‘The South says, that slavery is nothing to us at the North. But, through our trade, we are brought into constant contact with it; we grow familiar with it; still more, we thrive by it; and the next step is easy, to consent to the sacrifice of human beings by whom we prosper.’ (italics in original)
-William Ellery Channing, ‘Tribute to the American Abolitionists for their Vindication of Freedom of Speech.’, American Anti-Slavery Society (New York 1861) [1836]
1. Introduction
With trade we live in an interconnected world. We benefit from it. Our countries mutually gain from it. And we may be complicit in harms that underlie it. William Ellery Channing's 1836 Tribute to the American Abolitionists, following the murder of an abolitionist publicist by a pro-slavery mob, captures this basic dilemma.Footnote 1 What we trade and what we consume inextricably link us with how goods are produced, whether it be in ways that foster economic development, raise individuals out of poverty, and enable governments to spend more on environmental protection and technological adaptation to foster sustainability, or in ways that oppress labor, unravel the social fabric, and destroy the atmosphere on which life depends.
Trade agreements generally did not fundamentally address this dilemma.Footnote 2 Rather, they focused predominantly on removing barriers to foreign commerce and freeing up markets. Issues of sustainability and social inclusion were left to domestic policy, potentially complemented by non-trade agreements that were weak on enforcement.Footnote 3 The failure of members of the World Trade Organization to address these issues helps explain the regime's decline and challenges to its legitimacy.
Today, more trade agreements address the dilemma, however gropingly. In 2023, United States Trade Representative, Katherine Tai, invoked a paradigm shift in the way the United States formulates its trade policy goals. She contended that the United States has moved away from advancing the goal of ‘liberalization’ in ‘pursuit of efficiency’ toward supporting the goals of ‘resiliency’, ‘sustainability’, and inclusion.Footnote 4 This shift responds to multiple crises that the United States faces, from the material challenges of job precarity, wage inequality, and climate change, to social crises of polarization exacerbated by partisan media, to political crises of rising authoritarianism and geopolitical conflict.
This article assesses different flanking policies and institutional responses to the negative spillovers of liberalized trade. Flanking policies, as defined by Pauwelyn and Sieber-Gasser, are ‘policies that can mitigate negative effects of trade liberalization, or the concerns of domestic stakeholders regarding said effects, or both, and that are legally or factually linked to such trade liberalization’.Footnote 5 Flanking policies, in this sense, have both a substantive dimension and a political one. Substantively, they aim to ‘mitigate negative effects’ and thus contribute to social welfare. Politically, they respond to domestic stakeholder concerns about trade liberalization with the aim of making the agreements ‘more palatable to their domestic opponents’.Footnote 6
Flanking measures can be taken solely at the domestic level, independent of, or pursuant to, an international agreement. They can involve domestic adjustment policies, unilateral restrictions to trade, information sharing and peer review by transnational networks, or treaties imposing binding procedural and substantive rules. The last of these options involves positive obligations that combine trade, environment, and labor commitments in what we call a ‘package treaty’. A package treaty is ‘a legally binding treaty or other international convention or agreement, that includes legally binding commitments on both trade liberalization and mutually agreed flanking policies’.Footnote 7
Through a package treaty, countries mutually agree on flanking policies to mitigate the negative effects of trade, such as on labor, social inclusion, and environmental sustainability. Trade agreements increasingly include binding commitments that address the negative effects of trade liberalization. In some cases, however, the commitments are non-binding, as appears to be the case with the Biden administration's signature Indo-Pacific Economic Framework. Such agreements can be viewed as merely symbolic (such as forms of greenwashing), as potentially empowering of local and transnational stakeholders, or as potential steppingstones toward a package treaty that contains binding commitments. They are part of ongoing transnational normative development.
Conceptually, this article sets forth three justifications for flanking policies in terms of three categories of negative externalities from trade and economic globalization: material, moral, and social/political. Although these different justifications predominate in responses to particular policy issues, they also overlap. After explaining these justifications, the article notes that limiting analysis of them is unhelpful because, as Ronald Coase highlighted, externalities tend to be reciprocal and omnipresent.Footnote 8 In the case of trade in an interdependent world, while production of traded goods in one country can have external effects in the trading partner, the trading partners' regulation of those goods, including its application of trade restrictions, can have external effects in the exporting country. Each country will view the policies of the other as having an externality on it and its stakeholders.
The key questions for policy thus must extend beyond justifications for flanking policies and incorporate comparative analysis of institutional choices over the management of externalities. These institutional choices should be assessed in terms of participation and outcomes.Footnote 9 The article applies this analytic framework to developments of, and future options for, the governance of sustainability and climate change mitigation, on the one hand, and social inclusion and labor rights, on the other hand. In each area, the United States and European Union (among others) increasingly assert regulatory authority over the consumption of traded products based on how they are produced, while simultaneously driving the development of package treaties to address production processes that implicate labor, social inclusion, and environmental protection. These areas involve different forms of externalities and justifications for regulatory intervention as well as different tradeoffs among institutional choices.
This article makes three primary contributions to the growing literature on trade, social inclusion, and sustainability. First, it provides an original typology of negative externalities that justify trade restrictions, while noting the challenges posed for flanking measures given the reciprocal nature of externalities (Section 2). In doing so, it opens up analysis regarding the institutional choices and tradeoffs for addressing externalities compared to traditional approaches that contrast ‘free trade’ with ‘policy space’ or ‘protectionism’. The baseline for analysis becomes the management of externalities of one country's policies other countries. Second, it assesses different institutional choices for addressing such externalities, which can involve variation in reliance on, and deference toward, domestic measures, on the one hand, and the development of transnational and international measures, including through package treaties, on the other hand (Section 3). Third, and importantly, in assessing these choices, it shows how the concept of the flanking measure could, and arguably should, be flipped so that environmental sustainability and social inclusion become the core, and trade and other measures the flanking policies (Section 4). In each case, it applies this analytic framework to contemporary debates regarding the relation of trade to labor rights and social inclusion, on the one hand, and sustainability and climate change, on the other – which are two critical issues of our time.
2. On Negative Externalities: Three Categories and Their Reciprocal Nature
2.1 Three Categories of Externalities
The existential crisis of our age is climate change. The interrelated social challenges of our time (salient in the United States where I write, but not limited to the United States) are job insecurity and social inequality, which give rise to social and political polarization and distrust.Footnote 10 They threaten democratic governance and the rule of law, which are under pressure and showing declines around the world.Footnote 11 Liberalized trade, which was long trumpeted for the mutual economic gains it brought to countries, has been called into question because of its negative spillover effects, most prominently as regards social impacts but also as regards climate change.
Conceptually, three types of negative externalities (or negative spillover effects) justify the adoption of flanking policies to address them. Some might suggest that the term ‘spillover effects’ has a broader application than ‘externalities’, but many use the terms interchangeably, as do I.Footnote 12 I apply the term externalities because of its importance in the economic literature for justifying policy interventions. Economists have provided the dominant arguments in support of trade liberalization, and they have justified government intervention in the form of flanking policies in response to ‘market failures’ that give rise to ‘negative externalities’.Footnote 13
The concept of externalities nonetheless is a contested one. Classically, economists predominantly apply the concept in material terms to account for costs or benefits that are not incorporated in market prices.Footnote 14 Thus, if the production of a product gives rise to social costs in the form of air or water pollution, and those costs are not incorporated (and thus ‘internalized’) in the price of a product, they are viewed as negative ‘externalities’.Footnote 15 The externality is a material one because of harm to the environment, as in the case of air or water pollution, that is not incorporated in prices but which affects other people – referred to as ‘third parties’ because they are not parties to the market transaction.Footnote 16 Alternatively, externalities can be conveyed through the price system and materially affect third parties. In this sense, they are pecuniary, as when the wages of workers in one country are adversely affected by competition from imported goods produced in violation of labor rights in another country.Footnote 17
There are more than material externalities to a transaction. There are also moral ones.Footnote 18 Slavery and forced labor, for example, do not create a material social cost for those who consume products produced with them.Footnote 19 The consumer rather benefits materially from the products' lower prices. To conceptualize this cost as an externality, one must expand the concept to include moral ones. Where a society believes that slavery or forced labor is immoral, its importation and consumption of products produced elsewhere with such laborers renders it complicit in supporting such a system. Complicity with such practices through the purchase of products at low prices has moral costs. In the terminology of this article, it involves a moral externality. As Iris Marion Young writes, we are structurally interconnected in the sense that as consumers we ‘contribute … to the structural processes that produce injustice’.Footnote 20
In addition, there is a third category of externalities, which are not directly material or moral but social and political.Footnote 21 Trade and investment liberalization help empower capital in relation to labor and government.Footnote 22 Where capital can invest in another country with lower labor protections, it enhances its bargaining leverage over labor. If labor demands too much, capital can threaten to close a facility and move elsewhere. Similarly, capital can engage in tax arbitrage by structuring its investments in low tax jurisdictions, thereby pressuring governments to lower taxes and provide subsidies to attract and retain capital. The result is that labor and government are relatively disempowered in their relations with capital. These processes over time can exacerbate social inequality and job precarity, erode social solidarity and trust, and spur social and political polarization, emboldening authoritarian movements and undermining the rule of law. When the externalities are significant and extend beyond individual firms and workers, they can have systemic social and political impacts.Footnote 23 In the terminology of this article, they involve social and political externalities.
Negative political externalities extend beyond concerns over social inclusion and conventionally include national security. In the post-World War II era, trade was largely viewed as contributing to improved foreign relations and thus conducive to international peace.Footnote 24 However, countries can weaponize trade interdependence and they have increasingly done so.Footnote 25 These tactics have triggered calls for reducing reliance on trade, and thus make supply chains for the production of goods more ‘resilient’.Footnote 26 This article does not address flanking policies imposed on national security grounds, which have been extensively addressed elsewhere and at times overlap as when climate change or political polarization are conceptualized as security concerns.Footnote 27 Nonetheless, this article's framework of institutional analysis applies to them as well.
2.2 Distinction and Overlap of the Three Categories in Practice
Justifications for regulatory intervention predominantly highlight different types of externalities in different areas. Most of the economics literature regarding trade and the environment has focused on negative material externalities involving environmental harm, such as harm to human, animal, or plant life and health. In contrast, justifications for regulation that responds to trade's impact on labor rights, where raised, and are more likely viewed in social/political or moral terms. In practice, one can justify a governmental measure in terms of all three categories of externalities because all three overlap and can exist simultaneously. We illustrate this important point by starting with an analysis of externalities involving environmental harm, followed by those involving labor and social inclusion.
2.2.1 Environmental Harm
Environmental damage from a product's production that is not incorporated in the price of a product is classically conceptualized as a negative material externality. Where the harm is transboundary or global in scope – such as with transboundary pollution, the depletion of the ozone layer, or the negative effects of global warming – then the externalities are material for citizens outside the producing country. Some material environmental externalities, such as greenhouse gas emissions that trigger global warming, affect all life on earth.
Although most literature focuses on the material externalities of environmental harm, moral externalities also arise. When citizens of one country purchase and consume a product that is produced in environmentally harmful ways in another country, they and the importing country become complicit in that harm. The most famous example of trade restrictions successfully justified on moral grounds is the EC–Seal Products case before the World Trade Organization (WTO).Footnote 28 There the European Union justified a ban on the import of seal products under the ‘public morals’ exception set forth in GATT Article XX(a). The European Union maintained that its regulatory regime responded to the moral concerns of EU citizens regarding the inhumane killing of seals. The case, brought by Canada and Norway, involved animal welfare concerns, and not environmental ones. However, a similar logic applies to production processes that negatively impact the environment, such as climate change. The clearing of tropical forests for the production and export of agricultural crops, for example, not only eliminates critically important climate sinks, but it also destroys the livelihoods and life worlds of indigenous communities. More broadly, global warming-induced heat waves, floods, fires, storms, and droughts can result in catastrophic harm, particularly affecting the poor and most vulnerable. One can justify governmental measures to avoid moral complicity in such harms.
Some environmental harm, such as climate change, also can give rise to social and political externalities. Climate change can lead to droughts and water shortages that trigger societal conflict and even collapse. Such environmental and social stresses can erupt in violence. Governments have the duty to ensure social peace, and there is thus justification for governmental measures to address these transboundary social and political externalities.
The dispute over European trade restrictions on Indonesian palm oil exemplifies how all three categories of externalities may arise.Footnote 29 The consumption of palm oil produced through the clearing of rain forests contributes to climate change, and thus involves a material externality that, absent a tax or other sanction, is not internalized in the cost of the product. It also involves a moral externality to the extent that a society does not wish to be complicit in the consumption of products that will have severe effects on vulnerable people affected by climate change. It likewise can involve social and political externalities to the extent that climate change triggers mass migration that uproots livelihoods and social stability.
Table 1 summarizes the three categories of externalities in the environmental context.
Environmental externalities
2.2.2 Labor Rights and Social Inclusion Concerns
One can likewise conceptualize labor and human rights violations in the production process as a material, pecuniary externality for third parties in that the ‘costs’ to domestic workers from ‘unfair’ competition is not incorporated in the price of the product. W.E.B. Dubois made this key point in the chapter ‘The White Worker’ in his masterpiece Black Reconstruction in America. As he wrote, slavery in the South was not only a moral concern of Abolitionists in the North, but also a material one of white workers who risked being disadvantaged from competition with slave labor. In his memorable words, ‘[t]hese workers came to oppose slavery not so much from moral as from the economic fear of being reduced by competition to the level of slaves’.Footnote 30 Competition always will have pecuniary effects on producers, including labor, but it is only when that competition is deemed ‘unfair’ that ‘labor rights’ issues arise. Unlike with climate change, however, the economic ‘cost’ of competition because of lower labor and human rights standards is only indirectly transboundary and material in nature. Any harm to workers in the importing country applies only to the extent that employers in that country lay off workers or reduce wages and labor protections to remain competitive. That will be the case only where a sector is affected by trade and does not otherwise adapt. Labor in many services sectors, for example, has not been affected, and businesses can compete based on higher productivity and higher quality. Although labor rights advocates have successfully linked foreign labor rights violations to negative, pecuniary spillover effects on US workers, human rights advocates have been less successful.Footnote 31
Justifications based on moral arguments have been more salient in justifying governmental intervention to protect labor and human rights than to protect the environment. When citizens of one country consume a product that is produced in violation of fundamental labor and other human rights abroad, they and the importing country become complicit in that harm. The material harm from the rights violation is felt by the foreign victims, and the pecuniary harm and externality is felt by the domestic workers who lose their jobs or are otherwise harmed. Yet there is also a moral harm in being part of a society that consumes products produced in violation of the fundamental rights. Where one purchases products produced through slavery, for example, one becomes complicit, and thus responsible, in economically supporting that practice, as captured in Channing's protestation in this article's epigraph. tHowever, the prominence of labor rights issues, such as prohibitions against child labor, can vary in light of differing economic and social contexts in the Global North and Global South. The definition of the moral externality, in these cases, is (at least partially) socially and politically constructed. Those constructions are a result of historical, political struggles that may be national or regional in scope.
In addition, commentators more prominently justify trade restrictions to protect social inclusion in response to trade's social and political externalities, as compared with environmental protection. Economic globalization facilitated by trade agreements has, as noted above, empowered capital in its relationship with labor. It also has empowered capital in relation to governments since capital can threaten to locate elsewhere if states raise taxes, constraining a state's ability to fund social welfare and economic adjustment. The huge shift in manufacturing linked with trade with China, for example, has contributed to rising job precarity, social inequality, and social distrust in the United States, which political populists with authoritarian tendencies leverage.Footnote 32 The result has been a rise in economic inequality in the United States to levels not seen since the 1920s, and the erosion of the social fabric.Footnote 33 These processes arguably have contributed to declines in democracy and the rule of law globally.Footnote 34 In sum, where capital invests abroad and otherwise benefits from products produced under such conditions, it can weaken labor's bargaining position, heighten social inequality, distrust, and polarization, and undermine the social fabric and political governance unless adequate flanking measures are taken. The externality, in other words, is imposed on the entire society, as Karl Polanyi captured in his study of the inter-war period.Footnote 35
Table 2 summarizes the three categories of externalities in the labor context.
Labor and social inclusion externalities
In sum, policymakers can justify government intervention to protect labor, foster social inclusion, and support environmental sustainability in response to all three categories of externalities because they may simultaneously be in play. Figure 1 captures these categories and how they overlap in practice.
Overlap of different types of externalities
2.3 The Challenge of the Reciprocal Nature of Externalities: Who's Externality?
Given the presence of different types of externalities, it is easy to justify unilateral measures that ban or otherwise impede market access to foreign goods. However, ending the analysis here is problematic because externalities, in the real world, are both pervasive and reciprocal, as Coase showed. In his famous article The Problem of Social Cost, Coase provided a series of examples that illustrate the reciprocal nature of externalities. He begins the article by citing ‘the case of a confectioner the noise and vibrations from whose machinery disturbed a doctor in his work. To avoid harming the doctor would inflict harm on the confectioner’.Footnote 36 Concomitantly, to avoid harming the confectioner would inflict harm on the doctor.
Measures restricting trade illustrate the dilemma. Tariffs or trade bans imposed on products from a third country on account of labor rights violations from production processes can have negative effects on workers and economic development in that country. Indeed, workers in that country can be made even worse off by trade barriers since not only may they lose their jobs, but export industries generally pay at higher levels and provide better working conditions than purely domestic ones.Footnote 37 Vulnerable workers in poor countries may be pushed to the informal sector where working conditions are much worse.
Similarly, countries from the ‘Global South’ oppose policies in the ‘Global North’ that impose import taxes on their products on environmental grounds, such as climate change mitigation, because of the negative impacts on their development that affects their citizens. They acknowledge the need for climate change mitigation but contend that the primary costs should be borne by developed countries who are responsible historically for the buildup of greenhouse gasses and who have greater capacity to address the problem. They stress the principle of ‘common but differentiated responsibilities’ of developed and developing countries for climate change mitigation.Footnote 38 Because of the reciprocal nature of externalities, trade restrictions imposed by the Global North not only harm these countries' development prospects, but particularly negatively affect small and more vulnerable producers. Brazil, Indonesia, and other Global South countries thus attack unilateral trade measures threatened by the US and EU as ‘green’ colonialism.Footnote 39 Much of the analysis in the United States and Europe, in contrast, focuses on the need to address negative externalities experienced by constituents in the Global North from competition with products produced without adequate regulation in the Global South, and not those in the Global South from trade restrictions imposed by the Global North.
Trade negotiations traditionally were grounded in the principle of reciprocity, subject to ‘special and differential treatment’ for developing countries. The reciprocity principle, however, focused on market access, as in the case of tariff cuts.Footnote 40 The management of negative externalities raises new, more complex institutional challenges. It is because countries and their stakeholders are affected by the costs of different externalities, and because they may not agree on their nature, that the management of these externalities becomes critical for international cooperation and, ultimately, for social welfare.
3. Institutional Choice in Addressing Externalities: A Dynamic Perspective
Because of the reciprocal nature of externalities, whether resulting from production processes or from government regulation, the key question is institutional choice regarding their management. The baseline for analysis is no longer free trade and protectionism or national policy space. Rather, the reciprocal nature of externalities focuses attention on institutional choices that involve tradeoffs in the management of negative externalities. The choices should not be viewed in static terms, but rather dynamically as part of transnational processes that provide different participatory opportunities for affected parties and that result in different outcomes over time. Since tradeoffs in addressing material, moral, and social externalities are not commensurable, especially when involving countries and citizens living in vastly different economic and social contexts, one cannot simply aggregate total costs and benefits to derive the best outcome, as recommended by Coase.Footnote 41
Because domestic governments are most responsive to domestic social conditions, a great deal of policymaking should reside at the domestic level. As a result, leading trade policy experts opposed the incorporation of binding labor and environmental commitments in trade agreements. They advocated leaving these issues to the domestic level or to other international agreements (placing them in separate silos), and they contended that allowing unilateral trade barriers based on how goods are produced would wreck the trading system.Footnote 42 Others argued that trade agreements should not be expanded to new areas but that, in order to protect their social bargains and democratic choices internally, countries should be permitted to impose unilateral trade restrictions based on labor and environmental grounds.Footnote 43
Yet, because of the reciprocal nature of the externalities, international institutional arrangements are also appropriate. Among the options are package treaties that incorporate binding commitments on trade, environment, and worker protection. This treaty option is not a panacea given the political economy of international negotiations and enforcement. It rather should be viewed in relation to, and in interaction with, a range of other imperfect institutional alternatives, including domestic arrangements. The tradeoffs among these institutional choices have become more salient as the United States and European Union increasingly assert prescriptive jurisdiction to address labor and environmental harms occurring in other countries. This represents a shift of jurisdiction based on the location of production to one based on the location of consumption.Footnote 44
Analysis should examine the interaction of legal norms not only horizontally across international regimes (such as international trade and environmental regimes), but also vertically between domestic, regional, and international norm making, what I call, ‘transnational legal ordering’.Footnote 45 This interaction can give rise to the relative transnational settlement and unsettlement of legal norms dynamically over time. The transnational legal ordering approach differs from analysis of ‘international regime complexes’ because the latter focuses only on the international level.Footnote 46 The transnational legal ordering framework, in contrast, also includes analysis of domestic developments, such as ‘extraterritorial’ applications of domestic laws, transnational diffusion and borrowing of domestic laws, and private norm making and practice, as well as norms developed through international treaties, institutions, and networks.Footnote 47 For example, unilateral measures such as Europe's ‘carbon border adjustment mechanism’, which involves tariffs on imports produced with high greenhouse gas emissions (addressed below), can act as a catalyst for international negotiations and shape markets transnational
Table 3 sets forth key institutional alternatives that, in practice, can be combined in different ways and that interact.
Types of institutional alternatives
3.1 Labor and Social Policy
Economists stress the aggregate welfare gains from liberalized trade, while accepting that inevitably there are winners and losers from trade liberalization within countries. To address the concerns of the losers and thus maintain political support for trade liberalization, they advocate flanking policies. Different institutional options can be applied, whether alone, as complements, or as supplements. First, key flanking policies can be adopted at the domestic level without restricting trade. Second, the multilateral trade regime already provides rights and defenses for member countries to apply unilateral trade restrictions to protect vulnerable industries and workers, subject to conditions. Going further, recent bilateral and plurilateral trade agreements set forth positive obligations on parties to protect workers, constituting examples of ‘package treaties’. These policies can be expanded, although they are contested.
3.1.1 Domestic Policies
To address the concerns of those who lose their jobs on account of trade, most economists classically have supported domestic trade adjustment policies. The combination of trade agreements supporting trade liberalization with domestic adjustment constitutes a ‘two-step model’ in which trade liberalization commitments are made at the international level through treaties and are flanked by domestic tax and transfer policies that support those who must retrain and search for new jobs on account of trade-related displacements.Footnote 48 Trade adjustment assistance is the classic domestic ‘flanking policy’ to address a pecuniary externality.Footnote 49 That approach, reflected in the concept of ‘embedded liberalism’,Footnote 50 worked to some extent in a world not characterized by deep economic integration. However, as capital flows liberalized, trade rose as a percentage of global production, competition increased between high-wage and low-wage countries, and supply chain networks deepened, embedded liberalism became less tenable. It became less tenable because capital could more easily invest elsewhere, undermining the social welfare state and labor's bargaining power.Footnote 51 One challenge, particularly in the United States, was the asynchronous nature of trade adjustment assistance, which is approved periodically for fixed periods and then lapses under domestic law, while international trade treaties conventionally have no termination date.Footnote 52
Economists have long noted that technological change is a more important cause of job displacements than trade. In response, broader policies can address job displacements, whether they are caused by trade, technology, or other factors. These policies include active labor programs involving job retraining and conditional allowances, complemented by universal health care and other social welfare (which are used in northern Europe to help workers adjust). These broader policies constitute flanking policies even if they are not directly tied to trade concerns.Footnote 53 They can reduce the need for trade-specific flanking policies to support labor and social inclusion.
In addition, countries may impose trade restrictions as flanking policies. In doing so, they externalize the costs of internal adjustment to foreign producers, imposing externalities on foreign producers and their workers. WTO agreements already provide positive rights and specific defenses that can indirectly support domestic labor and social inclusion. The WTO Antidumping Agreement and the WTO Agreement on Subsidies and Countervailing Measures (SCM Agreement) create rights for countries to raise tariffs on foreign products on ‘fairness’ grounds, subject to conditions. The Antidumping Agreement permits WTO members to raise tariffs on products sold at ‘less than normal value’, which includes products sold in the importing country at a price lower than in the home country, constituting a form of price discrimination (US law uses the term ‘less than fair value’).Footnote 54 The SCM Agreement, in parallel, permits countries to raise tariffs on goods that receive subsidies that confer a ‘benefit’ from a ‘financial contribution’ by a government or other ‘public body’, subject to conditions.Footnote 55 In addition, under the WTO Agreement on Safeguards, governments may provide import relief for up to eight years to domestic industries that experience ‘serious injury or threat thereof’ caused by imports resulting from fair competition.Footnote 56 These three import relief mechanisms authorized by the WTO can be viewed as flanking policies, although they do not directly address labor rights and only indirectly and contingently address social inclusion concerns. In each case, they respond to pecuniary externalities that affect domestic producer interests. These policy measures, in turn, have external impacts on producers in the exporting country, reflecting the reciprocal nature of externalities. Because the United States provides a less generous social safety net than Europe, as well as less generous labor adjustment policies, US labor has relied much more on these trade defenses than labor in Europe.Footnote 57 The United States has externalized the costs of its failure to provide more generous domestic social safety policies on foreign producers whose products are subjected to higher tariffs. WTO jurisprudence interpreted the application of these import relief provisions in ways that reduced national policy discretion, raising the ire of the United States which depended on them as political and social safety valves.Footnote 58
The WTO General Agreement on Tariffs and Trade (GATT) provides additional affirmative defenses for unilateral trade restrictions on products that are made with ‘prison labour’ (under GATT Article XX(e)), as well as ones ‘necessary to protect public morals’ (under GATT article XX(a)).Footnote 59 In both cases, the imposition of such measures is subject to the condition that they must not be ‘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’. Neither has been the subject of a WTO panel decision regarding labor rights violations. GATT article XX(a), however, potentially could be widely used to address moral externalities given the open-ended concept of ‘public morals’. It also could be subject to significant protectionist abuse, which could negatively affect industries and workers in developing countries, thus creating a rational for a package treaty to manage the externalities at stake.
3.1.2 International and Transnational Policies: The Development of Package Treaties.
International and transnational networks have long addressed labor rights concerns. Countries created the International Labour Organization in 1919, which brings together networks of government officials with business and labor representatives to develop labor standards through conventions and recommendations, backed by supervision and review of implementation. In parallel, civil society networks press businesses to ensure labor protections, including of foreign suppliers in global supply chains. Companies develop corporate social responsibility policies, in parallel, in part to avoid being subject to boycotts and adverse publicity, which can especially affect those reliant on brand recognition.Footnote 60
A complement to reliance on domestic measures is to negotiate package treaties, as defined above, which combine binding labor commitments with trade commitments, and thus agree on how the externalities in question should be managed. At the multilateral level, developing countries have opposed the creation of positive obligations within trade agreements that directly address labor rights. The United States and Europe thus turned to bilateral and plurilateral agreements that incorporate them.Footnote 61 The United States has gone furthest under the United States–Mexico–Canada Agreement (USMCA) that replaced the North American Free Trade Agreement (NAFTA) in 2020. Under USMCA Chapter 23, the parties guarantee to adopt and maintain (i) freedom of association and the recognition of the right to collective bargaining; (ii) the elimination of all forms of compulsory or forced labor; (iii) the abolition of child labor; (iv) the elimination of employment discrimination; and (v) the protection of acceptable conditions of work with respect to minimum wages, working hours, and occupational health and safety.Footnote 62 These commitments replicate the language that the Obama administration negotiated in the TransPacific Partnership (TPP) and do not depart from the templates used in other US free trade agreements.Footnote 63 In addition to these five areas, the USMCA addresses additional labor concerns such as discrimination based on sex or gender, violence and threats against workers, and protection of migrant workers.Footnote 64 The TPP addressed these concerns, but the USMCA strengthened some of the language.Footnote 65 The USMCA also requires the parties to ban the import of goods from all forms of forced labor, again including tighter language than under the TPP.
Governments can use package treaties as leverage to press for legal changes in foreign countries. The EU–Vietnam trade agreement required Vietnam to ratify ILO Convention No. 98 on Collective Bargaining and revise its labor code as a condition for the agreement.Footnote 66 Relatedly, the USMCA required Mexico to substantively upgrade its labor law to protect the rights of workers to choose a union to represent them.Footnote 67 Before the USMCA was adopted, unions in Mexico were effectively recognized without employee approval and were often established before employees were hired. These ‘protection unions’ were criticized for protecting the interests of the facility and political party in power rather than those of the workers.Footnote 68 On 1 May 2019, Mexican President Lopez Obrador signed a law to ensure that workers can vote for union representatives by secret ballot to join a union of their choice, and to create an independent labor court to hear union claims.Footnote 69 In parallel, Mexico created a Labor Reform Monitoring mechanism to monitor implementation of these reforms. The United States required its approval of these legal reforms as a condition for US ratification of the USMCA, given the pecuniary externalities affecting US workers from economic integration with Mexico.Footnote 70
One key aspect of the US approach to package treaties is that these provisions are subject to binding dispute settlement that can give rise to sanctions, unlike the EU's traditional approach.Footnote 71 Under earlier US trade agreements that included labor requirements, it was difficult to enforce the labor provisions in practice because the complaining state had to show that the violations were not only sustained but they were done ‘in a manner affecting trade’.Footnote 72 Under the US–Dominican Republic–Central America Free Trade Agreement (known as CAFTA-DR), the United States challenged Guatemala's labor practices, but the panel found that the US had not met its burden of proof to show that the labor violations were in ‘a manner affecting trade’.Footnote 73 In light of this case, the USMCA reversed the burden of proof and created a presumption that violations are in a manner affecting trade.Footnote 74 It thereby facilitated the use of trade restrictions to address labor rights concerns in a package treaty.
The biggest innovation in the USMCA was a new facility-specific Rapid Response Labor Mechanism to address claims of denial of workers' collective bargaining rights.Footnote 75 Under this mechanism, a petitioner, such as a labor union, can submit a complaint to the US Department of Labor that a factory in Mexico has blocked unionizing efforts. The department, in turn, can ask Mexico to investigate. US customs officials may then suspend liquidation of customs accounts for the goods in question from the date of the request, which could give rise to significant tariffs. The suspension remains in effect until either the two countries settle the matter, or a binational panel of labor experts assesses the claim under a streamlined process of about four months. The remedy for a violation must be proportionate and may include a fine. This innovation is critical because the most important aspect of worker rights is empowering workers in negotiations over working conditions. The enforcement of this right constitutes the USMCA's biggest innovation on labor protection, enhancing worker participation in determining the terms governing the workplace. As of 1 January 2024, there had been 14 cases against Mexico that resulted in a formal course of remediation or other remediation measures being taken, with other complaints pending.Footnote 76
A drawback to the US approach is that the United States does not work through the International Labour Organization (ILO) and its tripartite system of government, business, and labor, in part because the US has ratified so few ILO conventions.Footnote 77 Rather, the United States uses its power to define core labor rights in a way that bypasses existing multilateral labor agreements and shields the US from criticism of its own failings, thus affecting outcomes. For example, the USMCA references both the 1998 ILO Declaration on Rights at Work and the 2008 ILO Declaration on Social Justice for a Fair Globalization, but its binding obligations only refer to the 1998 ILO Declaration, and not the 2008 Declaration. This is because the 2008 Declaration embraces a wider range of ILO conventions that the United States has not signed, including the Labor Inspection Convention, the Employment Policy Convention, and the Labor Inspection Agriculture Convention. Similarly, the USMCA Rapid Response Mechanism has been rendered useless for claims against US facilities (such as those located in ‘right to work’ states) because of a footnote in the text. For a ‘claim’ to be brought against the US, there must be an alleged denial of workers' rights ‘under an enforced order’ of the National Labor Relations Board (NLRB), which can take five years.Footnote 78 It is thus anything but ‘rapid,’ unlike claims against facilities in Mexico.
The institutional alternative of package treaties is thus highly imperfect, and criticism is often justified. Nonetheless, more can be envisaged in package treaties to protect labor rights while enhancing fairer, third-party oversight. As I have developed elsewhere, they could, for example, provide for a supplementary social safeguard provision in the form of a firm- or sector- specific safeguard where there are sustained labor rights violations (as defined in the agreement) and where an industry in the importing country has been materially injured or threatened with material injury.Footnote 79 The application of the mechanism would be subject to judicial review to protect against abuse. The USMCA (following NAFTA) innovatively provides for binational review panels of import relief measures, which could be extended to review social safeguard measures.Footnote 80 The process could incorporate detailed due process requirements and be subject to further review under an international dispute settlement mechanism, whether under a bilateral or plurilateral agreement (such as the USMCA), or potentially under the WTO's dispute settlement mechanism. The external reviews could incorporate ILO expertise and processes more directly, and they could enhance labor and civil society engagement in more transparent ways.Footnote 81 They would, in this sense, be largely procedural in nature, but be fairer in terms of participation and outcomes than unilateral determinations.
In sum, flanking policies are needed to address externalities from trade that implicate labor protections, labor adjustment, and social inclusion. Much can and should be done at the domestic level. However, given the structural forces of globalization that favor capital over labor, both unilateral trade restrictions and international measures under package treaties are justified. Given countries' varying economic and social contexts, harmonization of requirements is often inappropriate. Thus, policy space for unilateral measures to protect labor and the broader social bargain is necessary. Nonetheless, even though unilateral trade measures on labor rights grounds can be justified to provide protection from the pecuniary, moral, and social-political externalities of trade, there are legitimate concerns that they will be used in ways that prejudice countries in the Global South and their workers, and thus negatively impact the most vulnerable through externalizing adjustment costs on them. For that reason, package treaties negotiated with developing countries pursuant to which domestic trade measures are subject to procedural scrutiny and justification is a preferred, complementary institutional option. Package treaties hold the prospect of being more balanced in the management of reciprocal externalities than reliance on domestic trade restrictions imposed unilaterally.
Table 4 summarizes different domestic and international flanking policies to address externalities implicating labor and social inclusion, together with examples of them.
Institutional alternatives to address labor and social inclusion concerns
3.2 Sustainable Development and Climate Change
Assessing alternative and complementary policy responses to negative environmental externalities parallel and differ from those implicating labor and social inclusion. Capital can be invested elsewhere, which can put pressure on governments and businesses to reduce environmental protections in the importing country. Concern over ‘carbon leakage’ is a prime example. If one jurisdiction (such as the European Union) raises taxes or other regulations to reduce greenhouse gas emissions, production can shift abroad, not only displacing jobs, but also eliminating any positive effect on emissions from the domestic measures. Reduced emissions in such jurisdictions and their replacement by emissions in other jurisdictions are called carbon leakages. These risks may undermine national efforts to address climate change.
Concerns over labor and the environmental externalities, however, differ in at least three respects. First, they involve different constituencies. Labor and environmental advocates may work on opposite sides of climate change issues. Labor, for example, may collaborate with industry to lobby for reduced environmental requirements to protect existing jobs. Take the steel and coal industries. To reduce greenhouse gas emissions, industry needs to phase out the use of high-emission, coal-fired blast furnaces. Yet unionized labor may fight to maintain jobs in the industry, while environmental groups press for a rapid transition and closure of these plants. In other words, different domestic constituencies are affected by different externalities, reflecting once more the critical point that there is no single, agreed externality that can simply be ‘internalized’. Rather, managing externalities will involve institutional choices that affect participation and outcomes.
Second, environmental concerns are not just indirectly transnational in scope;they are also direct in their transboundary material effects. Transboundary environmental harm caused by the production of imported products directly and materially affects citizens in the importing country. It directly affects many more people outside the country of production than labor rights violations. The risks from climate change, resulting from greenhouse gas emissions, affect everyone. Successful climate change mitigation depends on massively reducing these emissions at their source, which requires changes in exporting countries.
Third, some environmental issues, such as the protection of the ozone layer and climate change, involve global public goods requiring global cooperation. They involve, in Scott Barrett's terms, ‘aggregate efforts public goods’ because producing the good requires the aggregate efforts of multiple countries. Reliance on domestic flanking policies that are not linked to trade, such as in the case of active labor and social welfare policies to help workers adjust, is simply not an effective option. Rather, mechanisms are required to provide assurance that other countries are doing their share to address climate change. Package treaties can play a critical role.
3.2.1 Domestic Policies
Climate change mitigation ultimately depends on domestic policies around the world. The policies will require a combination of carrots, such as in the form of subsidies for clean energy alternatives, and sticks, such as taxes and regulations imposed on greenhouse gas emitting production. The United States and China, for example, have massively subsidized clean energy alternatives.Footnote 82 The European Union, in parallel, is taking the lead in imposing taxes on emissions through an emissions trading system.Footnote 83 Many other countries and sub-national jurisdictions are taking initial steps toward developing them as well.Footnote 84 The state of California, for example, operates its own emissions trading system.Footnote 85
Countries may complement domestic policies with trade measures that impose taxes and other regulations on imported products. Although possibly illegal under current WTO rules, countries could impose duties on products produced with high greenhouse gas emissions to countervail the implicit subsidy in a country's failure to tax or otherwise regulate such emissions, thus aiming to ‘internalize’ the environmental externality through an import tax. The subsidy would be viewed as a ‘negative’ one in terms of a country's failure to regulate, as opposed to a ‘positive’ one based on a grant or other ‘financial contribution’. Although not labeled as a countervailing duty, the European Union adopted regulations that will impose a ‘carbon border adjustment mechanism’ (CBAM) on imports to complement its emission trading system. Under CBAM, starting in 2026, importers of products in six industrial sectors will have to pay a carbon tax that is equivalent to that imposed on EU producers in these sectors, after offsetting any carbon taxes paid in the country of production.Footnote 86 In this way, the European Union aims to address the material and pecuniary externalities of foreign countries' failure to regulate and tax carbon emissions.
The design of the border adjustment system will affect producer incentives.Footnote 87 If the calculation of the charge is based on plant-level emissions, then it could trigger a ‘reshuffling’ where plants with low emissions sell to the European Union, while those with high emissions sell elsewhere. If the charge is rather based on industry-level emissions, then individual plants have reduced incentives to lower their emissions. It thus may be preferable to set an industry-level baseline, but to permit individual exporters to ‘opt for self-certification’, if feasible, so that individual plants are incentivized to lower their emissions.
Such measures are unilateral,Footnote 88 but they may be permitted under WTO law under certain conditions, as recognized in WTO jurisprudence.Footnote 89 GATT articles permit trade measures that would otherwise violate WTO rules where the measures are ‘necessary to protect human, animal, or plant life or health’ (article XX(b)), or they relate ‘to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption’ (article XX(g)). In such cases, WTO jurisprudence requires that there be a ‘nexus’ between the regulating country and the harm (i.e., externality) resulting from the foreign production process.Footnote 90 In addition, under article XX's introductory clause (known as the ‘chapeau’), the measure must not be ‘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 risks of climate change clearly affect human, animal, and plant life, and they relate to ‘the conservation of an exhaustible natural resource’ (in this case, a stable climate), as interpreted under WTO jurisprudence.Footnote 91 In the case of the EU's CBAM, it is ‘made effective in conjunction’ with domestic restrictions, and there is a clear nexus between the regulation and harm within the European Union since global warming affects EU citizens. Thus, the legality of CBAM depends on whether its application will involve ‘unjustifiable discrimination’ on other countries. So long as the application is not discriminatory, CBAM should comply with WTO requirements. Even if aspects of CBAM are found to constitute unjustifiable discrimination or a disguised trade restriction,Footnote 92 the European Union should be able to adjust the mechanism to eliminate these aspects following such a ruling, depending on the ruling's scope.
3.2.2. International and Transnational Policies: The Development of Package Treaties
International and transnational networks have long addressed transnational and global environmental challenges. States have worked through conferences of the parties that adapt international principles, guidelines, and treaties, and that create and work through international environmental institutions, such as the United Nations Environmental Programme. Civil society groups participate in these meetings, and they network outside of them. They pressure companies, organize boycotts, and create certification and labeling regimes.Footnote 93
Starting in the 1990s, environmental issues became more salient in the trade regime. Because WTO agreements create disciplines on unilateral measures – such as the requirement that they not constitute unjustifiable discrimination or a disguised restriction on international trade – they already include international law conditions for addressing externalities. However, as yet they generally do not impose positive obligations to address environmental sustainability and climate change. The one exception is the 2022 WTO Agreement on Fisheries Subsidies (awaiting ratification by two-thirds of the membership before it enters into force), which prohibits subsidies in support of certain forms of fishing that threaten fish stocks.Footnote 94 In contrast, many bilateral and plurilateral trade agreements increasingly contain environmental provisions that go beyond ‘exceptions’, although only a minority involve binding dispute settlement.Footnote 95 Chapter 24 of the USMCA, for example, requires implementation of a number of environmental treaties, bans subsidies that negatively affect fish stocks, and maintains a Commission for Environmental Cooperation that can receive private submissions regarding a party's failure to effectively enforce its environmental laws. The EU's approach to preferential trade agreements also is progressively including more environmental commitments.Footnote 96
Package treaties offer many institutional advantages. They can ensure broader participation in the formation of requirements and standards, and wider acceptance of them. They also can facilitate coordination, harmonization, and mutual recognition of countries' domestic policies in support of more effective and equitable outcomes, compared to the alternative of unilateral trade measures. It would be preferable, for example, to determine measurements of greenhouse gas emissions and pricing in a coordinated way to foster cooperation and avoid bias in calculations that favor domestic interests.Footnote 97 Over time, countries can form ‘climate clubs’, which ideally should be open for membership and should include a broad range of countries at different levels of development, while taking account of developing countries' varying capacities.Footnote 98
Given the size of their markets, the United States and European Union exercise considerable leverage in asserting jurisdictional power when regulating products based on production processes. Countries with smaller markets wield no such power and their citizens' interests are not represented in the US and EU regulatory process. Moving regulation to the international level, as done with the Montreal Protocol on the Projection of the Ozone Layer, is thus preferable for them as an institutional option compared to unilateral, domestic measures. Multilateral treaties are more likely to respond to the concerns of developing country constituencies, even though these constituencies still exercise much less influence in international negotiations given power and wealth asymmetries.
For example, international environmental treaties recognize the ‘common but differentiated responsibilities’ of developed and developing countries, enshrined in Principle 7 of the Rio Declaration.Footnote 99 The first principle set forth in the United Nations Framework Convention on Climate Change correspondingly provides:
‘The Parties should protect the climate system for the benefit of present and future generations of humankind, on the basis of equity and in accordance with their common but differentiated responsibilities and respective capabilities. Accordingly, the developed country Parties should take the lead in combating climate change and the adverse effects thereof.Footnote 100
Application of this principle in trade law would account for countries' respective contributions to climate change and their respective capacities to address the challenges in light of their economic development levels.
In contrast, unilateral US and EU measures do not incorporate this principle when they address the material externalities of carbon emissions from foreign production processes. The European Union's CBAM is based on equality of treatment as opposed to equity of treatment, because it does not account for the different contexts of developing countries.Footnote 101 Moreover, the revenue collected through CBAM is to accrue to the European Union, and there is no guarantee that it will be used to finance technology transfers and climate change adaptation in developing countries whose populations are most vulnerable. Thus, even though CBAM may comply with WTO law, it would fail to adhere to this international environmental law principle. WTO judicial review, if available, could discipline discriminatory aspects of EU or US regulations, but WTO law would not directly apply this equity principle.Footnote 102
Even though there are strong critiques of CBAM, an important argument in support is that negotiations at the multilateral level have occurred for decades with little success in reducing emissions. This result contrasts starkly with the ozone regime where the Montreal Protocol incorporates trade sanctions that incentivize global compliance with the treaty's requirements.Footnote 103 CBAM could spur needed domestic regulatory change in third countries to adopt systems designed to reduce greenhouse gas emissions so that, over time, the negotiation of international package treaties addressing the catastrophic risks of global warming will be facilitated, including ones that incorporate required technological, knowledge, and financial transfers to developing countries.Footnote 104 For example, Vietnam and Indonesia have announced plans for a carbon tax and emissions trading scheme, and China will likely expand its existing one.Footnote 105 The United States is working with the European Union to see what can be done to reconcile their different approaches.Footnote 106 CBAM, in other words, has provided a useful provocation. It should be followed, however, by negotiations that give agency to affected countries, particularly from the Global South. These processes of normative and institutional development are dynamic and need to be seen as part of transnational legal ordering that goes beyond conventional international law and that can support more effective outcomes, which, in the case of climate change, is of existential importance. Hybrid structures are needed involving public–private collaboration, national and local innovation, and transnational information sharing, backed by international commitments, transparent accounting, reporting, reviews, and (where needed) sanctions.
Table 5 summarizes different domestic and international flanking policies to address environmental externalities, together with examples of them.
Institutional alternatives to address environmental concerns
4. A Comment on Flanking Measures: Conceptualizing the Core and the Flank
The concept of flanking policies in relation to trade liberalization is trade centric. If one does not start with trade liberalization as a goal, but rather with labor or environmental protection, then the terminology flips such that trade restrictions become the flanking measures and labor and environmental protections the core. As regards labor protections, ILO treaties and their requirements could form the core, and trade measures to enforce them constitute flanking measures. Likewise, as regards environmental protection, environmental requirements could form the core and trade restrictions provide flanking measures. When the starting point for trade policy is framed in terms of sustainable development as opposed to trade liberalization, then the notion of the core also shifts, such that trade and environmental measures can be integrated in new ways.Footnote 107
The Montreal Protocol reflects an approach where sustainability lies at the core. Central to the regime are sectoral committees consisting of government officials, scientists, and industry who work to develop new alternative technologies that will not deplete the ozone layer.Footnote 108 Technology transfers, time extensions, and trade sanctions form a key part of that regime.Footnote 109 They constitute flanking measures to account for different forms of economic harm when countries comply (or fail to comply) with environmental commitments. For example, developing countries and their constituents are harmed if they are not provided with technology and phase-in periods to adapt to new environmental requirements. Similarly, developed countries and their constituents are harmed if they lose industries to countries that do not comply with environmental commitments. Any trade or other sanction – or, in the words of Charles Sabel and David Victor, a ‘penalty default’Footnote 110 – is available as a last resort where there is no compliance, and the situation cannot be resolved otherwise.Footnote 111
Given the existential threat of climate change, its mitigation should lie at the core, and trade law should not impede, but rather contribute to addressing the problem. At the core, as with the earlier experience of combating depletion of the ozone layer, collaboration is needed among government officials, scientists, and industry that is sector-specific in orientation. Given the complexity of climate change, such a sector-specific, step-by-step, experimental approach is most practical to address the externalities at stake.
5. Conclusion: Externalities, Flanking Measures, and Package Treaties
The negative externalities arising from trade that implicate sustainable development and social inclusion can be addressed at multiple levels of governance through flanking policies. Given the material, moral, and social externalities at stake, there are strong justifications for domestic measures that set conditions for trade. These measures are permitted under WTO law, and countries increasingly use them. Since national and local governments are closer to their citizens and the social and environmental conditions in question, they are better positioned to respond to these contexts. But international institutions and transnational networks can be critical as well since one country's measures affect others, and since cooperation, coordination, and transnational processes of learning and experimentation that support domestic regulation are needed.
Package treaties can directly address social inclusion and sustainable development, and (in theory) they can give rise to institutions that facilitate deliberative problem solving. The United States and European Union are taking the lead in developing provisions to address these concerns in bilateral and plurilateral agreements. Given the market power they wield, however, some wariness is in order; they are in a stronger position to create provisions that reflect their regulatory priorities and to draft them in asymmetric ways, especially in bilateral settings. There is thus an important role for developing package treaties at the multilateral level to protect against abuse, including at the WTO, provided these agreements leave sufficient policy space for domestic experimentation in light of institutional, social, and normative diversity.
The question arises: What should lie at the core and what should be considered a flanking measure. Given the existential risks of climate change, the core focus should be on climate change mitigation, with trade sanctions serving as supportive flanking measures. With climate change, time is of the essence. Until multilateral efforts reach fruition, countries must be able to adopt unilateral measures in conjunction with their domestic initiatives in a non-discriminatory manner. These measures should be seen as part of broader transnational processes that focus on problem-solving over time in ways that equitably incorporate the concerns of publics around the world. Package treaties should form part of these transnational legal ordering processes as complements that do not replace, but that are supportive of domestic governance. Reimagining trade agreements as package treaties, including where sustainability and social inclusion concerns constitute the core and not the flank, will not be easy. Lawyers, economists, and others created the intellectual constructs and designs for the current trade legal order. They must do so for its redesign to respond to current global and transnational challenges.
This article has provided an analytic framework for assessing different types of externalities, how they overlap in practice, and institutional alternatives to address them dynamically over time. Future work should address how concerns over different externalities give rise to different institutional choices in light of the costs and implications for affected parties in particular contexts. There is no institutional panacea, including that of package treaties. Rather, package treaties are best viewed as institutional complements that may contribute to broader transnational legal ordering processes in addressing critical public policy concerns.
Acknowledgements
I thank Tomer Broude, Harlan Cohen, Ben Heath, Jeffrey Kucik, Nicolas Lamp, Desirée Leclerq, Tim Meyer, Joel Trachtman, one anonymous reviewer, and participants at workshops at the Graduate Institute, Geneva, and Georgetown Law School for their comments.
Open access
