Communitizing Transnational Regulatory Concerns

I. Introduction

In this Article we argue that the conventional command-and-control theories related to regulatory requirements fall away in the face of transnational regulatory governance. We therefore seek a new explanatory model that prioritizes and recognizes the discursive nature of international regulatory proceedings. Importantly, we contend that the way regulatory problems are discussed, and thus the way they are communitized, affects the way parties ultimately reconcile regulatory disputes. We conclude that collaborative and less confrontational regulatory exchanges tend to be significantly more successful than formal and inquisitive ones.

Globalization has added a transnational layer of complication to the conventional regulatory dyad of regulator (rule-maker) and regulatee (rule‑taker). Suppose that the Australian government regulates the proper labeling and packaging of tobacco products. It requires all cigarettes marketed in Australia to be packaged in a plain, standardized form without any branding, such as logos and images; however, the package must include a serious health warning.1 Also suppose that Ukrainian cigarette exporters refuse to comply with the labeling (packaging) requirement on the grounds that such a draconian rule is an unnecessary trade restriction and therefore violates the World Trade Organization’s (WTO) Agreement on Technical Barriers to Trade (TBT). Here, two regulatory relationships interface with each other. First, Ukrainian cigarette exporters, the rule-takers, are supposed to abide by the tobacco labeling requirement enacted by Australia, the rule-maker. Second, Australia, as rule-taker, must comply with WTO norms, such as the TBT Agreement, a rule-maker. Ukraine, on behalf of its cigarette exporters, may question Australia’s tobacco labeling regulation by invoking Australia’s obligations under the TBT.2

Thus, the conventional view of the regulatory process as a two-way game envisions two separate governance structures—domestic and international. Typically, both states and private businesses qua rule-takers will comply with rules if, and only if, behaving (compliance) is in their interest, or if violating (non-compliance) would entail certain disutilities, such as a penalty.3 Applying the conventional model to the aforementioned hypothesis, Ukraine cigarette exporters would likely elect to defy the Australian labeling requirement if compliance is too costly to them and their business partners, such as Philip Morris, who built cigarette factories in Ukraine. Likewise, in the event that the WTO invalidates Australia’s labeling requirement, Australia would likely agree to repeal the measure if a potential sanction is too painful to bear.

While the conventional view offers a simple yet powerful heuristic on regulatory governance, that view fails to embrace more nuanced aspects of regulatory relationships. The conventional preoccupation with law as coercion (enforcement) tends to dismiss important non-material dimensions of WTO norms, such as their rhetorical power. In other words, the two-way game regulatory model does not capture various discursive pathways provided by a regulatory intermediary. A regulatory intermediary, such as the TBT Committee, connects the domestic and international regulatory spheres and helps build a “compliance community” by hosting regulatory dialogue in a transnational setting.4

The ultimate rationale of the TBT Agreement is to achieve ostensibly conflicting goals of free trade and regulatory autonomy in a non-entropic, harmonious manner.5 A number of “procedural” disciplines under the TBT Agreement, such as transparency, notification, and reason giving, are geared toward these dual goals.6 These procedural disciplines are meta-regulation in that they regulate how each WTO member regulates in the domestic sphere. In this regard, the TBT Committee, in particular through specific trade concerns (STCs), provides a communal forum in which WTO members conduct peer review on other members’ domestic regulations in accordance with these procedural disciplines.7 The operational logic of the TBT Committee is not so much coercion as it is discourse by design. In other words, the TBT Committee provides both exporting countries that usually share the goal of free trade and importing countries that often represent the value of regulatory autonomy with unique discursive opportunities to engage in learning, perspective taking, mutual persuasion, and eventually the development of non‑binary regulatory solutions.8 The logic of discourse can be warranted particularly by the indeterminate nature of non-tariff barriers (NTBs), which restrict the flow of goods and services through means other than tariffs, and the inevitable governance gap they create. The reciprocal bargaining model prevalent in conventional trade negotiation simply does not work in tackling these NTBs, which are fraught with irreconcilable socio-cultural differences among trading nations.9 In this regard, the STCs mechanism under the TBT Committee has been broadly popular among WTO members in terms of frequency of its invocation.10 Yet its success, as well as the meaning of such success, still remains unarticulated. This Article attempts to both conceptualize and measure the success of the STCs mechanism.

In the preceding example, the TBT Committee may provide a forum in which Ukraine, on behalf of its domestic regulatees, cigarette exporters, may demand a justification from Australia for the latter’s labeling regulation.11 Ukraine may argue that the Australian labeling regulation lacks the scientific justification required under the TBT Agreement.12 It is, however, important to note that the WTO, in and of itself, is not a “world government.” The WTO cannot simply legislate away regulatory heterogeneity. Instead, peer review under a regulatory intermediary (the TBT Committee) promotes regulatory dialogue between regulators and regulatees. Such regulatory dialogue tends to result in both countries familiarizing themselves with each other’s regulatory regime. This mutually enhanced awareness between regulators and regulatees as well as the consequent regulatory cooperation is a necessary, if not sufficient, condition for any regulatory solution.

It remains an empirical, albeit often extremely complicated, question whether such a regulatory intermediary—the TBT Committee—is “effective,” in other words, whether it actually delivers any practical solution for the original regulatory dilemma. Confronted by these measurement challenges, one possible option might be to assess a regulatory intermediary’s effectiveness through various interactions (inquiries and responses) within the TBT Committee and their consequences. Note that these consequences do not necessarily mean any dramatic behavioral changes toward compliance. They may also include expressions of satisfaction, explicit or implicit, from inquirers. If the regulating country’s response indeed satisfies the original inquirer, there will be no more discussion or dispute. In this regard, the TBT Committee is a regulatory “clearinghouse” that channels and manages TBT-related inquiries that exist in the form of both a demand for clarification and a complaint about alleged illegality. Unlike adjudication through the WTO tribunal, the ultimate goal of the TBT Committee is to liquidate those demands or complainants by providing a discursive forum. As those STCs are circulated, shared, and peer‑reviewed (“communitized”) within the TBT Committee among whichever WTO members are interested,13 WTO members may familiarize themselves with the regulatory practices of their trading partners, better appreciate similarities and differences in regulatory approach on the same issue area, and therefore expand both shared regulatory grounds and a zone of regulatory tolerance. In this sense, the TBT Committee may function as a critical dispute “prevention” mechanism.

Against this background, our study investigates the TBT Committee data spanning the past twenty years. This study aims to achieve two main goals. First, it conceptualizes the WTO’s TBT Committee as an intermediary that assists regulatory compliance by facilitating regulatory dialogue in a transnational setting, crisscrossing domestic and international regulatory jurisdictions. Second, the study measures the effectiveness of such a regulatory intermediary by analyzing TBT Committee data for the past two decades.

The TBT Committee data concerns over 400 “specific trade concerns” (STCs) raised during the Committee meetings since the launch of the WTO system in 1995. We analyzed minutes of those meetings based on various patterns of interactions among WTO members in the TBT Committee, such as a clarification (fact-finding) request, a complaint questioning the legality of the measure in question, a response satisfying the inquiring country, and a vague response. We then developed the following hypotheses:

1. The higher the level of conflict an STC demonstrates, defined by the number of WTO members involved, the less likely it is to be eventually resolved or satisfied;
2. The more frequently WTO members discuss a particular regulatory issue (STC), the less likely it is to be eventually resolved or satisfied; and
3. A clarification request is more likely to produce a resolution or satisfaction than an illegality complaint.

After we completed coding the data, we evaluated these hypotheses. While the full statistical method and results are discussed in detail below, a brief summary follows. Ultimately, our data supports our hypotheses with respect to hypotheses (2) and (3), but does not support hypothesis (1). Our analysis supports significant statistical change in Resolution based on Frequency and Motivation. Ultimately, we conclude from the data that for each increase in Frequency, the likelihood of Resolution is reduced by approximately 2 percent. Intriguingly, each Legality inquiry is associated with an approximately 20 percent reduced likelihood of Resolution. These relationships are not strongly correlated, but they do evince the trend discussed in the hypotheses.14

Perhaps most interestingly, this analysis shows us how much we do not know, and cannot measure by merely evaluating, summarizing, and coding publicly available resources such as WTO and TBT committee meeting minutes. For example, we coded the Motivation variable using TBT meeting minutes. In doing so, we took into account the specific phrasing used in the minutes as well as the general “tone” of an inquiry to determine whether a complaint was primarily one of legality or clarification. These inquiries might be further complicated by the pre-existing relationship between States. It might be similarly complicated by the relationships among State representatives as well as between State and WTO staff. As is perhaps obvious, State-to-State—and subordinately, human-to-human—interaction is endlessly variant while remaining cumulative and iterative. Thus, the TBT Committee discourse can be highly complex. In this regard, our present data analysis demonstrates the potential challenge in presuming a rationalist model,15 which too often is based on monolithically prescriptive and proscriptive ideas related to regulatory governance. Instead, the complexity inherent in the interstate dialogue reflected by the TBT Committee minutes illustrates the need for more information and a more nuanced approach to evaluating regulatory intermediaries. Consequently, we will need to further contextualize this result by interviewing diplomats and WTO staff who actually participated in the TBT Committee meetings. This will enable us to gather more data and information related to the discursive processes inherent in the WTO generally and the TBT committee specifically.

In sum, this Article generates two broad implications. First, just as Newtonian-Einsteinian physics cannot explain ninety-five percent of the universe that consists of dark energy and dark matter, the conventional, rationalist model cannot fully capture complex state regulatory behaviors without probing the discursive and rhetorical exchanges that occur among states. Second, as NTBs have increasingly replaced traditional tariff barriers, conventional analytical models based on simplistic behavioral assumptions, such as utility maximization, can no longer effectively explain NTBs. This follows from the fact that most NTBs are embedded in local socio-cultural characteristics. Obviously, these types of NTBs may not necessarily be simply bargained away through a reciprocal give-and-take.16 Under these circumstances, communication must be prioritized over incentives when mediation between state parties occurs.

This Article proceeds as follows: Section II problematizes the conventional approach to international regulatory governance by spotlighting its rather monolithic assumption on the regulator-regulatee relationship based on a rationalist model. Although this regulatory conventionalism brings causality into relief, it largely fails to capture the rich social dynamic between regulators and regulatees. In response, Section III explores an alternative model of regulatory governance in terms of “regulatory intermediaries.” Section III then attempts to theorize “communitization” of international regulation. Section IV tests the theory of international regulatory communitization by applying it to the TBT Committee. The Article concludes that, while the effectiveness of the TBT Committee is not unlimited, it certainly signifies an innovative type of transnational regulatory governance.

II. Problematizing Transnational Regulatory Governance

International organizations (IOs), such as the WTO, suffer from a chronic mismatch between their ambitious goals and deficient capacities.17 These capacity deficiencies come in diverse forms. Most of all, sovereign countries are reluctant to grant direct and hierarchical regulatory authority to IOs.18 Likewise, most IOs lack adequate financial resources with which to hire and invest more in their corresponding regulatory areas to fulfill their mandates. While this material capacity gap within IOs has been widely discussed, another important gap, a “governance gap,” has been relatively underexplored and under-theorized.19 Part of the reason behind this paucity is the predominance of the conventional regulatory framework, which can be epitomized as a hierarchical, compliance-based model.20 It is based on a very strong assumption that a regulator is capable of directly formulating and executing regulatory policies vis-à-vis regulatory targets.21 Naturally, conventional studies on regulation also tend to highlight the rationalist dimension of effectiveness of, and compliance with, a given regulation, such as enforcement strategies.22 Here, regulatees are often viewed as “amoral calculators.”23 The conventional view related to amoral calculators, which is informed by neorealists and rational choice theorists, posits that a state would comply with regulatory treaties if, and only if, such compliance contributes to its utilities (material interest, power, or reputation) or avoids disutilities, such as sanctions.24

In the modern era, while globalization increases interdependency among nations, a domestic regulatory authority might still not have effective access to relevant regulatory targets in foreign jurisdictions.25 In this complicated transnational regulatory situation, the conventional assumption does not hold and therefore could not remedy the aforementioned mismatch.26 This conventional approach is particularly questionable in light of the WTO’s nature and design vis-à-vis regulatory action. In particular, the TBT committee is shielded from the IO design tendency to “institutionalize[ ] the nature of states’ control mechanisms, such as the frequency with which government representatives convene for oversight meetings.”27 Instead, as part of what Tana Johnson terms “IGO [intergovernmental organization] progeny,”28 the TBT’s subsidiarity relative to the WTO provides it with greater institutional authority and latitude. This is because, as Johnson contends, international bureaucrats factor into the design of regulatory IOs in ways that alter the typical cycle of State rationalism.29 In this sense, these international bureaucrats subvert the conventional command-and-control structure in favor of an increasingly discursive one. This tendency was built into the TBT committee in particular and many regulatory IOs generally.30 As such, international bureaucrats have “dampened formal mechanisms of state control,”31 evincing the important role played by these bureaucrats within their organization. In sum, the conventional view simply brackets another important dimension of regulatory governance in which a regulatee state’s action is guided not simply by calculation but by socialization. This socialization includes learning and persuasion often provided by international bureaucrats and other soft institutions including the TBT Committee.

Likewise, the conventional view often regards the WTO as a “contract” between Member States subject to ordinary rules of damages like efficient breach.32 However, the conventional view tends to misrepresent the genuine nature of compliance under the WTO system. For example, compliance can be viewed as having a dual nature embodied in first and second order compliance. First order compliance is similar to contractual remedy conceptualized by the conventional model, while second order compliance relates to the systematic development of rules and regulatory proceedings.33 Importantly, second order compliance is not a command‑and‑control relationship. Instead, as Gregory Shaffer notes, these frameworks are formed through a gradual and iterative process of adherence, break, and calcification.34

Indeed, the inherent ambiguity of key WTO terms, such as “discriminatory” or “unjustifiable,” has rendered rationalist interpretations of regulation (such as the “stimulus-response” model) unsatisfactory.35 These economic models make the deceptively clear assumption that these regulatory agencies and participants compete to create “optimal levels of enforcement.”36 In the context of collective discussion and debate surrounding what is discriminatory or unjustifiable, this assumption too often ignores the more subtle interpretive issues involved. Take, for example, a hypothetical TBT inquiry brought by China. Say China has misgivings on a new U.S. regulation related to acceptable lead levels in children’s toys. China points to the international standard and claims that the U.S. regulation is unjustifiable. The U.S. replies that it has its own misgivings about the validity of the international standard. Is the U.S. discriminating against Chinese toys? If so, is such discrimination justifiable? Answers to these questions are not predetermined; they are subject to rhetorical processes such as argument and persuasion within the institutional context of the TBT Committee, unless the question is eventually adjudicated through the WTO dispute settlement mechanism.37

The conventional model becomes even more problematic when faced with an iterative self-regulatory process such as the TBT Committee. The TBT Committee must develop delicate regulatory strategies that are both suitable to its limited authority (such as peer review) and flexible enough to be accepted by WTO members.38 In the preceding example, the U.S. may rebuff Chinese entreaties. But, what about the next instance? And the instance after that? As rounds of regulatory dialogue continue in the TBT Committee, the U.S. and China may be able to produce some kind of regulatory rapprochement in the form of a memorandum of understanding (MOU). With this hypothetical example in mind, we can conceptualize the discursive nature of the WTO. Here, a discursive process marks the most significant departure from the conventional model and requires a new analytical framework. For example, the most salient aspects of law or legality embedded in the regulation derive from its everyday operation.39 A domestic regulator, which may be an importing country in the TBT context, and regulatees, which may be foreign producers represented by exporting countries in the TBT context, interpret and reflect on each other’s behaviors in the collective yet uncelebrated routine interactions between them. Sometimes, they learn from each other and form a strong partnership based on shared regulatory grounds. Indubitably, this successful socialization augurs well in terms of effectiveness and compliance, although they may fail some other times.

In fact, the aforementioned emphasis on a discursive, rhetorical regulatory process is not new. Some scholars have already developed a new model of regulatory governance in the same line. For example, the concept of “responsive regulation” was created as a sort of “third way”: a “symbiosis between state regulation and self-regulation,” in the context of domestic regulatory agencies specifically.40 These domestic regulatory agencies were tasked with orienting their activities based on “two pyramids: a hierarchy of sanctions and a hierarchy of regulatory strategies of varying degrees of interventionism.”41 Obviously these types of hierarchies are unavailable in the increasingly transnational business age. With issues becoming increasingly intertwined with international politics, and business transversing the conventional boundaries of States, domestic regulators cannot reach out to all potential regulatees, such as foreign producers in foreign countries.42 For that reason, “responsive regulation” as an international regulatory concept has had to adapt to survive. The most important evolution of responsive regulation relates to activities known as “transnational regulatory standard-setting” (TRSS).43 These activities include “orchestration,” which entails an IO using its own capacities to influence, encourage, and sometimes even ‘enforce’ self or multi-stakeholder regulation.44

Furthermore, new literature has suggested that focusing on procedural aspects of regulation could provide a palliative to the woes of present-day regulatory bodies. Kristina Murphy et al. suggest that the “use of threat and legal authority, particularly when perceived as unreasonable, can produce the opposite behavior from that sought . . . [including] overt opposition.”45 Murphy et al. propose that parties who follow out of “obligation rather than out of fear” are more likely to comply with regulatory organizations.46 The authors conducted several studies on domestic regulatory agencies, including a study on the Australian social security system.47 They concluded that “procedural justice was found to be more important for nurturing compliance when people questioned the legitimacy of the laws they were being asked to comply with.”48 These findings related to procedural justice illustrate that IOs must also maintain a good sense of their own legitimacy in the eyes of their constituents. Certainly procedural justice is a solid goal to strive for in the IO regulatory context. In this regard, the TBT Committee operates as a forum in which such procedural justice is delivered through transparency, notification, enquiries, and reason giving.

III. Toward a Theory of Transnational Regulatory Communitization

As a result of the complicated transnational regulatory regime and the failures of the conventional model to offer solutions, the need to address such complexities through more innovative forms of regulatory discourse becomes apparent. In this Section, we discuss an alternative model of regulatory governance through “regulatory intermediaries,” including the role of the TBT Committee as a regulatory intermediary. We then explore a theory of “communitization” of international regulation, which hypothesizes that transnational participation in regulatory discourse creates a useful social framework for understanding and evaluating regulatory solutions.

A. Transnational Regulatory Intermediaries (TRI)

Facing the aforementioned governance gap left by the conventional model, we need an innovative regulatory framework under which we can experiment with various models based on discursive regulatory processes, such as argument, persuasion, deliberation, and learning. One promising way of thinking in this direction comes from the concept of “orchestration,” which Kenneth Abbott et al. developed.49 Breaking from the conventional command‑and-control model, Abbott et al. envision a middle actor (“intermediary”) that can bridge or coordinate between an original regulator (“orchestrator”) and a regulatee (“target”).50 In this orchestration model, a regulator (orchestrator) “creates, supports and integrates a multi-actor system of soft and indirect governance geared towards shared goals that neither orchestrator nor intermediaries could achieve on their own.”51 Importantly, certain professional intermediaries possess critical governance capacities, such as “technical expertise” and “direct access to targets.”52 By mobilizing these epistemic advantages of those intermediaries, IOs can orchestrate an indirect yet effective regulatory pathway in which members (targets) coordinate their preferences to meet IOs’ normative goals.

In this regard, the TBT Committee can be deemed a “transnational regulatory intermediary (TRI)” to the WTO (orchestrator). Notably, “[o]rchestration can also steer . . . internal forms of governance that reflect democratic principles and promote internal contestability, increasing the representativeness, deliberativeness, and legitimacy.”53 In the TBT context specifically, Article 13.1 of the TBT Agreement defines the mission of the TBT Committee as “[f]or the purpose of affording Members the opportunity of consulting on any matters relating to the operation of this Agreement or the furtherance of its objectives, and [the TBT Committee] shall carry out such responsibilities as assigned to it under this Agreement or by the Members.”

Note that the nature of discussion topics (“specific trade concerns”) in the TBT Committee is highly technical.54 Naturally, participants of the TBT Committee include sector-specific government officials from national regulatory and standardizing bodies who hold technical expertise in given regulatory areas.55 The TBT Committee’s characteristic “operational capacity”56 includes providing a peer review forum for WTO members, socializing them within the context of the TBT community, and eventually facilitating their implementation of the TBT Agreement. Importantly, the TBT Committee retains a direct—not necessarily diplomatic, yet more importantly professional—access to targets, such as issue-specific domestic authorities subject to the TBT disciplines.57 This epistemic connection is critical in securing the effective implementation of, or compliance with, the TBT Agreement by WTO members.58

Certainly the TRI model is based on certain assumptions. One of the key assumptions is that the parties are working—in good faith—within a structure that actually has the capacity to support the collaborative development of regulations. Christine Parker appropriately characterizes these assumptions as significant facial challenges to the theory of responsive regulation.59 Is this, as Parker herself asks, too optimistic a presumption to form the basis of a coherent and persuasive regulatory regime?60 Why would any State choose to be involved with, much less adhere to, such a complex system?

To answer these questions in turn, we first think this presumption might not be overly optimistic. Assuredly, some noncompliance occurs. Yet the open nature of promulgating these regulations, with repeated interactions allowing for group learning and consensus-building on a global basis, incentivizes continued participation. Second, the complexity of an IO regulatory system encourages further State participation precisely because it is discursive. Its variable nature ensures that States will always find one reason or another to work within the international system. In this regard, the TBT Committee tends to warrant the TRI model assumption.

As a regulatory intermediary, the TBT Committee does not directly command WTO members’ compliance with TBT obligations or discipline violations. Rather, the TBT Committee provides a discursive forum in which WTO members can freely exchange information on their regulatory experiences, raise questions on any problematic regulatory practices, persuade other members into better regulatory practices, clarify or defend their own regulatory decisions, and eventually nurture a culture of cooperation among members in the TBT area.61 This non-hierarchical mode of governance is effective in the sense that the TBT Committee can acculturate WTO members to the normative goals of the TBT Agreement, i.e., reconciliation between regulatory autonomy and free trade values. The TBT Committee often resolves disputes before they escalate and reach the WTO tribunal for formal adjudication. Indeed, this “forward-looking” nature, in other words, early detection of trade disputes and their diffusion in a preemptive manner via dialogue, is characteristic of the TBT Committee.62 The steady rise of the number of specific trade concerns raised in the TBT Committee annually, from four in 1995 to ninety-four in 2012, eloquently demonstrates members’ perception of effectiveness, and therefore their trust in this indirect, non-hierarchical mode of governance.63

Admittedly, the TBT Committee might diverge from typical intermediaries in that the WTO did not actually coopt or outsource (“enlist”) it, as it did in the case of the International Organization for Standardization (ISO) under the TBT Agreement.64 In contrast, the WTO “insourced” the TBT Committee as it formally institutionalized it in the TBT Agreement.65 In this regard, one might be tempted to characterize the relationship between the WTO and the TBT Committee as “delegation” in terms of a principal-agent relationship. Nonetheless, the WTO does not directly control day-to-day operations of the TBT Committee, nor is their relationship borne by a contract.66 The TBT Committee retains a fairly independent space in its operation from the direct control of the WTO (the General Council). Moreover, the WTO provides the TBT Committee with both “material” (budget and personnel) and “ideational” (institutionalization) support.67 In sum, orchestration appears to be a better characterization of the relationship between the WTO and the TBT Committee.

Interestingly, what distinguishes the TBT Committee from other types of intermediaries is its “(soft)-rulemaking” function. For example, in its triennial review mandated in Article 15.4 of the TBT Agreement, the TBT Committee developed a variety of targeted soft-law products, such as guidelines or recommendations, in areas of international standards and good regulatory practices.68 These guidelines and recommendations are voluntary and sector-specific. Admittedly, this soft lawmaking is not so much new rulemaking from scratch as it is a “translation” of existing norms, such as the TBT Agreement.69 In 2012, the TBT Committee agreed to establish a voluntary mechanism to promote good regulatory practice by “assessing policy options, including the need to regulate (e.g. how to evaluate the impact of alternatives through an evidence-based process, including through the use of regulatory impact assessment (RIA) tools).”70 Needless to say, this rulemaking function of the TBT Committee facilitates WTO members’ implementation of the TBT Agreement, which aligns with the WTO’s main goal. In this sense, the TBT Agreement tends to confirm the “intermediary availability hypothesis,” which predicts that “governance actors are more likely to orchestrate when intermediaries with correlated goals and complementary capabilities are available,” as well as the “orchestrator focality hypothesis,” which predicts that “governance actors are more likely to orchestrate when they are focal within the relevant issue area.”71

In sum, the orchestration model departs from the so-called “Old Governance”72 framework (that is, the hierarchical, top-down regulatory model via treaties and intergovernmental organizations) in another critical aspect. While the conventional model views an IO as a tool for Member States in pursuing certain regulatory goals, the orchestration model defines an IO as an autonomous actor that plans the orchestration and recruits intermediaries in a strategic sense.

B. TRI and International Regulatory Communitization

Departing from the conventional command-and-control (enforcement) model, transnational regulatory communitization focuses on “communication,” which connotes learning, tolerance, and acculturation among transnational regulators and regulatees within the TBT context. This social framework is particularly useful in understanding and evaluating regulatory solutions between transnational regulators and regulatees in a situation where the conventional centralized regulatory authority is absent.73

Notably, the TBT Agreement concerns meta-regulation in the sense that TBT aims to regulate WTO members’ technical regulations through various means. Most of these means are “procedural” in nature in that TBT requires WTO members to take a certain course of action or procedure, rather than prescribing substantive criteria to be met. As Ayres and Braithwaite described as the “minimal sufficiency” principle, “the more regulation relies on moral suasion rather than punishment, the more effective it will be, especially at inducing internalization and thus long-term compliance.”74 For example, the TBT agreement requires WTO members to publish and notify its new TBT measures in a timely manner as well as to respond to inquiries from other members concerning such measures.75 Indeed, many of those procedural obligations are performed in the peer review under the auspices of the TBT Committee.

The regulatory core of these procedural provisions—transparency—eloquently demonstrates a “public” nature of the TBT regime. Rather than bilaterally disposing of regulatory issues between complaining and responding parties, it shares and “communitizes” regulatory problems among WTO members. In this regard, a regulatory “solution” under this communitization model is distinguishable from that of a conventional solution. If the latter concerns a binary notion of full compliance (repeal of a measure in question) versus violation (maintenance of the measure), the former envisions a spectrum of constructive, if not dramatic, behaviors, including enhanced cooperation toward better understanding of one’s regulatory goals and information sharing. Indeed, many of the titular violations under the conventional view may be unintentional due to lack of information or capacity, as Chayes and Chayes emphasized in their “management” model of international regulation.76 Under these circumstances, cooperation based on dialogue can be a better solution than coercion or retaliation. Likewise, respondents (regulating countries) tend to value the opportunity to defend their positions within the institution (the TBT Committee) regardless of whether their justifications are accepted. These non-instrumental factors, which may be labeled “relational criteria,” enhance the general sense of fairness and legitimacy among WTO members and therefore contribute to the WTO governance.77

For the purpose of this project, we define “solution” as some level of understanding/satisfaction that would be enough to stop the original inquiring party from pursuing any further action in a given matter. By this, we might be able to claim that parties concerned have reached some kind of mutual understanding via communication. Then, we can identify what kinds of factors contribute to such solutions.78 This redefinition of solution may even accommodate regulatory tolerance, in addition to conformity.

The aforementioned communitization model and its new notion of regulatory solution gain particular traction in the era of uncertainty. While this uncertainty itself constitutes a daunting challenge to the global trading system, diverging modes of addressing such uncertainty among WTO members may function as serious trade barriers. For example, some WTO members may be more risk-averse than others in responding to scientific uncertainty in regulating genetically modified organisms or hormone-treated beef. In this situation, providing market participants (producers, consumers, investors) with adequate information is vital in adequately communicating regulatory risks involved in global trade.79 The communitization model corresponds to this emerging need in the trade and regulation nexus.

IV. A Case Study: The WTO TBT Committee

Given our departure from the conventional model as it becomes increasingly unsuitable to address contemporary trade realities, we must evaluate the alternative model of regulatory governance via regulatory intermediaries by attempting to measure the effectiveness, if any, of communitization. Here, we test the theory of international regulatory communitization by applying it to the TBT Committee and attempting to measure, and evaluate, the Committee’s effectiveness as a regulatory intermediary.

A. Measuring the Effectiveness of a Transnational Regulatory Intermediary

As discussed above, the TBT Committee’s unique epistemic advantages qualify the Committee to be characterized as a regulatory intermediary. Its technical nature is attested to by the fact that the WTO delegations in Geneva often bring experts from their headquarters to the Committee meetings.80 Also, its workman-like operation allows the Committee to often be chaired by second-level Geneva-based diplomats, while participants facilitate their communication by refraining from the use of legalese.81 Its professionalism, which might translate into de-politicization, tends to self-legitimate its operation, as evidenced by its ever-increasing use by WTO members.

Figure 1: Review of Measures by TBT Committee82

 

 

Despite its obvious popularity, measuring a TRI’s actual effectiveness still appears challenging. Notably, Abbott et al. aptly observe that:

[E]ffectiveness is difficult to evaluate, because one must determine what goals are being pursued and consider the counterfactuals: what might have happened without orchestration and what alternative governance strategies might have achieved. Such factors are easier to evaluate in specific cases, where experts with detailed case knowledge can consider such contextual circumstances.83

Therefore, in evaluating the effectiveness of the TBT Committee qua intermediary one must consider its discursive format and normative goal of transparency.84 In other words, within the context of the TBT Committee, one might attempt to assess its effectiveness through inquiries and responses and their consequences. Note that those consequences do not necessarily entail compliant behavioral changes from WTO members. They also include expressions of satisfaction, explicit or implicit through acquiescence, from inquiring members. Here, one must understand the fundamental “relational” nature of TBT regulation. Although there are certain prescriptive rules, both substantive and procedural, under the TBT Agreement, those rules are triggered if, and only if, a certain WTO member raises an issue concerning them, for example in terms of a “specific trade concern.”85 There is no centralized review mechanism governed directly by the WTO General Council or the Secretariat. If a particular response indeed satisfies the original inquirer, there will be no more discussion or dispute. In this regard, the TBT Committee is a regulatory “clearinghouse” that channels and manages TBT‑related inquiries that exist in the form of both a demand for clarification and a complaint about alleged illegality.86 Unlike adjudication through the WTO tribunal, the ultimate goal of the TBT Committee is to liquidate those grievances by providing a discursive forum. In this sense, the TBT Committee may function as a dispute “prevention” mechanism.87

Given that the main regulatory focus under the TBT Committee lies in procedural—rather than substantive—disciplines, such as transparency, it would be infeasible to define “regulatory specifications”88 that may function as baselines in evaluating effectiveness of the TBT Committee’s contribution to the attainment of such procedural disciplines. Rather, as a meta-regulation, the TBT regulation monitored by the TBT Committee can be characterized as “relational regulation” that aims to manage compliant behaviors within an “acceptable range of variation,” rather than completely eliminating the gap between an ideal regulatory status and actual performance.89 As an “ongoing production of organizational and material life through a network of interdependent human transactions,” the TBT Committee embraces the “impossibility of perfect conformity between abstract . . . rules . . . and situated action.”90 Here, regulatory dialogue, such as inquiries and responses, form an essential toolkit for governing compliance with those procedural regulations (such as transparency and notification) under the TBT Agreement.

B. The Data

Measuring the overall effectiveness of the TBT Committee requires a nuanced understanding of “resolution.” The only relevant reference available in this regard is Horn et al.’s definition of resolution under the TBT Committee. Their definition of resolution is rather liberal. They observe that:

In the . . . TBT Committee[ ], a Member, or a group of Members, engage in a dialogue with other Member(s) concerning a specific policy measure; there is an exchange of information and views, and concerned Member(s) can rest the case if they so desire, for instance if they are sufficiently convinced concerning the legality of the measure; or, they can request a change in the contested measure—or in the light of explanations and clarification the challenging Members may decide not to pursue the matter further. As a result of the information obtained, or of the change in the policy, the concerned Member may decide on its own that the matter has been resolved, even though similar decisions are void of any formalism. Thus, some form of settlement takes place also in the case of STCs.91

Although their definition of resolution appears plausible, its explanatory force remains limited as a proxy of effectiveness, in particular within the unique context of the TBT Committee as a regulatory intermediary. The TBT-specific effectiveness concerns not so much any radical convergence or harmonization toward a particular regulatory standard as it does a “communitization” of regulatory concerns. Such communitization involves a deep understanding of others’ regulatory situations, concretization of any differences, exploration of possible common grounds, and minimization, if not elimination, of negative trade impacts. Indeed, various procedural obligations under the TBT Agreement, such as transparency, notification, and reason giving, tend to warrant this communitization goal by collectivizing particular trade-regulatory concerns and sharing them among WTO members in a discursive fashion.92

Against this background, we coded over 400 specific trade concerns (STCs) collected from the WTO TBT website93 according to the following sequence and criteria: first, we categorized main dialogues (communications) on STCs within the TBT Committee.

 

Table 1: Communication Codes on STCs within the TBT Committee

[[{"fid":"396","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"3":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"3"}}]]

Based on this categorization, we classified all STCs into two groups: (1) resolved and (2) not resolved. For example, “S” or “SCO” can be classified as resolved, while “D” or “SC” is classified as unresolved. Some styled typologies can illustrate our classifications. First, “resolved” connotes the following patterns:

1.) State X requests State Y provide information. State Y agrees to provide a direct answer in the form of more recent information and to refer remaining questions to State Y’s authorities for further replies.94

     a. Example: Data Point (DP) 3 – Mexico

2.) State X requests State Y provide information. State Y, either immediately or at some later time, tells the Committee that State Y has provided a direct response to State X and will provide similar responses bilaterally.

     a. Example: DP 204 – EC

3.) State X requests State Y to provide information. State Y, either immediately or at some later time, tells the Committee that State Y and State X will continue discussions bilaterally.

     a. Example: DP 4 – U.S.

4.) State X requests postponement of standards implementation or other measures by State Y. State Y either [1] agrees in part, or [2] expresses a willingness to discuss the matter further bilaterally.95

     a. Example: DP 106 – U.S.

5.) State X voices a concern with State Y’s standard. State Y [1] notes that State X’s issue has been brought before State Y’s authorities; [2] provides some sort of explanation; and/or [3] informs State X that the specific concern will be addressed and a determination made soon.96

     a. Example: DP 255 – Israel

In contrast, “unresolved” connotes the following patterns:

1.) State X requests State Y provide information. State Y refuses to provide that information or says that State Y cannot provide that information at the present time.

     a. Example: DP 4 – U.S.

2.) State X voices a concern about State Y’s proposed or adopted standard. State Y counters by justifying State Y’s regulation or otherwise dismissing State X’s concern. (Optional: State X expresses dissatisfaction with State Y’s answer.)

     a. Example: DP 6 – EC

3.) State X voices a concern about State Y’s standard. State Y takes note of State X’s concern, but takes no further action.97

     a. Example: DP 155 – China

4.) State X voices a concern about State Y’s standard. After repeated discussion, State X still believes, or a third country (State Z) believes, that the standard is unfairly biased.98

     a. Example: DP 135 – Norway

5.) State X voices a concern about State Y’s standard. After some discussion, State Y merely reiterates that the regulation has not been fully decided upon yet.99

     a. Example: DP 303 – Indonesia

Furthermore, we classified all the STCs into two large groups in accordance with the “motivation” of initial inquiries: seeking information/clarification or remedial changes. In our database, the code for the first occasion is “CN”; the code for the second one is “CL.” However, some STCs may not have either code, in which case we should give them one. Our presumption is that there is both a legality and a clarification purpose to almost all TBT inquiries. Very few are purely one or the other, though we have observed among “friendly” states that there is occasionally a pure question by State X as to what State Y intends. So, in the typology below, we operate on the presumption that one or the other predominates. “Legality” connotes the following patterns:

1.) State X challenges State Y’s standard on the basis that it does not conform with international standards.100

     a. Example: DP 11 – Canada

2.) State X believes State Y’s standard is an unnecessary obstacle to trade under the existing rules/agreements.

     a. Example: DP 197 – Germany

3.) State X draws attention to State Y’s pending or proposed regulations and states that they believe they are unacceptable under present international standards/are illegal under present standards.101

     a. Example: DP 206 – EC

“Clarification” connotes the following patterns:

1.) State X requests information about the application of State Y’s standard.

     a. Example: DP 23 – Korea

2.) State X draws attention to State Y’s pending or proposed regulations and asks whether they will be altered based on State X’s or State Z’s comments.

     a. Example: DP 201 – China

3.) State X notes that it is waiting on State Y’s response to a previously raised issue.102

     a. Example: DP 24 – Mexico

Finally, according to Horn et al., “serious” STCs are those resulting in two or more committee meetings. However, our database focuses on the “frequency” or the number of relays or communications between inquirers and respondents.

Against this background, we developed the following hypotheses:

1.) The higher the level of conflict an STC demonstrates, defined by the number of WTO members involved, the less likely it is to be eventually resolved or satisfied

2.) The more frequently WTO members discuss a particular regulatory issue (STC), the less likely it is to be resolved.

3.) A clarification request from a potential regulatee is more likely to produce a solution, or resolution (satisfaction), than an illegality complaint.

C. Analysis

We begin with a notable caveat.

Figure 2: Pearson’s r Correlation:

[[{"fid":"401","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"4":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"4"}}]]

Figure 2 shows a Pearson’s r correlation output—a correlation typified by linear dependence—among all variables. As illustrated, L, F, and M are only weakly correlated with R. Nonetheless, this Figure does demonstrate the existence of linear relationships of varying degrees. This means that we can demonstrate some relationship between the variables, albeit not a strong relationship. Take, for example, the relationship between Motivation (M) and Resolution (R). We can see that there is a weak negative relationship between these variables. That is to say, from the correlation coefficient we can see that a “legality” motivation, as opposed to a “clarification” motivation, is more likely to produce a result that does not end in successful resolution, as defined by our typology.

Admittedly, this correlation alone does not tell us what amount of change these variables cause in one another. In essence, it still does not tell us the extent of their relationship. Nor does it tell us if this relationship is statistically significant. To accomplish that goal, we proceeded with single linear regressions and a multiple regression with Level of Conflict (L), Frequency (F), and Motivation (M) as X to our common Y, Resolution (R).

Figure 3: X = L (Level of Conflict)

[[{"fid":"406","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"5":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"5"}}]]

 

[[{"fid":"466","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"17":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"17"}}]]
 

[[{"fid":"416","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"7":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"7"}}]]

Figure 3 shows the output of a linear regression where X is Level of Conflict (L)103 and Y is Resolution (R). L represents the weakest relationship with R. As noted by Figure 2, L had an extremely weak negative correlation with R of approximately -.09. Linear regression illustrates how much change in R is due to L—here, approximately .008, or 0.8 percent. Interestingly, the coefficient value demonstrates that with each point increase (e.g., 1 to 2) in L, the likelihood of R actually increases by .05 percent. Additionally, the p-value of approximately .0622 is below our 90 percent confidence interval of .10.

At first blush, these results seem to show that we can be 90 percent confident, or greater,104 that Level of Conflict (L) is the variable actually causing the 0.8 percent of change we noted above. Specifically, it indicates that each single unit increase in Level of Conflict actually increases the likelihood of resolution by .05 percent. However, as will be demonstrated below in Figure 6, we cannot ultimately use these results because they are not statistically significant when compared alongside the other variables. For that reason, we cannot support our first hypothesis that the higher the level of conflict an STC demonstrates, defined by the number of WTO members involved, the less likely it is to be eventually resolved or satisfied.

Figure 4: X=F (Frequency)

[[{"fid":"421","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"8":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"8"}}]]
[[{"fid":"471","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"18":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"18"}}]]

[[{"fid":"431","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"10":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"10"}}]]

 

Figure 4 shows the output of a linear regression where X is Frequency (the number of times a given issue was raised) and Y is Resolution. In the above, Figure 2 illustrated that F also has a very slight negative correlation with R—approximately -0.16. Linear regression demonstrates that about 1.6% of the variance in R is explained by the F at which an issue is raised.

As an initial matter, the p-value of .005 shows that our results were not a chance occurrence. That is, these results seemed to indicate that we could be confident, above a 95 percent chance, that the 1.3 percent lesser likelihood in Resolution (R) was the result of an increase in the Frequency (F) with which a given issue was raised. Furthermore, we can see from our coefficient of approximately ‑0.02 that for each time an issue is raised, the likelihood of Resolution goes down by 2 percent. After conducting a multiple regression (see Figure 6) we note that the single regression above in Figure 4 was absorbing some explanatory value from untested variables. For that reason, we must reduce our confidence interval to 90 percent. However, we can still be confident that the data supports our second hypothesis.

 

Figure 5: X=M (Motivation)

[[{"fid":"496","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"19":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"19"}}]]

 

[[{"fid":"506","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"21":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"21"}}]]

[[{"fid":"451","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"14":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"14"}}]]

Figure 5 shows the output of a linear regression where X is Motivation (legality v. clarification) and Y is Resolution. As noted above, Figure 2 showed a slight negative correlation of approximately -.2 between M and R. Linear regression shows that M accounts for approximately .05, or 5 percent, of the change in R. Moreover, our p-value of 3.19E-06 shows that this result is unlikely to have occurred by chance.

Here, our p-value is substantially smaller than the 95 percent threshold discussed previously. For that reason, we can assert that this 5 percent observed negative change is statistically significant. Further, we can conclude due to our coefficient that a Legality motivation makes a Resolution approximately 22 percent less likely to occur. This means that we can support our third hypothesis: that a legality challenge, rather than a motivation desiring clarification of a given issue, is less likely to result in a resolution among the parties.

 

Figure 6: Multiple Regression (All Xs – L, F, M)

[[{"fid":"456","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"15":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"15"}}]]
 

 

 

[[{"fid":"461","view_mode":"full","fields":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""},"type":"media","field_deltas":{"16":{"format":"full","field_file_image_alt_text[und][0][value]":false,"field_file_image_title_text[und][0][value]":false,"field_caption[und][0][value]":"","field_photo_credit[und][0][value]":""}},"attributes":{"class":"media-element file-full","data-delta":"16"}}]]
 

 

 

 

The multiple regression provides further clarity as to the overall model we have composed. As shown by the p-value associated with L, approximately 0.79, we must change our initial conclusion. We cannot support our hypothesis that a higher level of conflict results in a lesser amount of issue resolution at any reasonable confidence level.

However, as discussed above, the multiple regression indicates that we can support our hypothesis that an issue being raised more often (frequency) results in a lower rate of resolution at a 90 percent, rather than 95 percent, confidence level. The associated p-value—approximately .08—has increased due to the multiple regression function, so our confidence level must decrease. Yet this is still a statistically significant value. We can therefore state, due to a coefficient of -0.02, that each increase in Frequency amounts to a 2 percent lesser likelihood of resolution.

The reason for these changes is that the multiple regression looks into the explanatory value of the variables together. In doing so, it eliminates the bias inherent in single regressions that the single independent variable being tested is “soaking up” explanatory value from independent variables not being tested; here, those would be Level of Conflict and Motivation.

Most substantially, the p-value of Motivation in the multiple regression is 1.8E-5. Therefore, we can confirm within a 95 percent certainty our hypothesis that a Legality inquiry, rather than a clarification inquiry, negatively affects the likelihood of resolution in a given instance. Our coefficient (approximately ‑0.2) indicates that a Legality claim results in a 20 percent lesser likelihood of Resolution of a given issue.

D. Explanations and Projections

From these results, we can make several projections about the use of a transnational regulatory intermediary (TRI) as a new model of transnational regulatory governance. Our results appear to support the new model in a few key respects.

We can infer from the first result that having prominent WTO members (such as the U.S. and the E.U.) involved may actually increase the possibility of resolving STCs. Our hypothesis that a greater level of conflict would contribute to a smaller likelihood of resolution was not supported by the data. We can posit that this result occurred because those members enjoy high regulatory capacity and therefore can bring useful information instrumental to regulatory deliberation to the TBT Committee. This may translate into a form of “communitization” of regulatory problems since what these developed country members offer eventually contributes to the resolution of potential disputes. At a micro level, social learning and capacity building may have happened between developed and developing country members in discussing particular STCs, which tends to generate resolution. Similarly, the level of conflict may also evidence a greater number of countries involved in a given issue. Therefore, in a strictly mathematical sense, the issue is being further communitized by greater group participation.

As to the second result, as discussed above, we can support our initial hypothesis that more frequent discussions on an issue means a smaller likelihood of resolution. However, that likelihood was only reduced by 2 percent, a fairly negligible number even among the complexity of modeling human behavior. For that reason, we pose a new hypothesis related to frequency of interaction. Rather than merely reflecting potential inextricability of underlying regulatory challenges, the frequency of interaction may further the communitization of these regulatory issues. Frequency of communication intuitively results in more persistent and productive avenues for dialogue. In point of fact, in an anecdotal sense, issues were sometimes “cut off” from resolution in the meeting minutes by a perfunctory denial of dialogue. Therefore, openness to further discussion contributes to the TBT process in a far more complex way than representing a failure to reach a resolution. Perhaps, we may speculate that the frequency of interaction demonstrates the depth of regulatory dialogue and understanding among WTO members over particular STCs, even in the absence of a satisfactory result.

Finally, the third result appears to be relatively clear. This clearly substantiates the effectiveness of the new model. Note that “legality” inquiries tend to symbolize the old model based on formalism and enforcement. States that make legality inquiries are relying on an outdated language system for bringing issues before the newly communitized regulatory regime. Conversely, “clarification” inquiries appear to create a better discursive sphere than legalization inquiries. Requests for clarification are perhaps best characterized as more collaborative and less confrontational. For that reason, inquiries under the new model appear to be significantly more successful.

V. Conclusion

Our conclusions illustrate the need for a new explanatory model. The conventional command-and-control theories related to regulatory requirements fall away in the face of IO regulatory bodies such as the WTO and the TBT Committee. Instead, a novel method must be developed that prioritizes and recognizes the discursive nature of international regulatory proceedings. Our findings that Motivation—Legality versus Clarification—makes a difference in the ultimate Resolution of an inquiry illustrates that the way regulatory problems are discussed, and thus communitized, affects the way parties ultimately reconcile. What remains is to continue pursuing, and improving, methods of analyzing this regulatory discourse.

In particular, given the highly complex nature of regulatory discourse, more qualitative methods, such as interviewing actual participants of the TBT Committee meetings, are called for in an effort to contextualize the quantitative result demonstrated in this article and to better understand why legality inquiries nevertheless persist as a method of discourse for certain states. Considering that most STCs peer-reviewed under the TBT Committee are those “in the pipeline,”105 one might postulate that by electing a legality inquiry, as opposed to a clarification inquiry, the inquiring member might intend to signal a more serious message to the regulating member, as the former would hope for a possible modification or repeal of the STC in question.

 

Annex: Coding Results and Statistical Analysis106

• 1Tobacco Plain Packaging Act 2011 (Act No. 148, 2011) (Austl.).
• 2See Australia—Certain Measures Concerning Trademarks and Other Plain Packaging Requirements Applicable to Tobacco Products and Packaging, Request for Consultation, DS434 (Mar. 13, 2012).
• 3Claire R. Kelly, Realist Theory and Real Constraints, 44 Va. J. Int’l L. 545, 585–87 (2004); Kenneth W. Abbott & Duncan Snidal, Hard and Soft Law in International Governance, 54 Int’l Org. 421, 424–26 (2000); Benedict Kingsbury, The Concept of Compliance as a Function of Competing Conceptions of International Law, 19 Mich. J. Int’l L. 345, 350–56 (1997); Andrew T. Guzman, A Compliance-Based Theory of International Law, 90 Cal. L. Rev. 1823, 1860–61 (2002). Regarding the critique on the conventional (rationalist) approach to compliance, see generally Sungjoon Cho, The Social Foundations of World Trade: Norms, Community and Constitution, 120–63 (2015).
• 4Regarding the “compliance community,” see generally, Karen Alter, The European Court’s Political Power: Selected Essays (2009); Beth A. Simmons, Mobilizing for Human Rights: International Law in Domestic Politics (2009); Laurence R. Helfer & Eric Voeten, International Courts as Agents of Legal Change: Evidence from LGBT Rights in Europe, 68 Int’l Org. 77 (2014).
• 5Agreement on Technical Barriers to Trade, 1994, 1868 U.N.T.S. 120, pmbl. [hereinafter TBT Agreement]; see also Sungjoon Cho, Linkage of Free Trade and Social Regulation: Moving Beyond the Entropic Dilemma, 6 Chi. J. Int’l L. 625 (2005).
• 6See Sungjoon Cho, The Social Foundations of World Trade, supra note 3, at 120–63; Sungjoon Cho, Of the World Trade Court’s Burden, 20 Eur. J. Int’l L. 675 (2009); Sungjoon Cho, From Control to Communication: Science, Philosophy and World Trade Law, 44 Cornell Int’l L.J. 249 (2011) [hereinafter Cho, From Control to Communication].
• 7See Sungjoon Cho, How the World Trade Community Operates: Norms and Discourse, 13 World Trade Rev. 685, 700 (2014) (observing that the institutionalization of peer review under the TBT Agreement requires WTO members to defend the legality of their regulatory positions if challenged) [hereinafter Cho, Norms and Discourse].
• 8In this Article, a “binary” regulatory solution refers to a dichotomy of compliance and non‑compliance (violation) in terms of regulatees’ behavioral responses based largely on utilitarian considerations, such as incentives and penalties. It epitomizes a conventional regulatory framework under international law. See supra note 3 and accompanying text.
• 9See Sungjoon Cho, The Social Foundations of World Trade, supra note 3, at 120–63.
• 10See Bryce Baschuk, WTO Panel Considers 500th Specific Trade Concern, Int’l Trade Daily (Bloomberg BNA), Mar. 11, 2016, available at (https://perma.cc/WG4P-2TNX).
• 11See Committee on Technical Barriers to Trade (WTO), Minutes of the Meeting of June 15–16, 2011, G/TBT/M/54 (Sep. 20, 2011) (discussing Australia–Tobacco Plain Packaging Bill 2011 (G/TBT/N/AUS/67)).
• 12Article 2.2 of the TBT Agreement requires WTO Members to consider available scientific and technical information when preparing technical regulations. “In assessing such risks, relevant elements of consideration are, inter alia: available scientific and technical information, related processing technology or intended end-uses of products.” TBT Agreement, supra note 5, at art. 2.2.
• 13Before the TBT Committee, any WTO member may challenge or request clarification on new technical regulations (“specific trade concerns”) notified and publicized under the “Technical Barriers to Trade Information Management System (TBT IMS).” WTO, TBT Information Management System, available at (https://perma.cc/3UXM-YLHE).
• 14“Resolution” refers to a situation in which both an inquiring party and a responding party reached a certain solution or agreement, explicit or implicit, regarding the original specific trade concern. “Frequency” indicates the total number of rounds of meetings of the TBT Committee in which the identical trade concern was raised. “Motivation” concerns the reasons behind the original inquiry: it can be either to challenge the WTO-consistency of the technical regulation in question (“Legality”) or to seek clarification of the content of such regulation (“Clarification”). Regarding further details on the definitions of these variables, see infra Section IV.
• 15The rationalist model of international trade presumes that, given the anarchic world order, WTO members participate in the WTO for the sole purpose of advancing their national interests and will eschew compliance when compliance is inconvenient. This view aligns neatly with the realist, binary view of international politics. See generally Sungjoon Cho, The Social Foundations of World Trade, supra note 3, at 49–59.
• 16See generally Cho, From Control to Communication, supra note 6.
• 17See Kenneth W. Abbott et al., Orchestration: Global Governance through Intermediaries 2 (Aug. 2012), (https://perma.cc/XD94-GQY8) (observing that international organizations “often lack the capabilities to perform the roles they have been nominally allocated”).
• 18Id. at 20 (“States are more likely to impose strict control on IGO independence in core areas of national sovereignty.”).
• 19Regarding notable exceptions, see Kenneth W. Abbott & Duncan Snidal, The Governance Triangle: Regulatory Standards Institutions and the Shadow of the State, in Walter Mattli & Ngaire Woods, The Politics of Global Regulation 44–88 (2009); Anne-Marie Slaughter, A New World Order (2004); Wolfgang H. Reinicke, Global Public Policy: Governing without Government? (1998); James N. Rosenau, Governance in the Twenty-first Century, 1 Global Governance 13–43 (1995).
• 20Under the conventional hierarchical regulatory model, “regulation is largely prescriptive and compliance is largely based on coercion, deterrence and sanctions.” Kenneth W. Abbott, David Levi‑Faur, & Duncan Snidal, Intermediaries in Regulatory Governance (Feb. 4, 2015) (unpublished manuscript) (on file with authors), at 1. In the American context, the conventional model was typified by a “control-and-command structure,” which is inappropriate and wholly unworkable in the voluntary‑compliance arena of international relations. See Orley Lobel, The Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought, San Diego Legal Studies Paper No. 07-27 (2004).
• 21Abbott, Levi-Faur & Snidal, supra note 20, at 4–5.
• 22Susan S. Silbey, Case Processing: Consumer Protection in an Attorney General's Office, 15 L. & Society Rev. 849 (1981).
• 23Robert A. Kagan & John T. Scholz, The “Criminology of the Corporation” and Regulatory Enforcement Strategies, in Keith Hawkins & John M. Thomas, Enforcing Regulation 67–95 (1984).
• 24See Sungjoon Cho, The Social Foundations of World Trade, supra note 3; Kenneth W. Abbott & Duncan Snidal, Hard and Soft Law, supra note 3, at 430.
• 25Kenneth W. Abbott & Duncan Snidal, Taking Responsive Regulation Transnational: Strategies for International Organizations, 7 Reg. & Governance 95, 96 (2013).
• 26Abbott, Levi-Faur, & Snidal, supra note 20, at 4–5. “Business is increasingly global, operating through affiliates in multiple countries, lengthy and opaque transnational supply chains, and other complicated structures that span regulatory boundaries.” See also Abbott & Snidal, supra note 25, at 96.
• 27Tana Johnson, Organizational Progeny: Why Governments are Losing Control over the Proliferating Structures of Global Governance 4 (2014).
• 28Id. at 6.
• 29Id. at 5.
• 30See Cho, Norms and Discourse, supra note 7, at 700–702 (discussing the “peer review” mechanism under the WTO system, including the one conducted within the TBT Committee).
• 31Johnson, supra note 27, at 23.
• 32Regarding the origin of the world trade contract, see Sungjoon Cho, The Social Foundations of World Trade, supra note 3, at 11–15; see also Gregory Shaffer, How the World Trade Organization Shapes Regulatory Governance, 9 Reg. & Governance 1, 2 (2015).
• 33See Shaffer, supra note 32.
• 34“The frames, however, are not simply given. They are rather shaped over time through recursive rounds of engagement among actors with differing epistemologies and interests at different levels of governance. In practice, the positions taken by state representatives in the WTO are often more of a mercantilist nature, as they defend both their export and their import-competing business interests. This provides an opening for contestation and argumentation in which officials must simultaneously look at their own practices when challenging others’. Government representatives before the WTO’s network of councils and committees engage in sustained deliberation, and, in this way, are subject to persuasion and learning.” Id. at 16.
• 35Julien Etienne, Ambiguity and Relation Signals in Regulator-Regulatee Relationships, 7 Reg. & Governance 30, 30 (2013).
• 36Id.
• 37Regarding the delicate relationship between peer review, as manifested in the TBT Committee, and adjudication in the WTO dispute settlement mechanism, see Cho, Norms and Discourse, supra note 7, at 685.
• 38Abbott & Snidal, supra note 25, at 97.
• 39Garry C. Gray & Susan S. Silbey, The Other Side of the Compliance Relationship, in Explaining Compliance: Business Responses to Regulation 123 (Christine Parker & Vibeke Lehmann Nielsen eds., 2011) (discussing the “habitual quotidian enactment”).
• 40Abbott & Snidal, supra note 25, at 96.
• 41Id.
• 42Id.
• 43Id. at 97.
• 44Id. at 97–98.
• 45Kristina Murphy, et al., Nurturing Regulatory Compliance: Is Procedural Justice Effective When People Question the Legitimacy of the Law?, 3 Reg. & Governance 1, 2 (2009).
• 46Id.
• 47Id. at 11–14.
• 48Id. at 18.
• 49Abbott et al., supra note 17, at 2; Abbott, Levi-Faur & Snidal, supra note 20, at 4–5.
• 50Abbott et al., supra note 17, at 2.
• 51Id. at 3.
• 52Id.
• 53Abbott & Snidal (2013), supra note 25, at 107.
• 54WTO, Technical Barriers to Trade, (https://perma.cc/4AK7-QEAA) (last visited May 9, 2016).
• 55WTO, The WTO Agreements Series: Technical Barriers to Trade 28 (2014).
• 56Abbott, Levi-Faur, & Snidal, supra note 20, at 6.
• 57Id.
• 58Shaffer, supra note 32, at 1 (highlighting the WTO’s transnational regulatory governance that affects “changes in professional expertise engaging with state regulation”).
• 59“[Responsive regulation] tends to assume that the kind of social relationships, opportunities for free and equal deliberation, and substantively just law that it seeks to create already exist. . . . That is, responsive regulation assumes that there are enough regulators, regulated businesses and third parties who are genuinely committed to the public interest, willing and able to communicate with one another to resolve problems, imaginative enough to come up with ‘win win’ solutions to make it possible and that they then have sufficient capacity and an appropriate substantively just law to enable them to implement those solutions.” Christine Parker, Twenty Years of Responsive Regulation: An Appreciation and Appraisal, 7 Reg. & Governance 2, 9 (2013).
• 60Id.
• 61See Abbott, Levi-Faur, & Snidal, supra note 20, at 1 (observing that intermediaries can play a role in “building communities of assurances, trust and compliance”).
• 62The WTO Agreements Series, supra note 55, at 32.
• 63Id. at 29; Record Number of New Trade Concerns Raised in Standards Committee in 2014, WTO News (Nov. 4–6, 2014).
• 64Abbott et al., supra note 17, at 8.
• 65TBT Agreement, supra note 5, at art. 13.
• 66For example, the TBT Committee is enabled to make an independent decision in granting developing countries “specified, time-limited exceptions” from the TBT obligations. Id. at art. 12.8.
• 67Id. at art. 13. (“The Committee . . . shall carry out such responsibilities as assigned to it under this Agreement or by the Members.”).
• 68The WTO Agreements Series, supra note 55, at 30.
• 69Abbott, Levi-Faur & Snidal, supra note 20, at 9.
• 70WTO, Sixth Triennial Review of the Operation and Implementation of the Agreement on Technical Barriers to Trade Under Article 15.4, G/TBT/32, 29 November 2012, ¶ 4.
• 71Abbott et al., supra note 17, at 15.
• 72Kenneth W. Abbott & Duncan Snidal, Strengthening International Regulation Through Transnational New Governance: Overcoming the Orchestration Deficit, 42 Vand. J. Transnat’l L. 520–32 (2009).
• 73Gray & Silbey, supra note 39, at 123.
• 74Abbott & Snidal (2013), supra note 25, at 107.
• 75TBT Agreement, supra note 5, at arts. 10, 13.
• 76Abram Chayes & Antonia Handler Chayes, The New Sovereignty: Compliance with International Regulatory Agreements (1995).
• 77Tom R. Tyler, Why People Obey the Law 276 (2006).
• 78Such factors include the subject matter, parties concerned (developing or developed countries etc.), number of parties concerned, whether a notification triggered the inquiry, duration of communication, etc.
• 79The Application of Risk Communication to Food Standards and Safety Matters, WHO (1998), (https://perma.cc/MS2F-ZQUG).
• 80Henrik Horn, Petros C. Mavroidis, & Erik N. Wijkstrom, In the Shadow of the DSU: Addressing Specific Trade Concerns in the WTO SPS and TBT Committees, 47 J. World Trade 729, 754 (2013).
• 81Id. It should be noted that the TBT does not ban the use of such legalese per se, but the TBT Committee does emphasize the importance of efficiency and streamlining TBT Committee discussions. In 2009, for example, in response to the “accelerated growth” in the number of STCs raised at TBT Committee meetings, as well as in the number of WTO Members raising concerns or substantively supporting other Member’s concerns, the TBT Committee emphasized the importance of making communications among Members more efficient. See WTO, Minutes of the Meeting of November 13, 2009, G/TBT/26 (Nov. 13, 2009), ¶¶ 67–68.
• 82 WTO, Sixth Triennial Review, supra note 70, at 10.
• 83Abbott et al., supra note 17, at 21.
• 84Id.
• 85See TBT Information Management System, supra note 13.
• 86Horn et al., supra note 80, at 730.
• 87Research Handbook on the WTO and Technical Barriers to Trade 573–74 (Tracey Epps & Michael J. Trebilcock eds., 2013).
• 88Ruthanne Huising & Susan S. Silbey, Governing the Gap: Forging Safe Science through Relational Regulation, 5 Reg. & Governance 14, 14 (2011).
• 89Id. at 15–16.
• 90Id. at 16.
• 91Horn et al., supra note 80, at 749.
• 92Cho, Norms and Discourse, supra note 7, at 700 (“[B]y framing their inquiries and responses within the context and terms of WTO norms, both an inquirer and a respondent transmit WTO norms to each other.”).
• 93WTO, Technical Barriers to Trade, supra note 54.
• 94This follows Horn et al.’s definition of “resolution” in that we are not looking at a formalistic “resolution” (i.e., not all problems are completely resolved), but rather that some agreeable progress has been made between the parties.
• 95This partial resolution follows our broader definition of “resolution.”
• 96Here, we illustrate that one of our primary considerations is the reconciliatory mechanism, rather than the ultimate policy change, inherent in the TBT process.
• 97Here, the response is in fact a “cutting off” of communication.
• 98Within our typology, this type represents an insufficient resolution.
• 99This primarily represents a delaying tactic distinguished from the “Yes” type that also includes invitation or solicitation for more discussion, either bilaterally or through the TBT committee.
• 100While the phraseology is relatively gentle, the fundamental issue is that there is an existing standard that Norway believes does not conform.
• 101Here, the underlying motivation is to merely challenge the proposed regulation rather than to get a response regarding an already challenged regulation.
• 102In this instance, even if the original basis was legality, the present motivation is clarification.
• 103This is coded per the above as the relative ‘weight’ given an issue by number of countries and their economic significance.
• 104This is the commonly accepted standard for statistical significance.
• 105WTO, Technical Barriers to Trade, available at (https://perma.cc/7LKQ-JSWV).
• 106Coding Results and Statistical Analysis (on file with authors and the Chicago Journal of International Law).