European Parliament Approves Controversial EU Copyright Directive

The European Parliament plenary voted 348 to 274 on March 26 to approve the Directive on Copyright and Related Rights in the Digital Single Market (Copyright Directive).[1] Originally proposed in September 2016,[2] the current text is available here. Before the Copyright Directive can go into effect, it must first be adopted by the Council of the EU,[3] which is expected to vote on the Copyright Directive on April 15, 2019.[4] If adopted, the Copyright Directive will be published in the Official Journal of the EU and then member countries will have two years to pass implementing legislation and regulations on a national level because the terms are not self-executing and leave room for interpretation.[5] Thus, as explained below, although the Copyright Directive sets forth a framework of rights and obligations, there are still many open questions about its application in practice that will be addressed at the national level.

The European Commission has described the framework in the Copyright Directive as “ensur[ing] the right balance between the interests of all players—users, creators, authors, press—while putting in place proportionate obligations on online platforms."[6] That said, certain provisions have been the subject of intense debate, leading some to describe the Copyright Directive as “the most controversial issue in EU history.”[7] The controversy, and this LawFlash, concern two articles in particular: Article 11 (now 15), the so-called “link tax,” and Article 13 (now 17), the so-called “filtering requirement.” Below is a brief explanation of key aspects of these provisions, which are expected to change the day-to-day operations of online news aggregation and content sharing services in the EU. Potential spillover effects into non-EU markets are described as well.

Article 11 (Now 15)[8]

Article 15 is often called the “link-tax” because it will impose liability on “information society service” providers[9] (referred to here as online news aggregators) that do not obtain permission to embed code that links to and displays more than “individual words or very short extracts” of press publications hosted on other websites.[10] Notably, Article 15 expressly exempts simple hyperlinking, which does not display content from third-party websites.[11] If the Copyright Directive is adopted, an online news aggregator that wants to embed a link displaying more than “a very short extract” of covered content will need to license such use. Article 15 grants publishers the rights necessary to authorize such uses while leaving ownership of the copyright with the original author. Press publications covered by the new “link tax” include online journalistic works, such as literary works, photographs, and videos on online news websites, but not publications published for scientific or academic purposes, such as scientific journals.[12]

Importantly, this “link tax” will only apply to covered content first published after the Copyright Directive enters into force, so links to current press publications are not impacted.[13] Further, the publishers’ new rights will also expire two years after the covered content’s publication, calculated from January 1 of the year following the date of publication.[14] Although this basic framework is clear, the most critical question remains: How will individual countries interpret the scope of “individual words or very short extracts”? While key stakeholders may push EU member states to adopt consistent standards, the level of disagreement surrounding Article 15 thus far cautions for possible inconsistent interpretations, which may encourage website owners to adhere to the strictest national law – or leave certain markets altogether.[15]

In the United States, there is currently no “link tax,” but a recent case found at least a hypothetical need for a license to embed links under US law. Goldman v. Breitbart, 302 F. Supp. 3d 585 (S.D.N.Y. 2018), illustrates by analogy what impact Article 15 rights could have in practice. In Goldman, a photographer alleged that media organizations infringed his copyright in a photograph by embedding a Twitter post containing that photograph on their websites. The court found that using embedding code to make the photograph “visible without the user having to click on a hyperlink” would constitute copyright infringement of the display right absent a defense.[16] That said, the Goldman court’s exploration of link-based liability remains hypothetical: the court noted that “there are [still] genuine questions about whether . . . [the] image [was] in the public domain . . . [and] a very serious and strong fair use defense.”[17] This decision also has received significant attention for declining to apply the US Court of Appeals for the Ninth Circuit’s well-known “server test.” The server test, introduced in Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146 (9th Cir. 2007), has effectively shaped the modern internet and requires that a work be stored on a website’s servers—not just hyperlinked—in order to constitute copyright infringement under US law. If the Southern District of New York ultimately concludes that the now-standard practice of embedding code, such as by embedding a Twitter post containing a photograph on a news website, can constitute infringement and is not otherwise excused by fair use, it may open the door to a “link tax” (or at least a market for link licenses) in the United States.

Article 13 (Now 17)[18]

The key feature of Article 17 is that it will shift the burden of reducing online copyright infringement to “online content-sharing service providers” (OCSSP)[19] by making such “internet platforms [] liable for content that users upload.”[20] In other words, notice and takedown will not be enough.[21] Currently, under Article 14(1) of existing Directive 2000/31/EC, OCSSPs are not liable for content uploaded by users if the service provider has no knowledge (actual or apparent) that the content is infringing and “acts expeditiously to remove or disable access” upon obtaining such knowledge (often through a notice). If the Copyright Directive is adopted, this safe harbor will no longer apply to OCSSPs[22] and such services must make “best efforts” to obtain a license to cover public access to copyrighted content uploaded by users on a non-commercial basis.[23] In the absence of a license, national laws will impose copyright infringement liability on OCSSPs for user uploads of unauthorized copyrighted content, unless the OCSSP has

• made best efforts to obtain authorization;
• made best efforts to ensure the unavailability of specific work(s) upon “relevant and necessary information” from rights holders;
• expeditiously disabled access to works upon substantiated notice from the rights holder; and
• made best efforts to prevent future uploads of identified work(s).[24]

Note, however, that Article 17 prohibits national laws from imposing a “general monitoring obligation.”[25] Within that context, to determine whether an OCSSP has satisfied these and other Article 17 obligations, national laws must apply the “principle of proportionality,” which will take into account factors such as availability of cost-effective technology and the audience and the size of the service provider.[26] As these conditions demonstrate, there are many practical nuances within the vague language to clarify in national laws. For example, what will constitute “best efforts” for entities of different size and will filtering be required? Even though the obligations above do not expressly include the word “filtering,” many view the obligation to prevent future uploads as a de facto filtering requirement.[27]

To address these ambiguities, there are plans for the European Commission to “draft guidance on how to apply” Article 17 in consultation with stakeholders and to organize additional stakeholders dialogues to discuss best practices for implementation of the new system.[28] These discussions may help clarify terminology and establish reasonable operating benchmarks to limit overly-burdensome interpretations in national laws. It is unclear whether EU member states will look to evidence of then-current industry practice to define “best efforts,” as in the current Section 512 safe harbor under US law, or will instead attempt to define objective standards. If the latter, efficiency and scalability concerns may again lead to application of the most restrictive standard across all countries.

US safe harbor law is under similar scrutiny. In February, the US Copyright Office announced another public roundtable as part of its study to determine whether the online service provider safe harbors under 17 U.S.C. § 512 (Section 512) need to be updated.[29] The April 8, 2019 roundtable was limited to just two topics: (i) recent developments in US case law; and (ii) “recent international legal and policy developments related to addressing liability for infringing content online”—including the proposed obligations on OCSSPs in the Copyright Directive.[30] In its announcement, the Copyright Office acknowledged “that such proposals have generated widespread debate” but requested discussion of “the effectiveness, ineffectiveness, and/or other impacts on online service provider liability.”[31] Whether the Copyright Directive is relegated to a footnote in the final Copyright Office recommendation or becomes a model for suggested updates to the Section 512 safe harbor remains to be seen, but the dedicated discussion at the roundtable indicates that the controversy could spill over into the United States.[32]

Lastly, while the Article 17 obligations are relevant to many online platforms, various other platforms are exempted from these obligations, including

• not-for-profit online encyclopedias;
• not-for-profit educational and scientific repositories;
• open source software-developing and sharing platforms;
• electronic communication service providers as defined in Directive (EU) 2018/1972,
• online marketplaces;
• business-to-business cloud services; and
• cloud services that allow users to upload content for their own use.[33]

Article 17(6) also limits the obligations on startup platforms in their first three years of operation.[34] There are also protections for users to create and share memes incorporating copyrighted content as a form of social commentary or parody.[35]

Next Steps

Morgan Lewis will continue to track developments under EU and US copyright laws relating to online content and safe harbors.

Contacts

If you have any questions or would like more information on the issues discussed in this LawFlash, please contact any of the following lawyers from Morgan Lewis’s copyright and media teams:

Boston
Jonathan M. Albano
Joshua M. Dalton
Rachelle A. Dubow

San Francisco
Brent A. Hawkins
Carla B. Oakley
Lindsey M. Shinn

Washington, DC
Kristin H. Altoff
Ron N. Dreben
Dana S. Gross
Anita B. Polott