Guest-blogged by Authors Alliance founding member Ariel Katz, Associate Professor of Law and Innovation Chair in Electronic Commerce at the University of Toronto.
Authors Alliance recently joined a coalition of research, science, and education organizations that called on the Association of Scientific, Technical and Medical Publishers (STM) to withdraw a set of New Model Licenses for purportedly “open access” publishing.
Beyond the flaws in those Model Licenses, the STM move raises some potentially serious antitrust issues. In other words, by adopting these set of model licenses and recommending that their members adopt them, STM and its member publishers might have broken the law. This is problematic for authors who write to be read, and who deserve a competitive publishing environment that allows them to find publishers who share their commitment to openness.
STM is an industry group of academic and professional publishers. According to its website, “it has over 120 members in 21 countries who each year collectively publish nearly 66% of all journal articles and tens of thousands of monographs and reference works.” STM members are competitors. For example, they compete downstream (in the sale of journals and subscriptions), and they compete upstream (competing for authors’ original content). While academic publishers typically do not pay authors for their content (and therefore do not compete on price in that market), they do compete on other features, such as the prestige of the publication, the quality and speed of the publication process, how widely the publication is disseminated, etc. Increasingly, as more and more authors consider the openness of a publication venue when they decide where to publish, publishers compete on this aspect too.
For antitrust purposes, when a group of publishers adopts a set of uniform licenses, or when it recommends that its members adopt them, they tread in the area of antitrust law’s core concern: “price fixing”. In antitrust lingo the term price fixing is not limited to coordinating on price, but applies to any coordination that affects the quantity, quality, or any other feature of the product. Indeed, “[t]erms of use are no less a part of ‘the product,’”[1] and competition between publishers is supposed to ensure optimal license terms just as it is expected to guarantee competitive prices. Therefore, when a group of publishers coordinates license terms, their concerted action is not conceptually different for antitrust purposes from a decision to coordinate subscription fees (downstream) or submission fees (upstream), and when the group represents the leading publishers and affects the majority of publications, antitrust concerns are further heightened.
But not any coordination among competitors is automatically illegal. Antitrust law permits some coordination if it is ancillary to what otherwise would be a pro-competitive activity and if it restricts competition no more than what is reasonably necessary to achieve the pro-competitive objective. For example, the benefits of standardization often justify some coordination among competitors. Setting and implementing common standards may be beneficial for consumers because it may decrease various transaction costs or facilitate compatibility between products and services. So, in principle, adopting a set of license terms might be pro-competitive, and STM might argue that what it’s doing is akin to what standard-setting organizations (SSOs) do.
This may be a fine point, but not a trump card because SSOs have and continue to give rise to antitrust concerns, and their activities often attract close antitrust scrutiny. The question that antitrust agencies or courts would have to ask is whether there are anti-competitive aspects of the STM initiative that go beyond what is reasonably necessary to achieve whatever pro-competitive goal coordinated set of model licenses attempts to achieve.
There seem to be at least three such aspects:
1. Undermining the benefits of standardization
While standardization is an often-cited benefit that requires coordination, it would be difficult for STM convincingly to argue that its initiative is required to achieve standardization. In fact, as the Coalition letter points out “Creative Commons licenses are the de facto global standard for providing users with legal confidence of their rights to reuse content. They are not perfect, but they have been applied to over a billion resources by millions of authors. Creative Commons licenses are the preferred option supported by major content platforms and Open Access publishers. They are recommended by governments in Australia, Europe, the United States and elsewhere.”
Indeed, one of the main problems with the STM Model Licenses is that, if adopted, they would reduce the benefits of standardization by making STM scholarly literature legally incompatible with hundreds of millions of Creative Commons licensed pictures on Flickr, videos on YouTube, articles on Wikipedia, and at least 1.2 million existing scholarly articles licensed on Creative Commons terms. Moreover, while Creative Commons licenses reduce transaction costs by using common terms and a common and established legal framework, the STM model licenses might increase such costs by creating legal uncertainty that could only be resolved by legal action, probably in multiple jurisdictions.
Moreover, not only do Creative Commons licenses already function as the standard, they also have been developed and adopted in a way that doesn’t give rise to any serious antitrust concerns. They may not be perfect, and in theory there might be a superior standard that serves the public interest better, but antitrust law is generally suspicious of claims by self-interested competitors that their concerted action benefits the public more than it benefits them.
2. Privately Re-writing Copyright Law
Unlike Creative Commons licenses that explicitly clarify that their terms do not have to be complied with when the use is permitted under any exception or limitation to copyright, some of the STM licenses purport to license (while at the same time control and limit) uses that do not require permission. For example, they give permission to cite a work only with an appropriate bibliographic citation, although permission to cite is not legally required, or they even prohibit acts that are generally legally permissible (e.g., linking, or text and data mining). Copyright law is a carefully balanced statutory scheme that allocates rights to owners, users, and the public at large, and the STM Model Licenses constitute a concerted effort by publishers to rewrite this balanced scheme by claiming rights that publishers do not have. Similar attempts to collectively extend the scope of IP rights or assert rights that do not exists, have been found to run afoul the antitrust laws in the past,[2] and this attempt may equally be found to be illegal.
3. Polluting Open Access
While it is certainly a positive development that many STM publishers begin—albeit reluctantly—to embrace open access, there is little doubt that many of them see open access publishing as a threat to their existing profitable business models. STM and many of its members have lobbied extensively against various open access mandates, and were behind the attempt to pass the failed Research Works Act.
Against this backdrop, releasing the set of Model Licenses is reminiscent of Microsoft’s “extend, embrace, and extinguish” strategy, for which, among other things, it was found to be liable for antitrust violations. In the late 1990s, Microsoft identified Java as a threat to the MS-Windows’ dominant position. Java threatened Windows because it allowed programmers to write application that could run on various operating systems. In response, Microsoft released and promoted its own Microsoft Java Virtual Machine, or “polluted java.” It heavily promoted it as a better Java, while deceptively concealing that fact that it was designed to be incompatible with any platform other than Windows.
Realizing that the they lost the battle to block open access, STM’s release of the Model Licenses and describing them as “open access” licenses—where in fact they are designed to be less than and incompatible with existing and widely used open access licenses—seems to be a version of a similar “extend, embrace, and extinguish” strategy. For authors who write to be read, such collusion is as harmful as collusion with respect to prices.
In response to the call to withdraw the Model Licenses, STM announced that it “is pleased to see such wide-ranging discussions taking place around the best licensing options available for Open Access” and it declared that it welcomes “all constructive debate and discussion around improving the options available to authors, funders and publishers within open access licensing.”
That is all nice and well. But STM will be wise to ensure that while it pursues such constructive debate, it is not simultaneously engaged in illegal concerted action.
[1] ProCD, Inc. v. Zeidenberg, 86 F.3d 1447, 1453 (7th Cir. 1996). [2] E.g., Standard Sanitary Mfg. Co. v. United States, 226 U.S. 20 (1912); Straus & Straus v. American Publishers’ Ass’n, 231 U.S. 222 (1913); Fashion Originators’ Guild of America v. FTC, 312 U.S. 457 (1941).