The Political Economy of WTO Exceptions


In a bid to save the planet from rising temperatures, the European Union is introducing a carbon border adjustment mechanism—essentially a levy on imports from countries with weak climate rules. The United States, Canada, and Japan are all openly mulling similar proposals. The Biden Administration is adopting new Buy American rules, while countries around the world debate new supply chain regulations to address public health issues arising from COVID-19 and shortages in critical components like computer chips. These public policy initiatives—addressing the central environmental, public health, and economic issues of the day—all likely violate World Trade Organization (WTO) rules governing international trade, as well as regional free trade agreements. This inconsistency poses a political problem domestically and a diplomatic problem internationally, to say nothing of potential consequences authorized by the WTO.

To ward off these consequences, governments will seek to justify their measures under a series of exceptions to trade obligations first drafted in 1947. Although governments have invoked these exceptions with increasing frequency in recent years, they have never been tested in the manner that they will be in the coming years. Indeed, a provision in the 2021 Infrastructure and Investment Act—the first major legislative piece of the Biden Administration’s economic agenda—contains a provision directing the government to invoke these exceptions to justify measures to manufacture personal protective equipment (PPE) in the United States.

This Article seeks to make sense of the exceptions and their role in the legal, political, and diplomatic proceedings that determine the fate of public policies that restrict trade. It distills three paradigms through which to view legal exceptions in international trade agreements. Under the Policy Space Paradigm, governments have the right to violate international obligations so long as the violation is necessary to pursue a public policy goal permitted by an exception. Under the Safety Valve Paradigm, exceptions excuse violations that are undeterrable, such as those motivated by overwhelming domestic political pressure. Both of these approaches, which are dominant in international legal practice, permit governments to invoke international legal exceptions only to the extent that the government acts with a single, permissible objective. In so doing, both paradigms rest on faulty assumptions about how domestic policymaking works.

I therefore introduce the Channeling Paradigm, which rests on the observation that international trade policies are the result of bargaining between domestic interest groups. Exceptions in trade agreements influence that bargaining process and the resulting domestic coalitions. Industries seeking economic protection will often ally themselves with groups pursuing “non-trade” public policy goals, such as environmental protection or public health. Both groups benefit. The domestic industry obtains protection from foreign competition by lending its political support to a public policy goal. Public policy advocates obtain important political support for policies that provide public goods that governments often undersupply, such as measures to protect public health, fight climate change, and address economic inequality. Counterintuitively, then, the domestic political bargaining that legal exceptions encourage serves the public interest by channeling protectionist pressure into the promotion of public goods.

The Channeling Paradigm has implications for dispute resolution under international trade agreements, as well as the drafting of new agreements. In short, existing tests for the application of trade agreements’ public policy exceptions unduly constrain domestic politics. This Article argues that trade tribunals and treaty negotiations should adopt a Predominant Motive test when interpreting and drafting exceptions clauses. Under this approach, a trade restrictive policy would benefit from an exception if the primary objective of the measure is a permitted goal under the exception. So long as it does not become the predominant purpose of the challenged policy, economic protection would not be fatal to invoking an exception. The WTO compatibility of a wide range of critical government policies that have mixed motives—including President Biden’s Buy American requirements that seek to address economic inequality within the United States; efforts to reshore critical U.S. supply chains with the goal of ensuring the United States has access to the components it needs to be a global leader in manufacturing; the European Union’s efforts to impose a carbon tariff in aid of its efforts to combat climate change; and public health restrictions on trade in medical supplies and the COVID-19 vaccine—all depend on a more flexible approach to international legal exceptions.


In July 2021, amidst the hottest summer on record and with wild fires burning from California to Turkey, the European Union (EU) announced that it would impose a carbon border adjustment—essentially a levy on imported products from countries that lack stringent climate rules governing production.[2] The move reflects a bid by the EU to encourage countries to adopt tougher climate rules, while also discouraging European consumers from consuming high-carbon products. The United States, Canada, Japan, and the United Kingdom are considering similar policies.[3] Since taking office in January 2020, President Joe Biden has expanded Buy American requirements in an effort to address economic inequality within the United States;[4] directed a review of critical U.S. supply chains with the goal of ensuring the United States has access to the components it needs to be a global leader in manufacturing;[5] and defended national security-related curbs on trade with Hong Kong in response to the suppression of democracy there.[6] The 2021 Infrastructure Investment and Jobs Act—the first major legislative piece of the Biden Administration’s economic agenda—contains an entire Title devoted to “Build America, Buy America.”[7] The EU imposed export restrictions on COVID-19 vaccines in early 2021,[8] a measure which follows extensive export restrictions on medical supplies during the early stages of the pandemic.[9]

These measures—each of which is likely to violate international trade rules, most notably those of the World Trade Organization (WTO)—are the most recent in a long line of trade restrictive measures governments have adopted that both serve a legitimate public policy objective and also benefit a domestic economic constituency. The United States, for instance, banned the import of shrimp caught in a manner that endangers sea turtles, a move supported by U.S. shrimpers eager for a market advantage.[10] To the delight of domestic tire producers, Brazil banned the importation of retreaded tires because when discarded they become breeding grounds for malarial mosquitos.[11] The EU refused to extend regulatory incentives for renewable energy to biofuels that lead to a risk of deforestation, a move hailed by the EU’s domestic biodiesel industry.[12] Russia restricted the transit across its territory of Ukrainian products due to the geopolitical tensions between the two nations since the former’s invasion of Crimea, but closer examination suggests that the specific restrictions imposed are motived by a desire to protect domestic industries.[13]

In each of these cases, a country imposed a trade restriction justified by an ostensible “non-trade” public policy goal—the environment, public health, or national security. And each of these measures was actively supported by a domestic interest group that stood to benefit from the resulting economic protection. This pattern, a common occurrence, presents an increasingly difficult challenge for the WTO and the world trading system more generally. More and more, governments clamor for the legal right to pursue policies that violate trade rules and burden international commerce. Many of these policies—carbon border adjustments to fight climate change,[14] restrictions on seal products to promote animal welfare,[15] domestic content requirements to spur the development of local renewable energy industries,[16] or restrictions on trade in goods made with forced labor[17]—address political and moral imperatives for which leaders and their constituents are willing to sacrifice the economic efficiency that accompanies trade liberalization. But building domestic political coalitions for these policies often involves using economic protection to benefit a domestic group. Governing is, after all, the art of compromise.

This Article argues that exceptions to international trade law’s general rules provide the incentives that lead to the formation of these domestic coalitions. This argument is counterintuitive. International trade law is commonly justified on the grounds that it restrains domestic political pressure for protection.[18] Yet trade law also operates through a norm-exception framework. International trade’s primary rules aim to ensure equality of competitive opportunities for global commerce. Policies that deviate from those rules in the name of some other policy objective require affirmative justification. This need to affirmatively justify policies is also an opportunity to justify economic protection. Unable to justify protection on its own terms, interest groups benefiting from protection will lend their political support to policies that fit within the exceptions. I refer to this dynamic as Channeling.

Channeling has two important effects that have been overlooked. First, the norm-exception framework harnesses protectionist pressures in the service of other legitimate public policy objectives. A common account of international trade law is that it operates to change domestic politics. Because international trade rules are predicated on a reciprocal exchange of concessions—that is, they link one state’s trade barriers to another state’s trade barriers—the international trading system turns exporters that otherwise might be uninterested in their own nations’ imports barriers into lobbyists for fewer trade barriers in their own countries. Trade’s primary rules, in other words, create countervailing domestic political forces to push back against protectionist pressure.

Exceptions work in a similar way.[19] Some degree of protectionist pressure is inevitable in a national economy. Bruce Lee famously said: “Be like water making its way through cracks. Do not be assertive, but adjust to the object, and you shall find a way round or through it.”[20] Protectionist political pressure is like water. It will seek to find a way around or through the barriers that face it. Exceptions respond to this reality by channeling protectionist pressure into a beneficial purpose.

Protection typically benefits the protected economic interest at the expense of the general public. But a number of public policy goals—climate change, public health—that benefit the public at large are underrepresented in domestic politics. The benefits from these policies, while large, are diffuse, meaning that no one group may have an incentive to strongly lobby for the policy. By linking the pursuit of those public policy goals to an element of protection, international trade law mobilizes domestic interest groups with protectionist goals to act in the public interest. In the language of public choice, international trade law’s exceptions can help states provide an otherwise undersupplied public good. In this way, Channeling serves a valuable public role.

Channeling also creates risks, though. The pressure to use legitimate public policy exceptions for protectionist purposes represents a threat to the legitimacy of the entire trading system. As anyone who has ever tried to fix a leak knows, sealing one crack can simply push the water into another crack. And when the same amount of water tries to fit through fewer cracks, the pressure on those that remain increases. Those cracks expand to accommodate the additional pressure. Pressure can erode the channels meant to constrain it.

In the context of trade law, protectionist pressures threaten to expand the gaps deliberately left by exceptions. Over time, protectionist domestic groups may increasingly try to fit more into exceptions that were meant for narrower kinds of claims. The increased use of antidumping duties by the United States and the European Union as a general response to nonmarket economies like China’s, as well as the novel uses of the national security exception by the United States to justify steel and aluminum tariffs, illustrate this dynamic.

This overclaiming can be inadvertently encouraged by case law that seeks to shut down virtually any discrimination in the invocation of a public policy exception. Due to the need to compromise, domestic lawmaking will often produce policies that weaken the pursuit of the measure’s primary objective in order to attract sufficient support to enact the measure. In invalidating these domestic political bargains, especially in situations in which there is little evidence that the political process had any animus toward imports, tribunals risk increasing protectionist pressure on the WTO, rather than defusing it.

This Article proceeds in four parts. Part I briefly describes international trade law’s primary rules and explains why governments have increasingly turned to exceptions to those rules to justify some of their most consequential policies. Part II describes the two conventional paradigms through which international trade lawyers and trade dispute panels have viewed and applied exceptions: what I call the Policy Space Paradigm and the Safety Valve Paradigm. This Article is the first to group these paradigms together in this way. The literature on exceptions in international law generally, and international trade law specifically, has tended to focus on the proper scope of an exception, as well as technical issues such as the burden of proof.[21] These issues are important for any tribunal asked to decide whether a WTO member is liable under international trade law for a particular action. This question is somewhat less useful, though, if we are trying to understand how exceptions shape government decisions, especially in times of political stress and crisis. It is even less useful if we are trying to design a legal regime that balances competing objectives and pressures, such as the twin imperatives of creating economic growth through trade liberalization while also responding to climate change or ensuring a just distribution of economic gains.[22]

Under the Policy Space Paradigm, exceptions mark the boundaries between two or more legal regimes or issues. For instance, international trade law should not foreclose governments’ ability to pursue noneconomic policy objectives, such as environmental or national security objectives. The General Agreement on Tariffs and Trade (GATT) famously contains a list of general exceptions for specific kinds of policies, such as those related to conserving exhaustible natural resources or those “necessary to protect human, animal, or plant life or health.[23] More recently, the push for a right to regulate in investment treaties seeks to “provide greater policy space for host states to take measures in the public interest” that might otherwise be thought to infringe an investor’s rights.[24]

The Safety Valve Paradigm, by contrast, is concerned with improving the efficacy of the primary legal regime by creating exceptions to undeterrable conduct. On this view, exceptions allow states to avoid losses arising from overwhelming domestic pressure to violate international trade law. Because most of international law lacks a system of expectation damages, most sanctions for the violation of international law are negative sum.[25] The violating party suffers sanctions, but other states do not receive those sanctions as compensation as they would in a domestic legal system. The only benefit from the sanctions is to deter future violations. But if violations are undeterrable, sanctions have no offsetting benefits. In such situations, legalizing or excusing the conduct at issue may make the most sense. It may, for instance, preserve the legitimacy of international law by not pretending that the law can prevent that which it cannot.[26] It may also encourage states to make greater concessions during the initial negotiations over a legal regime, knowing that they can safely escape their obligations in the future in response to specified conditions.[27] GATT rules on safeguards, which allow a government to impose trade restrictions in the face of a sudden surge in imports that causes serious injury to a domestic industry, offer the classic example.[28]

Part III makes the Article’s chief contribution: the development and analysis of the Channeling Paradigm. If the Policy Space Paradigm is concerned with the boundary between legal regimes, and the Safety Valve Paradigm is concerned with the efficacy of a primary legal regime, the Channeling Paradigm is concerned with the operation of domestic politics. The Channeling Paradigm starts from the observation that governments often have incentives to violate legal rules, but that they have a choice in how they do so and how they justify their violations. By legalizing some forms of violation but not others, exceptions can encourage states to violate the law in ways that advance some other objective. For instance, in the trade context, domestic producers often demand protection from competition with foreign imports. A government can aquiesce to that demand outright, although if it does it is likely in violation of trade rules. Alternatively, it can seek to graft its otherwise impermissible protection on to an existing exception in the law. In this way, the government’s protectionist action violates trade rules, but at least advances another permissible objective.

I argue that the Channeling Paradigm best describes the domestic political process that leads states to violate international rules in most cases. The Channeling Paradigm begins with the observation that legal rules do not eliminate domestic political support for policies that violate those rules. A domestic constituency seeking a government policy that violates international rules will likely seek to form a coalition with other groups in order to broaden the appeal of the policy within the government. For example, a demand for protection from imported shrimp becomes grafted on to an environmental demand to save sea turtles killed in the shrimping process.[29] A government may join protection from cheap imported biofuels to concerns about deforestation.[30]

Legal exceptions thus influence the kinds of domestic political coalitions that form. The presence of environmental and health exceptions in the GATT, for instance, encourages groups desiring protection to push for measures that pursue both a protection and an environmental or public health purpose. Indeed, the bipartisan infrastructure legislation that Congress recently enacted contains a provision that directs the government to justify Buy American rules by “invoking any relevant exceptions to [the WTO Agreements], especially those related to national security and public health.”[31] Likewise, government officials designing policies are, all else equal, likely to prefer policies that enjoy broader domestic support and are more likely consistent with international legal rules. The Channeling Paradigm best explains how legal exceptions shape government measures in light of political pressure.

Having explained how Channeling works descriptively, I turn to examining its implications for international dispute resolution and the negotiation of international trade agreements. The Policy Space and Safety Valve Paradigms have animated the application of exceptions by international trade tribunals. Under both approaches, “exceptional” policies can pursue but a single objective; any additional trade restrictiveness renders an exception inapplicable. The Channeling Paradigm, by contrast, suggests that most “exceptional” policies will have mixed motives.[32] Borrowing from scholarship in domestic law on mixed motive jurisprudence, I argue that WTO panels should adopt a Predominant Motive test—under which a WTO member is entitled to an exception so long as its measure’s primary objective is a permitted one—to determine whether the WTO’s general exceptions apply. I explain how such a test would work within the framework of existing WTO case law and also propose changes to the text of exceptions for use in future trade agreements.

A Predominant Motive test would strike a more appropriate and politically sustainable balance between WTO members’ right to regulate in the public interest and the objectives of trade liberalization. In so doing, it would shore up the legitimacy of international dispute resolution, which has been in crisis since the United States began blocking the appointment of WTO Appellate Body members, ultimately leaving the Appellate Body unable to operate due to a lack of members. The EU recently lauded the Appellate Body for its increasing willingness to permit the use of exceptions.[33] And to be sure, the Appellate Body’s application of the WTO exceptions has become more flexible, ultimately allowing a number of respondents to de facto prevail even if they nominally lose their cases. But the pace at which the Appellate Body adapted its jurisprudence has failed to keep up with the changing role of the trading system in foreign affairs and the domestic politics of key members, especially the United States. If the Appellate Body, or some other dispute body, is to resume operation in the coming years, an approach to exceptional jurisprudence that reflects these changed roles will be necessary.

I. Trade Rules and the Turn to Exceptions

Any discussion of exceptions should begin with a brief description of the primary rules from which states may wish to deviate, and why they may wish to do so. As I explain in this Part, the international trade regime and its primary rules have come under increasing stress in recent years. Trade policies aimed at creating robust labor markets within countries, securing critical supply chains, promoting human rights in places like China and Hong Kong, and addressing environmental concerns like climate change have all been challenged as inconsistent with international rules. Because in practice these policies are often hard to justify under trade’s primary rules, exceptions have taken center stage. This Part briefly describes the key rules of international trade law and how current events have prompted a turn to exceptionalism.

International trade’s core norm is nondiscrimination, which comes in two flavors. First, virtually all international trade agreements prohibit members from treating the commerce of one member state less favorably than they treat the commerce of another state.[34] For instance, the United States may not treat Mexican goods less favorably than it treats Canadian goods. This norm, referred to as the “most-favored nation” (MFN) principle, applies to trade in goods,[35] trade in services,[36] intellectual property,[37] and investment.[38] Perhaps no provision was as central to the GATT, the WTO’s precursor, as the MFN obligation. The GATT’s drafters enshrined the obligation in article I, reflecting in part the United States’ desire to take apart the preferential trading system that European countries had maintained in favor of their former colonies prior to World War II.[39]

Preferential trade agreements—such as the United States-Mexico-Canada Agreement (USMCA), the Comprehensive and Progressive Trans-Pacific Partnership (CPTPP), and the Comprehensive Economic and Trade Agreement between Canada and the EU (CETA)—have largely gutted the WTO’s MFN obligations today.[40] In the context of trade in goods, for instance, preferential trade agreements typically allow duty-free trade in goods among members. Imports from nonmembers, by contrast, are assessed under a nation’s WTO MFN tariff rate. Preferential trade agreements have become major sites for negotiating new trade rules at the same time that negotiations at the WTO have struggled to yield meaningful results. The shift to preferential trade agreements represents, in effect, forum shifting from multilateral institutions with high transaction costs to bilateral or minilateral fora with significantly lower transaction costs.[41]

The WTO agreements contain exceptions that permit preferential trade agreements and the resulting violation of the MFN principle.[42] These exceptions allow countries to grant preferential treatment to other nations so long as they abolish substantially all barriers to trade between the countries.[43] Due to the WTO’s exception for preferential trade agreements, a huge percentage of global trade today moves under terms set out by preferential trade agreements, rather than under WTO rules. This includes all trade between the EU’s twenty-seven members; trade across North America under the USMCA and the CPTPP (as between Mexico and Canada); and much of South America under Mercosur, to give but a few major examples.[44]

Nondiscrimination’s second form is “national treatment,” which requires that members of trade agreements not treat the commerce of other members less favorably than they treat the commerce of their own nationals.[45] The United States, for instance, may not charge higher sales taxes on Canadian products than it charges on U.S. products,[46] nor may it subject Canadian products to tougher regulations than those to which U.S. products are subject.[47] Like the MFN obligation, the national treatment obligation applies to goods,[48] services,[49] intellectual property,[50] and investment.[51] However, in recognition of the fact that governments may legitimately prefer their own commerce in some circumstances, some agreements—most notably the General Agreement on Trade in Services (GATS)—only apply the national treatment obligation to those economic sectors into which members voluntarily opt.[52]

The resurgence of industrial policy—policies that encourage resources to shift from one industry or sector to another[53]—combined with the tendency to use environmental, health, and safety regulations to protect domestic industries has put a significant amount of pressure on national treatment rules recently. One of President Biden’s first actions on taking office was to expand Buy American requirements, a form of a local content requirement that mandates the use of domestic over imported goods.[54] This commitment to Buy American, along with President Biden’s executive order mandating a review of critical supply chains,[55] suggests that the United States would like to see broader exceptions to the national treatment rule. Other nations’ industrial policies, such as “Make in India”[56] or “Made in China 2025”[57] similarly put stress on the national treatment obligation.

The use of industrial policy to favor domestic industries has become particularly controversial in the renewable energy sectors.[58] Renewable energy is key to facilitating an economy-wide shift to green energy. At the same time, nations also view the transition to a green economy as an opportunity to create and support manufacturing jobs and their agricultural sectors. Nations and provinces within nations, including Canada, the United States, the EU, and India, have used local content requirements to privilege their own domestic renewable energy sectors over foreign renewable energy technologies in violation of the national treatment rule.[59]

Beyond nondiscrimination, rules on market access are the other basic component of trade agreements.[60] In the context of trade in goods, the GATT and preferential trade agreements like the USMCA generally prohibit quotas or embargoes on imports of products from other member states.[61] Preferential trade agreements are also supposed to generally prohibit tariffs on imports.[62] The GATT, by contrast, permits tariffs as a general matter.[63] However, WTO members agree to “bind” their tariffs at certain limits in exchange for concessions on maximum tariffs from other WTO members.[64] As a consequence, WTO members each have their own product-specific limit on the tariffs they can levy. Similarly, in the services context, the GATS prohibits WTO members from placing certain limits on the provision of services by foreign providers within sectors of the economy that are individual to each nation.[65]

For years, policymakers and commentators offhandedly noted that trade policy was no longer concerned primarily with tariffs.[66] But tariffs, and the limits upon them, are once again central to trade policy. In part, President Trump’s tariffs on most Chinese imports, as well as his “national security” tariffs on steel and aluminum, a more high-profile and controversial example, explain this revival.[67] But even before President Trump took office, duties on imports designed to combat unfair trade practices had become increasingly widespread.[68] Some countries are also taking a hard look at using tariffs as a tool to fight climate change. Most notably, the EU is set to impose a “carbon border adjustment”—essentially a carbon tariff—on products from countries with carbon regulations that are more lax than the EU’s.[69] President Biden has promised a similar carbon border adjustment “fee.”[70]

These measures require justification under exceptions. Rules allowing duties on unfair trade practices are perhaps the most widely used exception to tariff limits. President Trump and now President Biden have defended U.S. national security tariffs under the GATT’s national security exception. And although the EU hopes to defend its carbon border adjustment measure as consistent with the WTO’s primary rules on tariffs and nondiscrimination, an economy-wide regulation like a carbon border adjustment would likely turn on the availability of exceptions to justify at least some portions of the regulation. If implemented, President Biden’s plan would also likely require reliance on WTO exceptions in order to fall within trade rules.[71]

* * *

International trade’s primary rules on nondiscrimination and market access have driven the incredible rise of global trade and wealth since the end of World War II. But the focus of modern international trade policy has shifted away from the core benefits of integration in favor of concerns around economic inequality within countries, competition among nations to develop new industries that can support good paying jobs, and using trade policy as an instrument to protect the environment, especially from the existential threat of climate change. In today’s world, international trade law’s exceptions have taken center stage. Policies to achieve the main political objectives of contemporary trade policy often violate trade’s primary rules. In order to be consistent with trade rules, these policies therefore must be justified under trade law’s exceptions. Understanding how those exceptions work in practice has thus never been more important.

II. Conventional Approaches to Exceptions

International trade law is riddled with exceptions. Those exceptions serve diverse and seemingly unrelated purposes. Some are for emergency use, such as national security exceptions[72] or the exception for measures to relieve a critical shortage of food.[73] Others can be used at any time, such as the exceptions for measures to protect public morals[74] or conserve exhaustable natural resources.[75] A WTO member can exceed the limits on its tariffs to stop unfair trade practices by other countries[76] or to enforce its own domestic laws.[77] Some exceptions are capacious, such as the one for measures necessary to protect human, animal, or plant life or health.[78] Others are specific, such as the exception for restrictions on the import or export of gold and silver[79] or the exception for measures to protect national treasures of artistic, historic, or archeological value.[80]

Despite the variety of exceptions within trade law, this Part argues that there is a conventional wisdom regarding exceptions. I distill that conventional wisdom into two approaches to exceptions: what I refer to as the Policy Space Theory and the Safety Valve Theory. Under the Policy Space Theory, governments invoke exceptions to accommodate objectives from policy domains other than trade policy, such as to fulfill domestic, social, or environmental objectives.[81] An unlawful trade measure might be justified, for instance, as necessary in order to take aggressive measures to combat climate change.

The Safety Valve Theory, by contrast, holds that governments invoke exceptions to defuse political pressure that trade liberalization itself creates. This pressure typically comes from domestic groups that are harmed by competition from imports. Exceptions for measures that target unfair trade practices, for instance, allow states to reintroduce some protection into the trading system in order to reduce attacks on the trading system as a whole from these interests.

In this Part, I unpack the logic of the Policy Space and Safety Valve Theories. Although many exceptions can be invoked for either Policy Space or Safety Valve reasons, I offer examples of the kinds of exceptions that are more likely to be invoked within each category. Critically, both approaches imagine exceptions as defining a limited unilateral right to deviate from trade obligations for a limited purpose. Exceptions themselves, and trade dispute panels that evaluate their application, must prevent these unilateral rights from being abused for ulterior motives.[82] In this way, both the Policy Space and Safety Valve Theories deny governments the ability to invoke exceptions in order to justify conduct taken from mixed motives—i.e., from both motives that are permissible under trade rules and motives that are not. As a result, governments and trade tribunals both end up applying trade law’s exceptions in a hypothetical vacuum that ignores the domestic political realities of how governments make policy.

A. The Policy Space Theory

Under the Policy Space Theory, exceptions mark the boundaries between policy domains and their associated legal regimes. The intuition is that states may have a difficult time satisfying their treaty obligations while also pursuing a range of reasonable social, environmental, or other public policy goals that are external to international trade policy. Under the Policy Space Theory, the role of exceptions is to allow states to unilaterally pursue those other public policy goals without incurring responsibility for violating their trade obligations.

1. The Idea

International trade agreements set forth their obligations in general terms. The obligation to treat imports no less favorably than domestic products (national treatment) or imports from other countries (MFN) applies almost always, regardless of circumstance.[83] The obligation to provide treatment no less favorable than concessions on market access contained in a tariff schedule or schedule of services commitments is likewise generally applicable.[84]

Yet, governments know at the time they negotiate trade agreements that they will wish to deviate from those general commitments in the future. Trade agreements, after all, are primarily concerned with constraining government policies that protect domestic industries from foreign competition. In effect, they seek to establish processes to reduce tariffs on imports and ensure that regulations are nondiscriminatory. These obligations, in turn, spur economic growth and can bring nations closer together as a matter of foreign policy.[85]

But governments have many other objectives beyond promoting economic growth and closer ties with foreign nations. Just as governments sacrifice economic growth in order to promote worker or consumer safety via domestic regulations, so too might governments be willing to sacrifice the economic growth that accompanies trade liberalization in order to establish policies that address climate change or fight the COVID-19 pandemic. Moreover, different nations have different models for government intervention in the market. State-led systems like China’s might have more interventionist governments than market-led capitalist systems like the United States’. Developing countries like India or Indonesia might push for more export-oriented growth than developed economies with large internal economies.

For years, “policy space” has been the concept that allows the WTO to accommodate this pluralism of goals and economic models. Professor Dani Rodrik, an early and prominent critic of excessive trade liberalization, has argued that globalization should “focus[] on enhancing policy space rather than market access. . . . A good argument can be made that it is lack of policy space—and not lack of market access—which is (or likely to become soon) the real binding constraint on a prosperous global economy.”[86] Much of this early debate focused on policy space for developing countries, which might require time to adjust to the liberalized rules required by the WTO.[87]

More recently, the rise of nationalist movements in developed countries has caused policymakers and commentators to wonder whether the United States and European nations have enough space to pursue their own domestic policies. Donald Trump’s election in 2016, the UK’s decision to leave the EU, and the rise of nationalist parties in continental Europe have all empowered politicians to push back against the WTO, the EU, and other global institutions. Professor Gregory Shaffer has distilled the argument for policy space into a concrete strategy for how states might bargain over market access and policy for domestic priorities.[88] Drawing on the work of Professor John Ruggie, Shaffer argues that the original GATT “provided a framework for embedded liberalism where countries could both liberalize trade and retain policy space for domestic social policy.”[89] He argues that trade officials have lost sight of this balance, and thus a recalibration in favor of policy space is necessary.[90]

In theory, this policy space need not come from exceptions. As Professors Julian Arato, Kathleen Claussen, and J. Benton Heath have argued, “[a] new jurisprudence could focus on identifying space for flexibility within the primary rules themselves.”[91] Yet as they also acknowledge, exceptions are an attractive way for governments to manage their legal exposure to trade and investment law claims.[92] Exceptions clearly exist to create some degree of flexibility within trade law, and prevailing intepretations of trade law’s primary rules focus on minimizing burdens to international commerce.[93] These interpretations encourage challenges to trade restrictive measures that pursue general public policy goals. Thus, absent significant changes to how primary rules are understood, exceptions bear the weight of reconciling states’ goals of trade liberalization with other public policy goals.

In this sense, the Policy Space Theory is closely tied to concerns about the fragmentation of international law.[94] Unlike domestic legal systems, where a generalist legislature often has plenary power over most, if not all, policy domains, the international system has separate institutions for each policy area, and often for discrete issues within a policy domain. The World Health Organization, for instance, is the lead international institution for global public health,[95] while a set of UN and regional institutions govern human rights.[96] International environmental law is a collection of largely unconnected treaties governing discrete issues, such as climate change,[97] ozone depleting substances,[98] or chemicals.[99] Moreover, whereas domestic legal systems usually have rules that prioritize some laws over others, such rules are considerably weaker in international law.

Exceptions offer an attractive solution to this problem of fragmentation. Instead of trying to embed trade rules within a larger governance framework that contemplates environmental, human rights, or public health policies, exceptions minimize the interaction necessary between legal institutions governing different policy domains. Trade lawyers and institutions can carry on within their own domain, and exceptions will tell them when they have strayed outside their competence. Put differently, trade law’s exceptions are the fences that aim to make good neighbors among fragmented and decentralized legal institutions.

2. The Policy Space Theory in Action

In practice, the most significant implication of the Policy Space Theory is that it prohibits states from justifying measures that rest on mixed motives. Under the Policy Space Theory, permission to deviate from trade rules only flows from identifying an objective within a permitted policy domain outside of trade. Policy objectives within trade policy, such as measures that seek to limit trade liberalization’s impact on economic inequality, are therefore not covered by the exceptions. Thus, in invoking an exception, states declare that the policy they are defending is, in fact, a policy designed to pursue some permissible public policy goal external to the trade regime. Other members and WTO panels then evaluate the veracity of that claim. Using objective criteria regarding the measure’s application they ask, in effect, whether the governmental measure in question really is pursuing the government’s stated purpose, or whether the measure is infected by an ulterior motive.

To see how a Policy Space approach to exceptions limits measures with mixed motives, consider the design and application of some WTO exceptions. Many of these exceptions announce policy objectives that WTO agreements “shall not prevent” states from pursuing. For example, GATT contains a national security exception upon which similar provisions in the GATS and TRIPS Agreement are modeled.[100] The GATT was drafted in the immediate aftermath of World War II and during the time of emerging nuclear competition between the United States and the Soviet Union. As such, the United States did not wish for a trade agreement to hinder its national security policies. As Professor Mona Pinchis-Paulsen has described, U.S. negotiators were the primary architects of GATT article XXI.[101] U.S. negotiators included selective, open-ended terms and phrases that “could, when reported to officials in Washington, seem to comply with the demands for total U.S. discretion over security policies in the future.”[102] In relevant part, Article XXI provides that a member may take “any action which it considers necessary for the protection of its essential security interests”[103] provided that “the action taken by a Member under Paragraph 1(b) must relate to fissionable materials, the supply of a military establishment or be taken in a time of war or other international emergency.”[104] Article XXI also excuses measures necessary to comply with obligations under the U.N. Charter, another exception that prioritizes a separate international legal regime.[105]

The WTO’s general exceptions in GATT article XX and GATS article XIV follow the same model, carving out a series of policy purposes that states are permitted to pursue even if their conduct violates trade rules. These purposes include: the protection of public morals;[106] the protection of human, animal, or plant life or health;[107] and securing compliance with domestic laws that are themselves WTO-consistent.[108]

The GATT also includes exceptions for a variety of other issue areas that are specific to trade in goods, including the conservation of natural resources,[109] limitations on products made with prison labor, and the preservation of artifacts.[110] The WTO’s general exceptions also carve out space for other international regimes to operate. The GATT, for instance, provides an exception for measures necessary to comply with certain commodities agreements.[111] The GATS excuses violations necessary to comply with international tax treaties banning double taxation.[112]

Other policy space exceptions may be more targeted. The GATS contains an exception for prudential measures to protect the financial system and participants within it.[113] The CPTPP (i.e., the renamed Trans-Pacific Partnership) contains a carveout in its investment chapter for tobacco measures.[114] The carveout puts tobacco control measures outside the scope of the treaty’s investor-state dispute settlement system. Australia insisted on this carveout to protect itself from what it viewed as spurious attacks on aggressive public health measures it had adopted to curb smoking, which tobacco companies had then challenged.[115] The carveout ensures that CPTPP members may implement tobacco control measures without fear of incurring liability to a foreign investor—a classic policy space rationale.

But while many exceptions identify public policy areas covered by exceptions, simply announcing that a measure pursues a permitted public policy goal is insufficient. The state invoking the exception could be claiming a public policy exception but smuggling in a measure designed to pursue an impermissibile motive, such as protecting a domestic industry. Thus, while trade tribunals have been deferential when asking whether a measure contributes to some permitted public policy objective that states have invoked,[116] they have been considerably more exacting in asking whether a measure only pursues the permitted objective, or whether the measure is trade restrictive in ways that go beyond the permissibile objective.

Tribunals apply a series of means-ends tests to decide whether a measure is sufficiently tailored to a permissible objective, or whether it entails additional restrictions on trade. For instance, most of the general exceptions are qualified by the word “necessary.” The WTO Appellate Body has interpreted the necessity test to require weighing the contribution the measure makes to the state’s public policy goal against the trade restrictiveness of the measure, in light of the importance of the value at stake.[117] Tribunals must also ask whether there were less trade restrictive alternatives reasonably available that make the same contribution to the public policy goal.[118] Finally, and most importantly, the Appellate Body asks whether any discrimination created by the challenged measure “can be reconciled with, or is rationally related to, the policy objective” in question.[119] Any trade restrictiveness must be justified in reference to the permissible policy objective.

A challenge to the EU’s ban on seal products illustrates how this approach can be problematic. The EU’s ban was motivated by concerns for animal welfare, because the process of hunting seal often involved painful deaths for the animals. The EU argued this concern qualified as a public moral under GATT article XX(a). But the EU had several other objectives in passing the measure as well. It wished to preserve traditional seal hunts by indigenous peoples, and it also wished to allow for the culling of seal herds in order to keep seal populations at a sustainable level. To make both of these activities economically viable, the EU allowed seal products derived from either activity. The WTO Appellate Body accepted that the EU’s primary objective—protecting animal welfare—fell within the scope of the exception and that the ban on seal products was necessary to achieve that objective. But the Appellate Body found that the way in which the EU pursued its other objectives—preserving indigenous hunts and seal population management—created unjustifiable discrimination. The EU, in other words, had the policy space to pursue its animal welfare objective, but could not combine that objective in the way that it did with other motives.[120]

WTO panels have been more deferential in applying the national security exception. Even there, though, panels have inquired into the sincerity of a state’s motives. For instance, in both panel reports interpreting the national security exception, the panels have held that the party invoking the exception must articulate its essential security interests.[121] The panels found that they could then ask whether, with appropriate deference to the state invoking the exception, the challenged measure might plausibly be thought to be necessary to further the protection of that interest. In a dispute between Saudi Arabia and Qatar, the panel actually rejected the defense as to one of Saudi Arabia’s challenged measures, finding that a desire to cut off contact with Qatar to minimize terrorism-related risks could not justify a failure to apply criminal penalties to a Saudi entity infringing on the intellectual property rights of a Qatari broadcaster.[122]

In sum, the Policy Space Theory has two key attributes. First, it envisions a de jure space for policies that are inconsistent with trade rules but can be excused because they pursue policy aims outside of trade policy. Second, in application this dichotomy between trade and nontrade purposes leads to an inquiry into whether protectionist features have infected a challenged measure. As I explain more fully in Part IV, by seeking to root out trade restrictiveness unrelated to a permissible policy objective, these tests rule out the possibility that states can use exceptions to justify policies that pursue a mix of permissible and impermissible objectives.[123] In so doing, the Policy Space Theory—an approach to exceptions that seeks to expand the opportunity to pursue public policy objectives without incurring trade liability—risks squeezing that very same space by insisting on unrealistic purity of motive.

B. The Safety Valve Theory

If the Policy Space Theory is concerned with the relationship between trade and nontrade policy objectives, the Safety Valve Theory is concerned only with trade policy. In short, the Safety Valve Theory justifies exceptions on the ground that they bolster a legal regime’s efficacy by reducing the amount of noncompliance in the system. Put differently, the Safety Valve Theory makes exceptions for certain kinds of explicitly protectionist trade policies. By excusing some protectionist policies, Safety Valve exceptions reduce noncompliance by definition. But this reduction in noncompliance has real consequences. It reduces the costs of sanctions that flow from noncompliance and which are usually negative-sum, it increases the compliance-pull the law has, and it reduces the domestic political pressure on governments to withdraw from, or engage in widespread violations of, the international trade system. The Safety Valve Theory has some overlap with the Policy Space Theory in the sense that states may not participate in a legal regime that does not contain exceptions for the pursuit of general policy objectives. But while the Policy Space Theory creates a dichotomy between trade and nontrade policy objectives, the Safety Valve Theory is concerned only with trade policies.

1. The Idea

The intuition behind the Safety Valve Theory is that no legal regime can deter all violations, and violations are costly to a legal system. If too many violations occur, the entire legal regime may unravel. States may be unwilling to join or remain within an international agreement. Alternatively, they may violate the agreement more frequently. In both cases, prior violations have set the stage for later noncompliance by giving cooperation-minded states little incentive to stay within the legal regime. The Safety Valve Theory suggests that undeterrable violations should therefore be legalized.

Domestically, legalizing undeterrable violations means that interest groups that advocate for economic protection do not need to oppose the entirety of the trading system. They can instead operate within the trading system’s rules. At the same time, nations do not observe rampant noncompliance by their peers. International law works most effectively on the margins, “putting a thumb on the scale in favor of compliance.[124] Reducing noncompliance makes law more effective in generating compliance in these close cases. It makes, in effect, the thumb a bit heavier.

In this sense, the Safety Valve Theory represents a second best. It acknowledges that governments sometimes face strong incentives not to comply with international rules. Furthermore, legal obligations will not be enough to overcome these incentives in every circumstance. When these circumstances are predictable in advance, therefore, the Safety Valve Theory suggests calibrating legal rules to minimize the costs of noncompliance. It recognizes the limits of legal regimes and asks them only to do what they can.

To unpack this intuition, notice that violations of the law are costly. In a legal system with perfect enforcement and a system of expectation damages, those costs can be imposed on the violator, such that a rational state will only violate the law if the gains exceed the losses.[125] But enforcement is far from perfect in international law. Moreover, with the exception of international investment law, international economic law does not have a system of expectation damages.[126] Instead, international trade sanctions take the form of raising trade barriers to a violating country’s imports.[127] But increasing trade barriers is typically costly to both the violator and the sanctioning state.[128] The violator loses market access, but the sanctioning state loses access to imports, raising the cost or lowering the quality of the products available to its own consumers. As a consequence, violations of international law tend to be negative sum. The violating state’s gains from flouting the law are less than the joint costs of sanctions, both to the violating and sanctioning states.[129]

At the time states make an agreement, they can be reasonably certain that some violations will be deterred, that some might be deterred, and that some certainly will not be deterred.[130] The possibility of sanctions—negative sum or otherwise—deters some set of violations. If a state is deterred from violating the law by the possibility of sanctions, the negative-sum losses never actually materialize. If, on the other hand, the violation does occur, states collectively suffer the loss imposed by the sanctions. Because states have not agreed to a system of expectation damages, they must instead design their legal agreements ex ante to balance the expected gains from deterring violations with the expected losses from undeterred violations.

Legalizing the last category—undeterrable violations of the law—can help in this balancing by minimizing the losses associated with international law’s negative-sum system of sanctions.[131] Making an undeterrable act unlawful, after all, does not change the costs associated with the act itself. If the act is undeterrable, the consequences of the action for other states—lost market access, environmental externalities, security consequences, etc.—will occur anyway. The only difference is that if the act is unlawful, both the violating state and any sanctioning states will suffer the costs associated with sanctions. For example, imposing tariffs in retaliation for a violation of trade law—as the United States did on a range of European products in retaliation for the EU’s WTO-inconsistent subsidies for Airbus—will increase the costs of affected products like wine and cheese for U.S. consumers without necessarily creating any additional market access for the U.S. exporters.

This “loss avoidance” rationale has implications beyond merely saving costs. Creating exceptions for undeterrable violations can facilitate the creation of a legal regime or save it from collapse. When states are negotiating on the front end, they may be reluctant to join an agreement if it appears that the costs from violations are likely to exceed the gains from more cooperative behavior.[132] Exceptions can change that calculus for states, making an agreement that may not be in the interests of states in general more valuable.

The modesty of ambition that comes with exceptions may also boost the legitimacy of a legal regime, and therefore its ability to generate compliance. Professor Ingrid Wuerth has described the “broken windows” theory of international law, which suggests “that widespread violations of some international legal rules likely make it more difficult to enforce others.”[133] On this view, repeated violations of the law send a signal about a lack of accountability for violations, and can also create a culture within and across states that compliance with the law is not an important policy consideration.[134] On this view, overlegalization can actually undermine the efficacy of a legal regime.[135] By reducing or excusing violations, exceptions prevent this kind of harmful overlegalization, thereby promoting norms of accountability within international law.

These considerations—loss avoidance and broken windows—apply to the value that states as a whole get from international law. But exceptions can also influence the position that domestic interest groups take toward compliance with a legal regime. International agreements typically impose both costs and benefits on member states. In the case of trade agreements, these costs are typically concentrated in industries or sectors that have to compete with cheap imports. These import-competing interests will lobby their governments for protection when cheap imports begin cutting into their market share.

Foreseeing this political pressure to impose economic protection in the future, governments may choose not to join in the first place. They also might choose to withdraw from the agreement, as former President Trump threatened to do with respect to NAFTA and the WTO.[136] Legalizing noncompliance in situations in which negotiators expect domestic political pressure to violate treaty commitments can persuade reluctant states to join an agreement and may even encourage them to make additional concessions at the outset of the agreement.[137] Doing so may reduce the overall value of the agreement to states, but it increases the value to the future violator and, importantly, to government decisionmakers who will actually face domestic political pressure to deviate from international rules.[138]

2. The Safety Valve Theory in Practice

Examples of Safety Valve exceptions are rampant in international trade law. Unlike Policy Space exceptions, which focus on measures that violate trade rules incidentally while pursuing some other purpose, Safety Valve exceptions apply to measures that are explicitly designed to be protectionist. As a result, Safety Valve exceptions generally do not have the kind of probing of a measure to see what a state’s real motive is. More so than Policy Space exceptions, though, WTO jurisprudence has severely limited the use of Safety Valve exceptions. As a result, protectionist motives increasingly doom any government measure.

The classic example of a Safety Valve exception is GATT article XIX,[139] which allows states to temporarily impose “safeguards”—restrictions on imports—in the event that domestic producers face serious injury from a sudden and unexpected surge in imports.[140] Professor Alan O. Sykes has argued that GATT article XIX defuses the pressure for protectionism at the time of negotiations. Government officials know that they are likely to face pressure to protect domestic industries in the future, and so might be reluctant to make trade liberalization concessions. GATT article XIX induces trade concessions by giving government officials some assurance that they can lawfully raise trade barriers in the future, thereby succumbing to domestic pressure without breaching international rules.[141] For similar reasons, Professor Sykes has been critical of Appellate Body rulings that have practically foreclosed the lawful use of safeguards.[142]

Antidumping and countervailing duties have also been compared to safety valves.[143] Like safeguards, antidumping and countervailing duties provide a process through which domestic industries can petition their governments for protection from imports. Antidumping and countervailing duties are essentially tariff surcharges imposed in response to unfair trade practices.[144] Antidumping duties are imposed in response to unfairly low pricing of imports,[145] while countervailing duties are imposed to offset foreign government subsidies.[146]

Antidumping duties have become the most important vehicle through which governments provide protection to their domestic industries.[147] They are used far more frequently than safeguards or countervailing duties.[148] Governments have also interpreted the antidumping rules flexibly to apply to a wide range of circumstances in which states do not agree that they are intended to apply.[149]

For instance, the European Union tried imposing antidumping duties on biodiesel fuel in order to offset foreign government intervention in the market for raw materials. Indonesia and Argentina used their tax system to reduce the cost of palm oil and soybean oil, respectively, in order to make the production of biodiesel cheaper in those countries. But because Indonesia and Argentina were not actually giving support directly to biodiesel producers, proving the existence of an unlawful subsidy was difficult.[150] The EU hoped antidumping duties would provide a permissible vehicle to protect its domestic biodiesel producers.

The United States has used antidumping and countervailing duties in a similar fashion. Arguably the longest-running trade dispute in the world focuses on U.S. claims that Canada unfairly intervenes in its softwood lumber market, effectively creating a subsidy.[151] The United States has also used both antidumping and countervailing duties to target Chinese producers that it feels benefit from extensive government support.[152] More generally, the United States has used a controversial technique known as “zeroing”—which involves calculating the average price of imported goods without including goods that are “fairly” priced, i.e., by zeroing out fairly priced goods in the calculation of average prices—to calculate antidumping duties.[153]

WTO panels and the Appellate Body have almost always taken a dim view of these efforts. In a series of cases, the Appellate Body condemned the practice of zeroing.[154] In response, the United States began blocking the reappointment of Appellate Body members, ultimately leading to the suspension of the Appellate Body’s work due to an insufficient number of members.[155] The Appellate Body and WTO panels have similarly found against the ability of nations to use antidumping or countervailing duties to counteract a range of market interventions that fall short of meeting the WTO’s narrow definition of a subsidy.[156] Along with the similar rulings preventing the application of safeguards, the result has been a significant constriction in the availability of exceptions designed to justify protection.

Safeguards, antidumping duties, and countervailing duties—collectively called “trade remedies”—are the best examples of Safety Valve exceptions because they are designed to co-opt specific domestic interests groups that lobby for greater economic protection. Legalizing that protection reduces the political pressure on the international trade system as a whole. As a result, safety valve exceptions counterintuitively support trade liberalization by authorizing a bit of protection.

Emergency exceptions work in a similar fashion. By legalizing noncompliance during an emergency, when the political pressure on governments to impose protection will be enormous, emergency exceptions reduce the political pressure on the trading system. The best example is the carveout in GATT article XI for “[e]xport prohibitions or restrictions temporarily applied to prevent or relieve critical shortages of foodstuffs or other [essential] products.”[157] GATT article XX contains several other examples, including exceptions for certain export restrictions on inputs as part of a government stabilization plan,[158] and for measures “essential to the acquisition or distribution of products in general or local short supply.”[159] GATT article XII authorizes import restrictions to address balance of payment crises,[160] while TRIPS Agreement contains exceptions for the compulsory licensing of patents, which is especially relevant to pharmaceuticals during a public health emergency.[161]

The primary difference between emergency exceptions and trade remedies is the origin of the political pressure. With trade remedies, political pressure originates from import-competing sectors that are harmed by increases in competition from foreign imports. In an emergency, the demand for protection may come from a wider range of interests, including basic consumers. For instance, many governments imposed export restrictions on medical supplies during the early days of the COVID-19 crisis.[162]

These restrictions did not arise from the need to protect domestic producers, but rather from the need to ensure a supply of medical equipment to consumers—doctors, hospitals, and ultimately patients. States did not challenge each other’s restrictions at the WTO, but the restrictions are likely justifiable under GATT articles XI or XX.[163] Indeed, emergency exceptions in general have not been subject to significant dispute resolution, as WTO members have tended to give each other leeway to respond to a crisis.

* * *

Descriptively, the Safety Valve Theory has some overlap with the Policy Space Theory. In democracies, the invocation of an exception will virtually always be at the behest of a domestic group, whether they do so to advance public policy goals or out of a desire for protection.[164] Likewise, domestic groups may oppose a trade agreement because the agreement does not contain enough policy space.

But the Safety Valve approach to exceptions differs in the justifications for the exceptions and the way they are invoked. Under the Safety Valve Paradigm, exceptions are about ensuring that the international legal regime’s ambitions are appropriately modest. The mantra is that the perfect should not be the enemy of the good. Under the Policy Space Paradigm, by contrast, exceptions do not create a second-best outcome. Instead, they create a first-best world by allowing states to pursue other public policy objectives.

Despite the difference in these approaches, in practice trade dispute resolution panels have taken a narrow view of the ability of exceptions to justify policy goals related to economic protection. As a result, the Safety Valve approach is applied narrowly by tribunals, limiting the usefulness of exceptions in promoting adherence to international trade rules. The Policy Space Paradigm is applied more widely. As I have explained, though, its focus on distinct policy domains effectively denies the use of exceptions to legitimize measures that have mixed motives. Given the prevalence of mixed motives in the real world and the pressure that the resulting measures put on trade agreements, it is time for a new paradigm.

III. The Channeling Paradigm

The Policy Space Paradigm focuses on the relationship between issue areas and legal regimes. The Safety Valve Paradigm focuses on the efficacy of the primary legal regime. Both of these paradigms, though, ignore how exceptions rewire domestic policymaking to produce mixed motive policies. Instead, practitioners operating in both of these paradigms assume that policies that deviate from trade’s primary rules should do so for a single purpose. Moreover, practitioners in both paradigms seek to limit the extent of economic protection that exceptions can justify. As a result, WTO law on exceptions is ill-suited to the contemporary challenges facing trade law.

I therefore introduce the Channeling Paradigm. The intuition behind the Channeling Paradigm is that a certain amount of domestic political pressure for protection from foreign economic activity will remain in any political system. That pressure will seek an outlet, which exceptions provide. Those interested in increasing barriers to foreign economic activity will therefore seek to form political coalitions in favor of policies that both create protection and serve some other objective authorized by the exception. The availability of international legal exceptions, in other words, channels domestic political pressure for otherwise impermissible objectives into policies that plausibly fit within the exceptions. As a result, “exceptional” trade cases are very likely to involve government policies that have mixed motives. A trade jurisprudence that does not have greater tolerance for governments’ mixed motives is one for which governments themselves are likely to have little tolerance.

This Part sets out the descriptive underpinnings of the Channeling Paradigm, while Part IV examines the doctrinal implications. Section A explains how the availability of international legal exceptions influences the design of domestic policies. Political science research has examined how domestic politics influence states’ international behavior, while another strand of research has looked at how international legal regimes constrain domestic politics. International legal exceptions play an important, but overlooked, role in this dynamic. By authorizing domestic policies that fit within the exception, international legal exceptions create incentives for practices similar to logrolling or the use of riders in legislatures. Policies are designed to appeal to multiple constituencies to improve their chances of both passage at the domestic level and being upheld as legal internationally.

Sections B and C examine the effects of this dynamic. I focus on two. First, exceptions promote coalitions between advocates for policies that are disfavored under international trade law. In many cases, those will be groups wishing protection from competition and groups advocating trade-restrictive measures in order to pursue a public good, such as environmental protection. Because policies supporting public goods are ones with diffuse benefits but concentrated costs, advocates of restricting trade in the name of public goods would likely need to ally with other interest groups in order to generate sufficient political support. Protection-seeking interests offer a natural alliance. By creating the incentives for such an alliance, exceptions foster the pursuit of public goods via trade policy.

Second, and for the same reason, exceptions cases are likely to be ones in which governments have mixed motives. They are both genuinely pursuing a permitted public good, while at the same time also seeking to protect their domestic economy. This fact, in turn, creates two dynamics that are harmful to the integrity of the international trade system: it creates incentives for governments to push the boundaries of how much protection they can justify with an exception; and it strains the legitimacy of international trade tribunals, like the WTO Appellate Body, which ultimately must pass on the legality of measures emerging from domestic political coalitions within member states.

A. How International Law Channels Domestic Political Pressure

The basic idea behind the Channeling Paradigm is that exceptions incentivize explicit or implicit bargaining among domestic interest groups. Groups with different potentially trade-restrictive policy objectives, such as an environmental group and a domestic import-competing industry, will unite around a single policy that plausibly fits within an exception. Such policies will be more likely to be enacted domestically and stand a better chance of an international tribunal finding that the measure falls within an exception. At the same time, though, because they are products of multiple policy objectives, “exceptional policies” are very likely to have mixed motives.

Internationally, exceptions make it “cheaper” on the margins for states to pursue policies that fall within the exception, as opposed to policies that fall outside the exception. If an otherwise unlawful policy can fit within an exception, then sanctions for violating international law are less likely.

Domestically, a lower likelihood of international legal sanctions reduces opposition to the policy. Domestic political groups that favor policies that fall outside of those permitted by international trade law would therefore look to craft their policies to fit within the exception. In so doing, they would often work with domestic political groups that favor policies that fall within the core of the exception to craft a policy of which both groups approve. In this way, the two groups generate domestic political support for their favored policy, by simultaneously uniting their political support while also reducing internal objections from trade lawyers and legal and diplomatic pressure from other nations.

To unpack these ideas, I first explain how international legal exceptions reduce the costs of policies that fall within the exception. I then explain how those reduced costs alter bargaining over the design of domestic economic policies.

1. How Exceptions Reduce the Risk of International Sanctions

Consider the point of view of a domestic industry that competes with low-cost imports. To be concrete, think of the renewable energy sector in the United States or the EU. The U.S. solar industry and the EU biodiesel industry, for instance, have both sought protection from their governments from cheap imports (primarily from China in the former case, and Argentina and Indonesia in the latter case).[165] That domestic industry would like to impose barriers, such as tariffs or quotas, that raise the cost of imports and/or limit their ability to enter the market.

Building political support for a measure reducing market access for imports may not be easy, however. As standard trade theory demonstrates, the losses to consumers within the protected market, who would have to pay more to buy the protected product, will exceed the gains to domestic producers.[166] These economic losses will often translate into political opposition to protection. To stick with our example, when the Trump administration imposed safeguard tariffs on imports of solar panels into the United States, the cost of installing solar panels in the United States increased.[167] This impacts end-use consumers, such as homeowners putting solar panels on their roofs. Such consumers tend to be poorly organized, meaning that they often are not able to mount successful political opposition to protectionism that harms them.[168] But the tariffs also impact other workers in the supply chain, such as the companies that import, sell, and install solar panels. Facing higher costs, these business groups will oppose trade barriers. In the United States, the Solar Energy Industries Association (SEIA) led the charge against the Trump administration’s solar tariffs, arguing that the tariffs would cost thirty-one U.S. jobs in solar installations for every job created in solar panel production.[169]

These directly affected domestic consumers—such as the SEIA and homeowners with solar panel on their roofs—may not have enough political clout to block protection. Enter international trade law. Violating WTO rules creates costs for violating states and for constituencies within those states. Where the state itself is concerned, those costs are reputational.[170] If a state (or, in practice, a government) fails to abide by its international commitments, other states may be unwilling to enter into future agreements with the violating state, or may demand better terms in order to do so.[171]

The Trump administration experienced this dynamic firsthand. Its use of “national security” tariffs on steel and aluminum imports, as well as its tariffs on Chinese imports, violate WTO rules in the view of most other WTO members. President Trump talked as if these tariffs would give him leverage to renegotiate U.S. trade relations with other countries. But while he was able to negotiate the so-called “Phase 1” deal with China, as well as renegotiate the North American Free Trade Agreement (NAFTA), the actual changes from the status quo contained in these agreements was quite modest. One commentator, for instance, described the Phase 1 deal with China as “simply restor[ing] the U.S.-China relationship to where it was pre-President Donald Trump, [and declaring] victory in areas that don’t matter.”[172] Republican Senator Charles Grassley described the revised NAFTA (the USMCA) as “95 percent . . . the same as NAFTA.”[173] Given the unlawfulness of the measures leading up to these negotiations, commentators have wondered why other nations would rely on the Trump administration’s commitments in any new agreements.[174] The Trump administration’s meager results from its negotiating strategy suggest that, in fact, other nations did not put much faith in its promises.

Beyond reputational sanctions, trade agreements create the possibility of economic sanctions. These economic sanctions mobilize domestic opposition within the violating country to the unlawful trade measures. They provide a reason for exporters of one product—say, oranges—to lobby against import protection for an unrelated product, such as solar panels.[175] In other words, trade law encourages the formation of coalitions between importers and exporters to lobby for lower trade barriers.

Trade rules create this linkage in two ways. When negotiating trade agreements, exporters in, say, the United States will urge the U.S. government to drop import barriers so that U.S. trading partners, like the EU, will drop trade barriers on U.S. exports. To take one recent example, U.S. exporters of lobster may urge the United States to reduce its tariffs on European glassware and cigarette lighters, so that the EU will reduce its tariffs on U.S. lobster.[176]

The same process works in reverse. If a country violates its trade commitments with unlawfully high trade barriers, the WTO can authorize complaining members to increase their trade barriers on imports from the violating country. The complaining countries will typically choose the products on which they levy higher duties based on political considerations. For instance, the WTO found that the EU had unlawfully subsidized Airbus, the European airplane manufacturer. When the EU refused to remove the subsidies, the United States received permission to retaliate against the EU. The United States did so by imposing tariffs on EU aircraft, but also on food, wine, and spirits. The United States targeted the latter group of products in a bid to hurt politically sensitive export industries that might lobby the EU to remove its subsidies for Airbus.[177]

International trade law thus operates as a kind of force multiplier. It provides domestic export-oriented interests an incentive to make a political common cause with domestic consumers that pay an economic price for their own government’s trade barriers. Trade law violations, such as unlawfully high tariffs or discriminatory domestic regulations, influence and encourage domestic political pressure. The system is designed to work that way.[178]

The political economy of exceptionalism alters the way this dynamic works. If the policy protecting the import-competing industry is designed to fall within one of the WTO’s exceptions, then any violation of trade rules is excused. The violation cannot be the basis for retaliation by other countries. Nor would it be the basis for reputational sanctions. Exceptions are bargained-for rights to deviate from legal rules. Exercising those bargained-for rights would not cause governments to reassess the reliability of legal promises made by other states.[179]

Invoking exceptions, in other words, neutralizes the force multiplier effect that trade agreements have. Exporters have no reason to oppose unlawfully high barriers on imports into their own country if those barriers fall within the WTO’s exceptions, because other countries cannot legally impose barriers to their exports in response. For instance, if the U.S. tariffs on solar energy could be excused as safeguards under GATT XIX, or under GATT article XX as measures related to the conservation of exhaustible natural resources (in this case, the atmosphere), our hypothetical U.S. orange exporters would no longer have a reason to oppose the solar panel tariffs. Domestic consumers are left as the only domestic political opposition to import barriers.

Finally, exceptions can also reduce diplomatic pressure from foreign governments. Governments, of course, can and do object to each other’s perfectly lawful trade policies (which therefore do not implicate reputational sanctions). Governments can convey these objections through committees at the WTO, as well as through bilateral negotiations. But the invocation of an exception in these less legalized contexts may mute the pressure on the importing country to remove its barriers. Bargaining in the shadow of a formal trade dispute, governments may back down from aggressive complaints about policies that seem prima facie justifiable under one of the WTO’s exceptions. The history of invocations of GATT article XXI, the national security exception, are a testament to this dynamic.[180] Governments have invoked GATT article XXI on several dozen occasions since 1947, but only in the last several years have dispute panels been called upon to adjudicate the exception’s meaning.[181]

2. Domestic Politics

From a political economy point of view, exceptions thus work to defang opposition to protectionist policies. By providing a legal excuse for an otherwise unlawful policy, exceptions reduce the domestic and international opposition to protection. In so doing, exceptions create an incentive to try to fit protectionist policies within exceptions. An otherwise unlawful policy stands a greater chance of being adopted if it plausibly falls within an exception. For this reason, exceptions can become important organizing principles for those designing domestic policies. Put differently, exceptions not only break up coalitions opposed to trade barriers; they also create incentives for new coalitions to form to advance trade barriers. The result is that “exceptional” policies are very likely to have mixed motives, a permissible motive authorized by an exception and an impermissible one that is not.

Consider three groups. First, take an organization that wants protection from cheap imports, such as the U.S. solar panel manufacturers or U.S. shrimpers. A simple tariff (or other trade barrier) on foreign imports may not generate the political support necessary to overcome the objections of consumers, export-oriented industries that face retaliation, and foreign governments. For this reason, the U.S. solar or shrimping sector may seek to craft a policy that fits within an exception. Some exceptions, those most closely associated with the Safety Valve Paradigm, focus on the protected industry itself and thus may not require designing a policy that appeals to other interest groups. In the United States, the Tariff Act of 1930 requires the government to impose antidumping or countervailing duties if the government concludes that unfair trade practices are injuring domestic industry.[182] Most exceptions, though, require a purpose unrelated to the protected industry, such as the conservation of natural resources, the protection of human life or safety, or national security. To fit within these exceptions, protection-seeking industries will design policies that arguably pursue these purposes.

In so doing, they often will support a policy that does not provide as much protection as they would like. Instead, they design a policy that they hope will attract more political support, delivering as much protection as is politically feasible. For example, the U.S. shrimp or tuna industry may support policies that limit imports that are captured in ways that harm sea turtles or dolphins, respectively, in order to join their political support for protection to the environmental community.[183]

Limiting imports to only those that are caught in ways that harm other marine life will not limit imports as much as U.S. shrimpers or tuna fisherman would like. But it makes the protection they do receive more politically viable. Domestic politicians may be willing to endorse laws that provide protection on environmental grounds, such as the Marine Mammal Protection Act in the United States. That Act requires the government to ban the import of commercial fish caught using methods that incidentally harm marine mammals.[184] Such laws can also be defended under GATT Art. XX(g)’s exception for the conservation of natural resources, reducing the likelihood of economic sanctions and the intensity of domestic and diplomatic pushback.[185]

Our second group is the environmental community. Its calculus is different. That community—or more generally the community interested in any particular policy area, such as national security or public health—wishes to build support for policies that advance its objectives. In today’s globalized world, those policies will often have a trade component. The business lobby in favor of low trade barriers is quite strong in the United States.[186] Every presidential administration in the United States since the Truman administration has either pursued or completed a new trade agreement.[187] Public policies that address issues with widely diffuse benefits, such as climate change, environmental protection, or public health, may have trouble generating political support in the face of business groups’ opposition to trade barriers. Public interest groups may thus calculate that policies that appeal to domestic business interests are more likely to pass.

Public interest groups, in other words, may look to craft policies that benefit one set of economic interest groups in order to offset the opposition from another set of economic interests. The environmental community and the U.S. seafood industry have very different incentives in supporting an import ban conditioned on environmental conditions. But the U.S. seafood industry understands that the United States is more likely to enact and sustain a ban on imports contingent on environmental policies than a straight-up ban on imports. And the environmental community knows those measures are more likely to pass when they enjoy support from a domestic industry.[188]

The EU’s push for renewable energy provides another case in point. The EU has issued a series of Renewable Energy Directives that seek to decarbonize the European economy. For environmentalists, pushing European industry to shift to renewable energy represents a win for the climate and thus for European citizens (and, in fact, citizens globally). But the EU explicitly justified its push for renewable energy in terms of both environmental objectives and local economic development objectives.[189] And when cheap imports of biodiesel began to undermine the EU biodiesel industry, the industry pushed for protection.[190] That protection first took the form of antidumping duties. But when the WTO ruled that the European antidumping duties did not fall within the WTO rules permitting such duties, the EU switched to an environmental justification: palm oil-based biodiesel, which happens to be from Indonesia and Malaysia, is produced in ways that lead to unacceptable levels of deforestation.[191] Now the EU faces a WTO challenge to its environmental ban on palm oil-based biodiesel, one that will likely end up turning on the application of GATT article XX(g)’s exception for measures related to the conservation of natural resources.[192]

The design of a policy to satisfy both members of an “exceptional” coalition can, of course, involve explicit negotiations between representatives of the two groups. Such negotiations could occur directly, or within a legislative or administrative body in which government officials represent the interests of the two groups. But the support does not necessarily have to take the form of explicit negotiations. The industry might, instead, simply choose to support the group pushing the public policy while remaining outside the limelight. Support of this kind could include financial support for nonprofit groups or to fund litigation, or behind-closed-doors lobbying in support of the policy. These kinds of nontransparent support may actually be more effective, as they avoid drawing attention to the role of import-competing industries in what might be called “greenwashing” their policies.[193]

Policies may also be designed to appeal to multiple constituencies even without explicit negotiations between the two groups. Government officials may act entrepneurially to design policies with widespread appeal. Litigation can also be used to shape policies. For example, waste tires from automobiles are an environmental and public health risk, in part because they serve as a breeding ground for malarial mosquitos. In order to cut down on the number of waste tires, the Brazilian government issued regulations blocking the import of foreign-sourced used and retreaded tires.[194]

The EU eventually challenged this policy, culminating in the Appellate Body’s 2007 report rejecting Brazil’s public health defense.[195] Prior to the EU’s challenge, though, Brazilian retread manufacturers obtained domestic court orders enjoining enforcement of the used tire import ban, while the embargo on finished foreign retreads remained.[196] As a result, domestic producers of retreaded tires were able to rely on a steady supply of cheap, foreign used tires to refurbish and resell.[197] They were also protected from competition from foreign producers of retreaded tires. The result was a mixed-motive measure, one that advanced Brazil’s public health aims by reducing the number of imported tires, but also protected Brazil’s domestic producers of retreaded tires. In a decision consistent with the Policy Space Paradigm, the Appellate Body found the latter motive unjustifiably discriminatory.[198]

That brings us to our third group: government lawyers. By appealing to government lawyers, exceptions can neutralize internal governmental opposition to protectionist policies. International trade, like virtually every other area of international relations, has become increasingly legalized in recent decades.[199] Commentators often describe the formation of the WTO and the creation of the Dispute Settlement Body as part of the “juridification” of international trade.[200] Lawyers and the logic of law have replaced diplomats and the methods of diplomacy as a means of addressing disputes. But government lawyers do far more than defend domestic policies before domestic courts and international tribunals. They also review domestic policies before they are ever implemented. Their concerns with the legality of a particular policy under international law can result in the policy being quashed.[201] Moreover, government lawyers are often prone to excessive caution. They may advise against policies that are plausibly legal in order to avoid legal risks.

The availability of exceptions makes it easier to persuade government officials generally, and lawyers in particular, not to oppose a policy during internal government deliberations. Exceptions reduce the risks that government lawyers are trained to spot and minimize. But the possibility of using an exception also gives lawyers a basis to help craft a policy. Instead of arguing that a policy will be unlawful, government lawyers can suggest modifications to the policy that make successful invocation of an exception more likely. In this way, exceptions engage legal expertise in support of getting to “yes” within internal deliberations.

Government lawyers’ use of exceptions to craft and justify policy internally was on display during the Trump administration. In 2017, President Trump reportedly held a meeting with his top advisers, including his new chief of staff John Kelly, in which he told Kelly to “[b]ring me some tariffs!”[202] Despite President Trump’s wishes, though, many of his advisers opposed large tariff increases.[203]

Government officials resolved this impasse between President Trump and his advisers by marrying the President’s protectionist inclinations to a legal exception for national security measures. In April 2017, President Trump directed Commerce Secretary Wilbur Ross to begin an investigation into whether imports of steel threatened the national security of the United States.[204] That report proceeded under Section 232 of the Trade Expansion Act, the domestic counterpart to GATT article XXI, allowing the president to raise trade barriers in response to national security concerns.[205] President Trump made clear in his public remarks that his concerns were about protecting the steel industry and steel workers’ jobs, not classic national security concerns. For instance, in a 2018 visit to Granite City, Missouri, he said, “Thanks to our tariffs, idle factories throughout our nation are roaring back to life.”[206]

Ultimately, the Commerce Department’s ability to identify national security as a legally available basis for the tariffs paved the way for the tariff hawks in the Trump administration to prevail. The Commerce Department produced a report concluding that “[t]he displacement of domestic steel by imports has the serious effect of placing the United States at risk of being unable [to] meet national security requirements.”[207] National security exceptionalism thus became the legal justification for tariffs that had little to do with conventional national security concerns.[208] Instead, legal exceptions shaped the process and justification for otherwise unlawful protectionist measures.

* * *

In sum, the Channeling Paradigm suggests that exceptions will act like magnets, attracting the exact kind of protection-seeking impulses that international economic law’s general rules are designed to constrain. As a result, governments’ “exceptional” policies are not likely to resemble the idealized measures envisioned by the Policy Space Paradigm in particular. Instead, measures are likely to meld permissible and protectionist objectives in a single measure in order to improve the overall measure’s political viability. The existence of exceptions drives the creation of mixed-motive measures.

The WTO’s list of major exceptions cases, many of which I have referred to above, provides the empirical basis for this claim. United States—Shrimp and United States—Tuna both involved environmental laws to protect marine life (sea turtles and dolphins, respectively) but that also benefitted from domestic support from the U.S. fishing industry. Brazilian courts refined the public health measure at issue in Brazil—Tyres, as discussed above, to make it friendlier to domestic producers seeking protection. In EU—Seals, the Appellate Body found the EU’s ban on seal products inconsistent with GATT article XX’s exception for public morals because the measure explicitly tried to pursue multiple objectives—protecting seals, permitting indigenous seal hunts, and allowing sustainable culling of seal herds—in a way that, in the Appellate Body’s view, produced arbitrary or unjustifiable discrimination. India—Solar Cells rejected a GATT article XX defense aimed at shielding subsidies for renewable energy programs that included local content requirements. And as discussed below, the United States’ ban on flavored cigarettes ultimately fell due to an exemption for domestically produced menthol cigarettes that emerged as a necessary compromise during Congressional debates.[209] In each of these instances, a government policy justified under an exception to general WTO rules fell because the policy pursued more than one objective, often a political necessity to ensure the policy’s enactment.

B. How Trade Exceptionalism Promotes Public Policy

In this Section, I argue that, within limits, the fact that exceptions attract protection is a feature, not a bug. The benefits of channeling flow from diverting protectionist pressure that will always exist in society to serve the public interest. Exceptions channel protectionists pressure into advocacy for policies that fall within the scope of exceptions. If the substantive scope of those exceptions is tied to important public policy values that are underrepresented in the political process—prominent examples include public goods like climate change or environmental protection—this channeling effect can serve a beneficial purpose. By yoking inevitable political pressure for protection to the pursuit of some other public goal, exceptions in international trade law can promote the pursuit of important “nontrade” values. Exceptions, in other words, act as the same kind of force multiplier for general public policy considerations that the primary rules of trade perform for trade liberalization—using reciprocal trade barriers as a means to incentivize the formation of domestic political coalitions.

To see how exceptions promote the pursuit of underserved public policy goals, notice that many of the public policy goals that are authorized by WTO exceptions are public goods. GATT article XX(a) protects public morals, which in practice states have invoked to protect animal welfare, a public good. GATT article XX(b) protects public health measures, while TRIPS articles 30 and 31 govern, in practice, compulsory licensing for pharmaceuticals, a major public health issue in the age of COVID-19. GATT article XX(g) involves the protection of natural resources. Even national security, covered by GATT article XXI (as well as similar exceptions in the GATS and TRIPS Agreement), is a public good, since anyone living within a nation benefits from the nation’s efforts to protect its citizens.

A public good is, by definition, nonrivalrous and nonexcludable.[210] Nonexcludability is particularly important. It means that people get the benefit of the public policy regardless of whether they pay the costs of the policy.[211] For instance, measures to address climate change, which would be covered by GATT article XX(g), are public goods.[212] The entire world shares in the benefits of greenhouse gas emissions reductions by the EU, regardless of whether they pay the costs of reducing their own emissions. Public health, covered by GATT article XX(b), is a similar kind of public good. Vaccination can be costly to the vaccinated, especially if pharmaceutical companies use their patent rights to charge exorbinant prices. But vaccination reduces the risk that the unvaccinated will contract the disease in question. Widespread vaccination against COVID-19, for instance, will reduce the risk to those who do not have the vaccine.[213]

Public goods are traditionally undersupplied by governments.[214] The reason is that public goods impose concentrated costs but, because public goods are nonexcludable, diffuse benefits available to many. Because those footing the bill for a public good cannot consume all of the benefits for which they pay, they will not pay for as much of the public good as would be socially optimal.[215] Moreover, because people can consume the good—a safe climate, avoiding disease, protection from foreign threats—without individually paying for it, they will often free ride. Public goods, in other words, impose concentrated costs on those who provide the good, with diffuse benefits that anyone can enjoy.

The traditional solution to this problem is to link the provision of public goods with individual, excludable benefits.[216] If access to an individual, excludable benefit is tied to contributing to a public good, people would be more willing to contribute to the public good. Taxes, for instance, solve this problem for many government-provided goods, such as national defense. Avoiding criminal prosecution for not paying taxes is a private benefit that induces citizens to contribute to the public goods the government provides. Market access via trade agreements is also a classic selective benefit that the United States has linked to human rights and national security objectives abroad, both during the Cold War and the second Bush Administration’s War on Terror.[217]

Just as market access—a reduction in trade barriers—can serve as an excludable benefit internationally, the creation of trade barriers can serve as an excludable benefit domestically. By permitting economic protection, exceptions can thus provide domestic economic interest groups a reason to lobby for public goods. Labor unions or domestic industries, for instance, may become concerned with animal welfare or climate change in order to secure protection from imports.

The organizational problem common to public goods that leads to underrepresentation in the political process can also apply to other forms of political failure. For instance, labor interests are frequently disadvantaged in developed countries.[218] In countries like the United States, the relative weakness of labor interest groups, like unions, stems in part from legal regimes that make organizing more difficult. But in part, it stems from the fact that a decline in trade barriers has allowed companies to locate production in countries with the lowest labor costs. This disadvantages labor interests in negotiations with capital. More generally, the poor and working class are often underrepresented in the political process.

Exceptional protection can solve this underrepresentation problem. Protectionist groups can justify their policies if they take up the banner of under- or unrepresented groups. This phenomenon is easy to see in environmental cases. When U.S. shrimping or fishing interests lend their support to import bans that protect sea turtles or dolphins, the unrepresented wildlife find an unlikely champion.[219] Likewise, the EU biodiesel industry has favored protection that has denied the benefits of the EU’s incentives for biofuels to products from countries that do not engage in sustainable land-use practices.[220] By arguing for measures that limit market access to biodiesel produced through deforestation in countries like Indonesia, Malaysia, and Argentina, the EU biodiesel industry may further climate change and biodiversity objectives that often get short shrift in the political process.

Indeed, the initial inclusion of labor and environment chapters in preferential trade agreements owes much to this logic. The United States first insisted on such chapters in NAFTA and has strengthened them in subsequent agreements.[221] The EU later began including similar requirements in its agreements. Although the United States initially pushed for these chapters to raise environmental and labor standards in Mexico for the benefit of Mexican workers and citizens, a significant reason for the higher standards was to make production in Mexico more expensive.[222] Increased costs would reduce the movement of jobs to Mexico from the United States, and thereby reduce imports into the United States from Mexico. U.S. labor unions thus became champions for labor and environmental rights in Mexico.

Channeling thus plays an important and welfare increasing role. To the extent that exceptions are tied to underserved public policy goals, exceptions redirect protectionist pressure into the service of those goals.[223] Doing so scrambles the usual welfare calculus. The ordinary story is that economic protection is inefficient and therefore welfare decreasing. Perhaps for this reason, pre-WTO GATT panels in particular were leery of giving broad scope to exceptions.[224]

But if economic protection results in the provision of public goods or other underserved public policy goals, then the welfare calculus is more complicated. If, for instance, local content requirements linked to renewable energy programs result in greater public investment in a transition to a green economy than would otherwise occur, the economic inefficiency created by the local content requirement may be offset by future reductions in greenhouse gas emissions.[225] Likewise, any discrimination that may arise in the EU’s carbon border adjustment would have economic consequences, but the benefits from making the EU’s climate change regulations more effective likely outweigh those economic costs. Buy American provisions, if effectively designed to promote job growth among those segments of the population who have lagged behind economically, could provide benefits to communities and families that greatly exceed any economic inefficiencies introduced into supply chains.

Although not commonly associated with exceptions, the benefits of channeling are well known within the GATT/WTO. The traditional understanding of the GATT/WTO rules is that they seek to ensure that all protection “take[s] the form of tariff protection”[226]—a process known as “tarrification.” Article XI bans all prohibitions or restrictions on both imports and exports, other than duties, taxes, or other charges.[227] Once inside a country, imports “should be assimilated to domestic products and be subjected to a regulatory regime identical to that applied to domestic products.”[228] Nondiscrimination guarantees this parity of treatment behind the border.[229] These rules forbid regulatory protection, while tariffs are permitted within the limits established for each country in its tariff schedule.

The incentives to channel protection into tariffs is thought to create benefits in the form of greater transparency[230] and incentives for WTO members to reduce their tariffs over time through reciprocal negotiations.[231] But this account of the tarrification of protection tells only half the story. Just as the GATT’s primary rules channel protection into tariffs, they also leave the possibility of channeling protection into the exceptions. Because the GATT’s primary rules close off regulatory protection, protection that cannot easily be channeled into tariffs can still be channeled into exceptions. Although the benefits of channeling protection into exceptions is different from the benefit associated with channeling protection in tariffs—the promotion of underserved public policy goals—they are no less important. Nor, as I argue in Part IV, are they any less deserving of solicitude from dispute resolution panels.

C. The Perils of Erosion

Channeling has significant benefits. The possibility of exceptional protection rewires domestic politics in a way that can promote underrepresented political interests. At the same time, channeling also presents significant risks for the trading regime. Like water coursing through a riverbed, channeling can result in erosion of the constraints imposed by trade law. Seeking more room for protectionist policies, governments may over time broaden the amount of room they seek for protection. Doing so can undermine the balance of rights established by the WTO Agreements and ultimately undermine the legitimacy of the trading system as a whole.

The sudden importance of the WTO’s national security exceptions illustrate the risk. Prior to 2017, no WTO panel had ever interpreted the national security exception. Governments invoked the exception in diplomatic disputes, but had shied away from putting it before a WTO panel. Since 2017, though, national security has been invoked in five sets of disputes. President Trump’s national security tariffs on steel and aluminum are the most well known. However, Russia’s successful invocation of the exception in a dispute with Ukraine,[232] and Saudi Arabia’s partially successful invocation of the exception in a dispute with Qatar[233] have actually produced the first two panel reports interpreting the exception. Japan has also invoked the exception in a dispute with Korea, while the United States has recently invoked the exception in a dispute with Hong Kong.[234]

To some extent, the sudden use of the national security exception represents an expansion of the concept of national security.[235] But to a significant extent, its more expansive use represents an erosion of the norm of restraint. As Professor Pelc has described, WTO members have generally exercised restraint in their use of exceptions in order to preserve both the viability of the trade regime overall and to make the signal sent by invoking an exception especially serious.[236]

The erosion of this restraint today is incontestable. Regardless of whether the United States’ steel and aluminum tariffs meet the capacious definition of national security contained in the U.S. domestic law purportedly authorizing them, section 232 of the Trade Expansion Act,[237] they almost certainly are not responding to any “emergency in international relations” as required by GATT art. XXI.[238] Yet the steel industry, and government officials in the Trump Administration with steel-industry ties (such as Commerce Secretary Wilbur Ross), wanted to impose protection and thus sought to fit concerns about overcapacity in global steel production into the national security exception.[239]

Likewise, while the use and abuse of antidumping measures has long been chronicled,[240] countries have tried to use antidumping duties to respond to an ever-broadening range of measures. The EU, as noted above, has tried to apply antidumping duties to respond to upstream tax incentives that have the effect of making domestic inputs cheaper.[241] Such a methodology might have broad application to nonmarket economies like China. The EU, though, first tested the methodology in the market for biofuels. Argentina and Indonesia imposed export taxes on crops used to make biodiesel that exceeded the export taxes on finished biodiesel. As a result, the crops became cheaper for biodiesel refiners in Argentina and Indonesia to acquire. It is not clear that such measures qualify as subsidies under the WTO agreements,[242] so antidumping duties became the EU’s preferred method of protecting its domestic producers of biofuels. But even that approach required bending the antidumping rules beyond what they plausibly say.[243]

While the erosion of restraint, especially as regards national security, is undeniable, the extent of the risk from erosion is less clear. Commentators have long expressed concern that any expansion of the amount of “exceptional” protection puts the enterprise of trade liberalization in peril.[244] In theory, of course, that is possible. But the lived experience of the GATT/WTO has not been thus. That gradual liberalization that has characterized the trade regime testifies to the fact that the current state of trade law is not a necessary endpoint; it is only one potential spot on a continuum.

More to the point, though, the pushback against a greater role for exceptions is the proximate cause of the current WTO crisis. The United States paralyzed the WTO’s Appellate Body—and thus in a real sense the possibility of mandatory binding dispute resolution for all WTO members, since an appeal “into the void” blocks the adoption of a panel report—in large part because it was unhappy with the Appellate Body’s narrow interpretations of antidumping and countervailing duty rules.[245]

Arguing that the erosion of restraint represents the real threat to the WTO thus involves a certain amount of myopia. To be sure, WTO members, especially the United States, are pushing for more room to pursue exceptional policies in ways that facilitate a domestic economic agenda that sits in some tension with the early twenty-first century style of trade liberalization. But doubling down on restraint has not shackled the protectionist pressures. It has shifted them from relatively well-constrained exceptions, like GATT article XX and antidumping rules, to claims of national security exceptionalism. It has, in other words, caused these pressures to shift from claims about how WTO exceptions should be interpreted into an existential crisis for the institution.

The way forward, then, has to begin with an approach to exceptions that acknowledges the political realities that underlie the trading system’s legal exceptions. Governments are demanding a new approach with respect to exceptions. Drawing on the insights of the Channeling Paradigm, the next Part turns to describing what such an approach might look like.

IV. Exceptionalism and Mixed Motives

As I have suggested above and explain in further detail below, trade panels have applied exceptions in ways that limit states to a single permissible objective. In so doing, tribunals have overlooked the reality that most state policies—especially those that invoke general exceptions—will be mixed motive cases. Alternatively, tribunals may have overestimated the extent to which states wish to use dispute resolution as a tool to constrain protectionist elements that domestic politics will often inject into otherwise permissible public policies.

Either approach to interpreting exceptions under WTO law is highly problematic. On February 21, 2021, in the context of a dispute over labeling exports from Hong Kong “Made in China,” the United States made clear that it will continue to press a broader scope for action under the GATT’s national security exception.[246] This stance throws cold water on the notion that the Biden administration might change from the Trump administration’s view on exceptionalism at the WTO.[247] With cases challenging the EU’s Renewable Energy Directive, and likely challenges to any carbon border adjustment the EU or the United States might apply, as well as possible cases stemming from emergency COVID measures, reinforcing the cramped notion of WTO exceptions seems a recipe for further instability in the multilateral trading system.

Trade law thus needs to rebalance its approach to exceptions. A new approach must recognize the need to constrain blatant attempts at protection, while also recognizing the fact that state policies that pursue legitimate regulatory objectives are likely to suffer from mixed motives in at least some degree. As the Channeling Paradigm suggests, allowing and even encouraging states to pursue those legitimate objectives requires greater tolerance for secondary motives that would be impermissible standing alone.

To that end, this Part proposes two changes in how trade law deals with exceptions: (1) a shift in how trade dispute panels approach mixed motive cases, and (2) revising the language of exceptions in new trade agreements. The goal of both reforms is to allow states the freedom to pursue mixed motive policies so long as the predominant motive is a permissible one.

A. Handling Mixed Motive in Trade Law Cases

In this Section, I argue that future WTO tribunals should adopt a “predominant motive” test for general public policy exceptions, such as those contained in GATT article XX and GATS article XIV. Panels should uphold the invocation of an exception motivated by a legitimate regulatory interest so long as the legitimate regulatory interest is the predominant motive. An impermissible objective would only be fatal to the invocation of an exception if the impermissible objective were the predominant motive.

1. Justifying a Predominant Motive Test

Mixed motive cases are common across the law, arising in areas as diverse as free speech, employment law, securities regulation, and tax.[248] The basic issue in these areas, as in international trade law, is how to analyze the legality of a measure that has two motives, one permissible and another impermissible. When does the impermissible motive so infect the permissible motive as to render the measure unlawful?

Happily, trade panels need not start from scratch. Courts and tribunals have applied a myriad of tests to answer this question.[249] Which test is appropriate depends on the context and the purposes of the rules in question. Professor Andrew Verstein has identified four tests that are most common: the Predominant Motive test, the But-For Motive test, the Sole Motive test, and the Any Motive test.[250] Under the Predominant Motive test, a challenged measure is only unlawful if the impermissible motive is the primary one. If the permissible motive is predominant, the measure survives despite the existence of an additional, impermissible motive.[251]

Under the But-For Motive test, a challenged measure is only unlawful if the impermissible motive is a “but-for” cause of the measure’s enactment. In other words, if the government would have enacted the measure even in the absence of the impermissible motive, then the measure is lawful.[252] Under the Sole Motive test, a measure survives unless the impermissible purpose was the only motive for the measure.[253] Intuitively, under the Any Motive test, the measure is struck down if there is any impermissible motive. In other words, the challenged measure must be entirely free from impermissible motives.[254]

A Predominant Motive test most closely aligns with the purposes of general public policy exceptions. Most telling is the fact that, unlike in many areas of the law in which mixed motives present themselves, we are here dealing with exceptions.[255] Public policy exceptions only arise in cases in which a tribunal has already found that a measure violates the primary rules. The entire purpose of an exception is thus to allow some measure of otherwise unlawful conduct to persist.

The context and preamble of the GATT in 1947 and the WTO Agreement in 1995 further make clear that member states did not mean to eliminate the possibility of pursuing objectives in ways that are at odds with the primary rules of the trading system. The GATT 1947, for instance, began by noting that international trading relations “should be conducted with a view to raising standards of living, ensuring full employment and a large and steadily growing volume of real income and effective demand, developing the full use of the resources of the world and expanding the production and exchange of goods.”[256] To this, the Marrakesh Agreement Establishing the World Trade Organization added sustainable development, environmental protection, and preservation.[257] The negotiating history of the exceptions further underlines the multiplicity of objectives member states had in mind. This negotiating history is most clear in regards to GATT article XXI, the national security exception. As Professor Pinchis-Paulson makes clear, U.S. GATT negotiatiors “did not seek to make exceptions that would create an open-ended, unchecked power” for member states to deviate from trade rules.[258] At the same time, negotiators “sought to prioritize U.S. national security [over trade policy considerations] in drafting the security exceptions.”[259]

In short, exceptions exist in trade agreements not just to create policy space but to allow states to prioritize other objectives. A panel’s review of an exceptions claim thus necessarily involves balancing the rights of members under the WTO agreements with the unilateral rights of the defendant state to pursue its other objectives.[260] In balancing those interests, though, panels cannot be blind to the way in which states actually make policy. Government policies are rarely, if ever, the product of a single mind or a single purpose. As the Channeling Paradigm suggests, policies are the product of compromises among a range of interests with varying aims.

Member states are, of course, aware of how their own political processes work. Thus, in excusing violations of trade rules for certain purposes, it stands to reason that states understood that they would have a margin of appreciation to pursue those objectives through the ordinary, messy, political process. Peer review by other members and ultimately dispute resolution would guard against abuses of the exceptions, but ordinary political bargaining should not constitute an abuse. A Predominant Motive test best captures this aim. It allows states to engage in ordinary politics, while also constraining efforts to completely hijack the use of exceptions for ulterior motives.

In so doing, a Predominant Motive test would address some of the legitimacy challenges that the Appellate Body and trade dispute resolution have continued to face. The number of major challenges coming down the pike that are likely to turn on exceptions is significant, including the U.S. invocation of the national security exception in response to the Chinese efforts to suppress democracy in Hong Kong and the likely reliance on GATT article XX to justify the EU’s 2018 Renewable Energy Directive and its forthcoming carbon border adjustment. These cases are likely to hit at a time when WTO dispute resolution’s legitimacy remains fragile, in light of the ongoing U.S. insistence on reform to the Appellate Body. Stiff reliance on the Policy Space Paradigm, with its insistence on single motive measures, would make the road back to a functioning dispute settlement system difficult indeed.

2. The WTO’s But-For and Any Motive Tests

Unfortunately, the Appellate Body’s existing case law ignores these realities. Like other areas of the law in which mixed motive cases appear, the Appellate Body’s doctrinal approach is not always clear nor consistent with prior applications. However, the Appellate Body’s various approaches often resemble an Any Motive test. Such a test is consistent with the Policy Space Paradigm, in which any trade restrictiveness must be in the service of a permissible objective. Any trade restrictiveness stemming from an impermissible objective renders the exception inapplicable.

To see the inflexible approach in existing WTO case law, notice that a WTO member’s regulatory objective is irrelevant to evaluating whether a violation of the WTO’s primary rules exists in the first place. WTO tribunals evaluate a challenged measure as a trade measure under the WTO’s primary rules. They then evaluate the challenged measure as a public policy measure under the exceptions, looking to see if there is evidence of trade restrictiveness unrelated to the claimed objective.

For a time, it looked like WTO tribunals might abandon this stark division between the economic and public policy aspects of a challenge measure. In the Dominican—Cigarettes case, the Appellate Body wrote that a detrimental impact on imports might be “explained by factors or circumstances unrelated to the foreign origin of the product . . . .”[261] In other words, a measure might not unlawfully discriminate if differences in treatment could be explained by a legitimate regulatory purpose.[262]

In later cases, however, the Appellate Body abandoned this approach for nondiscrimination cases under the GATT, holding that discrimination exists when like products are not accorded equal competitive opportunities.[263] Regulatory purpose thus plays no role in evaluating a measure’s consistency with the GATT. Regulatory purpose is only relevant to an exception analysis. This complete absence of a role for regulatory purpose is akin to an Any Motive test. Any protectionist element—defined with reference to the provisions of the WTO agreements alleged to have been violated—is too much.

Moving to exceptions, the picture is murkier, but the bottom line is similar. A substantial line of cases hew to an Any Motive approach in denying the application of exceptions. Start with nondiscrimination under the Technical Barriers to Trade (TBT) Agreement. The TBT Agreement does not have an exceptions clause, so the Appellate Body has found that—unlike for most other nondiscrimination provisions—regulatory purpose is relevant under the TBT.[264] The Appellate Body was explicit that this role for regulatory purpose is akin to that played by an inquiry into legitimate regulatory objectives, such as that conducted under GATT article XX.[265]

The Appellate Body made regulatory purpose relevant under the TBT Agreement by adding an additional step to the analysis of whether treatment is “less favorable,” and thus constitutes discriminatory treatment. Under the ordinary nondiscrimination test, a panel finds treatment “less favorable” when it finds that a measure operated to the detriment of the equality of competitive opportunities.[266] Under the TBT Agreement, however, such detrimental impacts do not amount to a violation if a panel finds that the “detrimental impact on imports stems exclusively from legitimate regulatory distinctions.”[267] Although not formally an exception, this additional requirement functionally operates as one. And by its express terms, this test is an Any Motive test. Unless discriminatory treatment is entirely explained by a legitimate objective, the measure is unlawful. In other words, any indication of an illegitimate objective is enough to doom a measure. The Appellate Body has directed panels to look for purity in how states make policy.

This search for purity has significant negative consequences. United States—Clove Cigarettes illustrates the problem.[268] In the Family Smoking Prevention and Tobacco Control Act of 2009, the U.S. Congress prohibited the production or sale in the United States of flavored cigarettes, but allowed the sale and production of menthol cigarettes.[269] Indonesia challenged the ban as discriminating among like products—clove cigarettes from Indonesia and menthol cigarettes from the United States.[270] As a matter of legislative history in the United States, the exception for menthol cigarettes appears to have been demanded by the Congressional Black Caucus on the grounds that menthol cigarettes were popular with its constituents.[271] Passing the broader health measure thus required weakening the measure somewhat through a concession to a politically necessary group in Congress.

This concession proved fatal at the WTO. Both the panel and the Appellate Body ultimately found that section 907 discriminated against Indonesian clove cigarettes in violation of Article 2.1 of the TBT Agreement.[272] Ignoring the domestic political process entirely, the panel and the Appellate Body each concluded that menthol cigarettes and clove cigarettes both appeal to youth and therefore could not be distinguished based on the United States’ purpose of preventing youth smoking.[273]

The decision in Clove Cigarettes aims for Solomonic wisdom. Although ultimately striking down the U.S. measure, the Appellate Body essentially read the possibility of an exception directly into Article 2.1’s nondiscrimination rule. The Appellate Body thus created the possibility of a future claim in which distinctions based on regulatory purpose might survive. On the other hand, by forcing any detrimental impact on imports to stem from that distinction, the Appellate Body’s decision effectively forecloses the possibility that mixed motive measures will survive. A measure that is the product of both public policy and protectionist forces would, almost by definition, never create detrimental impacts that stem exclusively from a legitimate regulatory distinction.[274] Rather, the detrimental impact will almost certainly stem at least in part from the desire to protect some domestic industry.

The difficulty with this approach is twofold. First, in Clove Cigarettes, the domestic political process in the United States was trying to protect menthol cigarettes from generally applicable domestic regulation, not from competition with clove cigarettes. Even with respect to youth smoking, Indonesia submitted that 1.1 million out of approximately 2.5 million youth smokers regularly smoked menthol cigarettes, while only 6,800 regularly smoked clove cigarettes.[275] This data is damaging to the U.S. argument about regulatory purpose, showing that the United States continued to allow youth access to a preferred cigarette. But it also suggests strongly that the United States was not animated by the purpose of discriminating against Indonesian cigarettes. They simply did not occupy a sufficient market share.

Second, these kinds of technical decisions that ignore the political realities in WTO member states create the risk of a backlash. Important constituencies within member states may not see the value in complying with the institution’s rules if those rules require a politics free of compromise. As Tomer Broude and Philip Levy argued in the context of this case, a consideration of regulatory purpose is virtually unavoidable in nondiscrimination cases.[276] But divorcing that consideration from the context in which national governments negotiate those purposes threatens to render hollow member states’ ability to pursue those public policy purposes.

Although the tests are more complicated, the picture is essentially the same in the context of the explicit general exceptions in GATT Article XX and GATS Article XIV. As noted above, most WTO general exceptions require that a measure be “necessary” to achieve a legitimate public policy. WTO panels evaluate necessity in two steps. First, they ask whether the contribution the measure makes to a legitimate regulatory objective outweighs the measure’s “trade restrictiveness, in light of the importance of the values at stake.”[277] This approach resembles a Predominant Motive (or more accurately, a Predominant Effect) test. Does the pursuit of the legitimate objective outweigh the pursuit of the illegitimate objective?

Second, though, panels ask whether there are reasonably available measures that make the same contribution to the regulatory objective but are less trade restrictive.[278] This element resembles an Any Motive test. Although the Appellate Body has not articulated this requirement in terms of intent, which remains technically irrelevant, in practice this test invalidates any trade restrictiveness above the minimum required for the state to achieve its objective. It denies, in other words, the state invoking the exception the discretion to choose among possible measures that differ in their trade restrictiveness.

Even where a measure is found “necessary” under this test, the chapeau of Article XX requires that any discrimination the measure creates not be arbitrary nor unjustifiable, and the measure must not constitute a disguised restriction on trade.[279] “[T]he relationship of the discrimination to the objective of a measure is one of the most important factors” in assessing whether a measure discriminates arbitrarily or unjustifiably.[280] The Appellate Body has said that “there is arbitrary or unjustifiable discrimination . . . when a Member seeks to justify discrimination resulting from the application of its measure by a rationale that bears no relationship to the accomplishment of the [permissible] objective . . . .”[281] This approach is essentially the same as that under the TBT Agreement: any discrimination must extend exclusively from a legitimate regulatory objective. Any secondary motive is fatal.

3. A Predominant Motive Test for the WTO

A Predominant Motive test is thus more consistent with the object and purpose of the WTO’s exceptions: balancing the government’s right to regulate with the objectives of trade liberalization. Such a test would need to make its way into WTO law in two ways: in evaluating the means-ends relationship between a measure and its purpose, and in the nondiscrimination element that the Appellate Body has read into the chapeau of the WTO’s general exceptions.

Converting the “necessity” from an Any Motive test to a Predominant Motive test is relatively straightforward. As discussed above, the “necessity” test is a two-stage inquiry. First, the party invoking the exception must show that the contribution the measure makes to its objective outweighs the trade restrictiveness, in light of the importance of the value at stake. Second, the party opposing the exception has the opportunity to show that there are reasonably available, less trade-restrictive alternatives. Simply eliminating that second step would convert the “necessity” test into a Predominant Motive test—one that asks whether the legitimate objective is the primary one that the measure pursues.

A similar shift in the nondiscrimination test would preserve the chapeau’s function and be consistent with its object and purpose, while granting states more room to engage in the normal bargaining that leads to public policy. Instead of insisting that any discrimination be justified with reference to the permissible policy objective, WTO panels should ask whether the predominant reason for the discrimination is the pursuit of the permissible objective. Such a shift would remain consistent with the text of the chapeau, which offers no guidance as to how panels should evaluate what constitutes “arbitrary” or “unjustifiable” discrimination.

Two objections to this approach are worth considering. The first is evidentiary: how would a panel assess motive? WTO panels, however, already have an approach that they use to evaluate the purpose of a measure developed in the context of the GATT’s affirmative nondiscrimination rules.[282] That approach does not call for looking at the subjective intent of government officials. Instead, it looks at the design, architecture, and anticipated application of a measure as objective evidence of what the government intended.[283] Beyond the measure itself, evidence can include information, such as studies about how a measure would apply, in the public record prepared by the lawmaking body.[284] WTO panels are, in other words, already assessing government motives and could continue to do so in the same manner. The change would be to the quantum of motive necessary to doom a measure, not to how motive is evaluated from an evidentiary standpoint.

Second, the scope of some WTO exceptions is potentially so broad that those exceptions could conceivably come to swallow the rules. The national security and public morals exceptions are perhaps the two most obvious candidates. As I argued above, the risk that these exceptions will be broadly invoked seems outweighed in the current moment by the risk that WTO rules will unduly hamper efforts to address systemic societal risks such as climate change or pandemics.[285] Moreover, though, safeguards against abuse remain in place. This concern about abuse goes to the permitted ends of the exception, but a Predominant Motive test—an evaluation of the means used to pursue those ends—would continue to rule out measures that are primarily driven by impermissible motives. Politics and diplomacy also remain an important check on overreach. The basic trade liberalizing infrastructure has many beneficiaries that would stop measures that wholesale destroyed those benefits before they were enacted. Finally, panels could also take a narrower approach to interpreting the scope of broad exceptions like public morals and national security. As panels have done in the national security context, for instance, panels could require more than simply a bare claim that national security or public morals is implicated, inquiring instead into the provenance of the claimed interest or moral.[286]

These modifications would require changes in WTO case law. But these changes should not be cause for heartburn among trade lawyers. Although it was often not transparent in its precedential practices, the Appellate Body not infrequently declined to follow its prior decisions.[287] Moreover, the Appellate Body’s technocratic adherence to and extension of precedent is a major reason that the United States paralyzed the institution, a trend that the Biden Administration has continued.[288] Changes to WTO case law on exceptions thus seem a critical part of restoring the Appellate Body (or replacing it with a new mechanism). These changes would go a long way toward that purpose.

B. Redesigning Exceptions Clauses

States continue to negotiate new trade agreements and therefore new exceptions clauses. The Channeling Paradigm suggests that these new clauses should reflect the reality that exceptional measures are likely to pursue multiple objectives. Most general and security exceptions currently in use were either drafted at the end of World War II or are based on provisions that were. But the prevailing approach to trade agreements at that time was one of embedded liberalism, which saw trade liberalization coexisting with substantial discretion for states to pursue social policies at home.[289] With a trade dispute system that increasingly infringes on states’ ability to engage in normal policymaking through coalitional bargaining, as the Channeling Paradigm explains, a new approach is necessary. To that end, this Section analyzes shifts (or lack thereof) in the drafting of general exceptions clauses and proposes textual changes to codify a Predominant Motive test.

The main text of the GATT has resisted change since 1947. That lack of change, though, obscures the expansion of exceptions into new areas of trade and economic law. This should hardly come as a surprise. Beginning in the 1980s, states have negotiated new trade agreements at a rapid clip. The Uruguay Round closed in 1994 with an expansion of trade rules into services (the GATS) and intellectual property (the TRIPS Agreement). Preferential trade agreements have exploded. The United States, Canada, and Mexico brought NAFTA into force in 1994, and a renegotiated version came into force in 2020. Overall, the United States has preferential trade agreements that cover trade with twenty nations.[290] The EU has forty-four preferential trade agreements covering trade with seventy-six nations.[291] The EU and its trading partners concluded a number of these agreements in the last five years. China reports sixteen preferential trade agreements covering forty countries (although some countries are members of multiple agreements).[292]

Each new trade agreement forces states to revisit the proper scope of exceptions, and thus is an opportunity to revise and tinker with the language in exceptions clauses. Some treaties have innovated in the use of exceptions, especially to deal with particularly thorny issues. The Trans-Pacific Partnership, as noted above, excepted tobacco control measures from the agreement’s investor-state dispute settlement procedures.[293]

Article 16 of the Northern Ireland Protocol between the UK and the EU provides another example. Perhaps the most difficult issue associated with the UK leaving the EU involved Northern Ireland. The Good Friday Agreement, a key component of peace in Northern Ireland, requires the absence of a “hard” border between Northern Ireland, which is part of the UK, and the Republic of Ireland, which is an EU member state.[294] When the UK was also within the EU, this absence of a border presented few problems. Once the UK left, however, a separate agreement was necessary to preserve the open border. That agreement, the Northern Ireland Protocol, keeps an open border on the condition that a regulatory border be established between Northern Ireland and the rest of the UK, in order to keep Northern Ireland from serving as a backdoor from the UK into the EU.[295] Article 16 provides an exception to this open-border requirement, though, for situations that cause “economic, societal or environmental difficulties.” The EU has already invoked the exception to prevent COVID-19 vaccines produced in the EU from making their way into the UK, although it reversed its decision after a public outcry.[296]

Outside of these specialized contexts, though, states have been relatively unambitious in how they have taken on treaty exceptions. For instance, the general exceptions clause in the GATS, concluded in 1994, largely mirrors the general exceptions clause in the GATT, concluded in 1947. Moreover, the differences go to the permissible regulatory objectives themselves.[297] The language regarding means-ends and nondiscrimination, which governs the freedom states have to design measures that pursue multiple objectives, is identical.[298] Other agreements concluded in the Uruguay Round, such as the TBT Agreement and the Agreement on Subsidies and Countervailing Measures, do not include general exceptions at all.[299]

More recent WTO negotiations have doubled down on the GATT’s approach, focusing on measures that are necessary and do not discriminate arbitrarily or unjustifiably. In February 2021, the draft negotiating text of a possible WTO agreement on e-commerce leaked.[300] Member states had made two suggestions regarding general exceptions. Both, though, essentially incorporated the GATT’s general exceptions approach, either by copying the text or by incorporating GATT Article XX by reference.[301] Proposals for the national security exception similarly either copied the text of GATT Article XXI, adapting it to e-commerce rather than trade in goods, or incorporated GATT Article XXI by reference.[302] As relevant here, the proposed exceptions would all reflect the GATT Article XX emphasis on the impermissibility of mixed motives by incorporating the necessity test and the nondiscrimination test developed under GATT Article XX.[303]

General and security exceptions have also made their way into investment treaties. Professor Caroline Henckels found that 43 percent of such agreements concluded between 2011 and 2016 included a general exceptions provision, while only 7 percent concluded from 1959 to 2010 included an exceptions provision.[304] These provisions “incorporate by reference or are modelled on the general exceptions in Article XX of the [GATT] and Article XIV of the [GATS] . . . .”[305]

An alternative approach within new trade agreements would borrow from the experience of investment treaties. In response to decisions that they perceived as overreaching, a number of states began including clarifications in the investment chapters of their trade agreements and standalone investment treaties. This language provides that “[n]on-discriminatory regulatory actions by a Party that are designed and applied to protect legitimate public welfare objectives, such as public health, safety and the environment, do not constitute indirect expropriations . . . .”[306]

This language was adopted to clarify the meaning of rules prohibiting expropriation, but other countries have adopted the means-ends relationship—“designed and applied to”—to their general exceptions clauses.[307] Specifically, they have modeled the scope of their general exceptions clauses on GATT Article XX, but replaced “necessity” with “designed and applied to.” They have also retained the chapeau of Article XX and its nondiscriminatory requirement.

This approach dramatically improves the workability of general exceptions clauses. Unlike the necessity test, the “design and applied to” standard does not preclude the possibility that a measure had more than one motive. It merely requires that one purpose—as evidenced by the measure’s design and application—is a permissible regulatory objective. The original context of this language’s development further underlines this point. The “designed and applied to” standard was created to provide defendant states greater discretion in how they pursue legitimate regulatory objectives. This negotiating history further makes clear to treaty interpreters that multiple objectives should not be fatal to invoking an exception. By contrast, new exceptions clauses that mirror or incorporate by reference existing WTO exceptions risk codifying, in the eyes of future treaty interpreters, existing WTO case law on exceptions.

To be sure, states have adopted the nondiscrimination requirement from the chapeau of GATT Article XX even in those treaties that have discarded the necessity test.[308] In recent cases like EU—Seals and Brazil—Tyres, the nondiscrimination requirement has been the stumbling block to successfully invoking the exception. Modifying the language of GATT Article XX in new agreements could make clear that a Predominant Motive test applies to the nondiscrimination test in the chapeau as well. Currently, that language provides that: “Subject to the requirement that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade . . . .” The chapeau could be rewritten to codify a Predominant Motive test in the following way: “Subject to the requirement that arbitrary or unjustifiable discrimination between countries where the same conditions prevail is not the predominant objective of such measures . . . .” This language makes two changes. First, it deletes the prohibition on “a disguised restriction on trade.” This language suggests that tribunals should look for hidden motives, a requirement that should be eliminated in favor of asking directly whether a legitimate objective constitutes the challenged measure’s primary motive. As the Channeling Paradigm suggests, multiple objectives are more likely the norm in policy making. Giving tribunals license to root out motives they deem “disguised” seems particularly unwise in the current context of fragility in international dispute resolution.

Second, this language makes clear that arbitrary or unjustifiable discrimination is only problematic if it is the predominant objective of discrimination among countries where like conditions prevail. This language preserves the fundamental nondiscrimination requirement of the chapeau. At the same time, though, it expands the discretion states have to choose among measures that might incidentally discriminate among countries.

In so doing, it will expand the margin of appreciation member states enjoy from second guessing by dispute panels. Trade panels woul still be able to strike down measures that are predominantly protectionist or discriminatory. But mixed motive measures that predominantly pursue permitted objectives will survive. That rebalancing between the right to regulate and the objectives of trade liberalization would bolster the legitimacy of trade law and its dispute settlement function at a time when such legitimacy is sorely needed.

V. Whither Public Policy Within the Trade System?

During most of the twentieth century, the United States treated international trade policy primarily as an instrument of foreign policy.[309] Trade liberalization helped rebuilding efforts after World War II and shored up support for the United States during the Cold War. In this context, the exceptions in international trade law primarily fit within the Policy Space and Safety Valve Paradigms. Exceptions demarcated the boundaries between different issue domains and created an outlet for domestic protectionist pressures.

By the end of the Cold War, however, U.S. trade policy was less and less moored to concrete foreign policy goals. To be sure, the George W. Bush administration entered into a number of trade agreements with countries that supported its war on terror, but those agreements were for the most part not economically significant for the United States.[310] Likewise, the Obama administration cited concern about China as a foreign policy justification for its Trans-Pacific Partnership.[311] But most of the terms in the agreement seemed to bear little relationship to this strategic concern.

Instead, trade agreements became increasingly about domestic economic policy and rent seeking by domestic interests. The intellectual property provisions of trade agreements offer one of the clearest examples.[312] Efforts to boost intellectual property rights overseas serve no foreign policy goal. They are motivated by efforts to increase the economic returns to multinational enterprises (usually U.S. based) that have business models based on the development of innovative technologies and medicines.[313] Likewise, efforts to boost regulatory harmonization promoted the economic interests of transnational business communities, but did little to promote foreign policy objectives.[314] Even the reduction in tariffs achieved through free trade agreements like NAFTA and the dramatic expansion of the WTO in the decades after the end of the Cold War had significant domestic economic effects. These agreements allowed multinational companies to disaggregate their supply chains across countries, moving jobs offshore. While the globalization of supply chains had beneficial effects for companies, and led to job growth in some communities within the United States, it also led to significant and long-lasting economic turmoil in communities in which jobs were shipped offshore.[315]

Despite this tremendous shift in the international trade landscape, however, the basic contours of trade agreements have been slow to change. The WTO has been a particularly difficult ship to turn. While many preferential trade agreements now at least include labor and environment provisions, the WTO still has no renegotiated settlement between trade liberalization interests and other public policy goals that may conflict with trade liberalization. Instead, WTO agreements create room to pursue those objectives through exceptions. It is thus critical that the interpretation of those exceptions accurately reflect the underlying political pressures that lead to their invocation.

Unfortunately, the WTO’s case law on exceptions is not up to the task. Today, the United States and the EU worry about whether WTO rules unduly constrain efforts to address domestic economic inequality that trade liberalization has exacerbated. Many developed countries worry that without aggressive action on climate change, in ways that are sure to be challenged at the WTO, humanity faces an existential threat. Policies that address these issues are sure to require the very kind of bargaining among domestic interests that the Channeling Paradigm predicts will occur around international legal exceptions.

Dispute panels and treaty negotiators must adapt to this reality. The hope, shared by many, that Donald Trump’s departure from office would mark a return to the status quo ante in the trading system is sure to be disappointed. The EU is moving forward with imposing a carbon border adjustment mechanism that will surely be challenged as discriminatory.[316] Similarly, President Biden has promised to impose a carbon border adjustment “fee,” which, if imposed as a fee, would be even more difficult to justify.[317] The United States seems likely to demand continued flexibility in how it responds to China and its highly interventionist economic model. And a more egalitarian trade policy—whether under the guise of “Trade Policy for All” or a “worker-centric” trade policy—remains central to how governments conceive the purpose of trade policy.

While in theory solutions to these problems exist that do not violate trade law’s primary rules, the Channeling Paradigm suggests that governments are not likely to choose those options. Instead, the most practical solutions to the pressing trade policy problems of the twenty-first century are likely to be mixed motive policies. In this context, the Appellate Body’s growth in recent years into a body that recognizes greater policy space for states to discriminate in the pursuit of a public policy objective, so long as they are not blatantly discriminating, may not be enough to reassure reluctant states, especially the United States. Instead, any path to a rejuvenated WTO dispute settlement system must include an approach to exceptions that accommodates the domestic political bargains necessary to tackle contemporary problems.

  1. * Professor of Law, Vanderbilt University Law School. For helpful comments and suggestions, many thanks to Julian Arato, Joseph Blocher, Anu Bradford, Rachel Brewster, Jessica Clarke, Kathleen Claussen, Harlan Cohen, Ben Heath, Larry Helfer, Simon Lester, Arti Rai, Suzanna Sherry, Ganesh Sitaraman, Todd Tucker, Matt Waxman, Ingrid Wuerth, and participants in workshops at Columbia Law School and Duke Law School. Thomas Hildebrand provided first-rate research assistance.

  2. . Steven Erlanger & Somini Sengupta, Europe Unveils Plan to Shift from Fossil Fuels, Setting Up Potential Trade Spats, N.Y. Times (Sept. 15, 2021), [].

  3. . See Timothy Meyer & Todd N. Tucker, A Pragmatic Approach to Carbon Border Measures, 20 World Trade Rev. 1 (2021) (cataloguing nations’ efforts to adopt carbon border measures); Rachel Morison & Anna Edwards, UK Considers Carbon Border Tax to Protect Domestic Industry, Bloomberg (May 28, 2021, 3:21 AM), [].

  4. . Exec. Order No. 14,005, 86 Fed. Reg. 7,475 (Jan. 28, 2021).

  5. . Exec. Order No. 14,017, 86 Fed. Reg. 11,849 (Mar. 1, 2021).

  6. . Statements by the United States at the Meeting of the WTO Dispute Settlement Body [hereinafter Statements], at 11 (Feb. 22, 2021), [].

  7. . Infrastructure Investment and Jobs Act, H.R. 3684, 117th Cong. §§ 70901, 70911 (2021) (enacted).

  8. . Matina Stevis-Gridneff, Amid Critical Shortage, E.U. Moves to Limit Vaccine Exports, N.Y. Times (Mar. 4, 2021), [].

  9. . Christopher A. Casey & Cathleen D. Cimino-Isaacs, Cong. Res. Serv., IF1151, Export Restrictions in Response to the COVID-19 Pandemic (2021).

  10. . See Earth Island Inst. v. Christopher, 19 Ct. Int’l Trade 1461, 1466 (1995) (quoting the affidavit of the president of the Georgia Fisherman’s Association (GFA), which stated “if everyone in the shrimping industry were [subject to the regulation], it would allow [GFA members] to compete evenly with foreign shrimpers”); see also Appellate Body Report, United States—Import Prohibition of Certain Shrimp and Shrimp Products, WTO Doc. WT/DS58/AB/R (adopted Nov. 6, 1998) [hereinafter United States—Shrimp].

  11. . See Appellate Body Report, Brazil—Measures Affecting Imports of Retreaded Tyres, WTO Doc. WT/DS332/AB/R (adopted Dec. 17, 2007) [hereinafter Brazil—Tyres].

  12. . See EBB Position: Red II Delegated Act on Low & High ILUC-risk Biofuels, Eur. Biodiesel Bd. (Feb. 2019), [].

  13. . Panel Report, Russia—Measures Concerning Traffic in Transit, ¶ 7.4, WTO Doc. WT/DS512/R (adopted Apr. 26, 2019) [hereinafter Russia—Traffic in Transit]. See also Pramila Crivelli & Mona Pinchis-Paulsen, Separating the Political from the Economic: The Russia-Traffic in Transit Panel Report, 20 World Trade Rev. 582 (2021).

  14. . Inception Impact Assessment, Carbon Border Tax Adjustment Mechanism (2020), [].

  15. . Appellate Body Report, European Communities—Measures Prohibiting the Importation and Marketing of Seal Products, ¶ 2.107, WTO Doc. WT/DS400/AB/R (adopted June 18, 2014) [hereinafter EU—Seals].

  16. . See, e.g., Panel Report, United States—Certain Measures Relating to the Renewable Energy Sector, ¶ 2.5, WTO Doc. WT/DS510/R (circulated June 27, 2019); Appellate Body Report, India—Certain Measures Relating to Solar Cells and Solar Modules, ¶ 5.46, WTO Doc. WT/DS456/AB/R (adopted Oct. 14, 2016); Appellate Body Report, Canada—Measures Relating to the Feed-in Tariff Program, ¶ 5.38, WTO Doc. WT/DS426/AB/R (adopted May 24, 2013).

  17. . Press Release, Dep’t of Homeland Sec., DHS Cracks Down on Goods Produced by China’s State-Sponsored Forced Labor (Sept. 14, 2020), [].

  18. . See Harlan Grant Cohen, Editorial Comment, What Is International Trade Law For?, 113 Am. J. Int’l L. 326, 329–332 (2019).

  19. . Following Jorge Viñuales, I use the term exceptions throughout this article “to refer to a wide variety of techniques that provide special legal treatment to certain situations otherwise governed by a general rule.” Jorge E. Viñuales, Seven Ways of Escaping a Rule: Of Exceptions and Their Avatars in International Law, in Exceptions in International Law 65, 66 (Lorand Bartels & Federica Paddeu eds., 2020). I do not limit the definition to exceptions in the strict sense. In the context of economic law, for instance, I treat antidumping and countervailing duties as exceptions to ordinary GATT rules on tariffs, even though they are not exceptions in the strict sense like those exceptions found in GATT article XX.

  20. . Tom Verducci, Astros Epitomize the Importance of Adjustments, Sports Illustrated (Oct. 13, 2021), []. See also Leonard Cohen, Anthem, on The Future (Columbia Records 1992) (“There is a crack, a crack in everything. That’s how the light gets in.”).

  21. . See, e.g., Exceptions in International Law, supra note 18.

  22. . See, e.g., Gregory Shaffer, Retooling Trade Agreements for Social Inclusion, 2019 U. Ill. L. Rev. 1 (2019).

  23. . General Agreement on Tariffs and Trade, art. XX, Oct. 30, 1947, 61 Stat. A-11, 55 U.N.T.S. 194 [hereinafter GATT].

  24. . See, e.g., Caroline Henckels, Indirect Expropriation and the Right to Regulate: Revisiting Proportionality Analysis and the Standard of Review in Investor-State Arbitration, 15 J. Int’l Econ. L. 223, 224–25 (2012) (“[T]he use of an appropriately deferential approach to proportionality analysis in the consideration of indirect expropriation claims . . . would provide greater policy space for host states to take measures in the public interest . . . .”); Barnali Choudhury, Recapturing Public Power: Is Investment Arbitration’s Engagement of the Public Interest Contributing to the Democratic Deficit?, 41 Vand. J. Transnat’l L. 775 (2008).

  25. . Andrew T. Guzman & Timothy L. Meyer, International Soft Law, 2 J. Legal Analysis 171, 192–93 (2010).

  26. . See Ingrid Wuerth, International Law in the Post-Human Rights Era, 96 Tex. L. Rev. 279 (2017) (discussing how too many violations of international law may undermine the legitimacy of the system as a whole).

    • . Alan O. Sykes, Protectionism as a “Safeguard”: A Positive Analysis of the GATT “Escape Clause” with Normative Speculations, 58 U. Chi. L. Rev. 255, 259 (1991) (“When self-interested political officials must decide whether to make trade concessions under conditions of uncertainty about their political consequences, the knowledge that those concessions are in fact ‘escapable’ facilitates initial trade concessions . . . .”).

  27. . GATT, supra note 22, art. XIX; Sykes, supra note 26, at 273. The persistent objector doctrine, which acts as a state-specific exception from rules of customary international law, offers a more general international law example of an exception that might be justified on loss avoidance grounds. See Andrew T. Guzman, Saving Customary International Law, 27 Mich. J. Int’l L. 115, 164, 167–68 (2005).

  28. . United States—Shrimp, supra note 9.

  29. . See Panel Report, European Union—Anti-Dumping Measures on Biodiesel from Indonesia, WTO Doc. WT/DS480/R (adopted Feb. 28, 2018) [hereinafter Biodiesel from Indonesia]; Appellate Body Report, European Union—Anti-Dumping Measures on Biodiesel from Argentina, WTO Doc. WT/DS473/AB/R (adopted Oct. 26, 2016) [hereinafter Biodiesel from Argentina]; see also Carolyn Fischer & Timothy Meyer, Baptists and Bootleggers in the Biodiesel Trade: EU–Biodiesel (Indonesia), 19 World Trade Rev. 297 (2020).

  30. . Infrastructure Investment and Jobs Act, H.R. 3684, 117th Cong. § 70952(6) (2021) (enacted).

  31. . Throughout this Article, I use the term “motive” interchangeably with the term “objective.” Both refer to a government’s purpose in enacting a measure as revealed through objective criteria such as the design, architecture, and application of a measure. This is consistent with how government purposes are evaluated by trade panels. See Appellate Body Report, Chile—Taxes on Alcoholic Beverages, ¶ 71, WTO Doc. WT/DS110/AB/R (adopted Jan. 12, 2000). I do not use the term “motive” to refer to subjective intent.

  32. . See Annex to the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions: Trade Policy Review—An Open, Sustainable and Assertive Trade Policy, COM (2021) 66 final (Feb. 18, 2021).

  33. . See generally Tony Cole, The Boundaries of Most Favored Nation Treatment in International Investment Law, 33 Mich. J. Int’l L. 537 (2012).

  34. . See, e.g., GATT, supra note 22, arts. I, XIII; Agreement on Technical Barriers to Trade, art. 2.1, Apr. 15, 1994, 1868 U.N.T.S. 120; Protocol Replacing the North American Free Trade Agreement with the Agreement Between Canada, the United States of America, and the United Mexican States art. 2.10, Nov. 30, 2018, 134 Stat. 11, (entered into force July 1, 2020) [hereinafter USMCA], []; Comprehensive Economic and Trade Agreement art. 2.4, Can.-EU, Oct. 30 2016, O.J. (L 11) 23 [hereinafter CETA],

  35. . See, e.g., General Agreement on Trade in Services art. II, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1B, 1869 U.N.T.S. 183, 33 I.L.M. 1167 (1994) [hereinafter GATS]; USMCA, supra note 34, art. 15.4; CETA, supra note 34, art. 9.5.

  36. . See, e.g., Agreement on Trade-Related Aspects of Intellectual Property Rights art. 4, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1C, 1869 U.N.T.S. 299, 33 I.L.M. 1197 (1994) [hereinafter TRIPS].

  37. . See, e.g., USMCA, supra note 34, art. 14.5; CETA, supra note 34, art. 8.7. There is no multilateral investment agreement, such as through the WTO. Instead, investment agreements are either included as chapters within preferential trade agreements, or are concluded as standalone agreements.

  38. . G.C. Hufbauer, J. Shelton Erb & H.P. Starr, The GATT Codes and the Unconditional Most-Favored-Nation Principle, 12 L. & Pol’y Int’l Bus. 59, 63 (1980) (“U.S. negotiators insisted on a strong statement in article I of the unconditional MFN principle. This insistence grew from U.S. hostility to colonial trading systems . . . .”).

  39. . See Jagdish Bhagwati, Termites in the Trading system: How Preferential Agreements Undermine Free Trade (2008).

  40. . See Chris Brummer, Minilateralism: How Trade Alliances, Soft Law, and Financial Engineering are Redefining Economic Statecraft (2014); Laurence R. Helfer, Regime Shifting: The TRIPs Agreement and New Dynamics of International Intellectual Property Lawmaking, 29 Yale J. Int’l L. 1 (2004) (arguing that countries shift negotiations among international institutions to achieve their negotiating objectives).

  41. . GATT, supra note 22, art. XXIV; GATS, supra note 35, art. V (providing that an economic integration agreement must have “substantial sectoral coverage” and “provide[] for the absence or elimination of substantially all discrimination”).

  42. . In fact, the WTO’s exceptions contemplate both “free trade” agreements and “customs unions.” The latter, which include the EU, also have a common external trade policy. See GATT, supra note 22, art. XXIV; GATS, supra note 35, art. V.

  43. . See infra Section IV.B for a longer list of such agreements.

  44. . See Mitsuo Matsushita, Thomas J. Schoenbaum, Petros C. Mavroidis & Michael Hahn, The World Trade Organization: Law, Practice, and Policy 179–92 (2015).

  45. . See GATT, supra note 22, art. III.2.

  46. . See id. art. III.4.

  47. . See, e.g., id. art. III; Agreement on Technical Barriers to Trade art. 2.1, Apr. 15, 1994, 1868 U.N.T.S. 120 [hereinafter TBT Agreement]; USMCA, supra note 34, art. 2.3; CETA, supra note 34, art. 2.3.

  48. . See, e.g., GATS, supra note 35, art. XVII; USMCA, supra note 34, art. 15.3; CETA, supra note 34, art. 9.3.

  49. . See, e.g., TRIPS, supra note 36, art. 3.

  50. . USMCA, supra note 34, art. 17.3; CETA, supra note 34, art. 8.6.

  51. . GATS, supra note 35, art. XVII.1 (footnote omitted) (“In the sectors inscribed in its Schedule, and subject to any conditions and qualifications set out therein, each Member shall accord to services and service suppliers of any other Member . . . treatment no less favourable than that it accords to its own like services and service suppliers.”).

  52. . Todd N. Tucker, Roosevelt Inst., Industrial Policy and Planning: What It Is and How to Do It Better 6 (2019), [].

  53. . Exec. Order No. 14,005, 86 Fed. Reg. 7,475 (Jan. 28, 2021).

  54. . Exec. Order No. 14,017, 86 Fed. Reg. 11,849 (Mar. 1, 2021).

  55. . Make in India, [].

  56. . Scott Kennedy, Made in China 2025, Ctr. for Strategic & Int’l Studs. (June 1, 2015), [ MF5D].

  57. . See, e.g., Timothy Meyer, Explaining Energy Disputes at the World Trade Organization, 17 Int’l Env’t Agreements: Pol., L. & Econ. 391 (2017) (discussing the raft of WTO challenges to clean energy subsidies).

  58. . See id.; Timothy Meyer, How Local Discrimination Can Promote Global Public Goods, 95 B.U. L. Rev. 1937 (2015) (cataloguing U.S. state laws that contain green local content requirements).

  59. . Trade agreements, especially preferential trade agreements, apply a host of more specific rules to trade in goods, services, technology (i.e., intellectual property), and capital (i.e., investment). A comprehensive cataloguing of these rules is well beyond the scope of this Article.

  60. . See, e.g., GATT, supra note 22, art. XI. There are market access rules for services as well, see GATS, supra note 35, art. XVI, but they apply only to sectors that states have opted in, and the rules are vaguer and have not been tested to the same extent through disputes.

  61. . See GATT, supra note 22, art. XXIV.8(a) (defining customs unions and free trade areas as groups of territories in which “the duties and other restrictive regulations of commerce . . . are eliminated on substantially all the trade between the constituent territories . . . .”).

  62. . See GATT, supra note 22, art. II.

  63. . Id. art. II.1(a) (“Each contracting party shall accord to the commerce of the other contracting parties treatment no less favourable than that provided for in the appropriate Part of the appropriate Schedule annexed to this Agreement.”).

  64. . GATS, supra note 35, art. XVI.

  65. . See, e.g., Edward John Ray, Changing Patterns of Protectionism; The Fall in Tariffs and the Rise in Non-Tariff Barriers, 8 Nw. J. Int’l L. & Bus. 285 (1987).

  66. . See Timeline: Key Dates in the U.S.-China Trade War, Reuters (Jan. 15, 2020, 5:08 AM), []; Ana Swanson & Peter Eavis, Trump Expands Steel Tariffs, Saying They Are Short of Aim, N.Y. Times (Jan. 27, 2020), [].

  67. . See, e.g., Chad P. Bown, The Global Resort to Antidumping, Safeguards, and Other Trade Remedies Amidst the Economic Crisis, in Effective Crisis Response and Openness: Implications for the Trading System 91 (Simon J. Evenett, Bernard M. Hoekman & Olivier Cattaneo eds. 2009) (linking economic downturns with increased usage of antidumping measures).

  68. . Kate Abnett, EU Sees Carbon Border Levy as ‘Matter of Survival’ for Industry, Reuters (Jan. 18, 2021, 11:41 AM), [ 649C].

  69. . The Biden Plan to Ensure the Future is “Made in All of America” by All of America’s Workers, Biden/Harris, [].

  70. . As discussed in Section IV.B, infra, investment treaties and investment chapters in trade agreements also include exceptions more frequently. See Caroline Henckels, Should Investment Treaties Contain Public Policy Exceptions?, 59 B.C. L. Rev. 2825, 2826 (2018) (footnote omitted) (“Exceptions have become an increasingly popular mechanism in investment treaties, appearing in 43% of investment agreements concluded between 2011 and 2016, compared to 7% of agreements signed between 1959 (when the first investment treaty was signed) and 2010.”). These exceptions are especially useful for states trying to justify deviations from obligations to provide “fair and equitable treatment” to foreign investments, see, e.g., USMCA, supra note 34, art. 14.6; CETA, supra note 34, art. 8.10, or to circumvent the prohibition on expropriation except upon payment of prompt, adequate, and effective compensation. See, e.g., USMCA, supra note 34, art. 14.8; CETA, supra note 34, art. 8.12.

  71. . See, e.g., GATT, supra note 22, art. XXI(b)(iii); GATS, supra note 35, art. XIV bis; TRIPS, supra note 36, art. 73.

  72. . See, e.g., GATT, supra note 22, art. XI.2(a).

  73. . See, e.g., id. art. XX(a); GATS, supra note 35, art. XIV(a); TRIPS, supra note 36, art. 27.2.

  74. . See, e.g., GATT, supra note 22, art. XX(g).

  75. . GATT, supra note 22, art. VI.3; Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, art. 2, 1868 U.N.T.S. 201 [hereinafter Antidumping Agreement]; Agreement on Subsidies and Countervailing Measures, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1869 U.N.T.S. 14 [hereinafter SCM Agreement].

  76. . GATT, supra note 22, XX(d); GATS, supra note 35, art. XIV(c); TRIPS, supra note 36, art. 3.2.

  77. . See GATT, supra note 22, art. XX(b).

  78. . See id. art. XX(c).

  79. . See id. art. XX(f).

  80. . My focus throughout the Article is on how states invoke, and tribunals apply, exceptions. However, since states are both the authors and subjects of their international commitments, states’ expected use of exceptions will often also serve as justifications for including exceptions in a treaty in the first place.

  81. . See generally Krzysztof J. Pelc, Making and Bending International Rules: The Design of Exceptions and Escape Clauses in Trade Law 6 (2016) (referring to Pascal Lamy’s description of the tradeoff between the flexibility created by exceptions and the possibility of abuse as trade’s “architectural challenge”).

  82. . WTO obligations do often contain some modest qualifications. For instance, GATT article I exempts some specific preferences from the MFN obligation. See GATT, supra note 22, art. I.2–4. Likewise, the national treatment obligation contains several qualifications, such as for government procurement. See id. art. III.8. Many, although not all, of these qualifications are transitional, applying to measures in existence at the time the GATT came into existence.

  83. . Scheduled commitments can themselves contain qualifications, but the treaty obligation is to provide treatment no less favorable than that contained in the schedule, which includes the qualifications. See GATT, supra note 22, art. II; GATS, supra note 35, art. XVII.

  84. . See generally Timothy Meyer & Ganesh Sitaraman, Trade and the Separation of Powers, 107 Cal. L. Rev. 583 (2019) (describing domestic economic and foreign policy and paradigms for viewing trade policy).

  85. . Dani Rodrik, How to Save Globalization from Its Cheerleaders 4 (Harvard Univ. John F. Kennedy Sch. of Gov’t, Faculty Research Working Paper No. RWP07-038, 2007).

  86. . See id. at 4 (“I will provide a range of evidence on trade and capital flows that indicates the obstacles faced by developing countries do not originate from inadequate access to markets abroad or to foreign capital.”); see also Bernard Hoekman, Operationalizing the Concept of Policy Space in the WTO: Beyond Special and Differential Treatment, 8 J. Int’l Econ. L. 405 (2005); Jörg Mayer, Policy Space: What, for What, and Where?, 27 Dev. Pol’y Rev. 373 (2009).

  87. . See Shaffer, supra note 21, at 29; see also Beth Baltzan, The Old-School Answer to Global Trade, Wash. Monthly (Apr./May/June, 2019), [ WGEE].

  88. . Shaffer, supra note 21, at 5 (footnote omitted).

  89. . Id.

  90. . Julian Arato, Kathleen Claussen & J. Benton Heath, The Perils of Pandemic Exceptionalism, 114 Am. J. Int’l L. 627, 635 (2020).

  91. . Id. at 633 (“This is not just a short-term problem. As governments forge divergent paths in the long-term response and recovery from COVID-19, their exposure to trade and investment claims is likely to increase.”); see also Suzanne A. Spears, The Quest for Policy Space in a New Generation of International Investment Agreements, 13 J. Int’l Econ. L. 1037 (2010) (discussing the turn toward exceptions in investment agreements to mediate the conflict between economic rules and other objectives).

  92. . Arato et. al, supra note 90, at 632.

  93. . See Int’l L. Comm’n, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International law, ¶¶ 5–7, U.N. Doc. A/CN.4/L.682 (Apr. 13, 2006); see also Gerhard Hafner, Pros and Cons Ensuing from Fragmentation of International Law, 25 Mich. J. Int’l L. 849, 849–51 (2004).

  94. . World Health Org., Constitution of the World Health Organization 1 (1946), [].

  95. . See, e.g., International Convenant on Civil and Political Rights, Dec. 16, 1966, S. Exec. Doc. No. E, 95-2 (1978), 999 U.N.T.S. 171.

  96. . See United Nations Framework Convention on Climate Change, opened for signature June 4, 1992, S. Treaty Doc No. 102-38 (1992), 1771 U.N.T.S. 107.

  97. . See Final Act of the Conference of Plenipotentiaries on the Protocol on Chlorofluorocarbons to the Vienna Convention for the Protection of the Ozone Layer, opened for signature Sept. 16, 1987, S. Treaty Doc. No. 106-10 (1999), 1522 U.N.T.S. 3.

  98. . See, e.g., Stockholm Convention on Persistant Organic Pollutants, opened for signature May 22, 2001, 2256 U.N.T.S. 119.

  99. . See GATS, supra note 35, art. XIV bis; TRIPS, supra note 36, art. 73.

  100. . See Mona Pinchis-Paulsen, Trade Multilateralism and U.S. National Security: The Making of the GATT Security Exceptions, 41 Mich. J. Int’l L. 109, 190–91 (2020).

  101. . Id. at 192.

  102. . GATT, supra note 22, art. XXI(b).

  103. . Pinchis-Paulsen, supra note 100, at 185 (quoting a memorandum prepared by executive branch officials for members of Congress).

  104. . GATT, supra note 22, art. XXI(c).

  105. . Id. art. XX(a); GATS, supra note 35, art. XIV(a).

  106. . GATT, supra note 22, art. XX(b); GATS, supra note 35, art. XIV(b).

  107. . GATT, supra note 22, art. XX(d); GATS, supra note 35, art. XIV(c).

  108. . GATT, supra note 22, art. XX(g).

  109. . Id. art. XX(f).

  110. . Id. art. XX(h).

  111. . GATS, supra note 35, art. XIV(e).

  112. . See GATS, supra note 35, Annex on Financial Services, para 2.

  113. . See Trans-Pacific Partnership Agreement, art. 29.5 [].

  114. . See Summaries of the Legal Challenges Against Plain Packaging Laws, campaign for Tobacco-Free Kids, [].

  115. . See, e.g., Appellate Body Report, Korea—Measures Affecting Imports of Fresh, Chilled and Frozen Beef, ¶ 158, WTO Doc. WT/DS161/AB/R, WT/DS169/AB/R (adopted Jan. 10, 2001) (“[D]espite . . . troublesome aspects, . . . [the Panel accepted that] the dual retail system was put in place, at least in part, in order to secure compliance with the Korean legislation against deceptive practices . . . .”).

  116. . See Brazil—Tyres, supra note 10, ¶ 156.

  117. . Id.

  118. . Appellate Body Report, European Communities—Measures Prohibiting the Importation and Marketing of Seal Products, ¶ 5.306, WTO Doc. WT/DS400/AB/R, WT/DS401/AB/R (adopted June 18, 2014).

  119. . Id.

  120. . See Russia—Traffic in Transit, supra note 12 (interpreting the national security exception in the GATT); Panel Report, Saudi Arabia—Measures Concerning the Protection of Intellectual Property Rights, WTO Doc. WT/DS567/R (circulated June 16, 2020) [hereinafter Saudi Arabia—IPR] (interpreting the national security exception in the TRIPS Agreement).

  121. . Saudi Arabia—IPR, supra note 120, ¶ 7.293.

  122. . See infra Section IV.A.

  123. . See Andrew T. Guzman, How International Law Works: A Rational Choice Theory 117 (2008).

  124. . See Guzman & Meyer, supra note 24, at 192; see also Warren F. Schwartz & Alan O. Sykes, The Economic Structure of Renegotiation and Dispute Resolution in the World Trade Organization, 31 J. Legal Stud. S179, S182 (2002).

  125. . See Guzman & Meyer, supra note 24.

  126. . See Understanding on Rules and Procedures Governing the Settlement of Disputes, art. 23, Marrakesh Agreement Establishing the World Trade Organization, Annex 2, 1869 U.N.T.S. 401.

  127. . Guzman & Meyer, supra note 24, at 193. Reputational sanctions can limit a state’s ability to negotiate favorable agreements in the future if it does not honor its present commitments. International law also makes reciprocal and retaliatory sanctions, such as suspending trade concessions, easier both by formally authorizing such conduct and by helping states coordinate their sanctions. Guzman, supra note 123, at 71–117.

  128. . Guzman & Meyer, supra note 24, at 193, 195–97.

  129. . If states behave rationally, states will legalize actions in this middle category when the expected costs from violations that end up not being deterred exceeds the gains from deterring those violations that are averted.

  130. . Guzman & Meyer, supra note 24, at 222.

  131. . A variety of other flexibility mechanisms, such as soft law and reservations, can be used to the same effect. See id. at 171, 179 n.11; Edward T. Swaine, Reserving, 31 Yale J. Int’l L. 307, 307–08 (2006).

  132. . Wuerth, supra note 25, at 283.

  133. . Id. at 325–26.

  134. . See Laurence R. Helfer, Overlegalizing Human Rights: International Relations Theory and the Commonwealth Caribbean Backlash Against Human Rights Regimes, 102 Colum. L. Rev. 1832, 1834 (2002).

  135. . Trump Threatens to Pull US Out of World Trade Organization, BBC News (Aug. 31, 2018), [].

  136. . Sykes, supra note 26, at 279.

  137. . Id. at 288.

  138. . GATT, supra note 22, art. XIX.1(a) (“If, as a result of unforeseen developments and of the effect of the obligations incurred by a contracting party under this Agreement, including tariff concessions, any product is being imported into the territory of that contracting party in such increased quantities and under such conditions as to cause or threaten serious injury to domestic producers in that territory of like or directly competitive products, the contracting party shall be free, in respect of such product, and to the extent and for such time as may be necessary to prevent or remedy such injury, to suspend the obligation in whole or in part or to withdraw or modify the concession.”).

  139. . In international law more generally, the persistent objector doctrine offers an example of a safety valve exception. Under the persistent objector doctrine, a state can avoid being bound by a rule of customary international law if, “whilst a practice is developing into a rule of general law, [it] persistently and openly dissents from the rule.” Int’l Law Ass’n, Comm. on Formation of Customary (Gen.) Int’l Law, Statement of Principles Applicable to the Formation of General Customary International Law § 15 (2000). Andrew Guzman has argued that the persistent objector doctrine can be justified on the grounds that it “provide[s] an exception for states that cannot be deterred[,]” thereby “avoid[ing] the systemic costs associated with ongoing violations, the cost to sanctioned states, and the costs to sanctioning states.” Guzman, supra note 27, at 167. For similar reasons, Professor Guzman has also called for an expanded role for a subsequent objector doctrine for states that develop an interest in, and deep-seated objection to, a customary rule after it has already formed. Id. at 171.

  140. . Sykes himself distinguishes his argument from what he refers to as the “safety valve” hypothesis on the grounds that the safety valve hypothesis focuses on reducing ex post protectionism, whereas his argument is the GATT XIX increases ex post protectionism as the price of reducing protectionism ex ante. Sykes, supra note 26, at 273. Others describe a bargain of ex post protectionism for ex ante tariff concessions as a “safety valve.” See Mark Wu, Antidumping in Asia’s Emerging Giants, 53 Harv. Int’l L.J. 1, 30 (2012) (“[The safety valve theory] suggests that antidumping acts as a way to secure and maintain domestic support for trade liberalization.”).

  141. . See Alan O. Sykes, The Safeguards Mess: A Critique of WTO Jurisprudence, 2 World Trade Rev. 261 (2003). U.S. domestic law contains a similar provision, section 201 of the Trade Act of 1974, which provides the domestic basis for the United States to invoke its rights under GATT article XIX. 19 U.S.C. § 2251. A number of commentators have argued that section 201, which provides an administrative process for imposing safeguards, is designed to prevent appeals for protection to Congress, which might impose even higher levels of protection. See, e.g., Robert Z. Lawrence & Robert E. Litan, Saving Free Trade: A Pragmatic Approach (1986).

  142. . See, e.g., Gunnar Niels & Adriaan ten Kate, Antidumping Policy in Developing Countries: Safety Valve or Obstacle to Free Trade?, 22 Eur. J. Pol. Econ. 618 (2006). But see Wu, supra note 140, at 36, 40 (finding minimal empirical evidence that India and China use antidumping duties in the manner predicted by the safety valve theory).

  143. . See GATT, supra note 22, art. VI.

  144. . See Antidumping Agreement, supra note 75.

  145. . See SCM Agreement, supra note 75.

  146. . Alexander Roitinger, Antidumping Reform, Trade Policy Flexibility, and Compensation 3 (Univ. of St. Gallen Dep’t of Econ., Discussion Paper No. 2002-18, 2002), [].

  147. . World Trade Org., Graphs Over Time, Integrated Trade Intelligence Portal, [].

  148. . Roitinger, supra note 146, at 4–9.

  149. . Meredith A. Crowley & Jennifer A. Hillman, Slamming the Door on Trade Policy Discretion? The WTO Appellate Body’s Ruling on Market Distortions and Production Costs in EU—Biodiesel (Argentina), 17 World Trade Rev. 195, 195–96 (2018) (discussing the difficulty of proving the existence of a subsidy on these facts).

  150. . See Panel Report, United States—Countervailing Measures on Softwood Lumber from Canada, WTO Doc. WT/DS533/R (circulated Aug 24, 2020); Panel Report, United States—Anti-Dumping Measures Applying Differential Pricing Methodology to Softwood Lumber from Canada, WTO Doc. WT/DS534/R (circulated Apr. 9, 2019).

  151. . See, e.g., Update: 1-U.S. ITC Finds Chinese Exports of Subsidized Engines Harm U.S. Industry, Reuters (Feb. 2, 2021, 2:53 PM), []; Randall Woods, US Imposing Anti-Dumping Duties on Chinese Aluminum Foil, Indus. Week (Oct. 29, 2017),

  152. . See Panel Report, United States—Laws, Regulations, and Methodology for Calculating Dumping Margins (Zeroing), ¶¶ 2.1–2.5, WTO Doc. WT/DS294/R (adopted May 9, 2006).

  153. . Chad P. Bown & Thomas J. Prusa, U.S. Antidumping: Much Ado About Zeroing 3–4 (World Bank, Policy Research Working Paper No. 5352, 2010).

  154. . Aditya Rathore & Ashutosh Bajpai, The WTO Appellate Body Crisis: How We Got Here and What Lies Ahead?, Jurist (Apr. 14, 2020, 7:16 PM), [].

  155. . See Panel Report, European Union—Anti-Dumping Measures on Biodiesel from Indonesia, WTO Doc. WT/DS480/R (adopted Feb. 28, 2018); Appellate Body Report, Anti-Dumping Measures on Biodiesel from Argentina, WTO Doc. WT/DS473/AB/R (adopted Oct. 26, 2016).

  156. . GATT, supra note 22, art. XI.2(a).

  157. . Id. art. XX(i).

  158. . Id. art. XX(j).

  159. . Id. art. XII.

  160. . TRIPS, supra note 36, arts. 31, 31bis, annex 2.

  161. . See COVID-19: Measures Affecting Trade in Goods, World Trade Org., (last visited Nov. 22, 2021).

  162. . See Joost Pauwelyn, Export Restrictions in Times of Pandemic: Options and Limits Under International Trade Agreements, in COVID-19 and Trade Policy: Why Turning Inward Won’t Work (Richard E. Baldwin & Simon J. Evenett eds., 2020); Timothy Meyer, Trade Law and Supply Chain Regulation in a Post-COVID-19 World, 114 Am. J. Int’l L. 637, 644–45 (2020).

  163. . See Pelc, supra note 81, at 6.

  164. . Panel Report, United States—Certain Measures Relating to the Renewable Energy Sector, WTO Doc. WT/DS510/R (circulated June 27, 2019); Panel Report, European Union—Anti-Dumping Measures on Biodiesel from Indonesia, WTO Doc. WT/DS480/R (adopted Feb. 28, 2018); Appellate Body Report, Anti-Dumping Measures on Biodiesel from Argentina, WTO Doc. WT/DS473/AB/R (adopted Oct. 26 2016).

  165. . See, e.g., Robert W. McGee, The Cost of Protectionism: Should the Law Favor Producers or Consumers?, 23 Ga. J. Int’l & Comp. L. 529, 555–56 (1993).

  166. . Alexander H. Schaefer & Frances P. Hadfield, Safeguard Tariffs Imposed by U.S. on Solar Panels and Washing Machines are First Since 2002 Steel Import Restrictions, Crowell Moring (Jan. 23, 2018), [].

  167. . Cf. McGee, supra note 165, at 544.

  168. . SEIA, Study: Solar Tariffs Cause Devastating Harm to U.S. Market, Economy and Jobs (Dec. 3, 2019), [].

  169. . See Guzman, supra note 123, at 71­–117 (describing a reputational theory of international law).

  170. . Id. at 33–35.

  171. . Zachary Karabell, Opinion, The U.S.-China Trade Deal Was Not Even a Modest Win, Politico (Jan. 17, 2020, 1:37 PM), [].

  172. . Heather Long, Winners and Losers in the Final USMCA Deal, Wash. Post (Dec. 10, 2019), [].

  173. . See, e.g., Simon Lester (@snlester), Twitter (Apr. 4, 2019, 4:41 PM), [] (“I’m not sure why anyone thinks the USMCA side letter will provide a stronger deterrent than [GATT] Article II.”).

  174. . See Joost Pauwelyn, New Trade Politics for the 21st Century, 11 J. Int’l Econ. L. 559, 559–60 (2008).

  175. . Press Release, Office of the United States Trade Rep., Joint Statement of the United States and the European Union on a Tariff Agreement (Aug. 21, 2020), [].

  176. . Andrea Shalal & David Shepardson, U.S. Leaves Tariffs on Airbus Aircraft Unchanged at 15%, Reuters (Aug 12, 2020, 5:01 PM), [].

  177. . See Mark L. Movsesian, Enforcement of WTO Rulings: An Interest Group Analysis, 32 Hofstra L. Rev. 1, 4–5 (2003).

  178. . Timothy Meyer, Power, Exit Costs, and Renegotiation in International Law, 51 Harv. Int’l L.J. 379, 394–95 (2010) (“A state that withdraws in accordance with an exit provision does so lawfully and thus does not, in theory, suffer a reduction in its reputation for complying with international agreements.”).

  179. . Russia—Traffic in Transit, supra note 12, app.

  180. . See id.; Saudi Arabia—IPR, supra note 120.

  181. . Tariff Act of 1930, §§ 701–783, 19 U.S.C. §§ 1671–1677.

  182. . United States—Shrimp, supra note 9; Report of the Panel, United States—Restrictions on Imports of Tuna, No. DS21/R, 30 ILM 1594 (1991) [hereinafter United States—Tuna].

  183. . Marine Mammal Protection Act, § 101, 16 U.S.C. § 1371(a)(2) (providing that the government “shall ban the importation of commercial fish or products from fish which have been caught with commercial fishing technology which results in the incidental kill or incidental serious injury of ocean mammals in excess of United States standards”) [hereinafter MMPA]. See also Nat’l Res. Def. Council, Inc. v. Ross, 331 F. Supp. 3d 1338, 1344–45 (Ct. Int’l Trade 2018) (applying the MMPA to ban fish caught in the waters of the endangered vaquita).

  184. . See, e.g., United States—Shrimp, supra note 9.

  185. . See U.S. Chamber of Com., Trade Works. Tariffs Don’t., [].

  186. . Arguably, the Trump Administration’s trade agreements, notably USMCA, actually raised trade barriers, rather than lowered them. Every other administration lowered trade barriers or sought to do so.

  187. . See Bruce Yandle, Viewpoint, Bootleggers and Baptiststhe Education of a Regulatory Economist, 7 Regul., May/June 1983, at 12, 13–14.

  188. . Council Directive 2009/28, pmbl. (4), 2009 O.J. (L 140) 16, 16 (EC) (“When favouring the development of the market for renewable energy sources, it is necessary to take into account the positive impact on regional and local development opportunities, export prospects, social cohesion and employment opportunities . . . .”).

  189. . See generally Crowley & Hillman, supra note 149; Fischer & Meyer, supra note 29.

  190. . Fischer & Meyer, supra note 29, at 307–12.

  191. . Request for Consultations by Indonesia, European UnionCertain Measures Concerning Palm Oil and Oil Palm Crop-Based Biofuels, WTO Doc. WT/DS593/1 (Dec. 16, 2019); Request for Consultations by Malaysia, European Union and Certain Member StatesCertain Measures Concerning Palm Oil and Oil Palm Crop-Based Biofuels, WTO Doc. WT/DS600/1 (Jan. 19, 2021).

  192. . See, e.g., Hajin Kim, An Argument for WTO Oversight of Ecolabels, 33 Stan. Env’t L.J. 421, 425 & n.12 (2014) (noting that greenwashing can involve deceit but can also involve win-win situations for industry and the environment).

  193. . See Brazil—Tyres, supra note 10, ¶ 2.

  194. . The Appellate Body found that Brazil’s measure was “necessary” to protect human life and health, but found the court-ordered exception for used tires constituted arbitrary and unjustifiable discrimination in violation of the chapeau of GATT article XX. Id. ¶¶ 182–83, ¶¶ 246–47.

  195. . See Panel Report, BrazilMeasures Affecting Imports of Retreaded Tyres, ¶¶ 7.300–7.303, WTO Doc. WT/DS332/R (adopted Dec. 17, 2007); see also Tamar Megiddo, Beyond Fragmentation: On International Law’s Integrationist Forces, 44 Yale J. Int’l L. 115, 134–36 (2019).

  196. . The panel found that the injunctions undermined the policy aims of the regulations to such a degree that it declared the regime to be unjustifiably discriminatory. Panel Report, Brazil—Measures Affecting Imports of Retreaded Tyres, supra note 195, ¶¶ 4.108, 7.306.

  197. . Brazil—Tyres, supra note 10, ¶ 247.

  198. . See, e.g., Jack Goldsmith, Power and Constraint: The Accountable Presidency After 9/11 122–201 (2012) (discussing the influence of lawyers within the intelligence and military agencies); see also Kenneth W. Abbott & Duncan Snidal, Hard and Soft Law in International Governance, 54 Int’l Org. 421, 421 (2000).

  199. . See, e.g., J.H.H. Weiler, The Rule of Lawyers and the Ethos of Diplomats: Reflections on the Internal and External Legitimacy of WTO Dispute Settlement, 35 J. World Trade 191, 192 (2001); William J. Davey, Non-discrimination in the World Trade Organization: The Rules and Exceptions (2012).

  200. . Goldsmith, supra note 198, at 167­–72. This is especially true if the policy originates elsewhere in the executive branch.

  201. . Jonathan Swan, Exclusive: Trump Vents in Oval Office, “I want Tariffs. Bring Me Some Tariffs!”, Axios (Aug. 27, 2017), [].

  202. . Id.

  203. . Memorandum on Steel Imports and Threats to National Security, 2017 Daily Comp. Pres. Docs. 259, at 1 (Apr. 20, 2017).

  204. . See Kathleen Claussen, Trade’s Security Exceptionalism, 72 Stan. L. Rev. 1097, 1115–21 (2020) (describing the domestic legal framework allowing the President to raise trade barriers for national security reasons); J. Benton Heath, The New National Security Challenge to the Economic Order, 129 Yale L.J. 1020, 1093–96 (2020) (discussing the expansion of national security as a framework for governing international economic issues).

  205. . Jacob Barker, President Trump Touts Tariffs at Granite City Steel Plant, St. Louis Post-Dispatch (July 27, 2018), [].

  206. . U.S. Dep’t of Com., Bureau of Indus. and Sec., The Effect of Imports of Steel on the National Security 57 (2018), []. Commerce made a similar finding for imports of aluminum. See U.S. Dep’t of Com., Bureau of Indus. and Sec., The Effect of Imports of Aluminum on the National Security 5 (2018), [].

  207. . The United States has also invoked GATT article XXI, the GATT’s national security exception, in defense of its tariffs on steel and aluminum in WTO cases. The thrust of the U.S. argument is that the national security exception is self-judging, meaning that a WTO panel cannot overturn the United States’ own determination that its national security interests are at stake, and thus the exception applies. See, e.g., U.S. Third-Party Submission, Russia—Measures Concerning Traffic in Transit, ¶ 2, WT/DS512 (Nov. 7, 2017), []. Those cases are still pending as of the time of writing.

  208. . See infra Section IV.A. As I explain below, United States—Clove Cigarettes is functionally an exceptions case, although as a strict doctrinal matter, it does not involve the invocation of a textual exception.

  209. . See Meyer, supra note 58, at 1972–74.

  210. . See id.

  211. . See id. at 2002.

  212. . Dep’t of Health and Hum. Servs., Vaccines Protect Your Community, [].

  213. . Meyer, supra note 58, at 1969.

  214. . See id. at 1972–73.

  215. . Mancur Olson, Jr., The Logic of Collective Action 51 (1965).

  216. . See Meyer & Sitaraman, supra note 84, at 598, 604, 623–24 (discussing the foreign policy paradigm of trade agreements).

  217. . See generally Bruce Western & Jake Rosenfeld, Workers of the World Divide: The Decline of Labor and the Future of the Middle Class, Foreign Affs. May/June 2012, at 88 (describing the rise and fall of the labor movement in the United States).

  218. . See United States—Shrimp, supra note 9; United States—Tuna, supra note 182.

  219. . See Council Directive (EU) 2018/2001, of the European Parliament and of the Council of 11 December 2018 on the Promotion of the Use of Energy from Renewable Sources, arts. 6, 9, 13, 2018 O.J. (L 328) 82; Biodiesel from Indonesia, supra note 29; Biodiesel from Argentina, supra note 29.

  220. . See Timothy Meyer, Saving the Political Consensus in Favor of Free Trade, 70 Vand. L. Rev. 985, 1002–03 (2017).

  221. . See id.

  222. . The scope of exceptions are, of course, key to these benefits. Exceptions that are not linked to underrepresented political groups are unlikely to generate the domestic political effects associated with linking economic benefits to the pursuit of general public policy goals. For instance, the GATT article XX(h) exception for measures undertaken to comply with commodities agreements, or XX(i)’s exceptions for restrictions on exports as part of domestic stabilization plan, do not obviously relate to underrepresented interests. Instead, they more plausibly are motivated by the Policy Space rationale.

  223. . See, e.g., United States—Tuna, supra note 182 (finding unlawful import restrictions on tuna caught in a manner that risked harm to dolphins).

  224. . Meyer, supra note 58, at 1972–74.

  225. . Petros C. Mavroidis, The General Agreement on Tariffs and Trade: A Commentary 128 (2005).

  226. . GATT, supra note 22, art. XI.1.

  227. . Mavroidis, supra note 225, at 128.

  228. . See GATT, supra note 22, art. III (providing for nondiscriminatory treatment between imports and domestic products); see also id. art. I (providing for nondiscriminatory treatment between imports of different countries).

  229. . See, e.g., Daniel Y. Kono, Optimal Obfuscation: Democracy and Trade Policy Transparency, 100 Am. Pol. Sci. Rev. 369, 370–71 (2006) (noting that the effects of tariffs are straightforward compared with other nontariff barriers).

  230. . See, e.g., Robert Pahre, Reactions and Reciprocity: Tariffs and Trade Liberalization from 1815 to 1914, 42 J. Conflict Res. 467, 467 (1998).

  231. . Russia—Traffic in Transit, supra note 12.

  232. . Saudi Arabia—IPR, supra note 120.

  233. . Statements, supra note 5, at 11.

  234. . See Heath, supra note 204.

  235. . Pelc, supra note 81.

  236. . See Am. Inst. for Int’l Steel, Inc. v. United States, 806 F. App’x 982, 983 (Fed. Cir. 2020) (rejecting the argument that section 232’s capacious definition of national security renders section 232 unconstitutional under the nondelegation doctrine), cert. denied, 141 S. Ct. 133.

  237. . The United States has thus far declined to say which specific subparagraph of Article XXI it believes covers the steel and aluminum tariffs, although XXI(b)(iii) would appear most likely. See Opening Statement of the United States of America at the First Substantive Meeting of the Panel, United States–Certain Measures on Steel and Aluminum Products, ¶ 3, WTO Doc. WT/DS556 (Nov. 12, 2019), [].

  238. . Other WTO members have challenged the U.S. steel and aluminum tariffs as inconsistent with WTO rules and exceptions. See, e.g., id. ¶ 10. As of the time of writing, no panel has yet ruled in those cases.

  239. . See, e.g., World Bank, Temporary Trade Barriers Database Including the Global Antidumping Database, [].

  240. . See Biodiesel from Indonesia, supra note 29; Biodiesel from Argentina, supra note 29.

  241. . Crowley & Hillman, supra note 149, at 209–212.

  242. . Id.; Fischer & Meyer, supra note 29, at 301–06.

  243. . See, e.g., Robert Kuttner, Balance of Trade, Harv. Kennedy School of Gov’t, (quoting economist Dani Rodrik as saying “Twenty years ago . . . economists would tell me, ‘Do you really want to say this in public—It will just feed the barbarians. Your arguments will be abused by protectionists.’”).

  244. . Jeffrey Kucik & Sergio Puig, Extending Trade Law Precedent, 54 Vand. J. Trans’l L. 539, 571 (2021).

  245. . See Statements, supra note 5, at 11 (“The WTO cannot, consistent with Article XXI of the General Agreement on Tariffs and Trade 1994, consider those claims or make the requested findings.”).

  246. . At the same time, the United States indicated that it is not prepared to revisit the appointment of Appellate Body members, an issue that emerged over the interpretation of Safety Valve exceptions like antidumping duties. See id. at 12.

  247. . See Andrew Verstein, The Jurisprudence of Mixed Motives, 127 Yale L.J. 1106 (2018) (documenting the areas of the law that have mixed motive jurisprudence).

  248. . Id. app. B (cataloguing motive standards by legal domain).

  249. . Id. at 1134–43.

  250. . Id. at 1134–36.

  251. . Id. at 1137–38

  252. . Id. at 1139–41.

  253. . Id. at 1141–43.

  254. . For purposes of this Part, I put aside carveouts, exemptions, and scope provisions that are, while functionally exceptions in the broad sense, are not exceptions in the sense of first requiring a showing of a violation. Having said that, insights from the approach I outline here are relevant to implementing limitations (such as temporal limitations) found in such provisions as well.

  255. . GATT, supra note 22, pmbl.

  256. . Marrakesh Agreement Establishing the World Trade Organization pmbl., Apr. 15, 1994, 1864 U.N.T.S. 154, (“allowing for the optimal use of the world’s resources in accordance with the objective of sustainable development, seeking both to protect and preserve the environment . . . .”).

  257. . Pinchis-Paulsen, supra note 100, at 118.

  258. . Id. at 117.

  259. . See, e.g., Appellate Body Report, United States—Measures Affecting the Production and Sale of Clove Cigarettes, ¶ 174, WTO Doc. WT/DS406/AB/R (adopted Apr. 24, 2012) [hereinafter United States—Clove Cigarettes] (“[T]he object and purpose of the TBT Agreement is to strike a balance between, on the one hand, the objective of trade liberalization and, on the other hand, Members’ right to regulate.”); United States—Shrimp, supra note 9, ¶ 159 (“The task of interpreting and applying the chapeau is, hence, essentially the delicate one of locating and marking out a line of equilibrium between the right of a Member to invoke an exception under Article XX and the rights of the other Members under varying substantive provisions . . . .”).

  260. . Appellate Body Report, Dominican Republic—Measures Affecting the Importation and Internal Sale of Cigarettes, ¶ 96, WTO Doc. WT/DS302/AB/R (adopted May 19, 2005).

  261. . See also Robert E. Hudec, GATT/WTO Constraints on National Regulation: Requiem for an “Aim and Effects” Test, 32 Int’l Law. 619 (1998).

  262. . EU—Seals, supra note 14, ¶¶ 5.105, 5.116.

  263. . United States—Clove Cigarettes, supra note 259, ¶ 174.

  264. . Id. ¶¶ 169–75.

  265. . EU—Seals, supra note 14, ¶ 5.108

  266. . United States—Clove Cigarettes, supra note 259, ¶ 174.

  267. . See id.

  268. . 21 U.S.C. § 387g(a)(1)(A).

  269. . Executive Summary of the First Written Submission of Indonesia, Annex A-1, United States—Measures Affecting the Production and Sale of Clove Cigarettes, ¶¶ 14–15, WTO Doc. WT/DS406/R, [].

  270. . Matthew Bigg, U.S. Tobacco Bill Puts Focus on Menthol Cigarettes, Reuters (June 17, 2009, 3:05 PM), [].

  271. . United States—Clove Cigarettes, supra note 259, ¶ 298(a); Panel Report, United States—Measures Affecting the Production and Sale of Clove Cigarettes, ¶ 8.1(b), WTO Doc. WT/DS406/R (adopted Apr. 24, 2012).

  272. . United States—Clove Cigarettes, supra note 259, ¶ 225.

  273. . Id. ¶ 174.

  274. . Panel Report, United States—Measures Affecting the Production and Sale of Clove Cigarettes, ¶ 7.390 n.708 WTO Doc. WT/DS406/R (adopted Apr. 24, 2012).

  275. . Tomer Broude & Philip I. Levy, Do You Mind If I Don’t Smoke? Products, Purpose and Indeterminacy in US—Measures Affecting the Production and Sale of Clove Cigarettes, 13 World Trade Rev. 357, 361 (2014).

  276. . Brazil—Tyres, supra note 10, ¶ 156.

  277. . Id.

  278. . GATT, supra note 22, art. XX.

  279. . EU—Seals, supra note 14, ¶ 5.321.

  280. . Brazil—Tyres, supra note 10, ¶ 246.

  281. . See supra note 31.

  282. . See Appellate Body Report, Chile—Taxes on Alcoholic Beverages, ¶ 71, WTO Doc. WT/DS110/AB/R (adopted Jan. 12, 2000).

  283. . Panel Reports, European Communities—Measures Prohibiting the Importation and Marketing of Seal Products, WT/DS400/R, WT/DS401/R, ¶¶ 7.156-7.163 (adopted June 18, 2014) (describing reports before the European Commission about the market impacts of the then-contemplated ban on seal products).

  284. . See supra Section III.C.

  285. . See Russia—Traffic in Transit, supra note 12 (interpreting the national security exception in the GATT); Panel Report, Saudi Arabia—IPR, supra note 120 (interpreting the national security exception in the TRIPS Agreement).

  286. . See Kucik & Puig, supra note 244, at 561 tbl.2 (finding that the Appellate Body narrows or distinguishes its precedent 13% of the time).

  287. . See Statements, supra note 5, at 12.

  288. . See John Gerard Ruggie, International Regimes, Transactions, and Change: Embedded Liberalism in the Postwar Economic Order, 36 Int’l Org. 379 (1982).

  289. . Free Trade Agreements, Off. of the U.S. Trade Rep., [].

  290. . Report from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on the Implementation of EU Trade Agreements 1 January 2019 – 31 December 2019, at 4, COM (2020) 705 final (Nov. 12, 2020).

  291. . China FTA Network, China’s Free Trade Agreements, Ministry of Com., People’s Rep. of China, [].

  292. . Trans-Pacific Partnership Agreement, supra note 113, art. 29.9.

  293. . Tom Edgington & Chris Morris, Brexit: What’s the Northern Ireland Protocol?, BBC News (Sept. 22, 2021), [].

  294. . Id.

  295. . Id.

  296. . Certain objectives that are relevant for goods are not relevant for services. Compare GATT art. XX, with GATS art. XIV.

  297. . Id.

  298. . Arguably, GATT article XX could apply to these agreements, which are themselves derived from some portion of the GATT. See Simon Lester, USTR’s View of GATT Exceptions in Relation to Non-GATT Goods Agreements, Int’l Econ. Law & Pol’y Blog (Feb. 22, 2021, 5:17 PM), []. This position has not carried the day, although the United States appears as if it may have tacitly adopted this view, at least in regards to the national security exception, in a dispute regarding labeling exports from Hong Kong as “Made in China.” Id.

  299. . World Trade Org., WTO Electronic Commerce Negotiations, Consolidated Negotiating Text–December 2020 (2020), wto_plurilateral_ecommerce_draft_consolidated_text.pdf [].

  300. . Id. at 85.

  301. . Id. at 86.

  302. . More targeted exceptions within the agreement also adopted the necessity test. See Simon Lester, Exceptions Proposals in the WTO E-Commerce Negotiating Text: Part 2, Int’l Econ. Law & Pol’y Blog (Feb. 16, 2021, 7:54 PM), [].

  303. . Henckels, supra note 70, at 2826 (footnote omitted).

  304. . Id. at 2828.

  305. . Trans-Pacific Partnership Agreement, Off. of the U.S. Trade Rep., Annex 9-B.3(b), []; see also CETA, supra note 34, at Annex 8-A.

  306. . See, e.g., Investment Agreement for the COMESA Common Investment Area, art. 22.1, May 23, 2007, [].

  307. . See, e.g., id.

  308. . See Meyer & Sitaraman, supra note 84, at 598.

  309. . See, e.g., Agreement Between the Government of the United States of America and the Government of the Sultanate of Oman on the Establishment of a Free Trade Area, Oman-U.S., Jan. 19, 2006, Hein’s No. KAV 8673 [hereinafter Oman Agreement]; see also Meyer & Sitaraman, supra note 84, at 623.

  310. . See Meyer & Sitaraman, supra note 84, at 624.

  311. . See, e.g., Oman Agreement, supra note 304, art. 15.

  312. . Margot E. Kaminski, The Capture of International Intellectual Property Law Through the U.S. Trade Regime, 87 S. Cal. L. Rev. 977, 980, 984 (2014).

  313. . See, e.g., The Dominican Republic–Central America–United States Free Trade Agreement arts. 6–7, Dec. 1, 2006, Hein’s No. KAV 7886; Oman Agreement, supra note 304, arts. 6–7; Agreement on the Application of Sanitary and Phytosanitary Measures, Apr. 15, 1994, 1867 U.N.T.S. 493; TBT Agreement, supra note 47.

  314. . David H. Autor, David Dorn & Gordon H. Hanson, The China Syndrome: Local Labor Market Effects of Import Competition in the United States, 103 Am. Econ. Rev. 2121, 2155–56 (2013).

  315. . See Inception Impact Assessment, Carbon Border Tax Adjustment Mechanism (2020), supra note 13.

  316. . The Biden Plan to Ensure the Future is “Made in All of America” by All of America’s Workers, supra note 69.

Published by Timothy Meyer

Professor of Law, Vanderbilt University Law School

%d bloggers like this: