

In so far as Prof. Geist intended to refer to a self-regulatory eligibility condition, his statement that this is a “three-strikes and you’re out systems that could result in thousands of people losing access to the Internet based on three allegations of copyright infringement” is also inaccurate.

The US has had ISP safe harbours since enacting the DMCA in 1998. Congress enacted them in an effort to resolve the unique copyright enforcement problems caused by the widespread use of the Internet. It recognized that tackling copyright infringement on the Internet required balancing the competing interests of several groups. The first set of competing interests includes those of copyright holders and end users. The DMCA intended to balance the need for rapid response to potential infringement with the end-users’ legitimate interests in not having material removed without recourse. The second set of competing interests were those of copyright holders and ISPs whose services may be used to infringe copyrights. The DMCA intended to balance the interests of these parties by creating a mechanism for rights holders to inform ISPs of potentially infringing conduct while, at the same time, providing greater certainty to service providers concerning their legal exposure for infringements that may occur in the course of their activities.  See Ellison v. Robertson, 357 F.3d 1072, (9th Cir.2004); In re Aimster Copyright Litigation, 334 F.3d 643, (7th Cir.2003);  Rossi v. Motion Picture Assoc. of America, 391 F.3d 1000, (9th Cir.2004).

This balancing effort resulted in a statute that creates incentives for service providers and copyright owners to cooperate to detect and deal with copyright infringements that take place in the digital network environment. For instance, a copyright owner who suspects that his/her copyright is being infringed may follow the notice and take down provisions set forth in § 512(c)(3) of the DMCA. Once properly notified, a service provider must respond expeditiously to remove, or disable access to, the material that is claimed to be infringing. If a service provider fails to take down the potentially infringing material, it exposes itself to full copyright liability.  Rossi,  Recording Industry Ass’n of America v. Verizon Internet Servs., 351 F.3d 1229 (D.C.Cir.2003).

These notice and take down provisions form part of the several safe harbors established by the DMCA that protect certain common activities of ISPs. They provide protection from liability for: (1) transitory digital network communications; (2) system caching; (3) information residing on systems or networks at the direction of users; and (4) information location tools.

To be eligible for any of the safe harbors, a service provider must meet a series of threshold conditions. One of them is the eligibility requirement that appears to be part of what is being considered for ACTA. Under § 512(i) of the DMCA for a service provider to have the benefit of the safe harbors it must show that it

“(A) has adopted and reasonably implemented, and informs subscribers and account holders of the service provider’s system or network of a policy that provides for the termination in appropriate circumstances of subscribers and account holders of the service provider’s system or network who are repeat infringers”.

This provision has been interpreted many times by US courts. Importantly, they have interpreted it as a flexible provision which does not mandate any specific circumstances e.g. “three strike” before an account is terminated. Rather the standard is “appropriate circumstances”. Further, it has been held that a “standard of reliability and verifiability” is required before an ISP would be required to terminate a repeat infringer’s account in order to claim safe harbor eligibility.  See, Corbis Corp. v. Amazon. com, Inc., 351 F. Supp. 2d 1090 (W.D.Wash. 2004), Perfect 10, Inc. v. CCBill LLC, 488 F.3d 1102 (9th Cir.2007), UMG Recordings, Inc. v. Veoh Networks Inc 2009 WL 3422839 (C.D.Cal. Sept 11, 2009).

Prof. Geist goes on to state that Spain has rejected graduated response. What he doesn’t disclose, however, is that Spain has passed legislation that will enable the authorities to shut down file-sharing sites rapidly. So it has taken another approach to dealing with massive online file sharing.

Prof. Geist then purports to suggest that countries that are pursuing graduated response have “faced formidable barriers”. He refers to New Zealand which he says “withdrew a three-strikes proposal in the face of public protests (a much watered-down version was floated at the end of last year)”. Infact, one of the main  reasons for the delay in implementing section 92A besides industry and subscriber concerns was the inability of rights holders and ISPs to reach an agreement on a voluntary ISP Copyright Code of Practice (the Code).  The Code would have provided guidance for ISPs and rights-holders on how section 92A would operate. The Government took the initiative in its December 2009 cabinet paper, to outline a full fledged (not watered down) graduated response regime.

Prof. Geist then states that the UK’s proposal has “been hit with hundreds of proposed amendments at the House of Lords”. It is not unusual for any complicated Bill to have amendments proposed to it. Further, many of the proposed amendments had nothing whatsoever to do with the graduated response portion of the Bill. The UK government is still firmly resolved to move ahead with the legislation.

Prof. Geist then refers to “France’s adventure with three-strikes” which he states includes “a Constitutional Court ruling that the plan was unconstitutional”.  He failed to disclose, however, that France has actually passed its legislation and that the French Constitutional Council ruled in October 2009 that the French bill was in conformity with the French Constitution.

Prof. Geist then goes on to argue against a graduated response system because of its costs which he says the UK Government estimated at 500 million pounds over ten years and because of a potential for 40,000 people to lose their Internet access due to anticipated increases in subscriber fees.

The UK impact assessment referred to by Prof. Geist actually found that the estimated costs would be between £290 – 500 million and that this “cost would have a relatively small but permanent effect of reducing demand for broadband connection between 10,000- 40,000”.  Further, the figures assume that all of the costs associated with the graduated response system would fall on the ISPs and would then be passed onto consumers.  However, it appears that the UK Government plans to pass on 75% of the costs to rights holders. If so, the impacts would be far less than what had been estimated.

Far more importantly, however, Prof. Geist examines only the costs of implementing a graduated response system. He never takes into account the benefits that would result from such a system. However public policy mandates that costs and benefits of any proposal be fully evaluated.

Prof. Geist’s references to the UK impact assessment omit entirely its analysis of the far greater economic benefits associated with graduated response. The UK assessment pegged the benefits to rightsholders of implementing a graduated response system alone at £1.7 billion.

The assessment also discussed at length the rational for graduated response. For example, the report states:

“The UK Government attaches particular importance to establishing the UK as a leading digital economy for several reasons:

• A Digital Britain can make a significant contribution to the Government’s New Industry, New Jobs agenda.
• A Digital Britain can play a crucial role in helping the government deliver a number of wider policy objectives.
• Broadcasting, the creative industries and the information and communication technology sectors are of major economic importance in the UK.”

“First, it can play a major part in helping the UK emerge from recession by encouraging innovation – one of the five drivers of productivity – in new digital and broadband technologies and content.  Second, it can provide the telecommunications infrastructure that UK businesses – particularly those in the content and creative industries – crucially rely on in order to compete effectively in the global economy.”

“Illegal file sharing of audio, video, data, or anything in digital format between users on a computer network has increased significantly in the last few years. This has served to reduce the incentive for the creative industries to invest in the development, production and distribution of new innovative content.”

The UK impact assessment then summarized the costs and benefits in the table set out below:

Policy Area	Policy Proposal	Benefits	Costs
Illegal peer to peer file sharing	Preferred policy option outlined in Government Response (January 2009) to previous Consultation (July 2008).This requires ISPs to take direct action against users  identified by rights holders as infringing copyright through peer-to-peer file sharing.	Benefits to rights holders of recovering displaced sales.(Total benefit: £1700 million.) Benefits to consumers in ensuring that investment in high quality and diverse creative content is at appropriate levels.	Costs to ISPs of complying with the legislation, including costs of notifying infringers, capital costs to ISPs, costs of setting up and running a call centre, annual capital and operating costs to mobile network  operators. Possibility of higher broadband costs for consumers. (Total cost: £290 – 500 million.) Costs to low income/low valuation digital product consumers who would stop consuming digital content altogether rather than purchase it; costs to rights holders of identifying infringing IP addresses and taking infringers to court.

 

The assessment also canvassed the various policy options available to address online file sharing and the rational for choosing graduated response over other options.

“The sheer scale of P2P file-sharing means it is not practicable to take all those involved to court: right-holders estimate there are some 6.5 million people in the UK who are active unlawful filesharers.”

“Further, due to the nature of the technology and the way in which individual infringements are identified, it is not possible for rights holders to identify who are the most frequent or serious file-sharers, making targeted legal action extremely difficult if not impossible.  Legislation is needed to require ISPs to notify subscribers that they appear to be engaged in unlawful activity so that they can alter their behaviour.  It is also needed to help rights holders to take targeted action about the most serious infringers.”

“An important feature of creative industries like the music, software and film industries is that they are characterised by strong intellectual property rights (IPR).  Strong IPR creates an incentive to invest in the development of new and more innovative products since it permits individuals to capture the gains from the new products it creates.”

“However with illegal file-sharing the incentive to invest in new and mainstream artists is undermined because industry cannot capture all the gains generated from its investment.  This is because the public good nature of file-sharing and the spillover effects which exist creates a free-riding problem whereby users may enjoy the benefits of file-sharing without paying the product’s price.  The disincentive to invest in artists as of result of free-riding is a particular problem in the music, film and videogames industries because they are characterised by large investment costs and a relatively high risk of failure.”

“Content companies spend vast amounts of money investing in the success of a product (e.g. film, song or videogame).  These costs are typically in production, marketing and promotion of creating and selling content to the consumer (advance payment to artists, advertising costs, retail store positioning fees, press and public relations to the artist, television appearances and travel, publicity and internet marketing). The industry is characterised by large fixed costs and low variable costs. The increasing trend for creative content to be traded digitally may have seen a change in the investment cost structure.  Overall, some costs have remained high like marketing costs but distribution and production costs have decreased with an overall effect of increasing variable costs relative to fixed costs which may give small, relatively less known artists more room for manoeuvre.”

Prof. Geist’s one sided analysis of the costs associated with graduated response has also been criticised by Prof. Olivier Bomsel for failing to take into account how graduated response can best internalize the externalities associated with copyright infringement.  Prof. Bomsel is a well known Professor of Economics and was involved in the design and the implementation of the French law on copyright enforcement over the Internet.  

 In support of his argument against graduated response, Prof. Geist then alleges that the “UK estimates are consistent with a 2006 Industry Canada commissioned study. However, the Industry study did not deal with any graduated response proposal. It focused on the costs associated with a notice and notice system, a notice and takedown system, and a compulsory license regime. Nevertheless, the report concluded that notwithstanding the costs associated with notice and notice or notice and takedown that “both approaches are economically beneficial for ISPs and other stakeholders in that economic uncertainty is removed.”

Other Criticisms of ACTA

Prof. Geist was not alone in trying to derail ACTA.  A consortium of groups consisting of the Washington-based Public Knowledge, the Ottawa-based Canadian Internet Policy and Public Interest Clinic (CIPPIC), and Quebec’s Union des consommateurs, also weighed in on the debate last week. Their views were given extensive media coverage including coverage in Canada from from CBC.ca, Canadian Press, and Radio-Canada.InternetNews.com.

The CBC article “Consumer advocates declare war on copyright treaty” stated the following:

“A number of these organizations on Monday launched a joint declaration of war on ACTA, which they say threatens the fundamental freedoms of the people living in member countries.”

“This agreement will restrain certain rights and freedoms, particularly freedom of expression and protection of privacy,” the declaration said.

“Leaked documents from the ACTA negotiations, going on for the past year and a half, show that a number of anti-consumer provisions are being discussed, the groups said.  They include:

• Allowing internet service providers to cut customers off if sharing copyrighted files is alleged.
• Limiting the interoperability of legally acquired digital content, such as music or movies.
• Authorizing border guards to search and seize laptops and MP3 players if copyright infringing material is found on them.”

These statements are inaccurate and misleading. They also potentially reveal a lot about who these ACTA critics are really trying to protect.

The statement that ACTA will allow “internet service providers to cut customers off if sharing copyrighted files is alleged” is inaccurate and misleading. This is essentially the inaccurate statement made by Prof. Geist which is addressed above.

There is nothing in the leaked documents aimed at “Limiting the interoperability of legally acquired digital content, such as music or movies”. The leaked documents suggest that ACTA would require that parties joining it to enact measures to protect technological measures along the lines of legislation already existing in most countries around the world, except Canada. A comment on Footnote 8 in the EU Analysis suggests that the US proposal “may not require” interoperability of content downloaded for one device to be played on another device. But, there is a big difference between a term that would limit or prohibit content interoperability (which as the EU Analysis pointed out does not exist in the EU; “Recital 48 to CISD uses the term ‘implies no obligation’”) and one that leaves it open to each country to decide whether or how to address interoperability.

Moreover, ACTA is not about authorizing “border guards to search and seize laptops and MP3 players if copyright infringing material is found on them”. Douglas George of DFAIT made this clear in his April 2009 presentation to Canadians when describing “What ACTA is NOT About”: “Seizing portable music players and laptops at the border”.

The statement that ACTA is “anti-consumer” potentially reveals a lot about its detractors. The groups that have coalesced to oppose ACTA seem to believe that graduated response is “anti-consumer”. I had always thought that a consumer was someone who legitimately acquires some good or service for that person’s own use. However, the “consumers” referred to by the coalition appear to be those individuals who engage in repeated acts of copyright infringement despite repeated warnings to stop. So, is the moral fight against the substance of ACTA’s proposed provisions related to graduated response by these detractions really about defending the right to take the creative labour and investments of creative individuals and industries for free?