11 - Self-Executing International Intellectual Property Obligations?

Summary

Self-execution is a matter of national rather than international law. While some countries regard international agreements as having direct effect, most do not consider international intellectual property agreements to be self-executing. This means that negotiators cannot assume that national law will be implemented in a manner that is entirely consistent with agreements as drafted. For intellectual property law, this situation is particularly problematic because the globalization of information, production, and manufacturing suggests that a high degree of integration is desirable. Nonetheless, there are many good reasons to preserve states’ sovereign authority in this arena. Intellectual property law involves balancing proprietary interests against public concerns. Because countries differ dramatically along the lines of culture, economics, technological capacity, and fundamental principles, it would be difficult to strike the same balance everywhere. Thus, consensus can often be achieved only through the use of “constructive ambiguities”– language that is unsuitable to direct application by judges but which allows for legislative tailoring to local needs, capabilities, and values. In addition, technologies and needs change over time. International lawmaking is too prone to capture, too shortsighted, and too cumbersome to deal effectively with such problems. In contrast, legislative intervention creates a degree of accountability to the public.

Table of Contents

1. A. Introduction 312

2. B. Self-Execution 313

   1. I. Overview 313

   2. II. The Mixed Approach 316

3. C. The Status of Intellectual Property Agreements 319

   1. I. The TRIPS Agreement 319

   2. II. The Paris Convention 320

   3. III. The Berne Convention 322

   4. IV. Free Trade Agreements 323

4. D. Normative Assessment 325

5. E. Alternative Approaches 330

6. F. Conclusion 334

A. Introduction

In theory, one might characterize an international instrument by asking whether or not it is self-executing. The obligations a nation undertakes in a treaty are regarded as self-executing when they apply as of the treaty’s own force in that nation’s courts. In contrast, a non-self-executing treaty must be implemented by the state’s lawmakers before it has direct effect, that is, before litigants can rely on its provisions in judicial proceedings. However, unlike many of the dimensions of international agreements discussed in this volume, the characterization of an agreement as self-executing is not itself a matter of international law. Rather, each party to the agreement decides the question whether the treaty has direct effect under its own national laws and policy. Although some countries have a rather rigid view on this issue, others take a mixed approach and decide it on a case-by-case basis (or, often more accurately, a provision-by-provision basis). Because the language of the treaty (or provision) and the intentions of the negotiating parties are determinative factors for these countries, the question arises whether the intended members of new intellectual property agreements should take steps to promote a particular view on the matter. Would it, in short, be beneficial – to member states and their citizens, to right holders, to the creative community, or to the international order – to draft agreements in a manner that most countries would regard as self-executing, and in this way, better ensure that states fulfill their international commitments?

This Comment begins in Part A with an overview of national views on self-execution, culminating in a description of the framework that countries adopting a mixed approach use when determining whether a measure has direct effect. This part focuses on the law of the United States, both because it has recently given considerable attention to this issue and because the decisions of any one party – particularly a large, economically powerful party – may influence the others. Part B considers how that framework plays out with regard to the core multilateral intellectual property agreements and free trade agreements (FTAs) that include chapters on intellectual property protection. Part C asks the normative question: as a general matter, is intellectual property appropriate subject matter for self-executing agreements? Concluding that it is not, Part D suggests ways to ensure that the goals of international intellectual property instruments are nonetheless fulfilled.

B. Self-Execution

I. Overview

Because self-execution is a matter of national law and policy, to the extent that something can be characterized along the dimension of self-execution, it reflects the approach that each state takes to the role its international commitments play in domestic disputes. Some states consider international law to be part of the domestic regime. In these monist jurisdictions, treaties are generally considered to have direct effect. An example, cited by Martin Senftleben, is a 1999 decision of the German Federal Court of Justice.¹ The decision, which required the Technical Information Library Hanover to pay the plaintiff equitable remuneration for its copying of scientific articles, was based directly on the three-step exceptions test in the Berne Convention, which permits certain unauthorized reproductions of protected works.²

In theory, a monist system has several advantages. It can save legislative resources because there is no need for implementing measures.³ Moreover, for countries new to the relevant field, adopting a self-executing treaty essentially imports a ready-made legal regime – often one that was drafted and adopted by countries with substantial experience in the area. A monist policy can also act as a signal that the country is prepared to adhere to the obligations set by the international community. For countries that see their future as enhanced by international cooperation and transnational business dealings, it can be highly beneficial to adopt what is essentially a pre-commitment strategy.⁴ In bypassing the need for implementing legislation, a monist jurisdiction ensures that its international obligations will not be derailed by corrupt officials, local lobbyists, or legislators who wish to pursue other objectives. In some monist states international law may even trump later-enacted legislation.⁵ Accordingly, a monist approach assures treaty partners that subsequent legislatures cannot easily undermine the nation’s commitments.

There are, however, countries that regard international and domestic laws as belonging to separate spheres. In these dualist systems, an international agreement is not internally binding until the national legislature transposes its provisions into domestic law. For example, under the UK doctrine of parliamentary sovereignty, Parliament must enact legislation to make a treaty domestically effective.⁶ Thus, a court in the United Kingdom could never require one party to compensate another based solely on a provision of the Berne Convention.

A dualist approach has much to recommend it. It provides the legislature with a chance to reconsider the outcome of negotiations conducted in locations remote to its capital. Thus, at least in theory, a dualist approach promotes local accountability.⁷ Furthermore, a dualist approach allows lawmakers to tailor the law to the specific circumstances of the country, to clarify the obligations set out in the instrument, and to put them into terms that can be more easily applied by judges and understood by lawyers and the laws’ consumers. Transposition also offers the opportunity to ensure coherence with other domestic legal regimes that use similar terminology or address related issues. Furthermore, making the change alerts the legislature that new administrative resources may be needed. When international and national measures are on a different footing, subsequent governments may retain more freedom to alter local law as needs change.

As the discussion of the advantages of monist and dualist systems suggests, both approaches also have disadvantages. The monist approach can be rigid. Moreover, circumventing the legislature can undermine democratic values. Since groups that can afford to lobby at the international level are often better heeled and more effectively organized than those that operate domestically, public choice theory suggests that a monist state’s legal regime will be more inclined than other systems to favor the rich over the poor and to favor concentrated business interests over the interests of dispersed consumers.⁸ Dualist systems do not have this problem, at least not to the same degree. However, they may have a harder time in negotiations because they cannot be fully trusted to implement the agreements they sign or ratify.⁹ And to the extent that their legislatures do fail to follow through and implement, they can find themselves in violation of international law.

Because of these problems, most systems are not entirely monist or dualist. For example, and as discussed further below, even monist systems do not generally regard the TRIPS Agreements as self-executing.¹⁰ By the same token, some dualist countries may give direct effect to human rights agreements.¹¹ Furthermore, many countries are neither monist nor dualist, but rather take a mixed approach to self-execution and examine a constellation of factors to determine whether a particular instrument (or provision) has a direct effect. The United States furnishes an example. Before relying on the Berne Convention to order a defendant to compensate a plaintiff, a US court would ask whether the Convention – or its three-step exceptions test – is self-executing.

Of course, the failure of a state to implement an agreement or regard it as self-executing will put that country in violation of international law. Nonetheless, even after a violation is found, local implementation will still be required. Experience suggests that this may not always be feasible. Consider, for example, the US-110(5) case in the World Trade Organization (WTO).¹² Although a WTO panel held that the United States had violated the three-step exceptions test of the TRIPS Agreement,¹³ the challenged exception remains good law in the United States. One reason may be that the provision was part of a legislative package: in exchange for extending the term of copyright generally, Congress enacted the challenged provision, which benefits certain access interests. To adhere to the WTO decision, the United States would have to unravel the sort of legislative compromise that is typical of democratic governance.¹⁴

II. The Mixed Approach

For countries that take a mixed approach, determining which international measures have direct effect is not an easy task. The experience of the United States is illustrative. The Supremacy Clause of the US Constitution specifies that “Treaties … shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby.”¹⁵ While that language implies that the Founders adopted a monist approach, it was clear by the early nineteenth century that the clause would not be interpreted that way. In an 1829 decision, Foster v. Neilson, Chief Justice Marshall limited its monist effect, reasoning that “[a] treaty is in its nature a contract between two nations, not a legislative act. It does not generally effect of itself, the object to be accomplished, especially so far as its operation is infra-territorial; but is carried into execution by the sovereign power of the respective parties to the instrument.”¹⁶

In that case, Chief Justice Marshall read the English language version of a land grant treaty between the United States and Spain as contractual and held that it was not self-executing. But as a subsequent case soon showed, Marshall’s analytic approach was unpredictable. The Supreme Court encountered the same treaty four years later, but this time it was presented with the equally authentic Spanish version. In that case, the Court found the agreement to be self-executing.¹⁷

This indeterminacy went on for many years. Typically, the treaties that reached the Supreme Court were found to be self-executing.¹⁸ For example, in a 1940 case, Bacardi v. Domenech,¹⁹ Chief Justice Hughes held that a Puerto Rico statute prohibiting use of certain trademarks was preempted by the national treatment provision of the Inter-American Trademark Convention, which the Court considered self-executing. Following the Court’s lead, in its 1956 decision, Vanity Fair v. Eaton, the Second Circuit held that the provision of the Paris Convention on unfair competition was self-executing.²⁰ But even though early Supreme Courts tended to interpret international agreements as having direct effect, lower courts did not understand the Court to have created a presumption in favor of self-execution. For instance, in Robertson v. General Electric, a patent case, the Fourth Circuit considered a provision of the instrument ending World War I that extended the time for acquiring a priority date under the Paris Convention. It found the provision was not self-executing.²¹

To a degree, matters changed with the Supreme Court’s 2008 decision in Medellín v. Texas,²² where the Court established a framework that focused on the instrument’s language and the events surrounding its adoption.²³ The question in that case was whether Texas was required to reexamine the conviction of a Mexican gang member for rape and murder. The defendant contended the conviction was defective because Texas had failed to notify Mexico of the defendant’s detention, as required by the Vienna Convention on Consular Relations and by a decision of the International Court of Justice (ICJ) holding that the Convention was meant to preempt a Texas procedural rule that barred further review.²⁴

To decide if the Vienna Convention or the ICJ decision had direct effect, Chief Justice Roberts began with the text of the agreement and looked for a “clear and express statement” that it was binding on courts.²⁵ Because the agreement to submit to the jurisdiction of the United Nations system used the phrase “undertakes to comply,” he reasoned that the agreement did not function as a directive to the judicial branch.²⁶ Rather, it called upon other arms of government to take specific actions.²⁷ To shore up this view, Roberts considered the negotiation and drafting history as “aids to … interpretation”²⁸ and analyzed the structure of the agreement as a whole. Under the UN Charter, a state aggrieved by noncompliance with an ICJ decision has, as its sole remedy, referral to the UN Security Council. Since the United States has a right to veto Security Council resolutions, the Justice reasoned it must not be automatically bound by the ICJ decisions.²⁹ Furthermore, he considered that the principal purpose of the Charter was to resolve disputes between governments, not to provide remedies to individuals like Medellín.³⁰ He also noted that in contrast to many of the agreements that have been found to be self-executing, this treaty raised questions that were primarily political in nature.³¹ Finally, he expressed willingness to give the government’s own interpretation of the non-binding nature of its obligations “great weight,”³² suggested that the negotiators’ understanding of what the United States had agreed to was relevant,³³ and noted that the post-ratification behavior of other member states indicated that none of them considered decisions of the ICJ to be binding.³⁴

In the last decade, the American Law Institute (ALI) revised its Restatement of Foreign Relations Law in light of Medellín and its progeny.³⁵ Although Medellín took a skeptical view of self-execution, and in remarks made out of court, Justice Scalia voiced even more adverse views to what he termed allowing foreigners to govern,³⁶ the ALI does not regard the United States as having moved into the dualist camp. Rather, it recommends courts consider whether the “treaty provision is sufficiently precise or obligatory to be suitable for direct application by the judiciary” and whether it was “designed to have immediate effect, as opposed to contemplating additional measures by the political branches.”³⁷ Additionally, the ALI suggests that courts should defer to Senate resolutions at the time when advice and consent were given and that they should consider whether implementing legislation is constitutionally required. Thus, treaties requiring the appropriation of money – which can only be accomplished by Congress – would never be regarded as self-executing.³⁸ In Comments, the Restatement goes on to caution that self-execution is distinct from the question whether the provisions of the treaty create rights and remedies.³⁹ That is, once a provision is found to be self-executing, it remains necessary to decide whether a litigant can obtain remediation for a violation of a commitment.

While not every state will follow the US analysis, it is not unlikely that states that take a mixed (or even a largely monist) approach will consider a similar set of issues: (1) the text of the agreement, including what it directs the parties to do and how precisely it delineates the obligations imposed; (2) the structure of the agreement as a whole, including whether it appears designed to have immediate effect and whether it contemplates action by other entities; (3) the negotiation history, to the extent it reveals the intentions of the parties; (4) the subject matter of the agreement and the branch of government responsible for (or constitutionally charged with) its oversight; (5) statements by the executive and legislative branches at the time the agreement was considered; and (6) the subsequent behavior of the other parties regarding the effect of the agreement.

C. The Status of Intellectual Property Agreements

The framework discussed above will have little application to strictly dualist or monist countries. But other nations will analyze international intellectual property instruments and provisions within them to determine their applicability in individual cases. How that analysis plays out will depend on the agreement in question.

I. The TRIPS Agreement

Although there has been debate on the issue,⁴⁰ the WTO Agreements are not generally regarded as self-executing.⁴¹ An examination of the TRIPS Agreement shows why. The Agreement starts with the admonition that “Members shall give effect to … this Agreement.”⁴² Although this language could have been inserted to deal with dualist regimes,⁴³ the provision goes on to state that members may provide more extensive protection and can “determine the most appropriate method of implementing the provisions.”⁴⁴ None of that language suggests that the parties intended the provisions to have direct effect.

The structure of the Agreement is consistent with this conclusion. It includes transition provisions for less and least developed countries, which appear designed to give those countries space to enact legislation suitable to their needs.⁴⁵ It also requires developed countries to provide technical and financial cooperation, including “assistance in the preparation of laws and regulations on the protection and enforcement of intellectual property.”⁴⁶ Moreover, it instructs the Council for TRIPS to “review the implementation of this Agreement.”⁴⁷ There is also nothing in the negotiation history that suggests it is self-executing, and it is difficult to see how certain provisions – such as measures that envision examination of advances to determine their suitability for protection⁴⁸ – could be effective without the legislature intervening to establish administrative agencies and procedures. Significantly, at the time TRIPS went into force, virtually all countries enacted implementing legislation. The United States even included in its implementation measure a statement that “[n]o provision of any of the Uruguay Round Agreements, nor the application of any such provision to any person or circumstance, that is inconsistent with any law of the United States shall have effect.”⁴⁹

II. The Paris Convention

The situation under the Paris Convention is not as clear as it is under TRIPS. In part, that is because the language of the Paris Convention has changed over the century and a half in which it has been in force, and in part, it is because each of the provisions of the Convention uses different language. Thus, each must be analyzed separately. For example, the provision on unfair competition reads: “The countries of the Union are bound to assure to nationals of [other Union] countries effective protection against unfair competition.”⁵⁰ The phrase “bound to assure” suggests a binding commitment. Moreover, “effective” is the type of standard with which courts are familiar. Furthermore, affording protection does not require the intervention of an administrative agency. Similarly, the telle quelle provision states that trademarks of one country “shall be accepted … as is in the other countries of the Union.”⁵¹ “Shall” implies immediate action. In contrast, the section on registration provides that the conditions for filing and registering “shall be determined in each country of the Union by its domestic legislation” and the well-known marks provision states that “[t]he countries of the Union undertake” to provide that protection.⁵² In both cases, it would appear up to the state to implement the provision.

As important, especially in countries that consider what the negotiators expected, GHC Bodenhausen, then Director-General of the World Intellectual Property Organization (WIPO), which administers the Paris Convention, produced a guide to the Stockholm Revision of 1967.⁵³ In it, he acknowledged that some provisions are addressed only to states and that others require national implementation.⁵⁴ But there is a long list of provisions that, he noted, “may directly govern the situation at issue,” depending on the position the relevant member state takes to self-execution.⁵⁵ Interestingly, the measures listed are not always the ones that recommend themselves as self-executing. To be sure, Bodenhausen included the telle quelle provision. However, he also included the provisions on well-known marks and the conditions of registration.⁵⁶

Courts in the United States have gone both ways on questions concerning the direct effect of the Paris Convention.⁵⁷ As noted earlier, in Vanity Fair, the Second Circuit held that the unfair competition provision was self-executing (but did not interpret the provision as providing the defendant with relief). Other courts disagree.⁵⁸ For example, in In re Rath, the Federal Circuit denied direct effect to the telle quelle provision, in a decision that suggested that it was the job of Congress to implement the Paris Convention – and not the court’s role to fix congressional failure to comply with international law.⁵⁹

The Supreme Court has yet to weigh in on the domestic effect of the Paris Convention, but a strong argument can be made that it is likely to agree with the Federal Circuit. As Justice Scalia’s comments about foreign rule suggest, views on self-execution have evolved and there is now considerably more skepticism about giving agreements direct effect than there was when the Paris Convention was promulgated and revised. Certainly, Medellín imposes a more stringent test than the one the Second Circuit applied in Vanity Fair. Moreover, experience under the Convention has demonstrated that the meaning of terms like “unfair competition” and “well known” vary quite significantly among jurisdictions.⁶⁰ Given these differences, the terms are not likely to be regarded as “sufficiently precise … to be suitable for direct application by the judiciary,” as required by the Restatement Foreign Relations Law.⁶¹

III. The Berne Convention

In the United States, the analysis of the Berne Convention is very different. The United States did not join Berne when it was first promulgated in the nineteenth century. By 1986, when it acceded, doubts about self-execution had grown (perhaps especially for an instrument closely associated with the droit d’auteur approach to protection with which the United States disagreed). Accordingly, in its implementation Act, Congress provided that “[t]he Convention for the Protection of Literary and Artistic Works, signed at Berne, Switzerland, on September 9, 1886, and all acts, protocols, and revisions thereto … are not self-executing under the Constitution and laws of the United States.”⁶² Thus, there are no US courts that have given direct effect to any provision of the Berne Convention.

For countries that do not have the clear guidance provided by the US Congress, the Berne Convention may be more easily considered self-executing than Paris. Because it eliminates formalities, there is no need for administrative support.⁶³ Moreover, the extension of protection to certain nationals of non-Berne Union countries suggests that at least some negotiators viewed authorial rights as natural, universal norms.⁶⁴ Nonetheless, like the Paris Convention, the Berne provisions read differently from one another and thus require a measure-by-measure analysis. For example, the provision on moral rights states that “[i]ndependently of the author’s economic rights, … the author shall have the right to claim authorship of the work and to object to any distortion, mutilation or other modification of, or other derogatory action in relation to, the said work, which would be prejudicial to his honor or reputation.”⁶⁵ (Again, a hint that human rights are at stake). But a provision on unauthorized uses provides that “[i]t shall be a matter for legislation in the countries of the Union to permit the reproduction … of articles … on current economic, political or religious topic ….”⁶⁶ Some provisions, such as the one on rebroadcast rights, are extremely specific and provide enough detail for judges to apply; others, such as the measure on adaptation, are fairly abstract.⁶⁷ As the German case about the Technical Information Library Hanover suggests, the three-step exceptions test is particularly difficult to parse. It leaves it to the legislation of each country to decide on exceptions, but specifies the limits of those exceptions in a way that lends itself to judicial action.⁶⁸

Additionally, the rapid changes in technologies relevant to the use of copyrighted works can make direct judicial implementation of much of the Berne Convention extremely difficult. For example, the Convention uses the term “communication to the public” multiple times.⁶⁹ Judges confronted with new technologies have had a hard a time parsing that phrase even when interpreting their own domestic law.⁷⁰ It is not insignificant that after the TRIPS Agreement essentially incorporated the Berne Convention with few updates, several new technology-related multinational agreements were adopted.⁷¹

IV. Free Trade Agreements

Many recent trade agreements include chapters that impose so-called TRIPS-plus obligations. These instruments arguably stand on a different footing from the TRIPS Agreement itself. To a large extent, they are directly aimed at clarifying ambiguities and open issues in TRIPS. Therefore, they tend to be extremely precise. For example, the TRIPS Agreement requires countries to protect data that is submitted for the clearance of pharmaceutical products for marketing purposes. The measure uses the terms “new chemical entities,” “considerable effort,” and “unfair commercial use,” none of which are defined.⁷² In addition, the provision fails to say how clearances based on approvals elsewhere should be treated. In contrast, many FTAs either omit these terms or define them. For example, the agreement between the United States and the Dominican Republic and Central America (CAFTA) eliminates the term “considerable effort,” and instead of “unfair commercial use” it imposes a requirement of five years of exclusivity. It deals with the issue of foreign approval by requiring each country to accord its own five years of protection, irrespective of its basis for allowing a pharmaceutical to be marketed.⁷³

Other examples abound. The TRIPS provision requiring patent protection for advances that involve an “inventive step” does not define that term, other than to say it is equivalent to “non-obvious.”⁷⁴ This has led to considerable controversy over whether new uses of old materials can be excluded from patentability. The agreement between the United States and Korea (KORUS) clears up that point by requiring “that patents shall be available for any new uses or methods of using a known product.”⁷⁵ As noted earlier, the Paris Convention and the TRIPS Agreement leave the meaning of a “well known” mark unclear; the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) references WIPO’s Joint Recommendation on Well Known Marks, which provides a detailed definition and expands trademark rights to include dilution protection.⁷⁶ As a textual and structural matter, these measures are therefore capable of having direct effect in all but the most dualist jurisdictions.

Admittedly, FTAs also contain language that suggests that implementation is required. For example, the intellectual property chapter of CAFTA and KORUS both state that “[e]ach Party shall, at a minimum, give effect to this Chapter.”⁷⁷ Similarly, the CPTPP provides that “[e]ach Party shall give effect to the provisions of this Chapter.”⁷⁸ That language suggests that legislative action is contemplated. However, the United States has developed an alternative to self-execution that may be equally effective at ensuring that the measures in an agreed instrument will be binding in the parties’ courts. That is, the United States chooses partners that it believes will implement the agreement, monitors how they plan to implement the agreement, and takes unilateral action when implementation fails to meet its expectations.⁷⁹

More important, the United States sometimes conditions its own implementation on a trading partner’s demonstration that it has already implemented the agreement to its satisfaction. For example, the US Act implementing CAFTA provides that: “At such time as the President determines that countries listed … have taken measures necessary to comply with the provisions of the Agreement that are to take effect on the date on which the Agreement enters into force, the President is authorized to provide for the Agreement to enter into force with respect to those countries.”⁸⁰ As Carlos Correa notes, the “certification” process entailed in making this determination not only ensures that right holders can seek relief in national courts but often also requires the other country to enact legislation that goes beyond the requirements of the agreement. In some cases, certification has led countries to provide right holders with more protection than is available to them under US law.⁸¹

D. Normative Assessment

As the previous part demonstrated, it is rare for states to consider the provisions of international intellectual property agreements to have direct effect. For the most part, they are minimum-standard regimes and afford member states leeway to implement the obligations in ways compatible with their own legal systems. But the international community is faced with many new challenges. Moreover, some sectors have expressed an appetite for further harmonization.⁸² It is therefore worth asking whether the global regime would benefit if future instruments were more often regarded as self-executing. Presumably, that would entail drafting measures that meet the standards of precision, clarity, and ease of application that most countries appear to require and that balance relevant interests in ways that potential members are willing to accept.

A case can certainly be made for this approach. It would be especially helpful to developing countries. Rather than work through all the complexities entailed in crafting exclusive rights regimes, those countries could simply adopt the systems constructed by their more experienced treaty partners. Moreover, pre-commitment may be especially attractive for intellectual property. Because these rights purport to promise long-term benefits at the expense of short-term costs, it might be difficult for poor countries to implement laws that may, over time, encourage local innovation, improve productivity, increase income, and yield social welfare gains, but which require the voting public to endure immediate sacrifices in the form of higher prices and reduced access.⁸³ Indeed, these considerations may be among the reasons why the Bacardi Court was persuaded that the Inter-American Trademark Convention – which involved the United States, Peru, Paraguay, Panama, Honduras, Haiti, Guatemala, Cuba, and Colombia – was self-executing.

Self-execution can also be to the advantage of developed countries. It ensures that they receive the benefits they expect from the trade-offs made during the negotiation process. For example, in the Uruguay Round, developed countries understood that if they opened their markets to imports, the manufacture of knowledge-intensive products would move to countries with lower labor costs. In exchange, they sought to capture returns on the innovations embedded in these products with stronger intellectual property protection.⁸⁴ Self-execution is also a direct way to overcome the problems of territorially limited rights. Harmonizing the level of protection available worldwide facilitates cross-border research, value chain production, and international distribution of creative products.⁸⁵ Furthermore, it enhances the incentives available to creators and aggregates the demand for products that appeal to small segments of dispersed populations. As concerns about developing and delivering vaccines and treatments to deal with COVID-19 have shown, nations are deeply interconnected, which makes an international approach highly desirable.

To be sure, negotiators may find that they must use some indefinite terms or measures in order to leave room for future developments. But even here, there are advantages. The dialogue generated when multiple courts consider the same open question is what US proceduralists call “percolation.” They view this process as a useful way to arrive at the best approach.⁸⁶ An example is the way in which Australia learned from US decisions on patenting products and phenomena of nature: it considered US caselaw and improved on it.⁸⁷ Or, as Christine Farley noted in connection with the “unfair competition” provision of the Inter-American Trademark Convention, the competing views of a multiplicity of courts might have led to a more refined understanding of what that cause of action ought to protect.⁸⁸

Perhaps the best way to convince countries to regard intellectual property agreements as self-executing is to argue that creators enjoy a fundamental right to control their intellectual efforts; that because these individual rights should not be subject to majority rule, they must have direct effect.⁸⁹ The categorization of intellectual property as fundamental is supported by several human rights conventions. For example, the Universal Declaration of Human Rights provides that “[e]veryone has the right to the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.”⁹⁰ And as Laurence Helfer observed, the European Court of Human Rights characterizes exclusive rights as property and protects certain aspects under the European Convention on Human Rights.⁹¹ Significantly, we saw a flavor of that approach in the Berne Convention’s extension of protection to certain authors in non-Berne countries and in its moral rights provision.

One problem is that there are other values – including free expression, health, and the “right freely to participate in the cultural life of the community, to enjoy the arts and to share in scientific advancement” – that are also regarded as fundamental.⁹² Since these values clash, they have been the subject of intense academic, legislative, and judicial debate. Plausibly, however, balancing them should also be handled at the international level. In fact, that may have been the thinking of the German court when it gave the Berne Convention’s three-step exceptions provision direct effect in the Hanover library case. And this may also be a reason why the United Nations appointed a Special Rapporteur in the field of cultural rights to submit a report on how fundamental values regarding intellectual property should be balanced.⁹³ Among other things, she noted that protection for authors does not necessarily require the recognition of exclusive rights.⁹⁴

That said, it would be difficult to persuade most countries that intellectual property agreements should have direct effect on the ground that they protect human rights. There is disagreement as to whether intellectual property rights are human rights.⁹⁵ Even if every country were to decide that some are, countries may not agree on which of those rights are fundamental or on how to strike the appropriate balance among them.⁹⁶ Thus, there are some countries in which free expression trumps moral rights, or patent protection gives way to health concerns, or privacy interests alter remedies for infringement.⁹⁷ Because countries see these values as situated at the core of their national identities, there is little likelihood that they would regard a regime that takes a different view as self-executing. To see the point, consider the evolution in the United States from the Supremacy Clause, to Foster, Medellín, and Scalia’s concern about rule by foreigners. According to Gráinne de Búrca, that development was repeated in the European Union as it matured. It began with monist impulses, but what emerged over time (in the same year as in Medellín) was Kadi⁹⁸ and the rejection of Security Council resolutions on the ground that they violated EU norms.⁹⁹

To be sure, self-execution can also be justified on the ground that it promotes global innovation. Seen that way, deep harmonization, which for many of the reasons expressed above is unlikely to lead to self-execution, is not necessary. Rather, if direct effect is desired, negotiators could concentrate on measures crucial to coordinating the worldwide intellectual property system to facilitate collaborative research and worldwide transactions. Nations could then retain flexibility in other spheres to further their own interests and values.

But even in this rather limited domain, there are forceful arguments against self-execution. As the structure of existing instruments suggests, a careful weighing of values and accurate identification of measures crucial to integration are unlikely to occur during multilateral treaty negotiations. For the most part, these agreements recognize the interests of right holders and, as interpreted by the WTO, leave members with little room to protect other values.¹⁰⁰ Many of these agreements are negotiated in secret; that negotiators consistently ignore recommendations to add user rights to these instruments suggests that public choice theorists are correct.¹⁰¹ Because right holders are better organized and funded, their demands overwhelm the dispersed interests of the public. Opportunity at the implementation stage for democratic engagement concerning the substantive level of protection accorded right holders is therefore critical. For intellectual property, accountability is particularly important, because infringement is largely self-policing – and as Jessica Litman succinctly stated, “[p]eople don’t obey laws that they don’t believe in.”¹⁰²

Self-execution is also problematic because countries are in very different positions economically, culturally, and technologically. TRIPS was sold on the claim that stronger protection would push developing countries to the creative frontier. However, over twenty-five years of experience has demonstrated that this was true only for some countries. For the rest, TRIPS – even as locally implemented – is proving to be an obstacle to development.¹⁰³ There are likewise differences among developed countries. Each nation’s legal regime reflects its own industrial needs and creative requirements. As Susy Frankel argued, the preferences of small economies can diverge from those of larger markets.¹⁰⁴ In addition, intellectual property laws are part of complex legal systems that differ greatly from one country to another. For example, some states use antitrust law to cabin overreaching by intellectual property owners.¹⁰⁵ Others may safeguard competition in their intellectual property laws. Similarly, the availability of discovery (i.e. legal procedures to obtain information from adversaries and other parties) can shape both patent and trade secrecy law. Tailoring may therefore be unavoidable.

Self-execution is also hazardous because needs change over time and international lawmaking is not as responsive as domestic courts and legislatures. Nor have international negotiators always demonstrated the foresight to deal with contingencies. The original version of the compulsory licensing provision in TRIPS is illustrative.¹⁰⁶ Despite the existence of countries that lag far behind others technologically, the Agreement initially failed to account for the possibility that particular nations may lack the capacity to manufacture pharmaceuticals and would therefore be unable to make use of the flexibility to award compulsory licenses to protect public health.¹⁰⁷ It took more than five years for the WTO to recognize the problem and another four years to solve it.¹⁰⁸ In order to reach agreement to allow one country to manufacture for another, some members opted out as potential importers – a decision that, in light of COVID-19, may prove to have tragic consequences.¹⁰⁹ Finally, changes in technologies can require adaptations in the law. Even when negotiators manage to react in a timely fashion, making new law at the international level may not be as successful as allowing states to experiment first. As Graeme Dinwoodie pointed out, WIPO’s solution to the digital distribution of copyrighted words was not a great success.¹¹⁰

E. Alternative Approaches

Even if promoting self-execution is not normatively desirable (or practicable), there are ways to fulfill the goal of coordinating the international intellectual property system. One approach comes courtesy of Chief Justice Marshall, who, prior to Foster, had stressed the role of statutory interpretation. Thus, in Murray v. The Schooner Charming Betsy, he opined that “an act of Congress ought never to be construed to violate the law of nations if any other possible construction remains.”¹¹¹ Because intellectual property is largely statutory even in common law countries, every nation with a doctrine akin to Charming Betsy gives judges considerable scope to fulfil its commitments.¹¹²

It is, however, questionable whether Charming Betsy remains good law, at least in the United States. While the doctrine allows judges to adhere to international law and yet tailor the domestic regime to local conditions, the outcome is much like self-execution in that it permits the legislature to avoid accountability. It is thus not surprising that when the Federal Circuit in Rath refused to give direct effect to the telle quelle provision of the Paris Convention, it also declined to rely on Charming Betsy.¹¹³ Significantly, the Supreme Court has occasionally ignored the doctrine.¹¹⁴ And then-Judge (now Supreme Court Justice) Kavanaugh has suggested that the doctrine did not survive Medellín.¹¹⁵

Megaregional agreements offer a somewhat different path to coordination. One reason that many countries may have balked at according direct effect to the WTO Agreement is that the parties did not have an equal voice in the Uruguay Round. As Susan Sell tells the story of TRIPS, twelve US-based multinational corporations held enormous sway over the US delegation, which pursued a divide-and-conquer strategy to undermine the leverage developing countries were mustering to counterbalance US demands.¹¹⁶ Free trade agreements can be equally problematic. Although there are often fewer parties, one party may have considerable control over the others.¹¹⁷ But megaregionals can present a sweet spot. The CPTPP is an example. The negotiating parties included developing countries, emerging economies, and a few that were highly developed. Among the latter, some enjoyed large internal markets; others relied heavy on import and export. Because the group was relatively small, the negotiating dynamics allowed the parties to identify positions that were true compromises (as the leaked texts suggest, this was particularly true after the United States withdrew from what had been the TPP). The final agreement includes TRIPS-plus provisions. Nonetheless, it is better balanced than the demandeurs of strong protection wanted or that many observers expected.¹¹⁸ Even though megaregionals still require implementation, it may be easier to convince parties to execute their obligations in an agreement produced though genuine give-and-take than one based on asymmetric bargaining power.¹¹⁹ Certainly, negotiators who know one another well are more able to anticipate the domestic reception of their agreements and can avoid including provisions that will not be implemented by one or more of the other parties.

Another possibility is to adapt the certification procedure we saw in connection with the CAFTA Agreement. That process, too, was asymmetric in that, as Correa noted, the United States used its clout to require of its trading partners more than it required of itself. However, one can imagine a system of reciprocal certification, where each party proposes implementing measures it believes will be acceptable to its legislature and then submits them for the approval of the other parties, with final implementation conditioned on joint approval. Although such a procedure is cumbersome, it might force the parties to focus harder on provisions that are necessary for coordination, rather than on demands that serve only the interests of right holders. The process would also help negotiators appreciate the problems that other parties face, such as inadequate competition laws or public health concerns.

Although top-down mandates through international agreement are one way to integrate legal regimes, it is also possible to coordinate from the bottom up, through the efforts of regulatory authorities and adjudicators. As Anne-Marie Slaughter and others have noted, in many fields, transnational networks of government officials have cooperated to produce effective solutions to jointly held problems.¹²⁰ Examples include the Basel Committee on Banking Supervision, the International Organization of Securities Commissions, and the International Competition Network.¹²¹ For patent law, efforts along these lines have been underway for some time.¹²² In 1983, the United States Patent and Trademark Office, the European Patent Office, and the Japan Patent Office created the Trilateral “to contribute to an increasingly efficient worldwide patent system.”¹²³ Joined by the Korean Intellectual Property Office and the National Intellectual Property Administration in China, the system now operates as IP5.¹²⁴ Much of its work is directed at improving the quality and speed of examination. However, the group also maintains lists of differing practices, categorized by whether the difference is attributable to office traditions, judicial decisions, or legislation.¹²⁵ While rule by administrative agency carries its own democracy deficit,¹²⁶ IP5 has no authority to create law or international obligations. However, where coordination is hampered by the participants’ own examination practices rather than legal obligation, IP5 can effectuate immediate change. Moreover, because these offices cater to right holders but exist, at least in theory, to protect the public domain, the group should be in a position to provide impartial advice to governments on how to change domestic laws in ways that improve global integration.

Judicial participation in this effort is more recent. At one time, the territoriality of intellectual property rights led courts to entertain multinational cases on a jurisdiction-by-jurisdiction basis.¹²⁷ But as the costs of piecemeal adjudication increased, judges began to consider the full geographic scope of these disputes and to develop tools for coordinating differing domestic legal regimes. The main approach is through private international law: rules that identify which court is most appropriate to hear a particular case and that determine the applicable law.¹²⁸ If sufficiently predictable, these rules allow the participants in multinational transactions to conform their conduct to the relevant law; if sufficiently supple, they allow countries to further their national interests and values effectively. Of course, to avoid over- or under-regulation, the rules themselves must be coordinated.¹²⁹ The ALI, the Max Planck Institute, groups in Asia, and the International Law Association have encouraged that effort, with recommendations on how courts should handle jurisdiction, choice of law, and enforcement questions.¹³⁰

Somewhat ironically, now that courts have agreed to hear transnational cases, they have occasionally achieved what centuries of international negotiations failed to accomplish: substantive harmonization. For example, in a 2017 decision, Eli Lilly v. Actavis UK, Lord Neuberger developed a view of claim interpretation that he found common to the laws of the UK, France, Italy, and Spain.¹³¹ In Unwired Planet Intl. Ltd. v. Huawei Techs. Co. Ltd., a UK appellate court imposed a worldwide royalty for the use of standard essential patents subject to a commitment to license under fair, reasonable, and nondiscriminatory (FRAND) terms.¹³² Cases involving secondary liability for copyright and trademark infringements on the internet similarly provide courts with opportunities to develop law for disputes involving intermediaries operating in multiple jurisdictions.¹³³

It is, of course, debatable whether substantive lawmaking through dispute resolution is superior to negotiating treaties. As with international instruments that are self-executing, there can be a democratic deficit. Depending on the terms of judicial appointments there may well be less control over adjudicators than over negotiators. And depending on the quality of the litigators, judges may be less versed in the relevant technology, less knowledgeable about the impact of particular rules on the creative community or the public interest, and too focused on the concerns of the litigants to consider broader issues. At the same time, however, adjudication is more nimble than international lawmaking and more responsive to domestic agendas. Furthermore, judicial decisions can usually be overruled. Since multiple courts will often consider the same issues, solutions will percolate – and that may be better than negotiation for finding the best solution to universally vexing problems.

F. Conclusion

Self-execution is a matter of national rather than international law. While some countries regard international agreements as having direct effect, most do not consider international intellectual property agreements to be self-executing. This means that negotiators cannot assume that national law will be implemented in a manner that is entirely consistent with agreements as drafted. Furthermore, leaving matters to the legislature can delay and interfere with the coordination (or harmonization) that many international agreements seek to achieve. For intellectual property law, this is particularly problematic, because the globalization of information, production, and manufacturing suggests that a high degree of integration is desirable.

Nonetheless, there are many good reasons to preserve states’ sovereign authority to implement international law for themselves. Legislative intervention creates a degree of accountability that is largely missing in the international sphere. Intellectual property law involves balancing proprietary interests against public concerns. Because countries differ dramatically along the lines of culture, economics, technological capacity, and fundamental principles, it would be difficult to strike the same balance everywhere. Thus, consensus can often be achieved only through the use of “constructive ambiguities”– language that is unsuitable to direct application by judges but which allows for legislative tailoring to local needs, capabilities, and values. In addition, technologies and needs change over time. International lawmaking is too prone to capture, too shortsighted, and too cumbersome to deal effectively with such problems.