In 1996, a database treaty that the European Commission had put forward, in connection with the WIPO negotiations on transmissions in cyberspace, ultimately failed to win the support of other regional groups [4]. Since then, the inability of the United States Congress to enact any form of database legislation has stymied further multilateral undertakings on this topic. This impasse may soon be broken, however, owing to the change of Administrations and to the appointment of new committee chairmen in the United States House of Representatives.

This article will discuss the prospects for an international regulatory framework for non copyrightable databases in the light of recent developments in the United States. Part 2 will locate the database problem within the larger context of international intellectual property protection, and it will demonstrate why the European Commission’s 1996 Directive on the legal protection of databases [5] represented a radical departure from basic tenets of the classical intellectual property system handed down from the nineteenth century. Part 3 will compare the existing E.U. model of database protection with the two proposed models currently under consideration in the United States, from which any compromise formula is likely to be drawn. It ends with some reflections on the deeper legal and economic implications of these proposals.

Part 4 will then explore the implications for the international intellectual property system likely to arise if the U.S. adopts a model of database protection that differs significantly from that of the E.U. It proposes an umbrella treaty to bridge the gap between high and low protectionist models. While a low protectionist outcome in the United States is by no means certain at the time of writing, a careful consideration of ways and means to reduce friction between countries that opt to provide different levels of protection in the global marketplace seems merited at the present juncture.

This solution attempts roughly to reconcile the needs of those who invest in publishing compilations of information with the needs of second comers to access and use that same information and to recombine it in follow-on information goods. The copyright approach thus strikes a balance between incentives to invest and free competition that tends to err on the side of second comers. In effect, by severely limiting the first comer’s derivative work rights, copyright law operates as a roving unfair competition law that protects investors merely against wholesale duplication of their information goods. In the United States, these limitations are thought to have Constitutional underpinnings, in keeping with First Amendment rights to freedom of speech and with the role of a robust public domain in sustaining democratic discourse [11].

This make-weight solution, however, always exposed commercial compilers of information to a risk of market failure whenever the end product consisted essentially of unprotectible data that a second comer could duplicate and rearrange at little cost in time or money. In such cases, the first comer prospects the market at his own risk, but finding himself exposed to instant copying, enjoys no period of natural lead time in which to appropriate the fruits of his investment or to recoup prior losses on unsuccessful essays. The second comer who free-rides on the information previously compiled may price the information good below the first comer’s marginal costs and drive him out of the market [12].

When “facts are piled on facts” [13], in other words, there is a classic public good problem with real risks of market failure [14]. That condition characteristically ushers in a tension between states of chronic overprotection, in which courts reinforce the compilers’ incentives to invest, and states of chronic underprotection in which users and second comers tend to prevail. Between these extremes, there lie unanswered questions about the economics of information goods in general and of the database industry in particular. This uncertainty then hampers the quest for legal solutions that could avoid suboptimal investment without impeding follow-on innovation and without impoverishing the public domain [15].

What changed in the 1990’s was the convergence of digital and telecommunications networks, which potentiated the role of electronic databases in the information economy generally, and which made scientific databases in particular into an agent of technological innovation whose economic potential may one day outstrip that accruing from the patent system [19]. The emergence of digitally networked environments “has generated a host of new value-added services and products, and appreciably increased the importance of this segment of the database market” [20]. In a previous article, Professor Samuelson and I emphasized two reasons why digital technology would cause the market for value-added data-based products to flourish in the near future. First, “digital technologies facilitate the disaggregation of value-added functions” and permit new forms of data aggregation and presentation that were unavailable in print media. Second, “digital technologies foster new functions, such as reformatting, filtering, and hot-linking, which have no counterparts in print media” [21]. These predictions have held up over time, and there is no doubt that the database industry as a whole, and its value-added components, have in fact flourished despite constant allegations of market failure [22].

Notwithstanding the robust appearance of the present-day database industry under free market conditions, it was logical to ask whether suboptimal investment in complex electronic databases would not inevitably hinder that industry’s long-term growth prospects if free-riding second comers could rapidly appropriate the contents of every successful new product without contributing to the costs of development and maintenance over time. If, in other words, there existed a gap in the law, which neither copyright nor residual unfair competition regimes adequately filled, then suitable regulatory action to enhance investment might produce positive social benefits [23]. At the same time, this utilitarian rationale raised new and delicate questions about the prospects for high and unintended social costs likely to ensue if intellectual property rights were injudiciously bestowed upon the building blocks of knowledge in general or on the raw material of the information economy in particular [24].

These uncertainties, in turn, raised a number of ancillary questions that required serious theoretical and empirical investigation. First among these was the extent to which any hypothetical impediments to investment were not being overcome by the database entrepreneur’s eagerness and willingness to invest anyway, in order to capture real world economic opportunities [25], and by the availability of self-help technical measures which, in combination with residual legal measures, sufficed to neutralize the risk of market failure. For example, electronic fencing through encryption devices, coupled with tagging or watermarking of the data, make it possible for online database providers to impose standardized contractual restrictions on all would-be users [26]. This restored power of the “two-party deal” in the networked environment [27] is so great, indeed, that critics fear it requires regulation lest online database providers abusively alter the pre-existing balance between public and private interests that the copyright law had previously sought to establish [28]. At the very least, these critics argue that, given the power of self-help remedies in the digital environment, contract and unfair competition law would suffice to close any regulatory gaps that were likely to ensue in the short or medium term, without further encumbering access to the public domain [29].

If the lack of any trustworthy empirical assessments of market failure under existing conditions thus made it difficult to ascertain the true need for regulatory action, the enquiry was further complicated by persistent anecdotal allegations that the database industry was dominated by sole-source providers who control niche markets and who seek to impede access by value-adding competitors [30]. The Feist decision in the United States, and many recent decisions under the E.U.’s database legislation [31], which is discussed below, fit this scenario. Given the opportunities for value-adding uses that digital technology makes available, any appropriate regulatory scheme should, in principle, seek to stimulate pro-competitive conditions that lower barriers to entry and encourage follow-on applications. It should not reinforce monopolistic tendencies that seem to plague key segments of the database industry at the present time.

Another question of capital importance is the role that databases currently play in both basic and applied sciences. Pre-existing legal regimes have treated scientific data as a common resource available from the public domain, and the ethos of science has been premised on a commitment to the free and open exchange of data to support scientific hypotheses and published findings [32]. The traditional and customary practice is accordingly for scientists to recombine data from existing databases into new databases to be used as electronic information tools to solve hard new problems. Any proposed regulatory solution must take these practices into account and avoid disrupting the worldwide scientific networks that depend on the sharing of essential data [33].

From a related but still broader perspective, vast quantities of technical information have always been freely available from the public domain, as a basic input of the knowledge economy, where the technology-exporting countries’ comparative advantages are most deeply rooted. In other words, information is both an input and an output of the information economy [34]. This economy has grown to its present magnitude under conditions in which entrepreneurs can only obtain exclusive rights in downstream aggregates of information that rise to the level of patentable inventions or copyrightable works of authorship.

Upstream flows of information have, instead, usually been subject to liability rules [35], such as trade secret laws and unfair competition laws sounding in the misappropriation rationale, rather than exclusive property rights [36]. These liability rules tend to repress market-destructive conduct without removing technical information from the public domain. Any regulatory action must thus logically take into account the unintended consequences that might flow from suddenly impeding innovators and inventors from access to upstream information that has, until now, been freely available as inputs into technological development.

These considerations about goals, in turn, should logically focus attention on the choice of legal instruments to remedy any market failure that survived rigorous empirical investigation. In recent years, for example, efforts to protect investors in small-scale applications of know-how to industry from free-riding duplicators have led either to broadening distortions of the patent and copyright paradigms or to a proliferation of hybrid regimes of exclusive property rights, loosely based on obsolete design protection and utility model regimes. These tendencies have caused the patent and copyright systems to evolve in dangerously overprotective directions while encumbering free competition with an array of miniature property rights that are cumulatively producing anti-competitive and anti-commons effects. My recent studies demonstrate the need for a new kind of intellectual property regime, based on liability rules rather than exclusive property rights, that could avoid market failure without creating barriers to entry and without impoverishing the public domain [37]. Such a regime, which I call a compensatory liability regime, solves the one key problem that all the hybrid regimes of exclusive property rights have so far failed to solve, namely that of deterring free riders without impeding follow-on innovation [38]. Any serious quest for an appropriate sui generis solution to the question of database protection would accordingly have engendered a serious investigation of the comparative economic advantages and disadvantages of regimes based on exclusive property rights as distinct from regimes sounding in liability rules. By the same token, this investigation would also have to factor in larger constitutional questions about the varying impacts of different legal regimes on freedom of speech and on the conditions of democratic discourse. For example, the Constitutional foundations of United States copyright law have always rested on a clear and sharp distinction between facts and ideas that were freely available to all and the author’s expression of facts and ideas, which could not be copied. Allowing exclusive property rights to cover aggregates of data and information which had been previously unprotectible must sooner or later pose fundamental Constitutional questions for countries that take freedom of speech seriously, questions that a creative use of liability rules might altogether avoid [39].

All of these questions, taken together, suggested the need for long and careful study and an extremely cautious approach to resolving a database protection “problem” whose very existence had yet to be clearly demonstrated. Instead, the European Commission, proceeding without answers to any of these questions, cut the enquiry short by bestowing the strongest intellectual property right ever conceived on publishers who contributed nothing more to the public good than “sweat of the brow” investments in noncopyrightable compilations of facts and information.

These memorials make little mention of the economic and political forces lobbying for protection behind the scenes or the countervailing interests that were likely to suffer according to the different solutions under consideration. We know only that the calls for protection to which the Commission was listening had grown ever stronger over time [41]; that a primary goal of the lobbyists operating behind the scenes was precisely to control follow-on applications of noncopyrightable data bases for the longest possible time; and that efforts by many of those who worked on the Directive – especially the German copyright officials – to include a compulsory licensing scheme as a quid pro quo for a strong property right were thwarted at the last moment by back door maneuvers that have never satisfactorily been explained [42]. The sui generis regime that the Commission ultimately adopted is like nothing we have ever seen before. It protects any collection of data, information or other materials that are arranged in a systematic or methodological way, provided that they are individually accessible by electronic or other means. This does not, however, imply that some organized form of storage is needed [43]. The criterion of eligibility is a “substantial investment”, as measured in either qualitative or quantitative terms, and the courts are left to develop this concept. That the drafters believed a relatively minimal level of investment would suffice appears from an explicit recognition that the qualifying investment may consist simply of verifying or maintaining the database [44].

In return for this investment, the compiler obtains exclusive rights to extract or to utilize all or a substantial part of the contents of the protected database. The exclusive extraction right pertains to any transfer in any form of all or a substantial part of the contents of a protected database; the exclusive reutilization right covers only the making available to the public of all or a substantial part of the same database [45]. In every case, the first comer obtains a powerful adaptation (or derivative work) right along the lines that copyright law bestows on “original works of authorship” [46], even though such a right is alien to the protection of investment under existing unfair competition laws.

The Directive provides no major public interest exceptions comparable to those recognized under domestic and international copyright laws. An optional but ambiguous exception concerning illustrations for teaching or scientific research is said to be open to flexible interpretation [47], and some member countries have implemented it in different ways. However, other countries have simply ignored this exception altogether, which contradicts the Commission’s supposed concerns about uniform law [48].

The Directive’s sui generis regime does exempt from liability anyone who extracts or uses an insubstantial part of a protected database. However, such a user bears the risk of accurately drawing the line between a substantial and an insubstantial part, and any repeated or systematic use of even an insubstantial part will forfeit this exemption [49].

Qualifying databases are nominally protected for a fifteen year period. In reality, each new investment in a protected database, such as the provision of updates, will re-qualify that entire database as a whole for a new term of protection. In this and other respects, the sui generis adaptation right is far more powerful than that of copyright law, which attaches only to the new matter added to an underlying, preexisting work [50]. As noted at the outset, the E.U. Database Directive thus breaks with the entire history of intellectual property law by allowing a property rule – as distinct from a liability rule – to last in perpetuity [51].

Finally, the Directive carries no national treatment requirement into its sui generis component. Foreign database producers become eligible only if their countries of origin provide a similar form of protection or if, in keeping with a goal attributed to the Commission, they set up operations within the E.U. [52]. However, non-qualifying foreign producers may continue to invoke the residual domestic copyright and unfair competition laws, where available, and the cases so far arising under the various member’ implementing statutes suggest that both regimes may often remain available to foreign parties [53].

Without going further into detail, it suffices to point out that the new regime embodied in the E.C.’s Directive on the legal protection of databases, adopted in 1996 [54], broke radically with the historical limits of intellectual property protection in at least three ways:

In this and other respects, the E.U. model abolished the concept of a public domain that had historically justified the grant of temporary exclusive rights in intangible creations from the start.

The Directive on Databases then took the further step of denying foreign producers protection unless their countries of origin had enacted comparable legislation that met a standard of material reciprocity, notwithstanding the drive for national treatment of intellectual property rights within the framework of the TRIPS Agreement [56]. Needless to say, this requirement of material reciprocity has only intensified the debate in the United States and in the rest of the world about the proper level of protection for noncopyrightable collections of data, which has impeded further consideration of a harmonizing international treaty in the intervening years.

The opponents’ ranks have swelled to include online service providers and certain telecommunication companies; online stockbrokers; major information technology companies, including dissident database publishers who license considerable amounts of data from others; the United States Chamber of Commerce; and an increasing number of diverse but powerful groups who fear rising costs for accessing data in the future. Also active in this coalition are the representatives of libraries, universities, and major scientific organizations. This “opponents lobby” prefers either no database regime or a soi-disant minimalist regime sounding in unfair competition law. It has found its legislative “champion” in the Chairman of the Commerce Committee of the House of Representatives, who has taken a public stand against enacting a strong exclusive property right to protect noncopyrightable databases while expressing favorable views about an unfair competition approach [57]. In July of 1998, these groups participated in negotiations among stakeholders that were held under the auspices of the Chairman of the Senate Committee on the Judiciary [58]. By that time, the opponents had also persuaded the Clinton Administration not to support efforts to launch an international treaty regulating databases, under the auspices of WIPO, pending a Congressional decision concerning the proper course of action. Once these negotiations failed, the opponents continued to block adoption by the Senate of bills emanating from the House of Representatives’ Judiciary Committee. In 1996, this Committee’s proposal had taken the express form of a sui generis intellectual property regime. From 1997 onward – as will be discussed below – the House Judiciary Committee adopted the tactic of couching a database right in “misappropriation” terminology, even though the regime it proposed to enact was as strong or stronger than that adopted in the European Union [59].

In August 1998, the Clinton Administration, which had been internally divided on database protection since the end of the 1996 WIPO Diplomatic Conference, issued a set of principles to govern the adoption of any database protection legislation. Jointly drafted by the Patent and Trademark Office and the Office of Science and Technology Policy, these principles represented a moderate compromise among the various agencies and took the form of a letter to Senate Judiciary Chairman Orrin Hatch and the Ranking Member Patrick Leahy [60] (making it clear that the Administration was prepared to bypass the House Judiciary Committee, if necessary). Over time, the Clinton Administration issued comprehensive, detailed analyses of the database protection bills in the House [61], it exerted a considerable restraining effect on the more extreme proposals, and in principle, its policies remain in effect, although the position of the Bush Administration had yet to be determined at the time of writing [62].

Numerous public hearings over the years have generated considerable publicity concerning the various legislative initiatives, and these forums have helped to raise the level of controversy over time. At one point, editorials against overly strong database protection appeared in the New York Times and the Washington Post. Besides a series of Congressional Hearings sponsored and largely controlled by proponents lobbying for strong protection, other public forums or inquiries were held under the auspices of the Copyright Office, the Patent Office, and the National Research Council (NRC). Major studies conducted by the NRC on this issue were published in 1997 [63] and, after extensive public debate and testimony, again in 1999 [64]. Most recently, in the Spring of 2001, the U.S. State Department and the National Research Council jointly sponsored a full-day’s Roundtable Discussion of pending legislative initiatives, which drew an impressive crowd of stakeholders and interested parties [65].

The proposals for database legislation have spawned a vigorous and ever growing academic literature [66]. Several economic studies have also appeared, which predictably reach different conclusions depending on who commissioned them [67]. There is also a spate of hard-hitting articles by respected Constitutional scholars, who by different routes have raised serious doubts about the authority of Congress to enact a sui generis exclusive property right at all and about the consistency of any such regime with First Amendment prescriptions on freedom of speech [68]. This literature reinforces earlier reservations expressed by the Department of Justice [69], and it raises questions about the ability of any high-protectionist regime to survive Constitutional scrutiny, even if enacted.

Meanwhile, the database bills have become ever more controversial as more sectors of industry discover that they, too, will be affected by the outcome, and the ranks of stakeholders have continued to grow, both in terms of numbers and political clout. Every new legislative initiative is, accordingly, subject to intense public scrutiny, to loud and often acrimonious debate, and to considerable academic analysis. However, none of this activity ensures that the U.S. will adopt a better database law than that enacted in the E.U., nor does it guarantee that the U.S. model ultimately enacted will differ in more than marginal ways from its European predecessor. It does mean that very little will escape public scrutiny under the pressures for transparency and domestic debate that have become too strong to resist. Against this background, the House Committee Chairmen mentioned above have recently pledged to reach some compromise solution during the present legislative session, and they have summoned the contending coalitions to participate in ongoing and relentless rounds of negotiations to this end. The outcome of these negotiations remains uncertain at the time of writing. It seems clear nonetheless that any viable database bill will be drawn from the two basic proposals that were still on the table at the end of the last legislative session, which ended in an impasse.

These proposals, as refined during that session, represent the baseline positions that each coalition carried into the current round of negotiations. One bill, H.R. 354 (as revised in January, 2000) [70], embodies the proponents’ last set of formal proposals for a sui generis regime built on an exclusive property rights model (although some effort has been made to conceal that solution behind a title that evokes unfair competition law) [71]. The other bill, H.R. 1858, sets out the opponents’ views of a socalled minimalist misappropriation regime as it stood on the eve of the current round of negotiations [72].

The bill in this form continues to define “collections of information” very broadly as “information… collected and… organized for the purpose of bringing discrete items of information together in one place or through one source so that persons may access them” [74]. Here the overlap with copyright law is so palpable that one can hardly conceive of any assemblage of words, numbers, facts or information that would not also qualify as a potentially protectible collection of information.

Like the E.C. Directive, this bill casts eligibility in terms of an “investment of substantial monetary or other resources” in the gathering, organizing or maintaining of a “collection of information” [75]. It then confers two exclusive rights on the investor, viz., a right to make all or a substantial part of a protected collection “available to others” and a right “to extract all or a substantial part to make available to others”. Here the term “others” is manifestly broader than “public” in ways that remain to be clarified [76].

H.R. 354 then superimposes an additional criterion of liability on both exclusive rights that is not present in the E.U. model. This is the requirement that, to trigger liability for infringement, any unauthorized act of “making available to others” or of “extraction” for that purpose must cause “material harm to the market” of the qualifying investor “for a product or service that incorporates that collection of information and is offered or intended to be offered in commerce.” The crux of liability under the bill thus derives from a “material harm to markets” test that is meant to cloud the copyright-like nature of the bill and to shroud it in different terminology [77].

Here a number of concessions were made to the opponents’ concerns in the last iteration of the bill (Jan. 11, 2000), some of them real, others nominal in effect. The addition of “material” to the market harm test [78], may, for example, address complaints that proponents viewed “one lost sale” as constituting actionable harm to the market. How much more trenchant a “material harm” test really is remains to be seen.

At the same time, the revised bill contains convoluted and tortuous definitions of “market” that the Administration hoped would reduce the scope of protection in the case of follow-on applications [79]. On closer inspection, however, these definitions provide a static picture of a moving target that amounts to a mostly illusory limitation on the investor’s broad adaptation right [80]. In other words, notwithstanding these socalled concessions, the bill effectively assigns most follow-on applications to any initial investor whose dynamic operations expand the range of potentially protectible matter with every up date, ad infinitum.

The bill then introduces a “reasonable use” exception that would presumably benefit the nonprofit user communities, especially researchers and libraries [81], and that is meant to convey a sense of similarity with the “fair use exception” in copyright law [82]. Once again, this resemblance turns out to become largely illusory on closer analysis, because under the proposed bill, the very facts, data and information that copyright law excludes have become the objects of protection, and there are no other significant exceptions. Hence, virtually every customary or traditional use of facts or information compiled by others that copyright law would presumably have allowed scientists, researchers, or other nonprofit entities to make in the past now become prima facie instances of infringement under H.R. 354. These users would in effect have either to license such uses or be prepared to seek judicial relief for “reasonableness” on a continuing basis. Because universities dislike litigation and are risk averse by nature, and this provision puts the burden of showing reasonableness on them, there is reason to expect a chilling effect on customary uses of data by these institutions [83].

The bill then recognizes an “independent creation” norm, which presumably exempts any database, however similar to an existing database, that was not the fruit of “copying” [84]. This provision codifies a fundamental norm of copyright law, and the European Commission made much of a similar norm in justifying its own regulatory scheme. In reality, this “independent creation” principle produces unintended and socially deleterious consequences when transposed to the database milieu precisely because the most complex and important databases become ever less susceptible of independent regeneration as their value grows over time.

Sometimes the database cannot be reconstituted because the underlying phenomena are one-time events, as often occurs in the sciences [85]. At other times, key components of a complex database will have gone lost or missing, and they can no longer be reconstituted with certainty at a later date. Any independently regenerated database suffering from these defects would necessarily contain gaps that made it inherently less reliable than its predecessors.

These problems point to a more general phenomenon that affects competition in complex databases generally. Even when, in principle, such databases could be reconstituted from scratch, the high costs of doing so – as compared with the add-on costs of existing producers – will tend to make the second comer’s costs so high as to constitute a barrier to entry. Meanwhile, the first comer’s comparative advantage from already owning a large collection that is too costly to reconstitute will only grow more formidable over time, an economic reality that progressively strengthens the barriers to entry and tends to reinforce (and, indeed, to explain) the predominance of sole-source data suppliers in the marketplace [86].

As more and more segments of industry come to appreciate the market power that major database producers could thus acquire under the proposed legislation, one after another has petitioned the subcommittee for special relief. Thus, the bill, which has now grown to some thirty pages in length, singles out various special interests who benefit, to varying degrees, from special exemptions from liability. At the time of writing the list of those entitled to such immunities included news reporting organizations; churches that depend on genealogical information, notably the Mormons; online service providers; and certain online stockbrokers [87].

Government-generated data remain excluded, in principle, from protection, in keeping with current U.S. practice [88], which differs from E.U. practice in this important respect. However, there is considerable controversy concerning the degree of protection to be afforded government-generated data that subsequently become embodied in value-adding, privately funded databases [89]. All parties agree that a private, value-adding compiler should obtain whatever degree of protection is elsewhere provided, notwithstanding the incorporation of government-generated data. The issue concerns the rights and abilities of third parties to continue to access the original, government-generated data sets, notwithstanding the existence of a commodified embodiment. At the time of writing, the proponents were little inclined to accept measures seeking to preserve access to the original data sets, but pressures in this direction were building [90].

H.R. 354 imposes no restrictions whatsoever on licensing agreements, including agreements that might overrule the few exceptions otherwise allowed by the bill [91]. Despite constant remonstrations from opponents about the need to regulate licensing in a variety of circumstances, and especially with respect to sole-source providers, the bill itself has not budged in this direction.

On the contrary, new provisions added to the last iteration of H.R. 354 in 2000 would set up measures that prohibit tampering with encryption devices (“anticircumvention measures”) and with electronically embedded or “watermarked” rights management information, in a manner that parallels the provisions adopted for online transmissions of copyrighted works under the Digital Millennium Copyright Act of 1998 [92]. Because these provisions effectively secure the database against unauthorized access (and tend to create an additional “exclusive right of access” without expressly so declaring) [93], they would only add to the database owner’s market power to dictate contractual terms and conditions without regard to the public interest. These powers are further magnified by the imposition of strong criminal sanctions in addition to strong civil remedies for infringement [94].

The one major concession that has so far been made to the opponents’ constitutional arguments concerns the question of duration. As previously noted, the E.C. Directive allows for perpetual protection of the whole database so long as any substantial part of it is updated or maintained by virtue of a new and substantial investment, and the proponents’ early proposals in the U.S. echoed this provision [95]. However, the U.S. Constitution clearly prescribes a limited term of duration for intellectual property rights [96], and the proponents have finally bowed to pressures from many directions by limiting the term of duration to fifteen years [97].

Any update to an existing database would then qualify for a new term of fifteen years, but this protection would apply, at least in principle, only to the new matter added in the update. In practice, however, the inability to clearly separate old from new matter in complex databases, coupled with ambiguous language concerning the scope of protection against harm to likely, expected, or planned market segments [98], may still leave some loophole for an indefinite term of duration.

H.R. 1858 begins with a definition of databases that is not appreciably narrower than that of H.R. 354, except for an express exclusion of traditional literary works that “tell a story, communicate a message”, and the like [100]. In other words, there is at least some attempt to draw a clearer line of demarcation between the proposed database regime and copyright law, in order to reduce overlap or cumulative protection as might occur under H.R. 354.

The operative protective language in H.R. 1858 appears short and direct, but it relies on a series of contingent definitions that muddy the true scope of protection. Thus, the bill would prohibit anyone from selling or distributing to the public a database that is 1) “a duplicate of another database… collected and organized by another person or entity” and 2) “is sold or distributed in commerce in competition with that other database” [101]. The bill then defines a prohibited duplicate as a database that is “substantially the same as such other database, as a result of the extraction of information from such other database” [102].

In other words, liability attaches only for a wholesale duplication of a preexisting database that results in a substantially identical end product. However, this basic misappropriation approach becomes further subject to both expansionist and limiting thrusts. Expanding the potential for liability is a proviso added to the definition of a protectible database that treats “any discrete sections [of a protected database] containing a large number of discrete items of information” as a separably identifiable database entitled to protection in its own right [103]. The bill would thus codify a surprisingly broad prohibition of follow-on applications that make use of discrete segments of pre-existing databases [104], subject to the limitations set out below.

A second protectionist thrust results from the lack of any duration clause whatsoever. In other words, the prohibition against wholesale duplication – subject to limitations set out below – could conceivably last forever. This perpetual threat of liability would attach to wholesale duplication of even a discrete segment of a preexisting database, if the other criteria for liability were also met. However, these powerfully protective provisions, put into H.R. 1858 at an early stage in order to weaken support for H.R. 354, are offset to some degree by other express limitations on liability and by a codified set of misuse standards to help regulate licensing.

To understand these further limitations, one should recall that liability even for wholesale duplication of all or a discrete segment of a protected database does not attach unless the unauthorized copy is sold or distributed in commerce and “in competition with” the protected database [105]. The term “in competition with”, when used in connection with a sale or distribution to the public, is then defined to mean that the unauthorized duplication “displaces substantial sales or licenses likely to accrue from the original database” and that it “significantly threatens… [the first comer’s] opportunity to recover a reasonable return on the investment” in the duplicated database [106]. Both prongs must be met before liability will attach.

It follows that even a wholesale duplication that was not commercially exploited or that did not substantially decrease expected revenues (as might occur from, say, nonprofit scientific research activities) could presumably escape liability in appropriate circumstances. Similarly, a follow-on commercial product that made use of data from a protected database might escape liability if it was sold in a distant market segment or required substantial independent investment.

H.R. 1858 then further reduces the potential scope of liability by imposing a set of well-defined exceptions and also by limiting enforcement to actions brought by the Federal Trade Commission (FTC). There are express exceptions for news reporting, law enforcement activities, intelligence agencies, online stockbrokers, and online service providers that are more or less comparable to those under H.R. 354 [107]. There is also an express exception for nonprofit scientific, educational, or research activities [108], just in case any such uses were thought to escape other definitions that condition liability on unauthorized uses in competition with the first comer. Still other provisions clarify that the protection of government-generated data or of legal materials in value-adding embodiments remains contingent upon arrangements that facilitate continued public access to the original data sets or materials [109]. A blanket exclusion of protection for “any individual idea, fact, procedure, system, method of operation, concept, principle or discovery” wisely attempts to provide a line of demarcation with patent law and to ward off unintended protectionist consequences in this direction [110].

The provision that conditions liability for infringement on an official FTC action [111] was a tactical expedient devised to provide the House Commerce Committee with some basis for asserting concurrent jurisdiction over database legislation, along with that of the Subcommittee on Courts and Intellectual Property. Most observers believe that the absence of any private right of action in H.R. 1858 as it stands constitutes a fatal flaw that would have to be removed in any final compromise decision to adopt an unfair competition approach. A vocal minority of supporters considers FTC supervision a necessary safeguard, especially in view of the First Amendment tensions that any database protection law is certain to generate in the United States.

A potentially more important set of safeguards emerges from the drafters’ real concerns about potential misuses of even this so-called minimalist form of protection. These concerns are expressed in a provision that expressly denies liability in any case where the protected party “misuses the protection” that H.R. 1858 affords. A related provision then elaborates a detailed list of standards that courts could use as guidelines in particular cases in order to determine whether an instance of misuse had occurred [112]. These guidelines or standards would greatly clarify the line between acceptable and unacceptable licensing conditions, and if enacted, they could make a handsome contribution to the doctrine of misuse as applied to the licensing of other intellectual property rights as well [113].

In summary, the underlying purpose of H.R. 1858 was to prohibit wholesale duplication of a database as a form of unfair competition. It thus set out to create a minimalist liability rule that prohibits market-destructive conduct rather than an exclusive property right as such [114], and in this sense, it initially posed a strong contrast to H.R. 354. Over time, however, different iterations of the bill, designed to win supporters away from H.R. 354, have made H.R. 1858 surprisingly protectionist, especially in view of its de facto derivative work right. The realities of the bargaining process are such that concessions unwisely made to the high protectionist camp at an earlier stage, for whatever tactical reasons, are unlikely to be withdrawn now.

The most fundamental question that these studies largely ignore is the extent to which any exclusive property right might a priori constitute the wrong kind of solution for a legal regime that aims to protect investment in compiling aggregates of data as such. One would, indeed, expect or prefer economic analysis to focus on the comparative advantages and disadvantages of using either exclusive property rights or liability rules [120] to address the underlying risks of market failure.

In this connection, a growing number of innovative proposals rooted in liability rules have been put on the table in recent years, in addition to the better known proposals for a more traditional unfair competition approach. For example, Wendy Gordon has proposed a tort of “malcompetitive copying” that would rest on specific economic criteria [121]. William Kingston has proposed a new type of liability regime that would transform intellectual property protection from a duration-based calculus of rights to an accounting-based calculus of rights premised on multiples of R&D costs [122]. I have elsewhere proposed a “compensatory liability” regime that would allow second comers freely to extract data from a protected database in order to compete with value-adding follow-on products, so long as adequate compensation was paid under an “automatic license” (not a compulsory license) for a specified period of time [123].

However, most economists engaged in this topic have so far ignored these and other proposals largely because their economic models and premises simply do not allow them to take liability rules into account. Others dogmatically castigate liability rules in the abstract and postulate their inherent inferiority to exclusive rights [124], without devoting any serious attention to the social costs that critics of strong database protection continue to fear.

As a result, formal economic analysis has so far taught us virtually nothing about how to craft a protective regime so as to avoid market failure without stifling competition and impoverishing the public domain. Small wonder that, amidst so much uncertainty, the most credible economic advice has been that of Scotchmer and Maurer, who advise against taking any premature action that might make the end result far worse than the predicament from which we started [125].

Disregarding copyright law, the ability of database producers to use self-help adhesion contracts and encryption devices to protect online delivery has greatly expanded [129]. However, such measures do not altogether close a gap in the law that opens when third parties not in privity of contract with the producer obtain access to the contents of the database [130]. It will not do either to exaggerate or to underestimate the extent of this risk.

That a codified, federal unfair competition law, sounding in the misappropriation rationale, could fill this gap remains a valid theory. It would constitute a minimalist response to a potential gap in the law whose true dimensions remain unknown, and it could also provide the uniform model needed for proper administration of the national system of innovation and for negotiating an international arrangement [131]. However, any legislative initiative in this regard risks being captured by special interests and converted into a high-protectionist exercise with serious unintended consequences.

The empirical evidence drawn from judicial application of the E.C.’s Directive on Databases so far sheds little light on the deeper issues. Most of the European cases have invoked copyright law, contracts law, or unfair competition law (especially the doctrine of parasitical copying) to reinforce or supplement conclusions reached under the sui generis database protection laws as such. Moreover, most of the extant European decisions deal with borderline subject matters under the old economy, such as telephone directories, television broadcast listings, and real estate listings, but not cutting-edge subject matter of the new economy, such as biotech databases [132]. The E.U. case law to date confirms the existence of all the hard problems that the literature has so far identified – the prevalence of sole-source providers; unreasonable restrictions on licensing; barriers to entry; and impediments to follow-on applications of data [133] – without a scintilla of evidence that the Directive has satisfactorily resolved any of these problems.

To see why critics in the United States fear the long-term consequences of the E.U.’s approach, it suffices to grasp how radical a change it would introduce into the domestic system of innovation and to consider how great the risks of such change really are. Traditionally, United States intellectual property law did not protect investment as such, a tradition that still has Constitutional underpinnings [136]. At the same time, the national system of innovation is premised on enormous flows of mostly government-generated or government-funded scientific and technical information (ST&I) upstream, which everyone is free to use [137], and on free competition with respect to downstream information goods.

The domestic intellectual property laws traditionally protect downstream bundles of information in two situations only: copyrightable works of art and literature, and patentable inventions. However, the following conditions apply:

As previously observed, the E.U.’s Database Directive changes this approach, as would the pending parallel proposal, H.R. 354, to enact strong database rights in the United States. Specifically, these sui generis regimes confer a strong and, in the E.U., potentially perpetual exclusive property right on the fruits of mere investment, without requiring any creative contribution. They also convert data and information – the previously unprotectible raw materials or basic inputs of the modern information economy – into the subject matter of this new exclusive property right.

The sui generis database regimes would thus effectuate a radical change in the economic nature and role of intellectual property rights (IPRs). Until now, the economic function of IPRs was to make markets possible where previously there existed a risk of market failure due to the public good nature of intangible creations. Exclusive rights make embodiments of intangible public goods artificially appropriable, they create markets for those embodiments, and they make it possible to exchange payment for access to these creations.

In contrast, an exclusive intellectual property right in the contents of databases breaks existing markets for downstream aggregates of information, which were formed around inputs of information largely available from the public domain. It conditions the very existence of all traditional markets for intellectual goods on:

This last constraint is perhaps the most telling of all. In effect, the sui generis database regimes create new and potentially serious barriers to entry to all existing markets for intellectual goods owing to the multiplicity of new owners of upstream information in whom they invest exclusive rights, any one of whom can hold out and all of whom can impose onerous transaction costs (analogous to the problem of multi-media transactions under copyright law). This tangle of rights is known as an anti-commons effect [139], and the database laws appear to be ideal generators of this phenomenon.

Under the new sui generis database regimes, in short, there is a built-in risk that too many owners of information inputs will impose too many costs and conditions on all the information processes we now take for granted in the information economy. At best, the costs of research and development activities seem likely to rise across the entire economy, well in excess of benefits, owing to the potential stranglehold of data suppliers on raw materials. This stranglehold will increase with market power if most databases are owned by sole-source providers. Over time, the comparative advantage from owning large, complex databases will tend progressively to elevate these barriers to entry [140].

The potential social gains of a strong database law cannot justify incurring these risks of disrupting or deforming the national system of innovation. It hardly seems logical to disrupt all existing markets for intellectual goods just to cure an alleged market failure for investments in a single type of intellectual good, i.e., noncopyrightable collections of information. At present, the U.S. dominates this market, and there is no credible empirical evidence of market failure that could not be cured by more traditional means [141].

The foregoing analysis reinforces the hypothesis that an exclusive property right is the wrong way to address the problem of legal protection for electronic databases, and it reconfirms the desirability of fashioning a modern liability rule that could avoid market failure without impoverishing the public domain. Supporters of strong database protection laws (and of strong contractual regimes to reinforce them [142]) believe that the benefits of private property rights are without limit, and that more is always better. They expect a brave new world in which these powerful legal incentives will attract huge resources into the production of electronic information tools [143].

In contrast, critics fear that an exclusive property right in noncopyrightable collections of data, coupled with the proprietors’ unlimited power to impose electronic adhesion contracts in the course of online delivery, will compromise the operations of existing systems of innovation, which depend on the free flow of upstream data and information. In place of the explosive production of new databases that proponents envision, opponents of a strong database right predict a steep rise in the costs of information across the global information economy and a progressive balkanization or feudalization of that economy, in which fewer knowledge goods may be produced as more tithes have to be paid to more and more information conglomerates along the way [144]. In the critics’ view, the information economy most likely to emerge from an exclusive property right in data will resemble models already familiar from the Middle Ages, when goods flowing down the Rhine River or goods moving from Milan to Genoa were subject to dozens, if not hundreds, of gatekeepers demanding tribute.

While the future of database protection in the United States has yet to be decided, it nonetheless seems clear that the present Administration will find itself compelled to support some form of international database protection no matter which regime Congress enacts in the end. In a worldwide digitally networked environment, the ability of free-riding duplicators to download commercially valuable databases in any territory that afforded them no protection whatsoever and to redistribute the contents online at very low prices to willing purchasers in the rest of the world would frustrate even a policy of soft protection for electronic databases (if it should ultimately prevail) [147]. In the new information economy, in other words, the very existence of an unregulated global market place puts purely territorial intellectual property policies at risk of extraterritorial subversion and fosters a compelling need for some form of transnational regulatory action.

How to meet this challenge without succumbing to high-protectionist demands for uniform intellectual property standards that could adversely affect economic growth in both developed and developing countries is a key challenge for international intellectual property relations in the post-TRIPS environment [148]. The question of database protection thus presents an opportunity to forge cooperative multilateral action in intellectual property law that advances the global public interest without the harmonizing excesses that have elicited intense criticism of the TRIPS Agreement and related high-protectionist undertakings [149].

In the rest of this article, I discuss both the risks of succumbing to a prematurely harmonized international regime of database protection and the need to avoid a “trade war” between high and low protectionists. I will then outline a proposal for a model umbrella treaty, concerning the international protection of databases, which is based on the solution earlier adopted in the Geneva Phonograms Convention of 1971 [150]. This proposal could enable all countries to cooperate in interdicting certain forms of market-destructive conduct without creating barriers to entry or otherwise disrupting their national systems of innovation.

On this scenario, a high-protectionist block installed at the heart of the international intellectual property system would enjoy significant advantages in developing any future blueprint for a multilateral regulatory solution. Only vigorous, persistent and entrenched opposition by countries opposed to a high-protectionist regime could then avoid an international framework modeled on joint E.U.-U.S. initiatives.

In theory, the developing countries could articulate an opponents’ coalition to this end, in keeping with their general need to acquire both foreign technology and the raw materials of the information economy at the lowest possible costs. In practice, the ability of the developing countries to oppose a common E.U.-U.S. harmonizing initiative, should they pursue it, seems doubtful.

As matters stand, these countries have showed little ability or inclination to master the intricacies and nuances of older, established intellectual property regimes – for example, the patent and copyright paradigms – so as to exploit the flexibility remaining in the still only partly harmonized international intellectual property standards of the TRIPS Agreement. Rather, they have been content, on the whole, to criticize the inequities of the TRIPS Agreement at the margins, especially with regard to so-called implementation issues, and to indulge in largely diversionary dreams about the potential benefits from exploiting traditional intellectual resources [153]. This inertia has left most developing countries dependent upon technical solutions that intergovernmental organizations controlled by the donor countries make available, and they are accordingly unprepared to play a leadership role in such a challenging area. Even if some developing countries were to muster sufficient expertise and initiative to form an international opponents’ coalition, they would still find it difficult to organize as a block and to ensure that major players did not cut separate deals with the advanced industrialized countries [154].

For this and other reasons, it seems likely that, faced with a common E.U.–U.S. approach, the developing countries would fall in line, in the hopes of gaining some compensatory trade advantages in other areas. In that event, a relatively high and uniform level of database protection would prevail at the international level, a result which reduces uncertainty and lowers transaction costs. If, however, critics are right in predicting that the social costs of strong database protection will greatly outweigh any benefits in the long term, a premature but successful harmonization campaign could mean that the entire world ended up with a socially harmful modality of protection.

Such a sobering possibility merits further reflection. If it turns out that the E.U. Directive embodies nothing more than a combination of ignorant tinkering and special interest lobbying, then a mandatory, globalized regime along the same lines would ensure that every national system of innovation would sooner or later have to digest the fruits of a poisoned tree. In that event, any social gains accruing from uniform law would gradually be offset by the social costs of diminished access to data and information and by a progressive suffocation of those upstream processes of scientific discovery and technical innovation that we now take for granted. To put it bluntly, if the high-protectionist approach turned out to be a colossal blunder that balkanized the worldwide flow of scientific and technical information, the fact that all countries now participated on an equal footing in the new feudalist information economy would only magnify the unintended social costs [155].

Aware of these risks, U.S. officials opposed to the E.U. model have taken the view that the U.S. must put forward a credible alternative in order to avoid the scenarios described above [156]. On this view, the benefits likely to accrue from articulating a second, less protectionist modality (whatever form it may take as a result of some compromise solution) would outweigh the costs of a socially unbalanced uniform law that produced universally harmful results. Because many influential stakeholders share this view, and it could prevail in the end, it is worthwhile to consider the possible implications at the international level that might flow from the existence of two different and competing models of database protection, one championed by the European Union and the other by the United States.

Such an initiative could have deleterious effects on the progressive development of international intellectual property law, however, and would almost certainly fail of its essential purpose in the end. The extent to which top-down harmonization projects, like that embodied in the TRIPS Agreement, will ultimately strengthen the global economic system remains to be seen [159]. It is well to remember, however, that the TRIPS Agreement mostly embodied backwards looking technical solutions that had emerged from the crucible of trial and error after long periods of divergent state practice. The one thing we do know from two hundred years of empirical results is that national treatment has been the sine qua non of bottom up, socially sound evolutionary progress in new fields of intellectual property law [160].

National treatment imposes a soft but critical economic discipline on all the stakeholders who participate in an integrated marketplace. While ostensibly allowing normative freedom in shaping any given state’s domestic intellectual property regimes, it exposes the domestic beneficiaries of unilateral protectionist incentives to the discipline of competition by foreign nationals who operate in less regulated, more open economic arenas. So long as all stakeholders benefit from a lowest common denominator that protects investors from market failure and that ensures a reasonable ability to appropriate the fruits of their investments, the requirement of national treatment acts as a cautionary brake or safety valve against domestic self-indulgence in unilateral protectionist experiments [161]. In other words, national treatment forces all the players to honestly evaluate different legal incentives and, by exposing any single player who exaggerates the level of unilateral incentives to global market discipline, it contributes to the experimental laboratory of trial and error at the national level from which sound international minimum standards of intellectual property protection may gradually and empirically emerge.

The absence of national treatment can, instead, produce an unhealthy protectionist environment with centripetal effects elsewhere. Exorbitant protection in one country, such as the E.U., on a condition of material reciprocity, can unleash latent free-riding interests in other countries, some of which may be emboldened to move to the opposite extreme and thus to legitimize local appropriation of foreign investments in this and other areas. Without the self-discipline of national treatment, the gradual crystallization of common measures to avoid market failure and to promote healthy competition everywhere are subtly undermined [162]. Attempts to corner the market for certain information goods by rigging legal monopolies in one group of countries may thus trigger retaliatory and self-serving measures to enhance the free-riding capacities of other countries, whose resort to nonreciprocity had otherwise been legitimated. In this situation, a determined group of countries could destabilize the database investment policies of others, who would find it difficult to prevent leakage from low to high-protectionist markets.

There is also a risk that developing countries will begin to articulate hybrid intellectual property regimes of their own in one field after another, including databases, and that they may progressively deny foreigners access to those regimes by invoking the E.U.’s own example. This is hardly a new risk. It was, indeed, the message that E.U. authorities conveyed to the U.S. in 1984, when they suggested that the reciprocity clause in the SCPA might turn out to be a grave political blunder [163]. Meanwhile, a material reciprocity clause will not seriously deter U.S. database companies from operating in Europe nor will it bring the U.S. into line. On the whole, European courts applying the domestic database regimes have gone out of their way to recognize alternative grounds of protection available from unfair competition and copyright laws [164]. These forms of protection remain available to all U.S. companies even if they are denied access to the sui generis database regimes. At the same time, E.U. companies may become vulnerable to the products of more competitive conditions in the U.S. and elsewhere, especially with regard to follow-on applications in which material taken from an existing database is combined with new matter and new investments to produce improved databases for the same or distant market segments.

The E.U. also remains heavily dependent on the U.S. for large amounts of data, especially government-generated and government-funded data. If the E.U. were too aggressively to enforce its material reciprocity clause and were to insist, for example, on charging non-profit users in the U.S. for kinds of data that are supplied to the E.U. gratis at present, the result would put severe strains on preexisting modalities for datasharing that have long been in place [165]. In such a case, authorities in the E.U. should expect that the U.S. government would begin to retaliate by charging E.U. users enough for its data to offset the costs imposed on U.S. users by E.U. suppliers of comparable data. By dint of such a “poison pill” defense, in other words, consumers in E.U. member states could end up defraying the costs of data that their own private or public sectors were unwisely imposing upon both the public and private sector in the U.S. Unfortunately, in such a retaliatory climate, the data-sharing ethos of the scientific and research communities would certainly suffer, and the gaps in worldwide database repositories that have already begun to appear are likely to grow larger [166].

The E.U., strong as it is, lacks the power to impose its model of database protection on the rest of the world by dint of its material reciprocity requirement. If good sense and self-restraint fail, the likely result of a database war would be a renewal of free-riding practices in many developing countries and a resort to “poison pill” retaliatory tactics by both public and private entities in the U.S. This scenario suggests that, unless one side succeeds in persuading the other of the virtues of its respective model, ways must be found to enable both models to coexist at the international level with a minimum of friction.

Apart from these deep-seated conceptual problems, differences of opinion concerning new subject matters of protection also reflect different social and political contexts, different cultural traditions, different economic approaches, and different value judgements about how production should be organized [170]. While the need to respect differences of this kind constitutes a truism of enlightened intellectual property discourse in general, that truism becomes compelling when the proposed object of protection turns out to be facts, information, and data – the raw material of the public domain, which U.S. courts and the Congress have consistently refused to protect under preexisting intellectual property laws [171].

The E.U. Commission, operating as it does under a well-known “democratic deficit”, may continue to downplay concerns about restraints on free c