News: Dr. Robert Rines Founder & Former President

 

Berkeley Technology Law Journal

Spring 2002

Symposium

Patent System Reform

Article

*685 HARMONY AND DIVERSITY IN GLOBAL PATENT LAW

John F. Duffy [FNd1]

 

 

 

 

Copyright © 2002 John F. Duffy

 

 

 

 

Abstract

The second half of the twentieth century saw the rise of a broad movement to harmonize patent laws across nation-states. The most recent, and most significant, manifestation of this movement is the 1994 TRIPS Agreement, which requires signatory nations to adopt uniform rules on many major issues of patent law. The TRIPS Agreement has now been implemented by well over one hundred countries, including almost all major industrial nations, and it heralds a new level of international uniformity in patent law.

 

This Article, while acknowledging the value of some harmonization of national law, explores the possible costs of the harmonization movement. Patent law itself owes its very birth not to harmony but to diversity of national law. The fifteenth-century Venetian patent statute was an experiment in law and a departure from the classical hostility to monopoly. Throughout the history of patent law, individual nations have varied their law and practice, and the results of these experiments have strengthened and improved patent practice. Diversity and experimentation continue today. As case studies of such experimentation, this Article examines business method patents and law governing the experimental use defense to patent infringement, an area in which both commentators and nations have split as to the proper approach for the law.

 

This Article concludes that the patent law of the twenty-first century would be enriched if national and international policymakers learn to value variety.

 

Table of Contents

I. Introduction .............................................. 686

 

II. Theoretical Reasons for Harmony and Diversity ............ 693

 

A. The Case for Harmonization .............................. 693

 

1. Jurisdictional Externalities .......................... 693

 

2. Economies of Scale in Governance ...................... 700

 

3. Preventing Destructive Protectionism .................. 702

 

B. The Case for Diversity .................................. 703

 

1. Matching Local Preferences ............................ 703

 

2. Competition as a Check on Government .................. 706

 

3. Permitting Experimentation and Innovation in the Law .. 707

 

III. Diversity and Innovation in Patent Law: A Case Study .... 709

 

A. Historical Examples of Innovation in Patent Law ......... 710

 

1. The Invention of Patent Law ........................... 710

 

2. Technical Examination ................................. 713

 

3. Early Publication of Patent Applications .............. 715

 

B. Ongoing Experiments ..................................... 716

 

1. Business Method Patents ............................... 717

 

2. The Experimental Use Exception to Infringement ........ 717

 

IV. Implications ............................................. 719

 

A. Interpreting TRIPS ...................................... 720

 

B. Beyond TRIPS ............................................ 723

 

V. Conclusion ................................................ 725

 

*686 I. INTRODUCTION

Uniformity of law has an undeniable intellectual appeal. It simplifies the law, makes it easier to learn and describe, and reduces administrative costs. Yet uniformity has its costs too. It makes the law unresponsive to local variations, eliminates interjurisdictional competition and decreases the possibilities for legal experimentation. The choice between uniformity and diversity is difficult and has, not surprisingly, generated great debate in numerous areas of law and social policy, including such diverse matters as corporate law, international antitrust law, local and international taxation, tort law, securities regulation, and environmental law. [FN1] The literature *687 in these areas frequently focuses on whether competing legal regimes will produce inefficient and socially destructive competition ("races to the bottom") or wealth maximizing competition ("races to the top"). But jurisdictional competition is only one of a number of relevant considerations; the debate encompasses a more general and fundamental inquiry into the social efficiency of harmonized law.

 

Curiously, the implications of this debate have generally not been considered in the area of patent law. With few exceptions, the international patent community has taken as a given the value of creating uniform patent law on a global scale. For example, in remarks concerning the future of patent law, the former head of the U.S. Patent and Trademark Office, Q. Todd Dickinson, simply presumed a consensus on the need for a global patent system:

I think most of us here [at the annual Fordham Conference on International Intellectual Property Law and Policy] would say that there definitely should be a global patent system of some sort by 2010. I think we can all list probable benefits of such a system: reduced costs for inventors and for their assignees, dramatically simpler protection, and uniformity of that protection throughout the world. [FN2] Mr. Dickinson's sense of his audience seems correct. The "profound public policy need for this global system" is a generally accepted postulate, and the task for policymakers is limited to sorting through the "diversity of the existing systems and the current proposals" and to "achieving a consensus on the nature of the global patent" system that should be created. [FN3]

 

*688 This attitude has produced concrete changes. The second half of the twentieth century saw the rise of a broad movement to harmonize patent laws across nation-states. The most recent, and most significant, manifestation of this movement is the 1994 Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS), which requires signatory nations to conform their patent laws to a uniform framework of international standards. [FN4] The TRIPS Agreement has now been implemented by well over 100 countries, including almost all major industrial nations, and heralds a new level of international uniformity in patent law. In the post-TRIPS world, harmonization continues to be a shibboleth in patent circles, and diversity a flaw to be remedied. [FN5]

 

Further harmonization on a global scale would, of course, provide certain benefits. As Mr. Dickinson notes, it could reduce administrative costs and provide simplicity and uniformity in application. But it would also preclude interjurisdictional competition and experimentation in patent law, among other things. The relevant policy question is to what extent interjurisdictional diversity and competition should be sacrificed to achieve global uniformity. This question is important not only for determining the optimal amount of harmonization to be pursued in the future, but also for understanding the proper limitations of the steps already taken toward global harmonization of patent law, particularly the TRIPS agreement. If jurisdictional diversity retains some merit, then the provisions in TRIPS *689 permitting variance should be interpreted much more broadly than if harmonization were an unqualified good.

 

To investigate the relative values of harmony and diversity, this Article draws on the rich literature concerning interjurisdictional legal diversity and competition that has developed in other areas. The law of patents can profit from this literature because it provides valuable insights into the forces driving patent harmonization and the potential costs of uniformity. Yet the benefits also flow in the other direction. Not only can patent law borrow from general theories of interjurisdictional diversity and competition, it can also enrich those theories.

 

Scholars studying interjurisdictional legal diversity tend to agree that such diversity has at least two potential values: First, it can help match the level of public goods to the tastes and resources of the local population. Second, under certain conditions, it can lead to interjurisdictional competition that checks inefficient government behavior. A third potential value of diversity-- that it can breed desirable experimentation and innovation in law--has proven more controversial. On the one hand, legal scholars have acknowledged the possibility of legal experimentation and innovation. [FN6] Indeed, this tradition dates back at least to the time of Justice Brandeis, who in 1932 famously analogized the states in our federal system to "laborator[ies]" that could experiment with new social and legal innovations. [FN7] Nevertheless, formal economic models of interjurisdictional legal diversity and competition often do not account for the possibility of legal innovation, [FN8] and some law and economics scholars have questioned the validity of Justice Brandeis's insight. [FN9] Indeed, one scholar, Frank Easterbrook, discounts Brandeis's views on legal experimentation as based on "Panglossian assumptions." [FN10]

 

*690 Moreover, even among scholars who study the process of legal innovation, the long-term value of legal innovation remains open to question. Legal innovation might be merely a product of ideological swings, with today's regulatory innovation being undone by tomorrow's deregulatory innovation. [FN11] If so, the long-term value of innovation might be slight. Alternatively, legal innovation might be thought necessary to adapt law to the unique conditions of a particular time and culture. Legal innovations would then have a significant value--much in the way that a new phonebook is more valuable than an old one--but there would not necessarily be any sense of progress, any sense that the law is getting better with time. [FN12]

 

Legal innovation would be most valuable, however, if it were like technical innovation--i.e., if it were a permanent, nonobvious addition to the storehouse of useful knowledge. Under this view, a diverse legal system has positive externalities for other legal jurisdictions precisely because it provides information to the other jurisdictions about the value of different legal rules. When that information shows the efficacy of a particular legal rule, other jurisdictions may change their laws to adopt the *691 new rule. In this way, the other jurisdictions benefit from the mere fact of difference, and the technology of law advances.

 

It is here that patent law can enrich the general theoretical discussion, for the history of patent law demonstrates not only the reality and value of progress in legal technology, but also the necessity of legal diversity in fostering that progress. Indeed, the entire field of patent law owes its birth not to harmony but to diversity of law. The fifteenth century Venetian statute that pioneered patent law was an experiment in law, and a departure from the classical hostility to government-sanctioned exclusive rights. [FN13] Since then, individual nations have varied their patent law and practice, with other jurisdictions following where the experiment was deemed successful. At least some of these innovations resulted in permanent legal advances. This process of experimentation and innovation continues today. For example, business methods patents and the experimental use defense have generated great differences of opinion among courts and legal commentators, and a diversity of approaches among nations. [FN14] These past and current experiments in law invigorate and strengthen patent law with new innovations.

 

None of this denies that consistency has merit too. For example, the United States has maintained a uniform, national patent system since 1790. Accepting the value of diversity does not lead to the conclusion that each of the fifty states should administer its own patent system. Nor does it even suggest that each nation-state should maintain its own patent system. Just as in private industry, a certain amount of consolidation may increase not only social welfare, but also competition. Thus, having four or five competing patent systems may be better than having one hundred.

 

But a complete international harmonization of patent laws--particularly, the institution of a single, integrated global patent system--would eliminate interjurisdictional competition and substantially stifle innovation in patent law. While the loss of competition occurs by definition, the reduction of innovation follows from several effects. First, in a unified global system, experimentation in law could take place only successively, with the entire world serving as the "laboratory" for a particular period of time. Because experiments would be spread out temporally, not geographically, the pace of innovation would necessarily be slower. Second, the resulting "data" from any particular experiment may be much harder to interpret because the experiment lacks a good "control." Thus, for example, the apparent success or failure of an experiment might be attributable *692 to then-prevailing world conditions; an innovation adopted at the beginning of a worldwide economic recession may appear to be less successful than it actually is. For this reason, experiments conducted on a global scale may produce inconclusive results and slow the rate at which innovations are adopted. Third and finally, conducting experiments on a global scale may be much more difficult than doing so within nation-states or groups of nation-states. Where a global regime is established by multilateral treaty, the process for modifying the agreement may require a worldwide consensus, a significant political barrier to legal experimentation. [FN15]

 

Aficionados of the patent system--a system built to reward risky experimentation--are particularly well-suited to appreciate the costs that such a development would entail. It would be both ironic and unfortunate if a legal system that owes its existence to experimentation, and that is designed to foster experimentation in technical areas, were modified to preclude substantial experimentation and further development of its own norms.

 

Part II of this Article provides a theoretical analysis of harmony and diversity in patent law. While this analysis finds strong reasons favoring a certain degree of harmonization, it also uncovers significant justifications for maintaining some diversity in patent law. At least one of these justifications for diversity--the value of diverse laws to match local preferences--has previously been mentioned by other patent law scholars. The value of diversity in fostering legal innovation has not. Part III explores this innovation rationale from a historical perspective and demonstrates that many valuable features of modern patent law began as controversial experiments in local systems. As shown in Part III.B, such experimentation continues today, with individual patent systems taking different approaches on controversial topics in the field. Part IV details the implications of these theoretical and historical analyses. If patent law's tradition of innovation is to continue, global patent law must maintain a degree of jurisdictional diversity. Part IV.A suggests ways to interpret TRIPS to achieve that goal, while Part IV.B suggests future steps in which a certain degree of additional patent harmonization can occur without compromising the value of diversity. Finally, Part V offers some concluding thoughts.

 

*693 II. THEORETICAL REASONS FOR HARMONY AND DIVERSITY

Finding the optimal balance between legal uniformity and diversity requires an analysis of the factors favoring each side. The analysis here is necessarily limited to qualitative assessments because, as is often true in other areas of law, quantitative measures of the costs and benefits of legal diversity are currently not possible. Nevertheless, even a qualitative analysis is valuable because it can explain the impulses for harmonization and identify circumstances in which the need for diversity is particularly great.

 

A. The Case for Harmonization

 

Legal harmonization--whether accomplished by consolidation of previously independent regimes or by less dramatic measures--is usually viewed as an appropriate response to three problems: jurisdictional externalities, economies of scale in governance, and destructive protectionism. The case for harmonization in patent law rests mainly on the first two of these.

 

1. Jurisdictional Externalities

The most compelling justification for harmonization in patent law mirrors the justification for creating a patent system in the first place, for both are efforts to account for the positive externalities associated with the creation of technical information. In a market economy, free competition between firms is thought to produce efficient outcomes provided that, among other things, each firm internalizes all the social costs and benefits of its own activities. [FN16] In the absence of a patent system, however, that condition does not hold because the production of easily appropriated knowledge will have positive external benefits.

 

*694 Consider, for example, the situation in which one firm produces a valuable, innovative product that can be easily reverse-engineered. Once it is marketed, the innovation will be quickly copied by the firm's competitors, and the price will be driven down to the marginal cost of manufacturing copies. Accordingly, the innovating firm will be unable to capture the full social benefit of its innovation--innovation will have positive externalities--and the incentives to invest in research and development will be inefficiently low. The patent system can be accurately described as a regulatory mechanism that attempts to correct this externality by more closely aligning the private and social value of producing new information. [FN17]

 

Just as the externalities provide a justification for the existence of a patent system, so too do they provide a reason for harmonization. Consider, for example, the situation in which one country maintains a patent system but its neighbor does not. Because of the incentives of the patent system in the first country, firms will invest resources in developing patentable innovations. Consumers in the first country will pay above-marginal-cost prices for those innovations and will thus bear the cost of the information necessary to develop the innovations. By contrast, consumers in the second country will, if competitive conditions prevail, pay only the marginal cost of reproducing the innovation; they will free-ride off the investments of their neighbors. The legal regime in the first country thus has a positive externality for the second country.

 

As the literature on jurisdictional competition consistently demonstrates, such an externality provides a good reason to distrust the body of law produced by diverse jurisdictions, [FN18] and also a justification for some form of transjurisdictional regulation. For example, in a seminal article on the competition between local jurisdictions to provide public goods, Charles Tiebout acknowledges that where "external economies and diseconomies are of sufficient importance, some form of integration [of the *695 competing jurisdictions] may be indicated." [FN19] Numerous other articles reach similar conclusions. [FN20]

 

In an era characterized by inexpensive communications and the free flow of information, externalities provide a particularly powerful justification for transnational patent harmonization because one nation's patent law can create a global externality. By comparison, the externality problem with most environmental controls is limited to some extent by geography; in many cases, stringent environmental laws benefit only neighboring or downwind jurisdictions. [FN21] In contrast, the disclosure of new technical information in Europe or the United States can travel to the other side of the planet almost instantaneously. [FN22]

 

The externalities argument in fact explains much of the most significant step in patent harmonization, the TRIPS agreement. The negotiations leading to TRIPS were essentially negotiations between industrially developed and less-developed countries. [FN23] Developed countries entered the negotiations with much stronger patent and intellectual property (IP) systems than those in the developing nations. In other words, the patent systems of the developed countries created positive externalities for the developing nations, which were free-riding on the technological information produced in more developed countries. [FN24] The goal of the developed nations (led by the U.S.) was to increase IP protections in developing countries. In exchange, developing countries obtained more open markets for *696 their textile and agricultural products in developed nations. [FN25] In effect, the TRIPS negotiations may be viewed as a form of Coasian bargain, with developing countries accepting valuable consideration in exchange for their agreement to adopt a legal system addressing the positive externalities problem.

 

The positive externalities associated with the IP systems also explain the overarching structure of the TRIPS agreement. If positive externalities were the chief concern of the parties in negotiating TRIPS, then the parties should be concerned only that a country may be providing too little, not too much, IP protection. A nation that decides to depart from an international norm and provide greater IP protection would only provide a benefit to other nations. In other words, the resulting treaty should mandate only minimum standards and, in fact, that is precisely what TRIPS does. [FN26] It provides only a harmonized floor; countries remain free to experiment with more stringent patent rights. For example, TRIPS Article 33 provides that the term of patents "shall not end before the expiration of a period of twenty years counted from the filing date." [FN27] Countries remain free to experiment with longer terms and, in fact, the United States recently exercised that option by providing patentees with certain patent term "adjustments" that extend the term where the PTO has failed to meet certain statutory goals during the prosecution of the application. [FN28]

 

TRIPS also provides little harmonization in areas where the laws of major industrialized countries diverge. For example, TRIPS makes disclosure of a "best mode" optional, does not require or forbid administrative opposition procedures, leaves nations free to choose a first-to-invent or a first-to-file patent priority rule, and imposes no obligation for countries to grant a "grace period" within which inventors can disclose their inventions without destroying their own novelty. [FN29] These unaddressed points represent areas where the laws of other developed countries differ but, more importantly, they are also areas where the externalities associated with choosing one approach or the other are indeterminate or insignificant: It is simply not clear whether nations with first-to-file rules create positive externalities *697 for nations with first-to-invent rules, or vice-versa. Since TRIPS was designed primarily to address an externality problem, it does not harmonize law on such matters. Similarly, TRIPS imposes only minimal regulation on the patent application process. It demands only that Member nations grant rights using "reasonable procedures and formalities" and that they do so "within a reasonable period of time so as to avoid unwarranted curtailment of the period of protection." [FN30] Thus, TRIPS produces no substantial savings on the administrative costs of obtaining worldwide patent rights. This approach is consistent, however, with the view that TRIPS is directed mainly to the pre-existing externalities created by the divergent substantive patent laws. [FN31]

 

Despite its theoretical strength and its power in explaining the TRIPS agreement, the externalities argument has significant limitations as a justification for comprehensive global patent harmonization. Most importantly, while positive externalities can lead regimes to free-ride by adopting suboptimal patent protection, externalities do not necessarily lead to a race to the bottom where each jurisdiction progressively reduces patent protection down to nothing. Even assuming that technical knowledge produced by a patent system benefits all individuals in the world equally (an extreme assumption), each country still has an incentive to adopt a patent system because its citizenry will benefit from the incremental increase in technical progress fostered by its patent system. Where the country is sufficiently *698 large (in terms of population, wealth, and inventive capacity), that benefit may outweigh the value of free-riding on other systems.

 

For example, consider a world consisting of only two countries: one large-- say, the United States--and the other small--say, Monaco. For the United States, the incentives to adopt a patent system are almost identical to those that would exist if there were no externalities. The external effects of the system on Monaco are sufficiently small so as to be ignored. The reverse, of course, is not true. Assuming that competition will drive price to marginal cost where no patent protection exists, the residents of Monaco will reap enormous benefits by free-riding on the inventions produced by the U.S. patent system. However, they would gain little by adopting their own patent system because Monaco's patent system would increase world technical progress only a small amount.

 

The historical development of patent law reflects the limited effect of externalities; many nations adopted patent systems even prior to any significant international cooperation. By the early nineteenth century, patent systems existed in the United States, England, France, Russia, Austria, Prussia, the Netherlands, and a number of the German states. [FN32] In some jurisdictions, most notably Venice, England and France, rudimentary patent systems date back to the sixteenth century. [FN33]

 

Other effects also curb the positive externalities associated with a patent system. Many modern inventions involve technologies that exhibit significant economies of scale. For those technologies, patents in a few large markets may be sufficient to confer a de facto worldwide monopoly because the market not covered by exclusive rights is too small to support an effective competitor. [FN34] In such circumstances, consumers in countries *699 without patent systems are not able to free-ride on innovations created elsewhere. Another important limitation on patent externalities is that the technical knowledge developed by a patent system does not benefit everyone in the world equally. Faster computer chips likely have less value to consumers in Bangladesh than in the U.S. and European Union. [FN35] Indeed, some innovations may be highly specific to a particular region. For example, a new composition of cleaner-burning gasoline may have high value to a single region plagued by automobile air pollution--say, California--but little value in other countries, or even in other regions of the same country. [FN36]

 

A more fundamental objection to using patent externalities to justify comprehensive harmonization is that externalities can be addressed as effectively by more limited measures. [FN37]The simplest way to prove this point is to imagine a world with fully harmonized patent standards across all jurisdictions. Now assume one jurisdiction changes its patent law by (1) narrowing the scope of patent rights somewhat, but also (2) expanding the patent term to compensate. If the legal modifications do not change the net expected value of rents from patents (ex ante), then the modified patent system will provide inventors the same incentives to innovate and will address the externalities problem as well as the harmonized system despite the difference in legal systems. Of course, the assumption here--that two patent systems could be said to generate the same net rents--may seem unrealistic because of the extraordinary difficulty of determining what incentives are actually created by any particular patent system. But in fact, that informational difficulty actually helps the argument because, ex ante, inventors will view the incentives created by two patent systems as identical provided that discerning any differences is sufficiently costly.

 

*700 Again, the structure of TRIPS seems consistent with this limitation on the externalities argument. TRIPS generally mandates that signatory countries provide broadly similar patent systems. It does not delve much into the details of systems in part because the precise effects of those details are not known with certainty. Thus, as previously mentioned, it remains uncertain which of the various technical differences between the patent systems of industrialized nations produces the better incentives to invent. TRIPS rightly leaves those matters open for each country's choice.

 

2. Economies of Scale in Governance

Patent systems exhibit economies of scale in certain respects--most obviously in the administrative examination of patent applications. The cost of examining each application does not change whether the patent office serves a jurisdiction covering six million or six billion people. In this respect, the administrative function of a patent office resembles a classic natural monopoly, with its average cost of service continuously declining as its jurisdiction expands. Thus, the efficient solution is monopoly--consolidation of the diverse patent offices into one. Indeed, the argument for consolidation here is much stronger than it is in private markets. Since existing patent offices are already government monopolies within their respective jurisdictions, consolidation will not have any efficiency losses associated with eliminating existing competition.

 

Like the externalities point, this economies of scale argument has a great deal of force. Indeed, it provides a good basis for former PTO Commissioner Dickinson's promise that a global patent system will deliver "reduced costs for inventors and for their assignees." [FN38] Because patentees pay filing fees to support the currently duplicative national examination system, they would benefit most from the reduction in administrative costs that would accompany global consolidation of patent offices.

 

Economies of scale also explain a number of developments in international patent law, including the European Patent Convention (EPC), Patent Cooperation Treaty (PCT) and even the nineteenth century Paris Convention. The EPC, which created the European Patent Office (EPO) and authorized it to serve as an examination office for all EPC member countries, [FN39] is the most obvious example of a reform motivated by the economies of scale associated with examination. The treaty allows patent applicants to prosecute their application through a consolidated patent office *701 with jurisdiction over most countries in Europe. The PCT also provides a very limited administrative consolidation, as it streamlines the early stages of patent prosecution on virtually a global scale, because all major industrialized nations are signatories to the treaty. [FN40] Even the Paris Convention effectively provides a very limited form of patent office consolidation by permitting applicants to file in any signatory country and thereby obtain a priority date in all countries. [FN41]

 

Like the externalities argument, however, an argument based on economies of scale has significant limitations as a justification for harmonization. First, the argument does not necessarily provide a reason to harmonize substantive law. While some substantive patent rules might be relevant to the examination process--e.g., the rules governing priority and grace periods (because they govern the universe of prior art used in examinations)--others are not. Most obviously, no administrative economy is realized by having patents run for a uniform term. Indeed, the point is demonstrated by the EPC, which authorized a unified examination system but not a unified patent right. Thus, a patent issued by the EPO is said to be a "bundle" of national patents, with the precise contours of the substantive rights governed by diverse national laws.

 

Second, the economies of scale argument explains few of the TRIPS reforms. The TRIPS agreement achieves no degree of administrative consolidation among patent jurisdictions. Moreover, TRIPS did nothing on the substantive issues that could assist in administrative consolidation (priority rules and grace periods), yet did require some degree of harmonization on an issue like patent terms, which cannot be justified by economies of scale.

 

Third, and perhaps most importantly, administrative examination systems exhibit economies of scale only in limited respects. While the administrative costs of examination do not increase where the jurisdiction gains additional people subject to the resulting patents, the costs do rise where the examination system processes additional applications. Indeed, in terms of processing applications, examination systems probably exhibit only limited economies of scale. Examining patent applications is a labor-intensive enterprise. Each additional application takes a certain number of examiner-hours, and the price of labor remains relatively constant.

 

*702 For such an administrative task, a single entity is probably not the optimal solution. A number of competing patent offices with overlapping, worldwide jurisdiction would be better. Patent applicants could choose a patent office for prosecution, and patent offices could compete with each other based on the level of their fees and the quality of their examinations. [FN42] To a very limited extent, we can already see the glimmerings of such a system. The jurisdiction of the EPO overlaps with that of national patent offices, and the EPO competes to attract patent applicants. Similarly, the Paris Convention permits any country to establish a world-wide priority date, and at least one nation has begun overtly advertising the efficiency of its patent office in providing this service. [FN43] A system of internationally competing patent offices would demand a certain degree of international cooperation. But it need not demand complete harmonization of patent laws, and it would preserve diversity of administrative approaches.

 

3. Preventing Destructive Protectionism

A final reason to favor some form of transjurisdictional arrangement is to restrain protectionist impulses. The reason is based on the view, supported by empirical data, that protectionism reduces overall social welfare. [FN44] However, this reason justifies only very limited transjurisdictional regulations that preclude discrimination against free trade. [FN45]

 

The desire to restrain welfare-decreasing protectionism can be seen in the contemporary American jurisprudence regarding the dormant Commerce Clause, which generally precludes state discrimination against interstate free trade and is justified as a means to achieve the desirable goal of a national common market. [FN46] It also accounts for the overarching structures of the General Agreement on Tariffs and Trade (GATT), which generally precludes discrimination against international trade between member *703 nations but otherwise leaves nations free to have diverse sets of regulatory laws. [FN47]

 

The anti-discrimination justification explains only a few parts of TRIPS-- e.g., Article 27's requirement that patent rights be "available and . . . enjoyable without discrimination as to the place of invention . . . and whether products are imported or locally produced," [FN48] and the more general requirement in Article 3 that "[e]ach [WTO] Member shall accord to the nationals of other Members treatment no less favourable than it accords to its own nationals with regard to the protection of intellectual property." [FN49] Some of the anti-discrimination rules in TRIPS were already imposed throughout much of the world by the Paris Convention. [FN50] But in any event, given the existence of the anti-discrimination norms in TRIPS, few if any additional transnational regulations could be justified as measures to preclude destructive protectionism.

 

B. The Case for Diversity

 

A review of the justifications for harmonization shows that, while no clear reason exists for a comprehensive integration of global patent law, there are reasons for supporting particular forms of transnational regulation. Yet these justifications for harmony must be balanced against the costs of harmonization--or, otherwise stated, the values of diversity.

 

1. Matching Local Preferences

The theoretical literature on jurisdictional legal variation posits that the primary reason for such variation is to permit each jurisdiction to match its laws to the unique tastes and preferences of its population. [FN51] The argument *704 has strong and weak versions. The strong version assumes individuals are free to move between jurisdictions and concludes that, under certain assumptions, the resulting diversity of laws between jurisdictions reflects an optimal provision of public goods. [FN52] While this version of the argument is important for diversity in local and state jurisdictions, it has little force in the international setting because changing nationality is relatively expensive for individuals.

 

More relevant in the international context is a weaker version of the argument, which assumes a stable set of preferences within each jurisdiction and concludes that the diverse laws of each jurisdiction more closely match the individual preferences within the jurisdiction than would a uniform set of laws imposed across all jurisdictions. Professors McGinnis and Movsesian rely on this argument in explaining why the GATT/WTO structure does not attempt to harmonize worldwide regulatory law:

Uniform health, labor, safety, and environmental regulations are unlikely to be appropriate for all members of the world trading community, as members of the WTO vary widely in their levels of development. As a result, they will rationally choose different regulatory standards. It is wrong to assume, for example, that Indian and American regulations on water purity should necessarily be the same. Indians may not be able to afford American water safety standards, just as they unfortunately cannot afford many other goods that Americans can. [FN53] A similar point is made in the patent context by Professors Dreyfuss and Lowenfeld, who advance the following argument for flexibility in the application of TRIPS:

More fundamentally, we are skeptical that there will always be a "best" rule for every problem that will arise under the TRIPS Agreement. Promoting innovation requires that care be taken not to raise the cost of knowledge to so high a level that it impedes further inventiveness. How that problem is best solved can depend on a country's intellectual and industrial development, its culture, and the types of creative work in which its citizens are engaged. Thus, the nature (and advantage) of a minimum standards regime is that where there is no "best" rule that will work *705 in every economy, each country can tailor the law to its own needs. [FN54]

 

Like McGinnis and Movsesian, Dreyfuss and Lowenfeld suggest that one obvious example of inappropriate uniformity would be applying rules from "highly developed countries" in the less-developed world. The concern here is consistent with the general theory, for the widest divergence of preferences might well be found between developed and less-developed countries.

 

TRIPS expressly recognizes the value of local diversity. Its very first article guarantees that "Members shall be free to determine the appropriate method of implementing the provisions of this Agreement within their own legal system and practice." [FN55] Similarly, Article 27, which generally mandates that patents shall be available in all fields of technologies, allows countries to create exceptions from patentability "necessary to protect ordre public or morality" [FN56]--a standard understood to "depend[] to a certain degree on the particular culture of a country or region." [FN57] Consistent with theory, the TRIPS Preamble explicitly recognizes "the special needs of the least-developed country Members in respect of maximum flexibility in the domestic implementation of laws and regulations in order to enable them to create a sound and viable technological base." [FN58] Special provision is also made for countries and nations "in the process of transformation from a centrally-planned into a market, free-enterprise economy," *706 both of which are likely to have preferences that widely diverge from those in developed, free-market nations. [FN59]

 

There are two significant limitations on this local preferences argument. First, harmonized patent law does not result in the same degree of uniformity that, for example, a harmonized minimum wage law does. The patent right does not mandate any particular price for an innovation. Patentees are free to sell their inventions and license their rights on different terms in different areas. In fact, the available evidence demonstrates that patentees often do engage in price discrimination--for example, by lowering the prices of patented drugs in poorer countries. [FN60]

 

Second, and more importantly for purposes of this Article, the local preferences argument is less compelling where the diversity occurs between nations having seemingly similar preferences (e.g., between two well-developed nations) than where preferences are likely to be widely divergent (e.g., between developed and developing nations). [FN61] If local preferences were the sole reason for maintaining legal diversity, a general harmonization of law among similarly situated nations, e.g., among developed nations, might be desirable. But the case for legal diversity does rest wholly on the local preferences argument.

 

2. Competition as a Check on Government

Another common justification for permitting jurisdictional legal variation is that tolerating variation will breed jurisdictional competition, which checks governmental inefficiency and abuse. This is sometimes referred to *707 as the "Leviathan" argument because the competition checks otherwise harmful tendencies of monopolistic governmental power. [FN62]

 

This argument is not, however, a very powerful reason for resisting global patent harmonization. The general limitation on the argument is that, even where harmonization is pursued in a number of legal issues, competition on other points can still provide an effective check on government. The point here is familiar to regulated industries scholars: Even where regulation constrains competition along one axis (e.g., by fixing price), firms can still compete with each other along other axes (e.g., by improving quality). Thus, harmonizing law in one particular area (e.g., IP) would leave jurisdictions free to compete for capital and, to a lesser extent, labor, through jurisdictional differences in other areas of law (e.g., tax policy, environmental standards, etc.).

 

A more specific problem with this argument is that, in the area of patent law, the current diversity of law is not imposing any significant check on government inefficiency. With few exceptions, the government of each nation still holds a monopoly on the power to issue patents within its borders, and thus government patent offices are not subject to any significant competition under the current state of affairs. To the extent that a patent office has incentives to be lazy or abusive, those incentives will not be checked by competition from other jurisdictions. Therefore, consolidation of national offices into a single world patent office would not necessarily have any significant costs in terms of sacrificing competitive checks on bureaucrats. [FN63]

 

3. Permitting Experimentation and Innovation in the Law

Tolerating legal diversity may also permit legal innovation to occur more rapidly. The point was made famous by Justice Brandeis, who observed that "[i]t is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory, and try novel social and economic experiments without risk to the rest of the country." [FN64] Legal experimentation and innovation can be valuable not *708 only to the jurisdiction that conducts the experiment, but also to other jurisdictions, which benefit from the information produced by the experiment. Thus, legal diversity has its own externality, which weighs against harmonization.

 

The innovation rationale has quite different implications than the other reasons supporting diversity. It provides a more substantial basis to oppose patent harmonization than does the Leviathan argument because, as previously explained, enforcing uniformity in one area of law may still permit significant legal competition to check government inefficiency. But any degree of harmonization necessarily removes the harmonized point from parallel experimentation by different jurisdictions and is thus likely to significantly retard further development as to that aspect of the law.

 

An experimentation rationale for opposing harmonization also has different implications than a rationale based on matching local preferences. Effective legal experimentation may occur in nations that have highly similar preferences. Indeed, a nation's experiment with a legal innovation provides the most direct benefits to nations with similar cultures and preferences, because such nations are likely to have similar experiences with the innovation. Moreover, legal innovations in an area such as patent law are probably more likely to occur in developed nations, which already possess a sophisticated understanding of the area, than in less-developed nations, which do not have long experience with patent systems. It may also be unwise for less-developed nations to undertake risky experiments with their embryonic patent systems, which may not be able to weather a failure. Thus, an experimentation rationale provides a reason for tolerating diversity between developed nations. Indeed, it may be sensible to tolerate more diversity between developed nations than between developed and less-developed nations.

 

 

The value of jurisdictional diversity in encouraging legal innovation has remained controversial in the literature. In one leading article, Professor Susan Rose-Ackerman reached "pessimistic" conclusions after analyzing the incentives of politicians in a federal system to take on risky projects and finding that "low-level governments remain flawed mechanisms to rely on in the search for new ideas." [FN65] But even if the incentives are flawed, experimentation and innovation are still possible; harmonization imposes a still greater constraint. Other commentators have been more sanguine than Rose-Ackerman about the contribution of jurisdictional diversity *709 to legal innovation. [FN66] The history of patent law provides a valuable lesson here, for it shows that experimentation--even experimentation conducted by large nation-states--can produce significant new ideas in law.

 

III. DIVERSITY AND INNOVATION IN PATENT LAW: A CASE STUDY

The desirability of a harmonized patent law cannot be fairly assessed without some understanding of the important role that legal diversity has played in the history of the field. As demonstrated in Part III.A infra, legal experimentation and subsequent change in prevailing legal norms have been continuing themes throughout the history of patent law. This process is responsible not only for building essential features of the modern law, but for creating the very subject of patent law. The process of experimentation and innovation continues today and, as shown in Part III.B, it extends to fundamental issues such as the range of patentable subject matter and the scope of exclusionary rights granted by the patent.

 

The point here is not that significant changes have occurred in patent law; mere change could represent random shifts in fashion, with little permanent value. The point is that legal knowledge has progressed through these changes-- that the process of local diversity and experimentation has produced permanent, valuable advances in our understanding of legal technology. Moreover, the experiments leading to these advances have been controversial. Though many of these legal variations were eventually adopted universally, they often went against prevailing norms and generated significant opposition in some nations. In other words, if the law had been globally harmonized at the time of these experiments, it would have been harmonized against the experiments. And if a global consensus were needed to engage in the experiments, that consensus may *710 not have emerged for a very long time, if ever. Global harmonization threatens to retard this process of innovation; indeed, the threat is already being realized.

 

A. Historical Examples of Innovation in Patent Law

 

The relative youth of patent law cannot be overstated. The law of contract, tort, crime, marriage and other areas can find antecedents dating back at least to Roman and Greek law. While specific doctrines and rules in these areas have evolved since antiquity, the fields have nonetheless been recognized for more than two thousand years. The same is not true of patent law. Legal protection of inventions (or, for that matter, other categories of intellectual property) simply did not exist in Roman or Hellenistic law. [FN67] Even in its most embryonic form, patent law can be traced back little more than five hundred years, and for all but the last two hundred years the area was so rudimentary as to be barely recognizable. As late as 1850, a structure central to modern law--the patent claim--was so unimportant that the leading treatise of the day did not include the subject in its index. [FN68] In a comparatively short period of time, patent systems have developed a complex body of law that defines property rights in many diverse fields of human creativity. The development of this law provides a case study in legal innovations pioneered by one jurisdiction and then copied by others.

 

1. The Invention of Patent Law

The most obvious example of innovation is the creation of patent law itself, which occurred in Renaissance Italy. Perhaps as early as the fourteenth century, isolated monopolies on industrial developments may have been granted in other European states, but these older grants were most *711 likely made as matters of discretion rather than of right. [FN69] They seem little more than exercises in industrial protectionism, rather than a considered policy of encouraging or rewarding technical prowess or innovation. Scholars typically trace the true origins of modern patent law to the fifteenth century Venetian Republic. [FN70] In the latter half of the fifteenth century, Venice granted monopoly privileges with increasing frequency for allegedly improved industrial devices and processes brought about by the applicant's "skill and experience," "pertinent thoughts and labors," or "efforts, study and ingenuity." [FN71] This practice was confirmed in a statute of March 19, 1474, which is the first known legislative statement of generally applicable patent principles. [FN72]

 

The concept of patent law quickly spread from Venice to Germany, France and England. [FN73] In Germany, patents on inventions began about ten years after enactment of the general Venetian statute, and some historical evidence suggests that the Venetian patent idea was imported by traveling German businessmen and immigrant Venetian glassmakers. [FN74] The idea first appeared in France in 1551 and, tellingly, the first French patent was *712 granted to an Italian. [FN75] Similarly, in 1559, an Italian inventor familiar with the Venetian system seems to have been responsible for introducing the concept of patents into England. [FN76] Modern evidence of borrowing patent law from other jurisdictions is even stronger. For example, many provisions of Japanese patent law are simply translations of their German counterparts. [FN77]

 

The creation of patent law was not, however, free from controversy. In fact, the possibility of providing some legal reward for innovation had been raised long before the Italian Renaissance, but the reaction recorded in Aristotle's Politics is typical of the classical hostility to the idea. [FN78] Aristotle considered the proposal by Hippodamus of Miletus that "some honour ought to be conferred on those who suggest an improvement which is of benefit to the city," but concluded that it "cannot be safely enacted, and has only a specious sound." [FN79] In detailing his objections to Hippodamus's proposal, Aristotle focused on providing rewards for improved laws, which he considered dangerous because "[t]he law has no other source of strength through which to secure obedience apart from habit." [FN80] Even if *713 Aristotle meant to criticize Hippodamus' proposal only in so far as it would encourage innovations in law, [FN81] the criticism still created a barrier to the development of patent law and intellectual property law in general. Intellectual property law would be a significant legal innovation and, under Aristotle's view, legal "change is a matter which needs great caution." [FN82] Caution in intellectual property would mean more than one and a half millennia without significant legal innovation in the field. The creation of the patent law would occur only after the dissolution of classical societies and the rise of numerous, advanced, competing jurisdictions, some of which were willing to try the unconventional idea.

 

The controversy surrounding patent law continued even as the idea spread throughout Europe. Indeed, in the nineteenth century (shortly after many European nations first adopted patent laws), a wave of opposition stopped the spread of law and, in Holland's case led to its repeal. [FN83] This tide of patent opposition did not finally recede until 1910, when Holland reenacted a patent statute. [FN84] Thus, even though the concept of patents had been pioneered in fifteenth century Venice, at no time before the twentieth century did all major European nations even have patent laws. [FN85] If legal change had required global consensus, patent law might have been delayed even longer.

 

2. Technical Examination

The modern patent examination system is another idea that appeared in one jurisdiction and migrated, haltingly, to the rest of the world. Many early patent systems included no regularized system for evaluating the technical merits of the asserted invention over the existing art. The early English patent law, for example, was based on a registration system, with the technical merits of the invention reviewed later in infringement suits. [FN86] *714 Institutional review of novelty and utility was first developed in the French patent system, which in 1699 authorized the French Academy of Science to examine patent applications and certify the novelty and utility of the inventions [FN87]

 

The French idea of patent examination spread to the United States when Congress enacted the Patent Act of 1790. Congress rejected the English registration system in favor of an examination process similar to the French model. [FN88] Congress broke with the French system by having government officials rather than a private organization (or learned society) conduct the examination. [FN89] The early U.S. system failed, however, because it imposed the examination duty on high-level government officials who had too little time to discharge this duty effectively. [FN90] Accordingly, Congress abandoned the system in favor of the English registration model within three years. [FN91] Still, the early American experience and the French examination system provided precedents, and as dissatisfaction with the American registration system grew in the early nineteenth century, leading figures such as Thomas Jefferson considered the examination system as a remedy. [FN92] When the U.S. returned to an examination system in 1836, the *715 development drew upon earlier experiences but also created a specialized bureaucracy to perform the task. [FN93]

 

The English moved toward an examination system slowly and, as they did so, looked to the experience of other nations that already had patent examination. Prior to the 1851 reform of English patent laws, "several Experts provided information on the various patent examination systems in Europe." [FN94] The movement toward a full examination model was delayed due to the English belief that the U.S. examination system was a failure; this belief generated "vigorous and well-organized opposition to examination on the United States model." [FN95] As the examination system became more entrenched in o