Kal Raustiala and Christopher Sprigman do a wonderful job alerting us to a phenomenon they describe as “copyright’s negative space.” Given the continuous clamor for ever-broader intellectual property protection, identification of this space teaches lawmakers an important lesson: intellectual property rights may not always be needed to promote robust levels of innovation. In the appropriate environment, competition can be quite a satisfactory way to spur creativity.
The trick lies in identifying the factors that produce the appropriate environment. As the authors tell it, negative space works for the fashion industry because the goods send status (positional) signals, these signals decay over time (obsolescence), and both the status and the decay are, in a sense, caused by the free-wheeling copying made possible by the absence of intellectual property protection (anchoring). At first blush, these appear to be rather unique conditions, raising the question whether there is less here than meets the eye. The authors suggest several other endeavors where a similar dynamic may hold (cuisine, furniture design, hairstyles, and perfume). But with all due respect to fashionistas, these fields are not of compelling social value.
On reflection, however, the factors Raustiala and Sprigman list may have more general application. Anchoring may be more akin to a network effect than the authors suggest: the value of the product depends on others having goods that look the same. While it is true (as they say) that this effect decreases as a particular style becomes obsolete, that just means there is no lock in; during the time the style is “in,” the goods increase in value the more popular they become. Further, the value can be easily extracted—first, by selling the fashion-setting items and second, through the sale of ancillary products—the shoes, bags, scarves, belts, blouses, jackets, and coats that complete the “look.” And since the initial entrant (or the conglomerate that owns it) is better positioned to quickly offer—or contract with others to quickly offer—these products, there are likely stronger first-mover advantages than might be supposed. Even after a style is passé, it may be the source of enduring and transferable reputational gains.
Indeed, this generalization can explain the low-IP equlibria in many of the authors’ examples: semiconductors and software experience network effects; database compilers enjoy enduring first-mover advantages.1 Open source software development is partly supported by companies like (the aptly named) Red Hat that provide ancillary services; individual programmers derive reputational benefits in the open source community that may be lacking in their day jobs, where credit is suppressed because their work is considered “for hire.”2 The authors also note that relationships in the fashion industry are cemented by trademark protection; tangential reliance on IP is replicated in other “innovation communities,” including semiconductors,3 databases,4 and software.5
Nor do Raustiala and Sprigman claim to have exhausted the list of equilibrium-supporting factors. In their investigation of cuisine, Emmanuelle Fauchart and Eric von Hippel posit three strong social norms underlying negative space in that arena—bars on exact copying and leaking, and a crediting requirement, all of which are enforced by community-accepted sanctions.6 A similar dynamic is likely operative in fashion. Seasonal shows may, for example, function as precommitment strategies: once a designer sends one set of styles down a runway, poaching a rival’s designs cannot be accomplished without detection. Viewed from this broader perspective, it is clear that there are many other endeavors functioning in negative space. In particular, academic science has traditionally operated with similar norms: peer review requires researchers to share papers, experimental results, and even tangible products. They understand that these materials are not to be distributed broadly, plagiarized, or used without giving proper credit. As in both cuisine and fashion, referencing is valued because it helps to establish priority.7 Similar practices are found in academia more generally and in various creative arts communities as well.
To me, then, the intriguing question is not why low-IP equilibria occur, but what makes them deteriorate. History reveals many examples of innovation communities that have moved from IP’s negative to positive space: the aforementioned university scientists, plant and animal breeders, business entities, and tax attorneys, all of whom are increasingly reliant on patent protection; architects, performers, and choreographers, who now tend to claim copyrights (or neighboring rights) in their work. Even the perfume industry (one of the authors’ examples) may be in the process of making the switch.8 A better grasp of the dynamics of migration from negative to positive space can, perhaps, identify ways to stabilize these equilibria.
Of course, one possibility is that not all negative spaces are socially desirable; those that are not may be destroyed by outside pressures. Thus, while Raustiala and Sprigman carefully refrain from making normative assessments, they do note that there may be supraoptimal turnover in fashion design. The interests of an innovation community may be misaligned with the interests of the public in other ways as well. The incentives created in low-IP regimes may be inadequate, leading to underproduction of goods that society values. In addition, there may be structural problems. Consider plant and animal breeding. Although farmers now complain that patenting has raised their costs, entry was far from free during the low-IP equilibrium era. According to Dan Kevles, before plant and animal patents, breeders maintained control over stocks through sharp limits on distribution, highly restrictive contracts, and rigorous monitoring programs. Animals were sometimes infected before they were sold to butchers so that they could not be used for breeding.9 These practices constrained husbandry decisions, endangered public safety, and led to severe in-breeding. Patenting may be raising farmers’ prices, but at least it gives them a measure of autonomy. Further, the patent system may better align private incentives with social interests. Ancillary-service support can similarly generate negative externalities. The contracts imposed by organizations like Red Hat may be as restrictive as intellectual property rights and they can endure for longer. More generally, because patents facilitate financing and licensing activities, they can be particularly valuable to new entrants, as empirical research demonstrates.10
Destructive pressures can also arise from within, for the norms undergirding negative space may not benefit all the individuals operating within it. In science, for example, those who are out of the “loop” of information interchange cannot easily stay at the intellectual frontier. In particular, research and teaching assistants may be excluded from the reputational pay-offs the low-IP regime generates; arguably, they would do better with rules that award explicit intellectual property rights to the contributions they make.11 Similarly, cases like Thomson v. Larson12 demonstrate how those in the creative arts —in that case, a dramaturg—can fall into the cracks between intellectual property law and the norms of their profession, leading them to make claims for novel forms of IP protection.13 Indeed, the movement here may represent a modern instantiation of a longstanding phenomenon: even authors and inventors once operated in negative space. The shift to a high-IP equilibrium was accompanied (or, perhaps effectuated) by an elevation in (or to some, a romanticization of) their status.14
That said, there are certainly low-IP equilibria that suffer from none of these problems: they provide social benefits, but are nonetheless imperiled. Further work needs to be done on why this is so; to begin the inquiry, I suggest five factors that may contribute to breakdown: bounded altruism, tipping, herding, failed leadership, and technological change.
a. Bounded altruism: Commentators focusing on negative space tend to doubt that altruism plays an important role in stabilizing it. Yet insiders regularly suggest that the impulse to share plays a key role in their participation. More important, descriptions of how these communities dissolve tend to feature accounts of participants defecting when they discover their work product was exploited commercially.15 Indeed, the Larson case displays exactly this dynamic: Lynn Thomson contributed to the revision of “Rent” without raising intellectual property claims—right up until the point where the play started earning enormous sums of money for the Larson estate.16
It could, of course, be argued that anyone giving her work away has no standing to complain about others’ commercialization. In fact, however, altruism appears to be bounded: it does not extend to allowing others to profit monetarily from one’s labor. Here, then, is an area where behavioral science might make an important contribution. A better understanding of the limits of altruism could illuminate the ways in which goods can be fully utilized without jeopardizing the negative space in which they are produced.
b. Tipping: Raustiala and Sprigman’s reference to the YSL/Ralph Lauren lawsuit involving point-by-point copying is reminiscent of the classic account of the INS v. AP case,17 which led the Supreme Court to create the tort of misappropriation. The story is that the negative space in which newspapers operated was supported by a norm allowing morning papers to appropriate stories from the editions sold the previous evening; similarly, evening papers could appropriate from the morning edition. The practice saved money and did no real damage because morning and evening papers were not sold to the same market. The dispute arose when INS took stories from papers on the east coast, wired them to the west, where they appeared in the same (competing) edition. The rest is history: the misappropriation tort created a new form of intellectual property and destroyed a successful low-IP equilibrium.18
The lesson here may be the obvious one: a community must be vigilant about upholding its norms. Further, the time to be vigilant is when a new money-making potential becomes evident. For example, academic patenting soared when the work of the academy began to have commercial value.19 Indeed, scientists who have collaborated with someone who holds patents are more likely than their peers to depart from the classic norms of science and be patentees themselves.20
In some instances, tipping may be caused by a collective action problem. That is, once some members of a community acquire formal intellectual property rights, an arms race ensues: every member feels the need to build a portfolio, to be used as counterclaims or for settlement in the event of an infringement suit. An example is the avalanche of applications for business method and software patents that were filed, largely for defensive purposes, after the Federal Circuit’s decision in State Street Bank & Trust Co. v. Signature Financial Group, Inc.21 Unfortunately, once this form of protection takes hold, the norms supporting the low-IP regime are likely to be replaced by lawyerly admonitions about such things as keeping information secret.
c. Herding The INS story also teaches another lesson: even a single lawsuit can be destructive. The reason is herding: there are significant first-mover disadvantages to developing a new theory of law, but once a theory is a winner, others may be quick to follow. Once again, lawyers may be playing a key role. Thus, § 43(a) of the Lanham was not utilized until 1954.22 Since then, there has been an explosion in claims.23 Similarly, State Street’s recognition of business method patents appears to have created a fad among tax attorneys, who are busily filing applications on tax-avoidance schemes.24 An analogous phenomenon is occurring in Europe, where Anheuser-Busch v. Portugal25 has generated a rash of intellectual property claims styled as human rights violations.26
The upshot is this: those of us who worry about the corrosive potential of a case like Madey v. Duke University,27 which diluted (or, perhaps, rejected) an experimental use defense to patent infringement for academic science, may not be far off the mark. Now that the hard work of destroying that defense has been accomplished, there is a strong possibility that other claims against academic institutions will follow. If they do, then the Federal Circuit’s view of universities as commercial players may seep into the academic culture and further endanger the norms in science.
d. Failed leadership: In their study of the open source movement, Josh Lerner and Jean Tirole note the importance of leadership.28 Good managers set the community’s vision, certify the quality of its work, and in projects where interoperability is crucial, prevent forking. Further, strong leadership can provide credible commitments to openness, thus abating the problems caused by the bounded nature of altruism.
To be sure, not every negative space requires this kind of control—it is not, for example, apparent in the fashion industry. But where Lerner and Tirole’s observations do apply, stabilizing a negative equilibrium would appear to require innovation communities to prepare for the loss if their founders, either by grooming successors or by organizing in a manner that promotes effective joint governance.
e. Technological change: INS can be viewed in yet another way. According to Doug Baird, the problem there was not so much a breakdown in norms as a change in the competitive environment. At one time, advantages in news reporting were largely attributable to ownership of large networks of leased telephone lines. Because the cost of maintaining these networks was so high, ownership of the distribution channels was enough to maintain a competitive edge. But after technology reduced distribution costs, that advantage disappeared and the industry shifted its attention to the ownership of content. According to Baird, the INS case was, basically, constructed in order to create a property right in news.29
Raustiala and Sprigman appear to discount this effect, arguing that although technological change has occurred, it has not endangered the fashion industry’s negative space. Their argument cannot, however, be generalized to all low-IP equilibria. First, it is not clear that technology has changed enough to affect the fashion industry. Designs may now be copied quickly, but they are not copied well—knock-offs are cheaper because they use inferior materials and less labor. Accordingly, they are unlikely to appeal to a designer’s own customers or diminish the potential for return (indeed, the accessories completing the “look” may be priced low enough to be purchased even by those who buy knock-offs of the principal items). If new innovations were to reduce the cost of manufacturing upscale fabric and adding intricate detail, we might see a rather different story. Second, even if technology does not endanger every innovation community, it may increase the need for explicit intellectual property protection in some fields. Indeed, it is clear that the Statute of Anne’s introduction of a copyright regime was caused, at least in part, by the invention of the printing press.30
Of course, even if technology requires new IP protection, there is a separate issue of proportionality. For example, the lobbying that led to the enactment of Digital Millennium Copyright Act arguably overstated the dangers of digitization, leading to an excessive level of protection.31
Raustiala and Sprigman have made a significant contribution by drawing attention to IP’s negative space. A comprehensive study of this space can illuminate a variety of issues, answering questions about the value of low-IP equilibria and the conditions threatening their stability. Such studies would, hopefully, also identify tools for preventing their deterioration. Most important, they can tell us when it is time to worry. Debates over reform often drift into discussions about who bears the burden of proving the need for it. If further research shows that my intuition is right and that these equilibria are valuable and fragile, perhaps single cases—like Madey v. Duke University—should be taken more seriously.
Preferred citation: Rochelle Dreyfuss, Fragile Equilibria, Va. L. Rev. In Brief (2007), http://.
[Back to this issue]