To encourage creative activity, the U.S. Constitution authorizes Congress to provide authors and artists the financial incentives to exercise their craft by granting them exclusive rights to their creations for a limited time. Doing so allows for the market exchange of intellectual property that would otherwise be copied and distributed at almost no cost and, consequently, offer little incentive to create new works. In U.S. copyright law, these exclusive rights are subject to some exceptions and limitations, including “fair use.” With a nod to the U.S. Constitutional principle of free speech and recognizing the importance of education and research, fair use allows for the secondary use of copyrighted material for purposes such as criticism, comment, news reporting, research and teaching without either permission from or compensation to the author or artists holding the exclusive right.
While it may sound straightforward, the courts have described fair use as “the most troublesome doctrine in the whole law of copyright.” There is no accounting for “fairness,” so disputes inevitably arise. And, there are those that will attempt to exploit any exception to a rule for profit or pleasure.
When a dispute over fair use does arise, the courts are called upon to make a determination. Section 107 of the Copyright Act provides four non-exclusive factors for a court to consider to determine whether a particular use is a “fair” use: (1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work. In practice, court decisions hinge primarily on the first and fourth factors, with the latter being described by the Supreme Court as “undoubtedly the single most important element of fair use.”
At least, it once was.
Guided by an influential 1990 law review by federal appeals court Judge Pierre Leval, judicial decisions on fair use have focused increasingly on the concept of “transformativeness,” which subsequent courts have interpreted as a key factor in analyzing the first factor—the purpose and character of the use. That is, fair use hinges, in large part, on whether the secondary user’s creation merely “supplants the original” or gives it a “new expression, meaning, or message” or else serves “a new purpose distinct from that of the original work.” (Translations and sequels do not normally qualify as transformative and remain the exclusive right of the author or artist as a “derivative” work.) The more transformative the work, the courts reason, the less the other factors “may weigh against a finding of fair use.”
Like “fairness,” “transformativeness” means different things to different people. Consequently, and perhaps inadvertently, the courts’ application of the concept of “transformativeness” has greatly expanded the boundaries of what constituted fair use, considering only whether the secondary work is somehow different than the original. In doing so, the adjudication of fair use cases in recent decades has weakened considerably the exclusive rights of authors and artists.
Hardly anyone contests the inclusion of fair use in copyright law (or its analogue “fair dealing” found in the copyright laws of other nations). In fact, many of the strongest advocates for strong copyright laws employ fair use in their own creations. Television news relies on fair use to cover stories but also relies heavily on copyright to monetize its content. The question is not whether fair use should exist or not, but rather over its boundaries. With the introduction of “transformativeness” to the analysis, the boundaries of fair use appeared in some cases to step beyond the limits imposed by copyright law. Two recent appellate judicial decisions suggest the courts acknowledge as much, and the course is now set to return fair use to its statutory limits.
The Need for a Coherent Economic Framework of Fair Use
Partly to blame perhaps for an expansion of the boundaries of fair use was the lack of a coherent economic framework for evaluating secondary uses under the “transformative” concept. A new academic paper entitled Fair Use in the Digital Age by scholars at the Phoenix Center for Advanced Legal & Economic Public Policy Studies and Auburn University’s Economics Department in the Journal of the Copyright Society of the U.S.A. (65 J. Copyright Soc’y U.S.A. 1 (2018)) offers a practical economic framework closely tied to the statute, judicial precedent, and Judge Leval’s “transformativeness” idea. The analysis in this article is rich, but it generally points to two key steps in analyzing fair use disputes.
First, does the secondary use qualify under the types of uses permitted by the statute —e.g., criticism, comment, news reporting, teaching, scholarship, or research? This is not an exhaustive list, but neither is it irrelevant. There are secondary uses that clearly do not qualify under fair use, including the translation of a work to a different language or a sequel. Fair use is not open ended. So, as a threshold question, is the nature of the secondary work compatible with the sort of things listed in the law as possibly qualifying for fair use? This question goes to the first factor—the purpose and character of the use.
Second, if the work survives this threshold question, the next consideration is whether the secondary work is sufficiently different from the original that it does not impinge on the market potential of the original work. Such markets may be actual or potential. This is the fourth factor—the effect of the use upon the potential market for or value of the copyrighted work (the factor which the Supreme Court described as “undoubtedly the single most important element of fair use.”). Here is where “transformativeness” becomes relevant. Transformation, to the economist, is simply another word for product differentiation. When two goods are highly differentiated—say toothpaste and motor oil—then they are not in the same market and do not compete with each other. Different brands of toothpaste, however, do compete. So, the question for a court to consider with regard to transformativeness is whether the secondary use reduces the sales of original work or forecloses the original creator from a potential market for its copyrighted creation? The second and third factors—the nature of the original work and how much is used in the secondary work—speak mainly to the question of transformation. A secondary work that takes large portions of the original will, in most cases, be a substitute for the original and thus impinge on the market for the original work.
When analyzing a fair use dispute, courts typically treat the four factors as independent, assessing each in turn and then determining which way the whole analysis leans. This approach is based on court precedent but has no statutory basis. It will, admittedly, be somewhat slow to change. Nonetheless, this approach is confusing because it lacks a sound conceptual framework. The four factors are not independent but interdependent. In fact, a careful read of judicial precedent reveals as much. And, advanced statistical analysis of judicial decisions reveals that the court’s assessment of the fourth factor is almost always determinative, with the first factor a close second in deciding the case (and the two are almost always in agreement).
Despite dicta suggesting otherwise, the four factors are not given equal weight. Indeed, judicial decisions admit that the second factor “has rarely played a significant role in the determination of a fair use dispute” and “typically has not been terribly significant in the overall fair use balancing.” Why? Because the court’s analysis lacks a sensible framework for understanding the relevance of the second factor. This newly published economic analysis gives the second factor a sensible role in the analysis as an input in determining whether the secondary work is sufficiently transformative to qualify for fair use by satisfying the fourth, dominant, factor. When put in its place, the second factor is seen to have an important and sensible role in fair use analysis.
A Return to Economic Principles?
The two recent court decisions turning the dial back on fair use are, in many ways, compatible with this new economic framework.
The first decision came out this February from the Second Circuit in the case of Fox News Network v. TVEyes, 883 F.3d 169 (2d Cir. 2018). There, the defendant offered a for-profit service that continuously records to its servers the broadcasts of 1,400 television and radio channels and allowed its customers to search transcripts of the recordings and then to view, archive, download, and even email relevant segments of the recordings. TVEye’s service consisted of two functions—Search and Watch.
Fox News did not contest the Search feature of the service, which the (majority opinion of the) court decided was “somewhat transformative.” Fox did challenge the Watch functionality of the service that merely re-played relevant portions of the original broadcast. The Second Circuit agreed with Fox News that the Watch function was not transformative and an infringement of Fox’s copyright. If TVEyes wanted to show the broadcast, then it must pay Fox a license fee to do so.
No doubt, much attention will focus on the issue of the transformation. More important than the court’s take on transformativeness (the first factor), however, was its analysis of the fourth factor (the effect on the potential market for the original work). The Second Circuit determined that the Watch function “usurp[ed] a market that properly belongs to the copyright-holder.” That is, the TVEyes service demonstrated that consumers are willing to pay well for a service that permits the search and viewing of select television broadcasts. Thus, there is a market for such a service and that market could be exploited directly by Fox News. The protection of copyright extended to this potential market, whether yet exploited by the copyright holder or not.
The next case came out just last month with a ruling by the Federal Circuit in Oracle v. Google, 886 F3d 1179 (Fed. Cir. 2018). In this case, the question was whether Google’s appropriation of Oracle’s open source Java code to develop its Android mobile platform constituted “fair use.” In a detailed 56-page decision, the court concluded that it did not. Transformativeness played a role in the decision and will no doubt receive much attention, but this part of the analysis was minor. More important to the determination was that the secondary use was not a “fair use” pointed to whether the nature of the use comported with the uses listed in the preamble to Section 107 (commentary, criticisms, news reporting, and so forth) and, as with the TVEye’s case, whether the secondary use usurped the copyright-holder’s potential.
As for transformativeness, the court determined that Google’s use of Oracle’s code was essentially plagiarism and served exactly the same purpose in the original and secondary works. The Federal Court observed that when “copying is verbatim, for an identical function and purpose, and there are no changes to the expressive content or message, a mere change in format (e.g., from desktop and laptop computers to smartphones and tablets) is insufficient as a matter of law to qualify as a transformative use.” Moreover, the court rejected Google’s argument that its use of Java code was de minimis. As the court noted, if it accepted Google’s argument, then it would mean that “verbatim copying could qualify as fair use as long as the plagiarist stops short of taking the entire work.”
The argument is plain enough. But it’s notable that the court anchored its analysis of transformativeness to the uses listed in the preamble to Section 107. That is, the court considered “whether the use is for criticism, or comment, or news reporting, and the like,” and decided that Google’s use of Oracle’s code “does not fit within the statutory categories.” In recent years, Google, among others, have described fair use as an open-ended concept completely divorced from Section 107’s preamble. This opinion puts an end to that philosophy, and rightfully so. By statute, there are some secondary uses compatible with fair use and some that are not. Oracle v. Google appears to be an attempt to re-impose some reasonable boundaries on fair use.
As in the TVEyes case, the Federal Circuit’s decision gave heavy weight to the fourth factor in an analysis of the potential market. The court considered the challenged use’s “impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets.” Google ported Oracle’s code to smartphones, but Oracle could have done the same. The court determined that Oracle “ha[d] the right to enter that market; whether it chooses to do so is entirely its business.” Indeed, Oracle had already licensed its code in the tablet market, thus providing potent “evidence of actual market harm.” In the end, the court found that Google’s “unrestricted and widespread” plagiarism of Oracle’s intellectual property had a significant impact both on market for Java as well as on the software’s derivative markets.
This same decision reached in these two cases can be set neatly inside this new conceptual framework for fair use rooted in law and economics. Take the TVEyes case. The Search function of the TVEyes service permitted interested parties to conduct research on copyrighted television and radio broadcasts. As such, this feature satisfies the threshold question of whether the secondary use is compatible with the sorts of uses consistent with a fair use (in this case, “research”). The Watch function, however, merely re-broadcasted the original work. In doing so, this feature usurped the market from Fox for licenses on secondary transmissions of its broadcasts. That is, the Watch feature was not sufficiently “transformative” to eliminate competition between Fox’s licensing rights and the TVEyes service.
In the Google case, the court was more direct with respect to the threshold question (the first factor), deciding Google’s use “does not fit within the statutory categories.” Deliberations could have ended here, in the same way a summary judgement would be issued if the secondary work was a mere foreign-language translation of the original (a clear “derivative” work). To the extent the court was uncertain about the threshold question, the second question of transformativeness left no further questions. The evidence clearly showed that Google’s unauthorized use of code foreclosed markets—actual and potential—to Oracle. As such, the fourth factor weighted heavily against Google’s secondary use. And while the court acted as if it considered the third factor—the amount used—independently, the dicta of the decision suggests the court was, in fact, subsuming the third factor into the analysis of the fourth factor, which economics suggests is the most coherent approach.
Many legal experts expect Google to seek certiorari from the Supreme Court. If the Court accepts the case, it will represent the first major look at fair use by the Supreme Court in almost 25 years. Assuming the Court grants certiorari, the Court will evaluate fair use with fresh attention to the four factors and Section 107’s preamble. Both the TVEyes and Google rulings represent a critical course-correction on fair use, returning the fair use doctrine to its statutory confines. If the Supreme Court upholds the Federal Circuit’s decision in Google, then future fair use cases will require a more careful analysis of harm to potential markets and an evaluation of whether the secondary use comports with the class of uses listed in statute.
* * *
Dr. George S. Ford is Chief Economist of the Phoenix Center for Advanced Legal & Economic Public Policy Studies (http://www.phoenix-center.org), a non-profit 501(c)(3) research organization that studies broad public-policy issues related to governance, social and economic conditions, with a particular emphasis on the law and economics of the digital age.