One often hears concerns that international judges may run amok, actively expanding international law in ways not explicitly based on state consent. There is a lot more rhetoric than reality to these concerns. In practice, all judges make law. Indeed, a key reason states delegate authority to international adjudicators is because governments know that agreements are incomplete and that legal commitments will need to be interpreted and gaps filled in.Footnote 1 Yet some international courts and tribunals (ICs) do extend international legal rules in ways that constrain national sovereignty beyond what states had intended. For governments and commentators concerned with such expansive judicial lawmaking, the European Court of Justice (ECJ) represents the problem in the extreme.
Decades of ECJ rulings transformed the Treaty of Rome into a de facto constitution for the European Community (EC).Footnote 2 The ECJ achieved this remarkable result by expanding the reach and scope of EC law and by enabling litigants to use the European legal system to promote key EC objectives. For example, the ECJ has condemned Germany's exclusion of women from combat-related military roles, Ireland's censoring of a student brochure locating abortion clinics in Britain, and the EU's implementation of a United Nations (UN) program freezing the assets of alleged terrorist supporters.Footnote 3 Proponents of effective international adjudication see the ECJ as an exemplar for other ICs.Footnote 4 Sovereigntists, by contrast, criticize the ECJ as riding roughshod over national autonomy,Footnote 5 suggesting a cautionary lesson in why states should avoid creating independent ICs.
This article reinvestigates the history and early decisions of the ECJ to explore how context shapes international judicial lawmaking. We do so by comparing the ECJ with its largely unknown cousin—the Andean Tribunal of Justice (ATJ). In 1969, five countries in the Andean region of South America imported from Europe the idea of building a common market through supranational institutions—but without an IC. Andean governments later concluded that the absence of a court undermined the uniform interpretation of and compliance with Andean law.Footnote 6 In 1984, they established the ATJ, explicitly modeling its design on the ECJ.
The ATJ is one of ten ECJ copies, and arguably the most successful.Footnote 7 It is the third-most active IC, having issued nearly 2,000 decisions through 2009. The ATJ has fewer rulings than the European Court of Human Rights and the ECJ, but far more than the World Trade Organization (WTO) dispute settlement system, the International Court of Justice, or Latin America's other ICs—the Inter-American Court of Human Rights and the Central American Court of Justice.Footnote 8 The ATJ has also, as we demonstrate elsewhere, influenced both Andean Community law and the behavior of its member states.Footnote 9
In this article, we focus on the ATJ's strikingly more modest judicial lawmaking as compared to the ECJ. The ATJ has incorporated the key ECJ doctrines of the supremacy and direct effect of community law. But unlike the ECJ, the ATJ overwhelmingly respects the discretion that Andean law leaves to state actors. Even when provided with opportunities to adopt broad interpretations, the ATJ is generally reluctant to expand its authority or the reach of Andean rules in ways that constrain national sovereignty.
The ECJ/ATJ's comparison reminds one of the “nature” versus “nurture” debate in child rearing. Is it the genes or the environment that shapes how an actor develops? Most international relations and international law scholars favor the nature side of this debate. They assume that judges are “hardwired” to increase their power. It is only limited opportunities or the fear of being sanctioned that inhibits judges' natural tendency to expand their authority. By contrast, nurture-based explanations assume that ICs typically apply the law in fairly straightforward and circumscribed ways. Expansionist lawmaking—which we define as broadening the reach or scope of international law and of IC authority at the expense of national political discretion—requires, according to this view, special nurturing and encouragement.
We investigate the relative influences of nature versus nurture by comparing the first twenty-five years of ATJ and ECJ decision making, periods when the regional organizations that created both courts had smaller memberships,Footnote 10 nascent supranational institutions, and lower trade volumes. We show that the ATJ has generally refrained from the sort of expansionist lawmaking that is the hallmark of its European cousin, and we offer a new explanation of the ECJ's extraordinary lawmaking.
Our analysis has three wider implications. First, we call into question the notion that international judges are power-seeking by nature. This assumption lies at the heart of international relations theories that focus on how specific design features of ICs facilitate or hinder expansionist IC lawmaking. Our analysis of two identically designed ICs suggests that scholars have placed too much emphasis on formal institutional rules and state control of judges.
Second, we argue that ICs are more likely to be expansionist where substate and societal interlocutors—government officials, national judiciaries, advocacy networks, administrative agencies, or the public more generally—encourage IC lawmaking and facilitate compliance with IC rulings. In emphasizing the preferences of these actors, we implicitly reject the dominant international relations approach that views the preferences of states as the primary factor shaping IC decision making.
A third broad policy implication of our analysis is that politically independent ICs can, contrary to sovereigntists' fears, be active but not activist, interpreting and completing international contracts while avoiding expansive judicial lawmaking that compromises state sovereignty.
In the next section, we define expansionist international judicial lawmaking and situate our analysis in existing scholarship on ICs. We show that the ECJ and ATJ are nearly identically designed institutions, creating a natural experiment to test nature versus nurture theories of judicial lawmaking. In the second section, we document the key trends in ATJ decision making. We show that Andean judges have copied several key ECJ doctrines. But we also show that Andean judges are, unlike their European counterparts, generally deferential to governments with the exception of the domain of intellectual property law—a puzzle we explain in the third section by comparing how different environmental factors have influenced ATJ and ECJ lawmaking. In the final section, we analyze the implications of this comparison for understanding how environmental and political context shapes IC lawmaking. Inasmuch as our data on the ATJ is new and largely unknown, the appendix explains our methodological choices and provides a guide for future research.
Theories of Expansionist Judicial Lawmaking by International Courts
Judging often involves lawmaking.Footnote 11 When faced with controversies involving indeterminate rules, judges must interpret those rules to resolve the parties' disputes. Most IC lawmaking fills gaps in existing legal rules, or, in the terminology of rational choice, completes incomplete contracts. Our focus, however, is on expansionist international judicial lawmaking, which we define as occurring when ICs identify new legal obligations or constraints not found in treaty texts or supported by the intentions of their drafters, and when these obligations or constraints narrow states' discretion. Typically, expansionist lawmaking significantly broadens the substantive reach of international law and asserts the IC's authority to interpret that law.Footnote 12 We do not include in this definition a requirement that international judges rule against governments. ICs can expand international law without ruling against governments; conversely, they can rule against governments without expanding international law.
Most international relations theories begin from the premise that judges are power-seeking by nature, using empowerment to explain why ICs expand judicial authority at the expense of state discretion. This assumption also underpins most claims about how institutional design features contribute to expansionist lawmaking.
There are two variants of this expansionist-by-nature theory. One set of scholars focus on how IC access rules shape opportunities to litigate, and thereby opportunities for expansionist lawmaking. Posner and Yoo claim that where states must consent to submit disputes to international adjudication, ICs will strive to make their rulings politically palatable to governments. By contrast, courts that do not require state consent to hear a particular dispute (ICs with compulsory jurisdiction) and those that allow private parties to initiate litigation are more independent and therefore more likely to issue expansionist rulings that displease states.Footnote 13 Other scholars concur that compulsory jurisdiction and private access lead to more active and independent courts that are more likely to expand the reach and scope of international law.Footnote 14 For example, Keohane, Moravcsik, and Slaughter argue that “transnational dispute resolution seems to have an inherently more expansionary character”; noting that “a steady flow of cases … allows a court to become an actor on the legal and political stage” and that “a court gains political capital from a growing caseload by demonstrably performing a needed function.”Footnote 15
A second set of scholars focus on the tools states use to thwart expansionist lawmaking. Building on the insights of principal-agent theory, these scholars argue that ICs are less expansionist where states can easily revise international rules, where judicial terms are short or subject to reappointment, and where design features allow states to punish judges for decisions they dislike.Footnote 16 Ginsburg adds that judicial lawmaking should increase with the number of parties to an agreement and the difficulty of amending the treaty or overruling judges, and decrease with the ease of exit from the regime.Footnote 17
The ECJ–ATJ comparison presents us with a natural experiment that holds constant these design features to explore how environmental factors and political context shapes judicial behavior. Andean governments replicated the ECJ's design, hoping to copy its success in enhancing respect for common market rules.Footnote 18 The ATJ's architects were well aware of the ECJ's history of expansionist lawmaking. Although neither the Cartagena Agreement nor the Treaty Establishing the Andean Tribunal explicitly stated that Andean law has direct effect in national legal systems or that it is supreme to national law, negotiators assumed that the Andean system would replicate these judge-made EC doctrines.Footnote 19 But the drafters also sought to hem in expansionist lawmaking by allowing only member states to bring noncompliance complaints, and by explicitly stating that the ATJ should not delve into the facts of preliminary rulings. The ATJ's faithful respect of these limitations contributed to the dearth of Andean cases in its early years. In 1996, member states relaxed both of these restrictions, so that after 1996, the ATJ resembles the ECJ's structure even more closely.Footnote 20
Both the ECJ and the ATJ provide the same mechanisms for challenging government behavior that conflicts with community rules. Both systems contain a noncompliance procedure that enables private actors and member states to inform the community secretariats about rule violations. The secretariats investigate the alleged violation, and, if necessary, file noncompliance complaints with the ECJ/ATJ. Both systems also contain a preliminary ruling mechanism in which private actors invoke community law in domestic litigation and national judges refer questions of interpretation to the ECJ/ATJ. Domestic courts then apply the ECJ/ATJ ruling to the case at hand. In both systems, sanctions can be imposed if a state fails to comply with the court's ruling.Footnote 21 As of 1996, the Andean system has one feature the ECJ lacks. If the Secretariat refuses to raise a noncompliance suit, a private actor can bring the suit directly to the ATJ.
Although the Andean Pact and the EC initially had different substantive goals,Footnote 22 the Andean Pact's Cartagena Agreement contained the same key elements of the Treaty of Rome that the ECJ used to expand its authority. Both treaties prohibit member states from creating new barriers to trade, require national treatment for products from other member states, and allow supranational bodies to adopt directly applicable secondary legislation.Footnote 23
Both regional systems also provide governments with the same formal opportunities to sanction expansionist lawmaking: judges are appointed to fixed, renewable terms;Footnote 24 member states can redefine each court's jurisdiction by unanimously amending its founding charter;Footnote 25 and they may in theory withdraw from both the EC and Andean Community.Footnote 26 As a practical matter, exit is unlikely in Europe, whereas states have exited from the Andean regime (for reasons unrelated to the ATJ).Footnote 27
Notwithstanding these nearly identical design features, the next section will show that the ATJ, unlike the ECJ, has not become an expansionist lawmaker.
Does Active Mean Activist? ECJ and ATJ Lawmaking Compared
This section compares ATJ and ECJ lawmaking over different twenty-five-year periods when each IC was establishing its legal and political authority. Both courts were similarly active during these periods, with a steadily increasing diet of preliminary references. Between 1960 and 1985, the ECJ issued 305 noncompliance decisions and 1,808 preliminary rulings (an average of 86.1 cases per year), whereas the ATJ, with a geographically and demographically smaller region to oversee, issued 85 noncompliance decisions and 1,338 preliminary rulings between 1984 and 2007 (an average of 71.5 per year).Footnote 28 Notwithstanding the large number of cases, European and Andean governments were only weakly committed to economic and legal integration during these periods. Pro-integration litigants urged the courts to overcome these political impediments with teleological interpretations that furthered the treaties' integrationist goals. The ECJ responded eagerly to these entreaties by engaging in expansionist lawmaking, whereas the ATJ was more circumspect.
We give greater attention to the ATJ, an IC whose activities have generated surprisingly little scholarship. We coded all 1,338 ATJ preliminary rulings available on the Andean Community Web site from the tribunal's founding through 2007. Where the ATJ broke new legal ground, we analyzed its decisions in depth. We focused on preliminary rulings because most ECJ lawmaking occurred in such rulings. We also expected ATJ lawmaking to occur in preliminary rulings. The General Secretariat raises most noncompliance cases, and we expected the Secretariat to be sensitive to state concerns and to allege noncompliance only where violations were manifest. However, our review of important preliminary rulings led us to a number of noncompliance decisions, several of which we analyze below. We also conducted more than forty interviews with lawyers, judges, and government officials in Peru, Ecuador, and Colombia.
Our account of ECJ lawmaking relies heavily on Weiler's seminal legal analysis, and on Burley and Mattli's political analysis, both of which argue that legal actors are power-seeking by nature.Footnote 29 Following Weiler, we group our study into time periods that correspond to varying levels of support for regional integration to capture dynamics between political and legal integration.
The Foundational Period: The ATJ During the Andean Pact (1984–95)
ECJ lawmaking was most expansive during what Weiler labels as the court's “foundational period” from 1962 to the mid-1970s. There was significant political turmoil in Europe during these years. Consumed by internal problems and divergent objectives, European governments seemingly turned away from supranationalism and the goal of building a common market.Footnote 30 Weiler argues that the ECJ responded to this political impasse by being remarkably activist, building through law what supporters of integration could not achieve through politics. In this period, the ECJ established the core doctrines of regional integration—the direct effect, supremacy, and pre-emption of community law over national law, and the implied powers of supranational institutions.Footnote 31
The ATJ's genesis shares some elements of the ECJ's experience, but not others. As in Europe, early on a handful of motivated litigants turned to the court to promote the objectives inscribed in the Cartagena Agreement.Footnote 32 The ATJ responded by mimicking the ECJ. Even though the ATJ's first case did not involve an actual or potential conflict between Andean and national law, the tribunal used its first preliminary ruling to declare the supremacy and pre-emptive power of Andean law (the direct effect of Andean law was assumed). It cited the ECJ's Costa v. Enel decision establishing the supremacy of European law and its Simmenthal ruling that obligated national courts to enforce the primacy of community law.Footnote 33 But whereas the ECJ had framed its analysis in constitutional terms, boldly asserting that “the Community constitutes a new legal order of international law for the benefit of which the states have limited their sovereign rights,”Footnote 34 the ATJ stressed the functional necessity and implicit state support for supremacy.Footnote 35
The ATJ's second preliminary ruling raised the question of what becomes of a national law that conflicts with Andean rules yet remains on the books. In the Simmenthal case, the ECJ had resolved this conflict with an unyielding rule: “Every national court must … apply Community law in its entirety and … must accordingly set aside any provision of national law which may conflict with it, whether prior or subsequent to the Community rule.”Footnote 36 The ATJ, although citing the Simmenthal decision, ultimately interpreted Andean law more modestly as “firstly settl[ing] for [a rule of] preferential application.” National law remains valid and on the books but is not applied. If the conflicting Andean law is later modified, however, the latent national law becomes compatible with community law.Footnote 37
A comparison of the two courts' pre-emption doctrines provides another example of the ATJ's greater deference to state autonomy. Without any textual support in the Treaty of Rome, the ECJ asserted that in fields such as the common commercial policy, Community powers were exclusive and precluded member states from legislating regardless of whether their actions conflicted with community law. In other areas regulated by European law, the ECJ concluded that member states could not legislate even where there is no EC rule on point. Not only do these rulings diminish state discretion, it is the ECJ that determines whether a particular EC rule or policy space is exclusive and pre-eminent.Footnote 38
In striking contrast, the pre-emptive force of Andean law is far more modest. In an early ruling, the ATJ announced the principle of complemento indispensable: even in areas where Andean law clearly governs, member states may enact domestic laws necessary to implement a community rule provided that the laws do not obstruct or nullify that rule.Footnote 39 Stated differently, whereas the ECJ both implied powers not explicitly delegated to the EC and asserted pre-emptive authority even where EC law was silent, the ATJ has not implied powers for the Andean Community, and it has allowed states to retain the power to legislate with the sole exception of national laws that directly conflict with extant community rules.
In a 1990 decision, the ATJ further cabined the pre-emption doctrine. Although citing ECJ case law to reaffirm that Andean laws can displace national rules, the ATJ also stressed that integration is a gradual, incremental process that limits the extent to which Andean rules pre-empt national authority: “Especially, when dealing with complex and vast issues, such as intellectual property, … it seems logical that many of these diverse issues, even if they have to be a matter of common regulation in the beginning, are still within the competence of the national legislator for an indefinite time until they are effectively covered by the community norms.”Footnote 40
Other ATJ rulings during the foundational period exhibited even greater deference. Consider, for example, the multiyear Reynolds Aluminum dispute. The ATJ first rejected the case in 1987 because private actors at the time lacked standing to raise noncompliance suits.Footnote 41 The case reappeared in 1990 as a preliminary ruling with facts strikingly similar to the ECJ's landmark Van Gend en Loos decision of 1962. As in Van Gend en Loos, the plaintiff claimed that the common market treaty created an immediate bar to increasing tariffs on imports from other member states. However, Andean governments had previously adopted, via secondary legislation, a Free Trade Program that allowed broad exemptions from regional free trade rules. The plaintiff nevertheless argued that the Cartagena Agreement should be read as freezing existing tariffs for products exempted by the Program. Colombia countered that the Free Trade Program should be interpreted as effectively amending the treaty.Footnote 42 The ATJ sided with the government, ruling that member states had free reign with respect to products included on the Free Trade Program's list of exceptions.Footnote 43 The ECJ's Van Gend en Loos ruling made the Treaty of Rome a constitutional document that created immediately enforceable rights for private actors and higher-order legal obligations for governments. By contrast, the ATJ in effect interpreted the Cartagena Agreement not as a fixed constitutional charter but only as a starting point for integration, one that member states could amend fairly easily.Footnote 44
The ATJ has also given more deference to national judges than its European counterpart. In a later iteration of the Reynolds Aluminum case, the tribunal considered which of three potentially applicable legal regimes governed the dispute. The ATJ defined the obligations under each regime, one of which gave states complete control over certain products “as long as exception and reserve lists exist.”Footnote 45 But rather than selecting the relevant rules from among the three options, the ATJ left that key question for national courts to answer.Footnote 46
In striking contrast, the ECJ has a long tradition of reserving to itself the authority to determine the extent of national exceptions to free trade rules.Footnote 47 Moreover, the ECJ directs national judges regarding the application of EC rules. Judge Federico Mancini acknowledged that the ECJ formally refrains from “overstepping” its authority to rule on whether a national law violates European law. But “having clarified the meaning of the relevant Community measure, the court usually indicated the extent to which a certain type of national legislation can be regarded as compatible with that measure. The national judge is thus led hand in hand as far as the door; crossing the threshold is his job, but now a job no harder than child's play.”Footnote 48 As previously explained, Andean governments had sought to limit the ATJ from similarly guiding the application of national law by initially instructing the tribunal not to consider the facts of referred cases, and later providing that it could only consider the facts “when essential for the requested interpretation.”Footnote 49
There is thus some textual support for the ATJ's reticence. But the extent of its deference to national governments and judges is surprising. There were a number of ways that the ATJ could have used the aluminum cases to expand the reach and scope of Andean law, and to create a larger role for itself in overseeing Andean legal integration. The ATJ could have found that Andean industrial programs preempted national discretion. It could have declared the exemption list a part of Andean law and itself determined which goods were included on the list. Or it could have allowed national judges to decide individual cases subject to well-defined Andean guidelines for determining the validity of government exemptions. Instead, the ATJ handed to national judges both the task of ascertaining which regime governed and whether a product was included on the list of exceptions, effectively removing itself from determining whether member states were complying with Andean free trade rules.
ATJ Lawmaking During a Period of Relative Political Harmony: 1996–2004
Would the ATJ become more of an expansionist lawmaker if there were greater political consensus among Andean member states over building a common market? The ATJ faced just such an environment beginning in the mid-1990s. During the earlier Andean Pact period, actors both within and across the member states disagreed vehemently about the import substitution strategy at the center of Andean integration efforts. By the 1990s, the import substitution theory had come into disrepute. Andean governments decisively embraced the neoliberal Washington Consensus, adopting sweeping legislative and institutional reforms at the national and regional levels with the renewed goal of building a common market.Footnote 50
In 1993, the Free Trade Zone became fully effective for Bolivia, Colombia, Ecuador, and Venezuela. In 1994, Andean countries agreed to a common external tariff. In 1997, they adopted the Sucre Protocol, which resembled the EC's Single European Act in that it committed member states (including Peru) to finally complete the common market.Footnote 51 Member states also reformed Andean institutions, replacing the ineffectual Andean Junta with a General Secretariat, increasing the size of the Secretariat's budget, and appointing a cadre of young lawyers eager to use those enhanced resources to promote regional integration.Footnote 52
Andean integration still faced challenges, such as Peru's withdrawal from the Common External Tariff and its delay until 2003 in ratifying the Sucre Protocol. But using the Andean legal system to promote compliance with Andean rules was not among them. To the contrary, revisions of the tribunal's jurisdiction in the 1996 Cochabamba Protocol suggested that member states wanted the ATJ to be more assertive in interpreting and enforcing Andean law.Footnote 53
We observe a marked change around this time in ATJ rulings concerning Andean intellectual property law. New Andean secondary legislation (Decisions 311, 313, and 344) expanded the region's intellectual property rules, which led to an increase in applications to register trademarks and patents with national intellectual property agencies. Although detailed, the decisions contained ambiguous provisions and lacunae that the agencies and private litigants sought to clarify in numerous references to the ATJ.
Andean judges responded to these repeated requests by interpreting regional intellectual property rules in ways not dictated by the texts of the Decisions. Some of the tribunal's decisions were relatively modest extensions of established legal principles. For example, the ATJ created judge-made doctrines concerning trademark coexistence agreements and pharmaceutical trademarks that instructed agency officials to balance the private rights of intellectual property owners against the need to prevent consumer confusion and promote public health.Footnote 54
Other decisions were far bolder and substantially cabined national government discretion. The first ATJ noncompliance ruling in 1996, which upheld a General Secretariat challenge to a bilateral agreement between Ecuador and the United States, provides an apt illustration. Adopted under pressure from the U.S. pharmaceutical industry, the bilateral treaty protected drugs in the production “pipeline” before an Andean law recognizing patented medicines took effect.Footnote 55 Ecuador defended the agreement by citing a clause authorizing member states, “in their own domestic legislation or under international treaties, [to] strengthen the intellectual property rights provided for in this Decision.”Footnote 56 The ATJ rejected Ecuador's reliance on this seemingly unambiguous text. It reasoned that Andean intellectual property rules were “one of the fundamental pillars for economic harmonization” and “an essential instrument” to “promote well-balanced, harmonious, and equal development” in the region. It then adopted a teleological interpretation of the word “strengthen” that allowed member states to enhance—but not contradict—Andean rules and precluded them from invoking bilateral treaties “as a reason to validate noncompliance with a prior Community obligation.”Footnote 57
In the same year, the ATJ was equally bold in a case referred by the Colombian Constitutional Court concerning the relationship between Andean intellectual property legislation and the Paris Convention—a multilateral patent and trademark treaty. In this ruling, the ATJ addressed the contested issue of national sovereignty. Whereas the ECJ found that EC members had transferred entire domains of sovereign authority to the European level by ratifying the Treaty of Rome, the ATJ reasoned that member states retained control over the pace and scope of the transfer via Andean secondary legislation. The ATJ was unequivocal in stating that sovereignty had in fact been shifted to the Andean Community. But it also reasoned that the more extensive the secondary legislation on point, the more “exclusive authority of the member countries” had been “sovereignly transferred” to the Andean level. Turning to the case before it, the ATJ relied on the existence of extensive and detailed regional intellectual property legislation to rule that significant national authority had been delegated to the community in that domain. The ATJ then asserted its exclusive authority to interpret Andean intellectual property rules and to define their relationship to other intellectual property treaties, making it clear that member states cannot deviate from “the common interests” of the community except by acting together through Andean institutions.Footnote 58
The ATJ extended this approach again in a 2001 decision involving second use patents, another intellectual property right sought by foreign pharmaceutical companies.Footnote 59 The plaintiff argued that the WTO's Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) required second-use patents. Adopting a position arguably bolder than analogous European rulings, the ATJ concluded that Andean law is supreme even over multilateral treaties such as those adopted in the WTO.Footnote 60
In litigation unrelated to intellectual property issues, by contrast, the shift in ATJ decision making was less pronounced. The court became more scrupulous in finding violations of clear Andean rules, in enforcing Andean procedures, and in reviewing the facts of preliminary references.Footnote 61 But it did not issue expansionist rulings that limited state discretion or pushed member states toward deeper levels of legal integration.
The Colombian alcohol monopoly cases illustrate the ATJ's approach during this period. In May 1991, Ecuador complained to the Andean Junta, arguing that municipal rules in Colombia impeded competition and discriminated against Ecuadorian alcohol products. The Junta settled the dispute out of court, but it reappeared in 1996 in a complaint by Venezuela. This time the Junta adopted Resolución 453, a legally binding decision that found fault with Colombian municipalities, and that required Colombia to fix the problem.Footnote 62 When Colombia ignored the resolution, the General Secretariat (which by then had replaced the Junta) filed a noncompliance suit with the ATJ.Footnote 63
Meanwhile, in 1997, a private citizen had asked the Colombian Constitutional Court to review the state's alcohol monopoly. One of the plaintiff's arguments was that the monopoly was incompatible with the Cartagena Agreement as applied in Resolución 453. In its May 1998 judgment, the Colombian court declined to enforce the resolution. It reasoned that, unlike human rights treaties that have quasi-constitutional status in Colombia,Footnote 64 Andean laws were equivalent to domestic legislation. Because such laws “and the Constitution do not share the same hierarchy, nor are [they] an intermediate legal source between the Constitution and ordinary domestic laws, … contradictions between a domestic law and Andean community law will not have as a consequence the non-execution of the [domestic] law.” The court also noted that Andean Community law has “primacy” over conflicting national law—a concept that the Constitutional Court interpreted to mean that Andean law “displaces but does not abrogate or render non-executable” conflicting national legislation.Footnote 65
The ATJ issued its noncompliance judgment six months later. The situation was remarkably similar to the landmark Costa v. Enel decision, in which the ECJ asserted the supremacy of EC law. But there was a key difference. Costa v. Enel had been simultaneously referred to the Italian Constitutional Court and the ECJ. The constitutional court ruled first, finding that European law was inapplicable to the case at hand and not supreme over national law.Footnote 66 The ECJ, by contrast, found that European law is supreme and that national courts were obliged to apply it instead of conflicting national law, but that the Italian law at issue did not conflict with EC law.Footnote 67 The ATJ was less politically fortunate in that the case involved a Colombian practice that had been challenged by two member states and condemned by a General Secretariat Resolution. Thus, whereas the ECJ found Italian law compatible with EC law, in the alcohol case, the law and facts were such that the ATJ could not avoid ruling against Colombia.
In the noncompliance decision, the ATJ went out of its way to agree with the Colombian Constitutional Court that there was no inherent conflict between the Colombian alcohol monopoly and Andean law. But the implementation of the monopoly was a different matter. Although the national government had tried to introduce a common system of alcohol taxation, local policies that created barriers to trade persisted. Because of these municipal practices, Colombia remained in violation of Andean law.
The same litigant who had filed the Constitutional Court challenge later asked another Colombian court—the Consejo de Estado—to review the municipal policies. To nullify municipal acts, especially after the Constitutional Court ruling, would have been a radical step. National courts in Europe had taken just such a step when they embraced the supremacy of European law. But as of 1999, no court in the Andean Community had shown much willingness to overturn domestic statutes or doctrines to help enforce Andean law. The Consejo referred the case to the ATJ as required under Andean law. The ATJ reiterated that Colombia was obligated to modify practices that conflicted with Andean law.Footnote 68 Yet the ATJ refused to extend its earlier reliance on the European supremacy doctrine and, as the ECJ had done, instruct national judges to do whatever was necessary to give effect to community law. Instead, the ATJ simply declared (again) what Andean law required without asking national judges to help it to enforce that law. Although the ATJ did not explain its reluctance to follow the ECJ, its ruling is difficult to divorce from the very real concern that national judges might find that they lacked the legal authority or the political will to heed the ATJ's request.Footnote 69
Another case further illustrates the ATJ's unwillingness to expand its authority or the reach of Andean law during this period. In 2003, a former staff attorney of the General Secretariat raised a case involving Andean rules regulating the use of pesticides in the hopes of eliciting a ruling that would push Colombia to increase national protections. The ATJ found that regional rules required states to improve the quality of health in rural and agricultural industries, and it described the administrative procedures for registering imported and domestic pesticides set forth in the relevant Andean Decisions. But the tribunal refused to go beyond the letter of those Decisions or to guide national authorities in exercising their discretion so as to achieve the Andean Community's broader health goals.Footnote 70
The ATJ in Times of Crisis: 2005 to the Present
In the mid-2000s, the Andean Community entered a new period of crisis triggered by political and economic schisms between the member states. Colombia and Peru remained true to market liberalization, but Venezuelan President Hugo Chavez advocated a different approach. The election of Evo Morales and Rafael Correa as the leaders of Bolivia and Ecuador in 2006 created a three-country block against new market-oriented initiatives. Policy disagreements led Venezuela to withdraw from the Andean Community in 2006, taking with it a significant portion of the organization's budget.
This political crisis has sapped community resources, slowing the filing of noncompliance suits by the General Secretariat and curtailing efforts to build political support for Andean law within national judiciaries.Footnote 71 It has not, however, noticeably affected ATJ preliminary rulings. The ATJ continues to issue mostly narrow, technical rulings concerning Andean intellectual property law. But the ATJ has also reaffirmed established precedents, even when doing so required confronting governments. In 2005, for example, a private litigant challenged a Colombian tariff that had previously been rejected by the General Secretariat and condemned by the ATJ in a noncompliance case. Extending its earlier ruling regarding the Paris Convention, the tribunal reasoned that regulating the common external tariff was one of the powers that, like intellectual property, had been transferred to the community level. The relevant Andean law permitted countries to derogate from common rules in cases of emergency. But only the General Secretariat could determine whether such derogations satisfied Andean rules. Since the Secretariat had nullified the Colombian tariff, the national court was required to do so as well.Footnote 72 In 2007, the ATJ asserted the supremacy, direct applicability, and pre-emptive power of Andean transportation rules, unequivocally stating “there is only one Andean policy in relation to passengers, equipment and transportation.”Footnote 73
Comparing Twenty-Five Years of Expansionist Judicial Lawmaking by the ATJ and ECJ
The above analysis encompasses ATJ lawmaking across the universe of ATJ preliminary rulings and important noncompliance decisions. To summarize our findings: early ATJ rulings mostly emulated key ECJ doctrines, making Andean law directly effective and supreme to national law and pre-empting national governments from enacting conflicting domestic legislation. The ATJ stressed that governments had agreed to these developments, which by all appearances is true. The ATJ also enforced clear Andean laws and required national judges to give priority to those laws, particularly in the area of intellectual property. In these ways, the ATJ generally followed in the ECJ's footsteps as a builder and strong defender of a supranational legal order.
In numerous other ways, however, the ATJ has exhibited far more deference to national authority. It has reiterated that Andean legal commitments are a product of member state consent and has scrupulously respected the discretion that Andean secondary law reserves to national governments. The ATJ has also eschewed opportunities to expand the reach and scope of Andean law. While the ATJ formally treats the Cartagena Agreement as higher-order law, the tribunal readily defers to collective decisions concerning the scope and pace of Andean integration when interpreting the treaty. This approach gives member states broad discretion to revise Andean rules to reflect the waxing and waning of their collective commitment to integration.
In contrast to the ATJ, the ECJ is pervasively expansionist. Stone Sweet has examined ECJ decision making involving three substantive areas of European law—the environment, sex discrimination, and free movement of goods. He concludes that “through its rulings [the ECJ] has acted—relatively systematically—to reduce the domain of national autonomy, to expand supranational modes of governance to the detriment of intergovernmental modes, and to create the conditions for the gradual Europeanization of national administration and judging.”Footnote 74
Analyzing ATJ and ECJ lawmaking in discrete historical periods revealed another important difference between the two ICs. Weiler observed that in Europe, legal supranationalism advanced furthest when political steps toward integration were in retreat.Footnote 75 The ATJ also developed its key doctrines during its foundational period, but the tribunal's lawmaking tends to reflect rather than counterbalance political support for integration.
Explaining the Divergent Supply of Expansionist Lawmaking
What explains these striking differences in ECJ/ATJ lawmaking? The ECJ and ATJ possessed the same potent combination of wide access rules, self-interested litigants, swelling dockets, and repeat player legal entrepreneurs who sought out cases to promote legal integration.Footnote 76 In other words, both ICs contained the ingredients needed for judicial empowerment. But unlike the ECJ, the ATJ often thwarted litigant efforts to use Andean law to dismantle national policies contrary to their economic interests. The ATJ's refusal to help litigants achieve their goals sets up a vicious circle that inhibits the filing of additional cases that might have expanded community law. It is a striking fact that of the 1,338 ATJ preliminary rulings between 1984 and 2007, only thirty-five involve subjects other than intellectual property.Footnote 77 The abstract and repetitive nature of the ATJ's legal analysis also contributes to a sense among lawyers that preliminary rulings have little practical benefit.
Nature-based theories would seek to explain these differences by looking for judicial design features and political constraints that inhibit judges from issuing expansionist rulings in response to litigant demands. But there are no clear differences between the two regional courts on these scores.
One cannot, for example, explain the ATJ's behavior by the judicial appointment choices of governments. In the Andes, as in Europe, governments nominated supranational judges whom they expected to have some expertise in the legal issues the court would hear. Andean governments submit a ranked list of three nominees, which, like their European counterparts, typically hail from government, academia, the foreign service, and the national judiciary. Other member states choose from this list. None of our interviews suggested that governments seek out nominees with particular judicial philosophies or approaches to their job.Footnote 78
Once appointed, ATJ judges appear no more vulnerable to retribution than are ECJ judges. Both courts issue rulings unanimously. In neither context have governments attempted to remove judges from office, although they have elected new judges at the end of their terms. Judges rotate in Europe and in the Andes mostly for professional reasons and because new governments want to exercise their power of appointment.Footnote 79
Nor did we find evidence that the ATJ is institutionally more vulnerable than the ECJ. Both European and Andean judges have been criticized for decisions governments did not like. But member states have only rarely overturned unwanted legal rulings.Footnote 80 The ATJ decreased the likelihood of such outright reversals by interpreting the Cartagena Agreement in light of Andean secondary legislation. By contrast, there have been numerous political challenges to the ECJ and national judiciaries regarding the supremacy of European law.Footnote 81 Yet member states have repeatedly extended both courts' jurisdiction.Footnote 82
We next consider whether distinctive features of the Andean context inhibit ATJ expansionist lawmaking. One might argue that Latin American judges are more reluctant to challenge political authority as compared to those in European countries, where constitutional democracy is more firmly entrenched. Yet it is both too sweeping and ahistorical to say that judges in Europe are independent and expansionist, whereas judges in Latin America are politically penetrated and timid. Few judges in Europe in the 1960s and 1970s engaged in judicial review, let alone expansionist lawmaking.Footnote 83
More recently, across Latin America judges are increasingly confronting political power by expanding national and international law, notwithstanding episodes of judicial repression.Footnote 84 In Argentina, Chile, Colombia, and Peru, for example, judges have investigated violations of international human rights law and government and military corruption at the highest levels.Footnote 85 Numerous decisions of the Inter-American Court of Human Rights and the Colombian Constitutional Court demonstrate that neither court avoids expansionist lawmaking or confronting political power in high stakes cases.Footnote 86 This reality only deepens our puzzle.
Some observers may focus on the extent to which member states are stable liberal democracies.Footnote 87 Andean countries score reasonably well on measurements of democracy.Footnote 88 But these countries have also experienced significant economic and political crises.Footnote 89 The ATJ's greater willingness to let member states determine the scope and pace of integration arguably reflects the region's unstable political and economic climate. We should not, however, forget that the ECJ too undertook expansionist lawmaking in the face of waning political support for integration, huge student and labor protests in 1968, and mass labor unrest in the 1970s.
A third potential explanation is the differing levels of intra-community trade in the two regions. Intra-European trade was relatively small at the EC's founding, constituting less than 3 percent of the member states' gross domestic product.Footnote 90 Although perhaps not as economically central to European countries in the 1960s, trade among EC member states has always constituted a larger proportion of the total trade as compared to intra-regional trade in the Andes. In 1958, between 23.7 and 27.5 percent of French, Italian, and German trade went to other European member states. By 1972, intra-European trade constituted between 40 and 60 percent of total member states' trade.Footnote 91 Trade among Andean countries increased over the period we study, beginning from a low of 3 percent of total trade during the Andean Pact, peaking at about 10 percent of total trade in 2002. For Bolivia and Ecuador, common market trade is relatively more important (between 15 and 24 percent).Footnote 92 Nevertheless, internal trade has never been as significant in the Andean region as in Europe.
What conclusions can be drawn from these differences? Stone Sweet and Brunnell built on neofunctionalist theory to suggest that increasing trade flows create a demand by private actors for ECJ litigation, which in turn contributes to judicial lawmaking.Footnote 93 Pitarakis and Tridimas respond that Stone Sweet and Brunnell's data cannot determine if trade drives litigation, or litigation drives trade.Footnote 94 In a later analysis, Stone Sweet observes that in Europe, internal trade levels, secondary legislation, and ECJ preliminary rulings rise in tandem, making it impossible to know if legislation drives litigation and trade, or trade drives legislation and litigation.Footnote 95 Scholars are only beginning to untangle these linkages. Early results involving a far more precise coding of ECJ preliminary rulings suggest that ECJ litigation contributes to greater levels of European trade.Footnote 96 This finding tracks what scholars have found with respect to the WTO, namely that law influences trade levels.Footnote 97
Although we lack precise comparative data, we know that the EC has more secondary legislation to promote the free movement of goods, services, capital, and people as compared to the Andean Community. The existence of secondary legislation clearly affects litigation. For example, there is ECJ litigation involving social security benefits for migrant workers, agricultural policies, and common customs classifications for which there is no Andean legislative analogue and thus no Andean litigation. Andean legislation also sometimes contains loopholes that make it hard for litigants to challenge conflicting national policies.Footnote 98 These differences in secondary legislation influence our dependent variable, since the ATJ is more willing to assert its authority in domains where there are extensive Andean rules.Footnote 99 But is trade the driving factor, or does the extent of secondary legislation determine the volume of trade and thus the extent of expansive lawmaking?
It is likely that economic realities influence the extent to which European and Andean governments invest in building a regional economic community. Nonetheless, we find it implausible that different trade levels explain the divergent lawmaking by the two regional courts. It is far from clear, for example, that European governments in the 1960s and 1970s saw economic integration as vital to their economies. One could easily argue that European governments saw maintaining national barriers to trade as necessary for the larger project of protecting the embedded liberal compromise of postwar European societies.Footnote 100
We therefore proceed to interrogate the power-seeking assumption of nature-based theories. When we shed this assumption, the ATJ's prudent decision to let member states set the pace of integration seems rather unsurprising. What is surprising is that the ECJ repeatedly chose to be an engine for regional integration even when European governments, by all appearances, had largely abandoned the integration project. Jettisoning the assumption that international judges are power-seeking by nature thus calls into question existing explanations of ECJ lawmaking.
Recent historical scholarship has revised previously settled understandings of early European integration by exploring the role of diplomats, lawyers, and judges who supported the ECJ's legal revolution. This new scholarship explains how pro-integration activists wrested the subject of European law from traditional international law scholars, who believed that legal lacunae should be interpreted to preserve state discretion.Footnote 101 Activists founded scholarly and practitioner associations that united pro-integration advocates and worked to educate the larger legal community about the ECJ's view of European law. Pro-integration lawyers then used their positions as judges and court officials to identify test cases and facilitate preliminary references. Pro-integration activists also regularly met with ECJ judges, European Commission members, and pro-integration government officials at events organized by Euro-law associations. At these meetings, which often focused on key unresolved doctrinal issues, ECJ judges learned how different legal interpretations would be received by academics, judges, and officials in key government ministries, and these actors, in turn, encouraged ECJ judges to be bold.Footnote 102
New historical assessments of key decisions in the ECJ's legal revolution reveal the decisive influence of supportive advocacy movements. For example, with respect to the issue of European law supremacy, the European Commission had identified a safer interpretive route grounded in the traditional rules of public international law. These rules indicated that, because the EC treaty did not directly grant legal rights to individuals, European law should not be given direct effect. The rules further indicated that the only way to make European law supreme to national law was to find a basis for supremacy in national constitutions. This traditional international law approach was unsatisfactory, however, because the constitutions of a few member states did not recognize treaties as trumping national law.
The problem became more acute after the Italian Constitutional Court ruled that European law was not supreme to subsequent national legislation. After the Italian ruling, key actors heatedly debated how the ECJ could find a legal basis for EC law supremacy.Footnote 103 Members of the Commission and judges on the court initially skirted the issue. The ECJ asserted the direct effect of European law in its 1962 Van Gend en Loos ruling, but it did not claim that European law was supreme to conflicting national law. ECJ Advocat General Maurice Lagrange also hesitated to claim a basis for supremacy in the Treaty of Rome, suggesting instead that national constitutional interpretation should evolve to recognize EC law primacy. A 1964 meeting of the German Academic Society for European Law focused intensively on this topic. Representatives of the EC legal secretariat, the ECJ, and the German Ministry of Justice were in attendance. The meeting occurred after the Italian Constitutional Court had rejected the supremacy of European law, after Advocat General Lagrange had delivered his reasoned opinion in the ECJ's famous Costa v. Enel case, but before the ECJ had issued its ruling. At the meeting, Hans Peter Ipsen, a highly regarded pro-integration legal scholar, made an impassioned plea that the ECJ go beyond what Lagrange had suggested and assert that the Treaty of Rome itself implied that European law was supreme to national law.Footnote 104 Issued shortly thereafter, the Costa decision closely tracked Ipsen's argument.
Once the ECJ ruled, members of pro-integration advocacy movements wrote scholarly articles championing the ECJ's far-reaching decision, often shedding their affiliations with the integration project so that it appeared that the authors were writing as a random assortment of individual lawyers and scholars.Footnote 105 These actors also helped to persuade sympathetic national judges to issue path-breaking rulings supporting ECJ jurisprudence.
The ATJ, by contrast, has not received the support of legal advocacy networks or national judiciaries. In our interviews in the region, we looked for Andean analogues of the academics, attorneys, and interest groups who spearheaded European legal integration. We identified a few legal entrepreneurs within the Andean system, but no infrastructure or institutions to connect them to each other or to a larger pro-integration movement. Unlike efforts to promote human rights in Latin America, we found no evidence that the Andean integration project has attracted the support of national judges, nongovernmental organizations (NGOs), and government officials.Footnote 106
The ATJ does, however, enjoy strong support of national intellectual property agencies. We explore this puzzle in depth elsewhere.Footnote 107 These agencies were restructured in the early 1990s, as part of a larger reform of the administrative state spurred by the Washington Consensus.Footnote 108 The new intellectual property agencies were staffed with Western-trained lawyers who received the financial and material support of the World Intellectual Property Organization and international and regional financial institutions. Andean administrators built an informal regional network to share information and develop criteria to resolve common legal problems. We document how and why national intellectual property agencies looked to the ATJ to interpret ambiguous provisions of Andean law and to provide support in their conflicts with foreign firms and national governments. The result is a symbiotic relationship in which the ATJ develops detailed procedures to help administrators to resolve disputes between private parties, and the agencies, in turn, influence Andean legal doctrines to reflect their own policy preferences.Footnote 109 Unlike the ECJ, however, the ATJ has not seen its support extend beyond the domain of intellectual property.
Conclusion
International judicial decision making that clarifies ambiguities and fills gaps in treaties is an inherent part of judging. But expansionist lawmaking by international courts and tribunals, in which judges broaden the reach of international legal rules and their authority at the expense of state discretion, may well be the exception, not the rule. Our comparison of the ATJ and ECJ suggests that such lawmaking requires active nurturing by substate and societal actors, which may—but need not—include government officials.
A few conclusions follow from this analysis. First, our comparison of two identically structured ICs suggests that too much weight has been placed on the design of ICs as an explanation of judicial behavior. To be sure, IC design is a relevant factor. For example, both the ECJ and the ATJ are highly active ICs because private litigants can trigger national court references in litigation involving EC or Andean Community legislation. But our study demonstrates that ICs can be active without being activist. We also find no evidence that differences in state sanctioning mechanisms meaningfully predict when IC expansionist lawmaking is more or less likely to occur.Footnote 110
Second, although judicial power seeking and self-interest may be necessary for expansionist lawmaking, they are not sufficient. In reopening the question of whether power-seeking explains the ECJ's behavior and in identifying how advocacy movements shaped ECJ lawmaking, our study echoes the findings of Sikkink, Keck, Simmons, and others who find that advocacy networks help build support for international human rights law.Footnote 111 We add to these accounts that actors within states—national judges and national administrators—also provide ICs with constituencies that encourage judicial lawmaking and promote compliance with expansionist rulings. If this explanation is correct, we should expect ICs to supply expansionist decisions when they anticipate that domestic advocacy networks and compliance constituencies will pressure states to change their treaty-incompatible conduct. The support of these domestic interlocutors facilitates IC lawmaking, while the prospect that states may exit from an IC's jurisdiction provides a counterweight.Footnote 112 The reality that advocacy movements and substate actors can encourage expansionist IC lawmaking may be of concern to sovereigntists and defenders of national executive power. Yet we have also shown that the ATJ hewed closely to the mandate given it by the member states, revealing that IC lawmaking does not necessarily compromise national sovereignty.
Finally, our analysis suggests a starting point for future studies of how environmental and political factors shape IC decision making. We compared the behavior of two identically designed ICs in different regions and the behavior of those courts across time. One could also compare ICs within the same region, such as lawmaking by the ATJ in comparison to the Inter-American Court of Human Rights. We expect IC behavior to be influenced by whether substate actors support the larger objectives of the international law regime in which these courts are embedded, and whether the international rules in question are seen as serving larger social and political purposes in addition to promoting the narrow interests of the lawyers and litigants.
It may be that more ICs will resemble the ATJ than the ECJ in their penchant for expansionist lawmaking. Whether or not we are right in this speculation, the essential point is that scholars can gain insights by comparing IC behavior across political contexts, and thus investigating when, why, and to what end ICs become expansionist lawmakers and active enforcers of international rules.
Appendix: Research Note
Given the paucity of literature on the ATJ, we had to make blind choices when we began our investigation. We focused on ATJ preliminary rulings because those decisions have been the most important venue for expansionist ECJ lawmaking, and because preliminary references from national courts are the largest source of ATJ cases. Our coding revealed that 97 percent of ATJ preliminary rulings concern intellectual property issues. Our coding also revealed significant cross-national variation in reference rates to the ATJ. But our interviews suggest that cross-national variation primarily reflects differences in demand for intellectual property rights. In light of the subject matter, concentration of preliminary references, and the fact that cross-national variation in references is highly affected by variations in intellectual property applications, we do not believe that regression analysis of reference patterns would yield useful insights for our dependant variable. Elsewhere, we discuss the patterns of references in detail.Footnote 113
Unlike in Europe, Andean governments rarely offer observations in preliminary reference cases, and ATJ rulings generally provide scant clues as to how national judges should apply the law to the facts. Instead, preliminary rulings mostly contain abstract interpretations of specific provisions of Andean law. For all of these reasons, one cannot conduct the type of analysis undertaken by Carrubba, Gabel, and Hankla to probe for state influence over judicial decision making in preliminary rulings.Footnote 114 Government and General Secretariat positions are, however, discussed in noncompliance cases.
ATJ decisions are reported and referenced by the Secretariat and ATJ using the case number. The letters connote the type of procedure (for example, IP = interpretaciones prejudiciales, N = nulidad, AI = Acciones de Incumpliemiento). The last number (for example, 87) corresponds to the year the case was filed (for example, 1987). Access to ATJ rulings has evolved over time. We found the General Secretariat to be a more reliable source, and we relied on their Web site to access decisions.Footnote 115 The number of cases available for download, however, is slightly fewer than the number of cases reported in the ATJ's statistics. There is also a significant mismatch with the data reported on the Integrated Data Base of Trade Disputes in Latin America and the Caribbean, which has its own metric of what qualifies as a dispute.Footnote 116 At the time of our research, ATJ rulings could be downloaded as Word documents. Anticipating that this system may change, we note section headings rather than the page number of the rulings we discuss in detail. Judges are not entirely consistent in how they use section headings or in whether they number their conclusions, which is why sometimes we refer to letters, and other times to numbers.
