The 2016 presidential election was one of the most divisive in re­cent memory, but it produced a surprising bipartisan consensus. Donald Trump, Hillary Clinton, and Bernie Sanders all agreed that U.S. trade agreements should be, but are not, “fair.” Although it achieved broad con­sensus only recently, the critique that U.S. trade agree­ments are un­fair has been around for decades. Since 1992, much of this fairness cri­tique has focused on ensuring that trade liberaliza­tion does not undermine noncommercial values, such as environmental pro­tection and labor conditions. Beginning with the negotiation and ratification of the North American Free Trade Agreement (NAFTA) in the early 1990s, governments have responded by including in their trade agreements a prohibition on the selective enforcement of environ­mental and labor laws. This ban—a central component of efforts to make sure that free trade agreements are fair—aims to prevent a global race to the bottom in environmental and labor standards.

These efforts have fallen wide of the mark. This Article makes two novel contributions. First, it demonstrates empirically that selective en­forcement is considerably more pervasive than commonly thought. But contemporary selective enforcement is the reverse of the kind of selective enforcement that has traditionally concerned trade critics. Instead of selectively enforcing environmental and labor laws to gain a trade advantage, governments selectively enforce trade laws in ways that undermine environmental and labor interests. This Article presents data from trade enforcement actions in the energy and fisheries sectors to demonstrate this claim. In both sectors, trade laws are enforced exclu­sively against natural resource substitutes, such as renewable energy and farmed fish. The natural resources with which these products com­pete, fossil fuels and wild fish, benefit from the same allegedly unlawful conduct but are not targeted for enforcement.

Second, this Article presents a theory of how selective enforcement of trade law distorts markets to the detriment of the environment. It argues that selective enforcement is an implicit subsidy for products that are not targeted for enforcement but benefit from the same allegedly unlawful conduct as targeted products. This Article presents evidence that selective enforcement in the energy and fisheries sectors has indeed caused these effects.

This Article concludes by suggesting how governments can reform trade law enforcement to address the pernicious effects of selective enforcement. Governments have acted to address other kinds of selective enforcement in the past, so reform is politically feasible. Nevertheless, given the current political climate, reforms should concentrate on increasing trade law enforcement across the board.


    1. Selective Enforcement of Environmental and Labor Laws
    2. Selective Enforcement Against Weak Countries
    1. Selective Enforcement in the Energy Sector
    2. Selective Enforcement in the Fisheries Sector
    1. The Financial Costs of Selective Enforcement
      1. Litigation Costs
      2. Liability
        1. Withdrawn Government Support
        2. Higher Duties on Natural Resource Substitutes
      3. Lost Investment and Higher Costs of Capital
    2. What Kind of Relationship Between Products Makes Selective Enforcement Pernicious?
    3. The Social Costs of Selective Enforcement
    1. The Effects of Selective Enforcement in Energy
    2. The Effects of Selective Enforcement in Fisheries
    1. Reforming the WTO Enforcement Process
    2. Reforming Trade Remedy Investigations



The 2016 presidential election was one of the most divisive in recent memory, but it produced a surprising bipartisan consensus: U.S. trade agreements should be, but are not, “fair.” Shortly after taking office, Republican President Donald Trump told Congress, “I believe strongly in free trade, but it also has to be fair trade.” 1 Donald J. Trump, Remarks by President Trump in Joint Address to Congress (Feb. 28, 2017), []. During the Democratic pri­mary, Senator Bernie Sanders argued for “fair trade which works for the middle class and working families.” 2 Bernie Sanders on Free Trade,,
2016/Bernie_Sanders_Free_Trade.htm [] (last updated Aug. 18, 2016).
By the general election, Democratic candidate Hillary Clinton, who initially supported many ex­isting U.S. trade agreements, also embraced “smart and fair trade” as a criticism of existing U.S. policies. 3 Jared Bernstein, Hillary Clinton Calls for ‘Smart and Fair Trade.’ What Is That?, Wash. Post (Oct. 3, 2016),
03/hillary-clinton-calls-for-smart-fair-trade-what-is-that/?utm_term=.9c508926d19a (on file with the Columbia Law Review).
The notion of “fair” trade implies that trade agreements should protect values other than pure trade liberaliza­tion. But which values must be protected in order for trade to be “fair”?

Since 1992, much of the fairness critique of U.S. trade agreements has focused on ensuring that free trade does not undermine environ­mental and labor standards in developed and developing countries alike. 4 See Timothy Meyer, Saving the Political Consensus in Favor of Free Trade, 70 Vand. L. Rev. 985, 1002–08 (2017) [hereinafter Meyer, Saving the Political Consensus] (recounting the history of labor and environmental chapters in trade agreements). Beginning with the negotiation and ratification of the North American Free Trade Agreement (NAFTA) in the early 1990s, 5 North American Free Trade Agreement, Can.-Mex.-U.S., Dec. 17, 1992, 32 I.L.M. 289 (1993). critics have worried that reducing trade barriers will cause companies to relocate production facilities to countries with lower environmental and labor standards. 6 Douglas Seay & Wesley Smith, Why the Governors Support the NAFTA (and Washington Doesn’t), Heritage Found. (June 15, 1993),
reports/1993/06/bg946nbsp-why-the-governors-support-the-nafta [] (discussing NAFTA’s opponents’ strategy of “portray[ing] the NAFTA as leading to a massive loss of jobs, the de-industrialization of the U.S., and a lowering of labor and environmental standards to those of the Third World”).
This dynamic would, in turn, pressure countries to lower their environmental and labor standards in order to keep or attract jobs. 7 See Anthony DePalma, Law Protects Mexico’s Workers but Its Enforcement Is Often Lax, N.Y. Times (Aug. 15, 1993), (on file with the Columbia Law Review) (discussing “concerns that American workers would lose jobs and receive lower wages under [NAFTA]”). To prevent this race to the bottom, the NAFTA parties agreed on rules prohibiting the selective enforcement of labor and environmental laws. 8 North American Agreement on Environmental Cooperation, Can.-Mex.-U.S., art. 5, Sept. 8–14, 1993, 32 I.L.M. 1482 [hereinafter NAFTA Environmental Side Agreement] (“[E]ach Party shall effectively enforce its environmental laws and regulations through ap­propriate governmental action . . . .”); North American Agreement on Labor Cooperation, Can.-Mex.-U.S., art. 3, Sept. 8–14, 1993, 32 I.L.M. 1502 [hereinafter NAFTA Labor Side Agreement] (“Each Party shall promote compliance with and effectively enforce its labor law . . . .”). This ban on selective enforcement soon became standard in trade agreements worldwide and a central component of efforts to make sure that free trade agreements could, indeed, be fair. 9 See Cathleen Cimino-Isaacs, Labor Standards in the TPP, in Trans-Pacific Partnership: An Assessment 261, 265–66 (Cathleen Cimino-Isaacs & Jeffrey J. Schott eds., 2016) (discussing the evolution of labor provisions in U.S. free trade agreements ); Meyer, Saving the Political Consensus, supra note 4, at 1003–04 (discussing the negotiation and adoption of NAFTA side agreements, which “required the NAFTA parties to enforce their existing labor and environmental laws” and authorized “monetary fines if a country failed ‘to effectively enforce its occupational safety and health, child labor or minimum wage technical standards’” (quoting NAFTA Labor Side Agreement, supra note 8, art. 39)).

These efforts have fallen wide of the mark. This Article makes two novel contributions. First, focusing on the environmental context, it de­monstrates empirically that selective enforcement in trade law today is pervasive. But contemporary selective enforcement is the reverse of the kind of selective enforcement that has traditionally worried trade critics. Instead of selectively enforcing environmental laws to gain a trade ad­vantage, governments selectively enforce trade laws in ways that hurt environmental interests. Second, this Article argues that this new kind of selective enforcement slows the development of competitive environ­mentally and labor-friendly products. In effect, selective enforcement of trade law acts as an implicit subsidy for products with large social costs—the epitome of an “unfair” trade practice.

To illustrate these two arguments, this Article evaluates how gov­ernments enforce rules that limit government subsidies and other forms of government financial support for industries against, on the one hand, programs benefitting natural resources (such as fossil fuels) and, on the other hand, against programs benefitting their substitutes (such as renewable energy). 10 Trade rules include a definition of a “subsidy” that is narrower than the colloquial meaning of the term and that does not capture all forms of government financial support. See Agreement on Subsidies and Countervailing Measures art. 1, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1869 U.N.T.S. 14 [hereinafter SCM Agreement]. Governments, however, can and do use a broad range of trade rules to challenge government financial support that might not technically count as a subsidy under trade rules. See infra Part II. Except for where this Article discusses the SCM Agreement itself, it uses the term “subsidies” in the colloquial sense of “government financial assistance,” rather than its technical meaning under the SCM Agreement. I define selective enforcement in the commercial context as the systematic enforcement of laws against some producers but not others that (1) compete with the targets of enforcement and (2) engage in or benefit from the same allegedly unlawful conduct. (I refer to producers that meet these criteria as “similarly situated.”) The energy sector provides the most glaring example of this phenomenon. Govern­ment financial support for fossil fuels is about seven times larger than support for renewable energy: $934 billion to $135 billion in 2014. 11 See infra notes 286–293 and accompanying text. Yet in the last eight years, governments and the World Trade Organization (WTO) have, within the energy sector, cracked down exclusively on ef­forts to support renewable energy.

Since 2008, governments have launched over 20 domestic trade investigations and WTO challenges to other nations’ renewable energy support programs, 12 See infra Part IV. disrupting investment in the renewable energy sec­tor. Indeed, as this Article goes to print, President Trump has just imposed thirty percent “safeguard” tariffs on solar cells imported into the United States, a move some estimate could divert investment away from solar energy and jeopardize 23,000 jobs in the renewable energy sector. 13 Press Release, Office of the U.S. Trade Representative, President Trump Approves Relief for U.S. Washing Machine and Solar Cell Manufacturers (Jan. 22, 2018), []; Oliver Milman, Trump’s Tariffs on Panels Will Cost US Solar Industry Thousands of Jobs, Guardian (Jan. 24, 2018), [] (citing the Solar Energy Industries Association’s figure for potential job loss). Ontario canceled the largest part of its renewable energy Feed-in Tariff Program, which paid preferential rates to electricity generators that relied on renewable sources, following an adverse WTO ruling. 14 See infra notes 109–111 and accompanying text; see also Appellate Body Report, Canada—Certain Measures Affecting the Renewable Energy Generation Sector, para. 6, WTO Doc. WT/DS412/AB/R (adopted May 6, 2013) [hereinafter Canada—Renewables Appellate Body Report]; Appellate Body Report, Canada—Measures Relating to the Feed-in Tariff Program, para. 6, WTO Doc. WT/DS426/AB/R (adopted May 6, 2013) [hereinafter Canada—Feed-in Tariff Program]; Paul Gipe, Two Steps Forward, One Back: Ontario Cancels Feed-in Tariffs for Large Projects, Renewable Energy World (June 10, 2013), []. Argentina initially saw its biofuels industry dry up after the European Union imposed antidumping duties that effectively closed the European market to Argentinian biofuel producers. 15 See Council Implementing Regulation (EU) 1194/2013 of 19 Nov. 2013, Imposing a Definitive Anti-Dumping Duty and Collecting Definitively the Provisional Duty Imposed on Imports of Biodiesel Originating in Argentina and Indonesia, 2013 O.J. (L 315) 2, 8 (imposing duties on Argentinian and Indonesion biodiesel); En Argentina Cae 40 % la Producción de Biodiesel [Biofuel Production Falls 40% in Argentina], Biodiesel Argentina Noticias Sobre Biodiesel y Energias Renovables (May 14, 2013), [] [hereinafter Biodiesel Argentina Noticias]. Governments have challenged their trade partners’ support for renewable energy despite the fact that the industry offers the promise of environmentally sustainable growth in energy to meet the needs of both industrialized countries and the developing world.

By contrast, no WTO member has ever initiated proceedings directly targeting government support for domestic fossil fuel industries. This contrast is both puzzling and troubling. Subsidies for fossil fuel consump­tion lower the cost of energy to consumers. Lower costs incentivize en­ergy-intensive industries like steel production to relocate to countries with fossil fuel subsidies, which disadvantages the countries from which those industries flee. 16 Why governments have not used trade rules to challenge fossil fuel subsidies is beyond the scope of this Article, but several recent articles have addressed this question. See Dirk De Bièvre, Ilaria Espa & Arlo Poletti, No Iceberg in Sight: On the Absence of WTO Disputes Challenging Fossil Fuel Subsidies, 17 Int’l Envtl. Agreements 411, 413 (2017); Timothy Meyer, Explaining Energy Disputes at the World Trade Organization, 17 Int’l Envtl. Agreements 391, 393 (2017) [hereinafter Meyer, Explaining Energy Disputes]. Moreover, fossil fuel subsidies run counter to global carbon reduction objectives. In 2012, Faith Birol, then the chief economist for the International Energy Agency (and now its head), esti­mated that removing subsidies for coal, oil, and natural gas could pro­duce half of the greenhouse gas emissions reductions necessary to keep global temperatures from rising more than two degrees Celsius above preindustrial levels, 17 Duncan Clark, Phasing Out Fossil Fuel Subsidies ‘Could Provide Half of Global Carbon Target,’ Guardian (Jan. 19, 2012),
2012/jan/19/fossil-fuel-subsidies-carbon-target []. A 2015 study from the Nordic Council of Ministers and the International Institute for Sustainable Development estimated that removing just consumption subsidies for fossil fuels could reduce greenhouse gas emissions by somewhere between 6% and 13%. Laura Merrill, Melissa Harris, Liesbeth Casier & Andrea M. Bassi, Glob. Subsidies Initiative, Int’l Inst. for Sustainable Dev., Fossil-Fuel Subsidies and Climate Change: Options for Policy-makers Within Their Intended Nationally Determined Contributions 7 (2015), [].
the limit to which nations have agreed. 18 See Paris Agreement on Climate Change art. 2, Dec. 12, 2015, T.I.A.S. No. 16-1104 (entered into force Nov. 4, 2016) (noting nations’ agreement to hold global temperature increases to “well below 2°C above pre-industrial levels”). Hence, the logic of both liberalized trade and environmental protection suggests we should see challenges to fossil fuel subsidies.

When natural resource substitutes such as renewable energy are more likely to be targeted for enforcement, they suffer a competitive dis­advantage. 19 See infra Part IV. Governments may withdraw support to avoid retaliation by other nations, as they have in the renewable energy sector. 20 See infra section III.A.2.a. Invest­ment in the sector may be withdrawn as investors seek higher returns else­where, dampening development and innovation. Natural resource substi­tutes may face higher import duties. 21 See infra section III.A.2.b. Companies producing natural resource substitutes incur financial costs in defending them­selves. 22 See infra section III.A.1. Taken together, these costs allow natural resources to sell at a discount relative to their substitutes, which boosts natural resource consumption. Selective enforcement is, in other words, a subsidy for natural resource consumption. 23 See infra section III.C.

This Article proceeds in five parts. Part I provides a historical ac­count of the concern with selective enforcement in trade law. Govern­ments and scholars have identified and taken steps to address several kinds of selective enforcement, including the selective enforcement of national environmental and labor laws 24 See infra section I.A. and the disproportionate en­forcement of trade obligations against weak states. 25 See infra section I.B. These efforts demon­strate that efforts to reform selective enforcement can appeal to governments, pro­vided they are aware of the selective enforcement in the first place.

Part II provides the first empirical account of selective enforcement of trade laws in ways that harm environmental interests, with special focus on the energy and fisheries sectors. In both of these markets, natural resources compete with natural resource substitutes. And in both of these markets, there are strong environmental reasons to slow the con­sumption of the natural resource (fossil fuels or wild caught fish). Yet in both, governments around the world provide significant financial sup­port to encourage development and consumption of the natural resource. In both cases, government supports for natural resource substi­tutes (renewable energy and aquaculture) are considerably more likely to be targeted for enforcement than similar measures supporting the natu­ral resources themselves. The presence of selective enforcement in these two sectors suggests a broader pattern in which trade enforcement sys­tematically disadvantages environmentally sustainable products.

Part III offers a theoretical explanation for how selective enforce­ment in a commercial context can distort competition among products. The central insight is that selective enforcement acts as an implicit sub­sidy, imposing costs on some products but not other similarly situated products. These costs come in at least three forms: litigation costs, liabil­ity, and increased costs of capital. These costs limit the targeted product’s ability to compete in the marketplace and may slow the investment in innovation necessary to make environmentally friendly products compet­itive in the marketplace. In the environmental context, the effect of this selective enforcement is to reinforce the market position of natural resource–intensive industries. These industries create significant negative externalities in the form of environmental degrada­tion. The effect of selective enforcement in the natural resource context is thus the reverse of what law and economics scholars have long called for: Selective en­forcement subsidizes products with large social costs at the expense of products with large social benefits.

Part IV moves from the general to the specific, providing evidence of how selective enforcement in the energy and fisheries sectors does indeed protect fossil fuels and wild fisheries from competition with envi­ronmentally sustainable alternatives. Part V argues that ongoing efforts to renegotiate trade agreements to make them fair should include measures that limit selective enforcement of trade laws in order to ensure that similarly situated products are regulated in the same fashion. WTO members should consider creating an administrative enforcement pro­cess as an adjunct to the WTO’s dispute-focused method of regulation. The administrative process would identify any products that compete with products at issue in formal WTO disputes and prima facie benefit from the same kind of unlawful conduct. A weak version of this adminis­trative process would involve the distribution of the list of such products to member states and inclusion of the list for discussion at WTO meet­ings. A stronger version would allow the WTO Secretariat or a new pros­ecutorial office to bring a claim against nations maintaining identified measures. Past efforts to address other forms of selective en­forcement suggest that at least the weak version of selective enforcement is politi­cally feasible.

I. Selective Enforcement in Trade Law

Virtually all laws are underenforced. Governments typically lack the resources to pursue all violators, and it is by no means obvious that striv­ing for perfect enforcement would make sense. 26 See Kevin M. Stack & Michael P. Vandenbergh, The One Percent Problem, 111 Colum. L. Rev. 1385, 1393 (2011) (“For decision makers, whether private businesses or government regulators, it is often more efficient to focus on the most significant contributors to a problem.”). For instance, pursuing all violators may not be justified from a cost–benefit perspective. 27 Id. at 1393–94 (“Given economies of scale, it is often the case that with small-percentage contributors the costs of regulation exceed the benefits. As a result, a focus on high-percentage factors often concentrates effort and resources on the sources that may make the most difference at the least expense.”). For these reasons, governments have discretion in choosing the violators against whom they will enforce the law. In the criminal context, prosecu­tors may choose to prioritize going after the perpetrators of particularly heinous crimes. 28 See Richard S. Frase, The Decision to File Federal Criminal Charges: A Quantitative Study of Prosecutorial Discretion, 47 U. Chi. L. Rev. 246, 303 (1980) (concluding that the “prosecutorial discretion of the U.S. attorney is vast and unchecked by any formal, external constraints or regulatory mechanisms”). In the environmental context, regulators may choose to pursue the largest polluters in order to achieve the greatest environ­mental benefit. 29 See Stack & Vandenbergh, supra note 26, at 1393–94.

Selective enforcement, at least here, does not refer to this ordinary exercise of prosecutorial discretion, nor does it refer to random variation in how the laws are enforced. Instead, this Article defines selective enforcement as the systematic imposition of the law against one class of actors but not another that is similarly situated. For purposes of this defi­nition, two actors are similarly situated if (1) they compete with each other in the marketplace and (2) both engage in or benefit from the unlawful practice to similar degrees. Selective enforcement of this kind suggests that prosecutorial discretion is being exercised for reasons other than merely ensuring compliance with the law. Once selective enforce­ment has been demonstrated, the question becomes whether its causes or effects can justify targeting only some legal subjects. 30 Cf. Shoba Sivaprasad Wadhia, The History of Prosecutorial Discretion in Immigration Law, 64 Am. U. L. Rev. 1285, 1288–89 (2015) (discussing the Obama Administration’s decisions to prioritize the enforcement of immigration laws against certain classes of illegal immigrants); P.S. Kane, Comment, Why Have You Singled Me Out? The Use of Prosecutorial Discretion for Selective Prosecution, 67 Tul. L. Rev. 2293, 2300–01 (1993) (discussing empirical evidence that demonstrates that prosecutorial discretion can result in discriminatory behavior).

Trade lawyers have worried about at least two kinds of selective en­forcement. First, they have worried about countries’ selectively enforcing their own national environmental and labor laws in order to attract busi­nesses. 31 See infra section I.A. Second, they have worried that WTO dispute resolution may result in selective enforcement of WTO rules against weak countries but not strong countries. 32 See infra section I.B. This Part addresses these forms of selective en­forcement in turn.

A. Selective Enforcement of Environmental and Labor Laws

The early 1990s marked a new period for trade liberalization. Until that point, parties to the General Agreement on Tariffs and Trade (GATT) had engaged in “rounds” of trade negotiations that typically lasted a number of years. 33 Dani Rodrik, The Globalization Paradox: Democracy and the Future of the World Economy 71 (2011). Most of these rounds focused on reducing tariff rates among members. 34 See Chad P. Bown, Self-Enforcing Trade: Developing Countries and WTO Dispute Settlement 11–14 (2009) (providing a history of the GATT negotiating rounds). In the Tokyo Round in the 1970s, governments took up the issue of nontariff barriers, while the subsequent Uruguay Round concluded with the negotiation of the WTO Agreements. Id. at 13 tbl.1-2. In the early 1990s, nations took two dra­matic steps to increase economic integration. First, they created the WTO. The WTO increased economic integration by expanding trade rules to cover trade in services and intellectual property, elaborating rules on nontariff barriers, creating a functioning dispute resolution sys­tem and ultimately paving the way for major economies such as China to join. 35 See id.; see also Marrakesh Agreement Establishing the World Trade Organization, Apr. 15, 1994, 1867 U.N.T.S. 154. Second, nations began to create free trade agreements (FTAs) and customs unions, both of which require countries to abolish “substantially all” trade barriers between members. 36 General Agreement on Tariffs and Trade art. XXIV.8, Oct. 30, 1947, 61 Stat. A-11, 55 U.N.T.S. 194 [hereinafter GATT]. Notable FTAs and customs unions created in the early 1990s include NAFTA; 37 North American Free Trade Agreement, supra note 5. the Treaty of Asunción, which created Mercosur in South America; 38 Treaty Establishing a Common Market art. 1, Mar. 26, 1991, 30 I.L.M. 1041. and the Treaty of Maastricht, which deepened European integration to form the modern European Union. 39 Treaty on European Union, Feb. 7, 1992, 1992 O.J. (C 191) 1.

This dramatic removal of trade barriers—especially between devel­oped and developing countries—allowed companies to locate their pro­duction facilities in countries in which they could most cheaply pro­duce their goods. 40 Transportation costs are also a constraint on offshoring, although a declining one and not necessarily meaningful between neighbors like the United States and Mexico. Robert C. Feenstra, Integration of Trade and Disintegration of Production in the Global Economy, 12 J. Econ. Persp. 31, 34 (1998) (estimating how much trade growth stems from trade liberalization versus falling transportation costs). Government policies, in turn, can be the key determinant of a firm’s costs of production. Environmental regulation, labor laws, and taxes all increase production costs.

In the presence of relatively high trade barriers, countries did not necessarily need to compete over these domestic policies. High tariffs, for example, may prevent a company from moving its production over­seas, even if production costs are lower there by virtue of different gov­ernment policies. 41 See Rodrik, supra note 33, at 20 (noting that “government-imposed tariffs . . . constitute clear transaction costs on international exchanges”). The removal of trade barriers, however, incentivizes countries to reduce their domestic regulatory standards in order to at­tract business. 42 See Lawrence Summers, A Strategy to Promote Healthy Globalization, Fin. Times (May 4, 2008), (on file with the Columbia Law Review) (arguing that “the mantra of needing to be ‘internationally competitive’ has been invoked too often as a reason to cut back on regulation” in the financial and labor sectors).

Partially to counteract these incentives, governments have agreed on minimum standards in a wide range of areas. For example, nations have adopted a range of international labor standards in treaties negotiated under the auspices of the International Labour Organization (ILO). 43 See Labour Standards, Int’l Labour Org.,–en/index.htm [] (last visited Oct. 10, 2017) (describing the ILO’s instruments and associated labor standards). They have also agreed to a variety of environmental treaties, such as the Montreal Protocol on Substances that Deplete the Ozone Layer. 44 Montreal Protocol on Substances that Deplete the Ozone Layer, Sept. 16, 1987, 1522 U.N.T.S. 3. Institu­tions like the Basel Committee have sought to impose common standards in financial regulation. 45 See Chris Brummer, Minilateralism: How Trade Alliances, Soft Law, and Financial Engineering Are Redefining Economic Statecraft 99–100 (2014).

Yet the mere existence of these international standards, or even their commitment to domestic law, does not ensure enforcement. Each time the United States has negotiated a major free trade agreement with a developing country, environmental and labor interests have expressed concern that the country will not effectively enforce its laws, preferring the trade advantage that comes with lower de facto standards. For instance, when he first sought authority to negotiate NAFTA, President George H.W. Bush told Congress that “Mexico’s labor standards are comparable to those in the United States . . . .” 46 Jerome I. Levinson, World Policy Inst., Unrequited Toil: Denial of Labor Rights in Mexico and Implications for NAFTA 2 (1993). During the 1992 presidential campaign, however, then-Governor Bill Clinton charged that NAFTA did “nothing to reaffirm our right to insist that the Mexicans fol­low their own labor standards, now frequently violated.” 47 Jerome I. Levinson, Econ. Policy Inst., The Labor Side Accord to the North American Free Trade Agreement: An Endorsement of Abuse of Worker Rights in Mexico 1 (1993), [] (internal quotation marks omitted). Fair environ­mental and labor standards meant not only the law on the books, but also whether and how that law was enforced.

After he became President, Clinton took steps to address this form of selective enforcement. He negotiated what is known as the NAFTA Side Agreements (or, more formally, the North American Agreements on Environmental and Labor Cooperation) to address these concerns. 48 NAFTA Environmental Side Agreement, supra note 8; NAFTA Labor Side Agreement, supra note 8. The agreements require that NAFTA parties “effectively enforce [domes­tic] environmental laws and regulations through appropriate govern­mental action” 49 NAFTA Environmental Side Agreement, supra note 8, art. 5. and “promote compliance with and effectively enforce [domestic] labor law,” 50 NAFTA Labor Side Agreement, supra note 8, art. 3. respectively. These commitments re­main, however, subject to ordinary exercises of prosecutorial discretion in the face of limited resources. 51 See NAFTA Environmental Side Agreement, supra note 8, art. 45 (“A Party has not failed to ‘effectively enforce its environmental law’ . . . where the action or inaction . . . by agencies or officials . . . reflects a reasonable exercise of their discretion.”); NAFTA Labor Side Agreement, supra note 8, art. 49 (“A Party has not failed to ‘effectively enforce its . . . labor standards’ . . . where the action or inaction by agencies or officials . . . reflects reasonable exercise of . . . discretion.”). Partially as a consequence, concerns about selec­tive enforcement of environmental and labor laws have per­sisted to the current day. During negotiations on the Trans-Pacific Partnership, for instance, critics expressed doubts that countries like Vietnam or Mexico would comply with the heightened environmental and labor standards contained in that agreement. 52 Alana Semuels, The TPP’s Uneven Attempt at Labor Protection, Atlantic (Jan. 22, 2016), [] (citing John Sifton, Human Rights Watch’s Asia advocacy director, as saying “[c]ountries such as Vietnam would have to completely revolutionize their legal systems to comply with the labor-union requirements, which doesn’t seem likely”).

B. Selective Enforcement Against Weak Countries

Over the last ten or so years, some scholars have suggested that international trade law suffers from a second kind of selective enforce­ment. They have argued that trade rules tend to be enforced more often than one would expect against weak countries (those with relatively small economies) and less often than one would expect against powerful countries (those with major economies). 53 See, e.g., Chad P. Bown, Participation in WTO Dispute Settlement: Complainants, Interested Parties, and Free Riders, 19 World Bank Econ. Rev. 287, 287–88 (2005) [hereinafter Bown, WTO Dispute Settlement] (showing that a country’s retaliatory and local capacities affect whether it chooses to settle trade disputes); Chad P. Bown, Trade Disputes and the Implementation of Protection Under the GATT: An Empirical Assessment, 62 J. Int’l Econ. 263, 292 (2004) (finding that GATT parties are more likely to implement unlawful protectionist policies when the target country lacks the capacity to retaliate). But see Andrew T. Guzman & Beth A. Simmons, Power Plays and Capacity Constraints: The Selection of Defendants in World Trade Organization Disputes, 34 J. Leg. Stud. 557, 558–59 (2005) (finding that weak countries do not avoid bringing WTO disputes against powerful countries but are constrained in the number of cases they can file); Henrik Horn, Petros C. Mavroidis & Håkan Nordström, Is the Use of the WTO Dispute Settlement System Biased? 2 (Ctr. for Econ. Policy Research, Discussion Paper No. 2340, 1999), (on file with the Columbia Law Review) (arguing that the evidence does not support a conclusion of bias in the WTO dispute settlement system). To understand the root of this concern, consider how the WTO structures dispute settlement. Neither the WTO nor any other centralized body enforces WTO rules against member states. Instead, under WTO rules, states initiate trade disputes against each other. 54 Understanding on Rules and Procedures Governing the Settlement of Disputes art. 6.1, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 2, 1869 U.N.T.S. 401 [hereinafter DSU] (“If the complaining party so requests, a panel shall be established . . . .”). In this sense, states act as private attorneys general, bringing actions on their own to enforce obligations created by treaty. 55 Alternatively, one can think of the WTO treaties as contracts among nations, which bring private actions to enforce their contractual rights. But if a complainant prevails on its claim, it does not win compensatory damages. 56 See DSU, supra note 54, art. 22.1 (providing that “[c]ompensation is voluntary”). Instead, it wins the right to impose retaliatory sanctions if the losing party does not comply with its trade obligations. 57 See DSU, supra note 54, art. 22.2 (“[W]ithin 20 days after the date of expiry of the reasonable period of time, any party having invoked the dispute settlement procedures may request authorization from the [Dispute Settlement Body] to suspend the application to the Member concerned of concessions or other obligations under the covered agreements.”). The lack of a centralized enforcement system means that the WTO regime—a body of multilateral, public laws—is enforced as a series of bilateral arrangements through the use of self-help.

Given this arrangement, weak countries might reasonably fear that they will not be able to enforce a victory because any retaliatory trade measures they are able to impose will have a negligible effect on the more powerful state. For instance, in 2005, Antigua won a WTO dispute against the United States after the latter banned online gambling services from overseas. 58 See Appellate Body Report, United States—Measures Affecting the Cross-Border Supply of Gambling and Betting Services, para. 373, WTO Doc. WT/DS285/AB/R (adopted Apr. 7, 2005) (finding that U.S. laws prohibiting internet gambling while permitting in-person gambling violate the General Agreement on Trade in Services). To date, the United States has not complied with the ruling. 59 See Panel Report, United States—Measures Affecting the Cross-Border Supply of Gambling and Betting Services, Recourse to Article 21.5 of the DSU by Antigua and Barbuda, para. 7.1, WTO Doc. WT/DS285/RW (Mar. 30, 2007) (“[T]he Panel concludes that the United States has failed to comply with the recommendations and rulings of the [Dispute Settlement Body] in this dispute.”). Antigua, for its part, is not a major destination for U.S. products or services and can do little to punish the United States for its continued failure to comply with the law.

Weak states might also lack the capacity to bring WTO disputes in the first place. 60 See Marc L. Busch, Eric Reinhardt & Gregory Shaffer, Does Legal Capacity Matter? A Survey of WTO Members, 8 World Trade Rev. 559, 563–66 (2009) (surveying the staffing and resource decisions of WTO members to assess their capacity to participate in WTO dispute settlement); see also Bown, WTO Dispute Settlement, supra note 53, at 287–88; Guzman & Simmons, supra note 53, at 558–59. They may not have, for example, a large governmental legal office specializing in international trade, as the United States or the European Union has. 61 Cf. Guzman & Simmons, supra note 53, at 566 (“States that commit more people to trade issues and have more qualified individuals working in the area, more mature and sophisticated institutions to handle trade matters, and more financial resources to address trade disputes are higher capacity states.”); Horn, Mavroidis & Nordström, supra note 53, at 2 (“[W]e find no evidence that large countries target small countries dis­proportionately, or that small countries bring fewer complaints against large countries than expected.”). They also may not be able to afford the substan­tial financial costs of bringing a WTO dispute, or the political and diplomatic costs of dealing with an upset trading partner. 62 See Bown, WTO Dispute Settlement, supra note 53, at 308 (presenting empirical evidence that a country is less likely to participate in WTO litigation “if it has inadequate power for trade retaliation, if it is poor and does not have the capacity to absorb substantial legal costs, [or] if it is particularly reliant on the respondent country”).

Such a pattern of enforcement raises concerns about fairness and distributive justice. Weak defendants may be targeted for enforcement because they lack counsel, the resources to obtain it, or the ability to defend themselves. Powerful states, for their part, avoid being named as defendants because prevailing against them is expensive and sanctioning them to induce compliance is even more costly. Thus, the legal system does not treat like conduct alike. And the resulting case law reflects the interests of strong, repeat players, not the interests of weak, infrequent litigants. 63 See Catherine Albiston, The Rule of Law and the Litigation Process: The Paradox of Losing by Winning, 33 Law & Soc’y Rev. 869, 869 (1999) (expanding on the idea that “repeat players influence the development of law by settling cases they are likely to lose and litigating cases they are likely to win”).

Just as with selective enforcement of national environmental and labor laws, governments (and the WTO itself) have taken steps to ame­liorate the effects of selective enforcement against weak states. These steps have primarily focused on capacity building so that developing states in particular can draw on legal expertise to help them defend or even bring WTO disputes. For instance, WTO rules require the WTO Secretariat to conduct training for governments on how the dispute reso­lution process works and to provide a legal expert to developing coun­tries to assist them. 64 DSU, supra note 54, art. 27.2–27.3. Perhaps most importantly, a number of WTO mem­bers created the Advisory Centre on WTO Law, which is tasked with “provid[ing] legal training, support and advice on WTO law and dispute settlement procedures to developing countries . . . and to countries with economies in transition.” 65 Agreement Establishing the Advisory Centre on WTO Law art. 2, opened for signature Nov. 30, 1999, 2299 U.N.T.S. 249 (entered into force July 15, 2001).

*   *   *

Selective enforcement has thus been a persistent concern for gov­ernments and trade lawyers. Nations have repeatedly come together to address selective enforcement that undermines the fairness and legiti­macy of trade law. They have renegotiated trade agreements like NAFTA to respond to selective-enforcement concerns and devoted resources to ameliorating selective enforcement’s effects. Yet governmental efforts have fallen short, in part because of a blind spot about other kinds of rampant selective enforcement.

II. Selective Enforcement of Trade Laws

This Part provides empirical evidence of a heretofore unremarked-upon kind of selective enforcement in trade law: selective enforcement that disadvantages environmentally beneficial products. This is the reverse of the usual concern that environmental and labor laws will be selectively enforced to gain a trade advantage. It raises, though, the same question of fairness: Is trade law in its current form undermining im­portant nontrade values? This Part examines cases challenging gov­ern­ment financial support and “unfair” trade practices in two sectors—energy and fisheries. In both sectors, trade law is actively used to regulate government support for natural resource substitutes (renewable energy and farmed fish) but rarely, if ever, used to challenge government sup­port for production and consumption of the natural resources themselves.

These two sectors are two of the most important to the global economy, with fuels constituting the most traded commodity in the world and fish the most traded food commodity. 66 WTO, International Trade Statistics 2015, at 72 (2015),
english/res_e/statis_e/its2015_e/its2015_e.pdf [] (noting that fuel is the highest traded major commodity with 16.6% of global exports); Martin D. Smith et al., Sustainability and Global Seafood, 327 Science 784, 784 (2010) (noting that “[s]eafood is . . . the most highly traded food commodity internationally”).
The existence of selective enforcement in these two sectors is therefore important in and of itself. It also, however, suggests a broader pattern in which trade enforcement systematically disadvantages products that compete with established natural resource–intensive industries.

A. Selective Enforcement in the Energy Sector

Energy production can be broken down into two types, broadly speaking: (1) energy produced from fossil fuels, such as oil, gas, and coal; and (2) renewable energy, including solar, wind, and biofuels. 67 Renewable energy also includes more established forms, such as hydroelectric (and arguably nuclear) generation, although renewable energy disputes have not focused on these sources. The con­sumption of fossil fuels to produce energy releases greenhouse gases linked to climate change. 68 Intergovernmental Panel on Climate Change, Climate Change 2014 Synthesis Report: Summary for Policymakers 5 (2014), [] (“Emissions of CO2 from fossil fuel combustion and industrial processes contributed about 78% of the total [greenhouse gas] emissions increase from 1970 to 2010 . . . .”). The threat posed by climate change has made reducing greenhouse gas emissions a central objective of both gov­ern­ments and the private sector. 69 See Paris Agreement on Climate Change, supra note 18, at 1–2; Michael P. Vandenbergh, Private Environmental Governance, 99 Cornell L. Rev. 129, 140–41 (2013) (introducing the scheme of public and private governance dedicated to environmental protection, including greenhouse gas reduction). Renewable energy furthers this objec­tive through low- or zero-emission energy production.

Both fossil fuels and renewable energy benefit from government support. In 2014, governments subsidized fossil fuels to the tune of approximately $934 billion. 70 See infra section IV.A; infra notes 284–290. Renewables received only $135 billion. 71 See infra section IV.A; infra notes 291–292. In recent years, governments have enforced trade law vigorously against re­newable energy subsidies. No government has ever, however, brought a direct enforcement action challenging another’s support for fossil fuels and its effects on pricing. 72 See Dispute Settlement Gateway, WTO,
dispu_e/dispu_e.htm [] [hereinafter WTO, Dispute Settlement Gateway] (last visited Oct. 21, 2017) (listing and providing information on each WTO dispute); see also De Bièvre, Espa & Poletti, supra note 16, at 412–13 (discussing and analyzing the lack of trade cases challenging government financial support for fossil fuels); Meyer, Explaining Energy Disputes, supra note 16, at 394–97 (same). The WTO collects and publishes information on its cases and their subject matter, making it relatively simple to determine that no WTO cases have involved direct challenges to government financial support for fossil fuels. Similarly, governments report information on domestic trade remedies investigations, which can be searched for evidence of direct challenges to fossil fuel subsidies.

Since 2008, governments have initiated at least 25 challenges, either directly before the WTO or through domestic mechanisms such as trade remedy investigations, to other nations’ support for renewable energy. 73 See infra Table 1. Table 1 provides an overview of trade disputes involving government support for renewable energy.

Table 1: Renewable Energy Trade Disputes

Date 74 The date listed refers to the date a government initiated a dispute. For WTO disputes, this means the date on which a party filed a request for consultations. For domestic disputes, this means the date on which a government initiated an investigation. Hence, Table 1 understates the number of domestic trade remedies complaints filed by including only those that result in an investigation. Moreover, Table 1 lists all WTO complaints but lists only the first domestic investigation if subsequent investigations appear linked, rather than including each subsequent investigation or review. WTO trade remedy disputes, of course, require a domestic trade remedy dispute in the first instance. See infra text accompanying notes 115–116 (explaining the “trade remedies” governments impose upon others under national law and the procedure under which nations may challenge these trade remedies before the WTO). Nonetheless, Table 1 lists only the WTO dispute in such instances. Thus, for WTO cases the initial imposition of trade remedies will have been earlier. Type of Dispute (Legal Claim) 75 Because antidumping and countervailing-duty investigations often overlap, Table 1 uses “AD/CVD” as shorthand for a domestic trade remedy case, although not all trade remedy disputes listed included both antidumping and countervailing-duty investigations. “AD” refers to an antidumping investigation or claim; “CVD” to a countervailing-duty investigation or claim; “SCM” to a claim brought directly under the Agreement on Subsidies and Countervailing Measures; and “NT” to a national treatment claim brought under GATT Article III or TRIMs Article 2. For an explanation of these terms, see infra notes 102–107, 115–124 and accompanying text. Complainant 76 “Complainant” and “Respondent” refer to the countries involved. In trade remedy investigations, however, the actual petitioner and respondent will be private parties from the named countries. Note that this means that, for WTO cases, the respondent is the party imposing duties. Respondent 77 See supra note 76. Industry/ Program Targeted
June 2008 78 See Investigations: Initial Investigation AD531, European Comm’n, [] (last updated Oct. 10, 2017). Domestic (AD/CVD) European Union United States Biodiesel
March 2009 79 Peru: Definitive Antidumping Duty on Imports of Biodiesel from the United States of America, Glob. Trade Alert (Mar. 26, 2009), []; Peru: Extension of Definitive Countervailing Duty on Imports of Biodiesel from the United States of America, Glob. Trade Alert (Mar. 26, 2009), []. Domestic (AD/CVD) Peru United States Biodiesel
June 2010 80 Michael Darby, U.S. Dep’t of Agric. Foreign Agric. Serv., Australia: Biofuels Annual (2010),
Canberra_Australia_07-06-2010.pdf [].
Domestic (AD/CVD) Australia United States Biodiesel
August 2010 81 Commission Regulation 720/2010, 2010 O.J. (L 211) 1, 1 (EU). Domestic (AD/CVD) European Union Canada, Singapore Biodiesel (from U.S.)
September 2010 82 Request for Consultations by the European Union, Canada—Measures Relating to the Feed-in Tariff Program, WTO Doc. WT/DS426/1 (Aug. 16, 2011); Request for Consultations by Japan, Canada—Certain Measures Affecting the Renewable Energy Generation Sector, WTO Doc. WT/DS412/1 (Sept. 16, 2010). WTO


European Union,


Canada Ontario Province’s Feed-in Tariff
November 2011 83 James O’Toole, U.S. to Investigate Chinese Solar ‘Dumping’ Claims, CNN Money (Nov. 9, 2011),
index.htm [].
Domestic (AD/CVD) United States China Solar Panels
November 2011 84 Joanna I. Lewis, The Rise of Renewable Energy Protectionism: Emerging Trade Conflicts and Implications for Low Carbon Development, 14 Global Envtl. Pol., no. 4, 2014, at 17, 22 tbl.2. Domestic Investigation (NT) China United States State-level Renewable Energy Support Programs
November 2011 85 See Investigations: Initial Investigation AS581, European Comm’n, http:// [] (last updated Oct. 30, 2017).
Domestic (AD/CVD) European Union United States Ethanol
January 2012 86 U.S. Int’l Trade Comm’n, Utility Scale Wind Towers from China and Vietnam 1 (2012), []. Domestic (AD/CVD) United States China, Vietnam Wind Components
July 2012 87 White & Case LLP, General Trade Report—JETRO 4 (2012),
ext_images/theme/wto-fta/news/pdf/w_c_monthly_report-2012_November.pdf [http://].
Domestic  (AD/CVD) China United States, South Korea, European Union Polysilicon
September 2012 88 See Investigations: Initial Investigation AD590, European Comm’n, http:// [] (last updated Oct. 11, 2017).
Domestic (AD/CVD) European Union China Solar Panels
September 2012 89 Request for Consultations by China, United States—Countervailing and Anti-Dumping Measures on Certain Products from China, WTO Doc. WT/DS449/1 (Sept. 20, 2012). WTO


China United States Wind Components (among other products)
November 2012 90 Marc Roca, India Plans Levy on US, Malaysia, and Taiwan Solar Panels, Livemint (May 23, 2014), [] (last updated May 24, 2014). Domestic (AD/CVD) India China, Taiwan, Malaysia, United States Solar Panels
November 2012 91 Request for Consultations by China, European Union and Certain Member States—Certain Measures Affecting the Renewable Energy Generation Sector, WTO Doc. WT/DS452/1 (Nov. 7, 2012). WTO


China European Union, Greece, Italy Certain EU Member States’ Feed-in Tariffs
May 2013 92 Request for Consultations by Argentina, European Union and Certain Member States—Certain Measures on the Importation and Marketing of Biodiesel and Measures Supporting the Biodiesel Industry, WTO Doc. WT/DS459/1 (May 23, 2013). WTO


Argentina European Union Biodiesel
February 2013 93 Request for Consultations by the United States, India—Certain Measures Relating to Solar Cells and Solar Modules, WTO Doc. WT/DS456/1 (Feb. 11, 2013). WTO


United States India India’s National Solar Mission
December 2013 94 Request for Consultations by Argentina, European Union—Anti-Dumping Measures on Biodiesel from Argentina, WTO Doc. WT/DS473/1 (Jan. 8, 2014). WTO


Argentina European Union Biodiesel
June 2014 95 Request for Consultations by Indonesia, European Union—Anti-Dumping Measures on Biodiesel from Indonesia, WTO Doc. WT/DS480/1 (June 17, 2014). WTO


Indonesia European Union Biodiesel
December 2014 96 Archived — Statement of Reasons: Concerning the Initiation of Investigations into the Dumping and Subsidizing of Certain Photovoltaic Modules and Laminates Originating in or Exported from the People’s Republic of China, Can. Border Servs. Agency (Dec. 19, 2014), [
RK7E-HA22] (last updated July 30, 2015).
Domestic (AD/CVD) Canada China Solar
April 2015 97 Peru: Definitive Antidumping Duty on Imports of Biodiesel B100 from Argentina, Glob. Trade Alert (Oct. 26, 2016), []. Domestic (AD/CVD) Peru Argentina Biodiesel
January 2016 98 Erin Voegele, China Takes Action Against US DDGS, Ethanol Imports, Ethanol Producer Mag. (Jan. 11, 2017), []. Domestic


China United States Biofuels
September 2016 99 Request for Consultations by India, United States—Certain Measures Relating to the Renewable Energy Sector, WTO Doc. WT/DS510/1 (Sept. 19, 2016). WTO


India United States Subnational Renewable Energy Measures
April 2017 100 Fact Sheet: Commerce Preliminary Finds Countervailable Subsidization of Imports of Biodiesel from Argentina and Indonesia, Int’l Trade Admin., http:// [] (last visited Oct. 10, 2017).
Domestic (AD/CVD) United States Argentina, Indonesia Biodiesel
February 2018 101 Brian Bradley, EU, China Request WTO Consultations Over U.S. Safeguards on Solar Cells, American Shipper (Feb. 8, 2018),
news/eu-china-request-wto-consultations-over-us-safegua-70552.aspx?source=Little4 [http://].
WTO (Safeguards) China, European Union United States Solar Cells


Both international and national trade rules apply to government financial support for products. Internationally, this Article focuses on WTO rules, which have been applied to government support for energy in several ways. The most direct way is through the WTO’s Agreement on Subsidies and Countervailing Measures (SCM Agreement), which allows nations to challenge discriminatory or injurious subsidies. 102 SCM Agreement, supra note 10, art. 4. The SCM Agreement’s rules, however, are somewhat technical and, in practice, impose a difficult evidentiary burden on complaining states. 103 In order to obtain relief under the SCM Agreement, a complainant must first show that there has been a subsidy within the meaning of the SCM Agreement. See id. arts. 1, 4.2, 7.2, 11.2 (defining a subsidy and requiring proof of a subsidy’s existence to obtain relief and initiate investigations under the GATT). Such a showing requires (1) a financial contribution—broadly defined to include, inter alia, tax credits and the transfer of goods—by the respondent government (2) that has conferred a benefit on the recipient, meaning that the government made its financial contribution on better terms than those available in the marketplace. See id. art. 1.1. Therefore, in a subsidies case before the WTO’s Dispute Settlement Body (DSB), the complaining party bears the burden of demonstrating the existence of a subsidy. Demonstrating a benefit can be difficult in markets with significant government intervention, such as energy markets, because no free market comparison exists.
Moreover, after demonstrating the existence of a subsidy, the complainant still must show that the subsidy either is “specific” to an industry or enterprise or is otherwise prohibited. See id. arts. 1.2, 2, 8.1 (subjecting a subsidy to the SCM Agreement’s prohibitions and liabilities only if it is “specific,” defining when a subsidy is “specific,” and explicitly rendering non-specific subsidies “non-actionable”). General subsidies, such as those available to consumers or through generally applicable provisions of the tax code, do not count. Id. art. 2.2. The SCM Agreement also prohibits two narrow classes of discriminatory subsidies: (1) those contingent on export performance and (2) those contingent on the use of domestic products. Id. art. 3. Such subsidies are per se specific and the respondent government must remove them, id. art. 2.3, lest the complaining government gain the right to retaliate, id. art. 4.10.
If the SCM Agreement does not outright prohibit a subsidy, however, the complaining party must show that the subsidy causes “adverse effects” before the respondent government is obliged to remove the offending subsidy. See id. arts. 5, 7.2 (defining adverse effects and exemplifying the showing of injury a member must make to request consultations under the SCM Agreement). In this way, subsidies cases differ from other kinds of trade cases. In the run-of-the-mill WTO dispute, “nullification or impairment” (the WTO term for damage) is presumed. DSU, supra note 54, art. 3.8. In most subsidies cases, however, the complainant must demonstrate harm before the obligation to remove the subsidy applies. This additional showing makes it much more difficult for the complaining party to prevail in an SCM dispute.
The diffi­culty of succeeding on an SCM claim pushes governments to chal­lenge other governments’ support for industry through generally applicable GATT rules. 104 Programs challenged in this way are often subsidies in a colloquial sense but may not constitute “subsidies” within the meaning of the SCM Agreement. For this reason, this Article uses the term “government support” to refer to the broader class of measures that governments adopt that provide financial assistance to their domestic industries. For example, discriminatory government-support measures may be more easily challenged under Article III of the GATT, which creates a nondiscrimination rule known as the national treatment (NT) obligation. 105 The Agreement on Trade-Related Investment Matters (TRIMs Agreement) incorporates the national treatment obligation, and it holds that governments may not treat foreign investors less favorably than similarly situated domestic investors. See Agreement on Trade-Related Investment Measures art. 2, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1868 U.N.T.S. 186 (prohibiting trade-related investment matters that are “inconsistent with the provisions of Article III”). Furthermore, although the SCM Agreement prohibits discriminatory subsidies, SCM Agreement, supra note 10, art. 3, challenging a discriminatory government support program under the national treatment rule avoids the need to prove that a “subsidy” within the meaning of the SCM Agreement exists in the first place, see id. arts. 1–2. GATT Article III provides that a nation may not treat foreign goods less favorably than it treats its own “like” prod­ucts. 106 GATT, supra note 36, art. III. Local content requirements—rules that condition a benefit, such as a subsidy, on use of locally produced materials or equipment—violate the NT rule. 107 Holger P. Hestermeyer & Laura Nielsen, The Legality of Local Content Measures Under WTO Law, 48 J. World Trade 553, 566 (2014).

Six disputes have challenged government support for renewable energy directly before the WTO on SCM or NT grounds. 108 A seventh dispute between Argentina and the European Union involves a local content requirement on biofuels in Spain. This dispute, however, does not appear to involve financial support, so it is omitted. See Request for Consultations by Argentina, European Union and a Member State—Certain Measures Concerning the Importation of Biodiesels, WTO Doc. WT/DS443/1 (Aug. 17, 2012) (requesting consultations over an order providing for computation of biofuel targets based only on biodiesel produced in EU member states). In the most important such case, the WTO Appellate Body upheld a finding that Ontario (and therefore Canada) violated the NT obligation in its Feed-in Tariff Program. 109 Canada—Feed-in Tariff Program, supra note 14, para. 5.85 (finding that Canada’s local content requirement (LCR) programs—“Minimum Required Domestic Content Levels”—violate Article III, section 4 of the GATT 1994 standards). Under that program, electricity producers qualified for preferential rates from the government if they produced a certain amount of their electricity from renewable sources, provided that the equipment used to generate the renewable energy was manufactured in Ontario. 110 See id. paras. 1.1–1.4 (describing the Ontario Feed-in Tariff Program and its “Minimum Required Domestic Content Levels”). The WTO’s Dispute Settlement Body (DSB) found that such a local content requirement violates the NT obligation by disadvantaging foreign products that compete with the locally produced goods. 111 See id. para. 5.85. In 2016, the WTO Appellate Body upheld a similar finding about a local content requirement in India’s national solar support program in a case brought by the United States. 112 Appellate Body Report, India—Certain Measures Relating to Solar Cells and Solar Modules, para. 6.2.c, WTO Doc. WT/DS456/AB/R (adopted Oct. 14, 2016) [hereinafter India—Solar Cells and Modules]. The United States also challenged an allegedly discriminatory wind subsidy in China, 113 Request for Consultations by the United States, China—Measures Concerning Wind Power Equipment, WTO Doc. WT/DS419/1 (Jan. 6, 2011). although China agreed to remove the subsidy without further proceedings. 114 Doug Palmer & Leonora Walet, China Agrees to Halt Subsidies to Wind Power Firms, Reuters (June 7, 2011), (on file with the Columbia Law Review).

National trade laws also allow governments to respond to “unfair” trading practices of other countries, most notably government support. 115 See Judith Goldstein, Ideas, Institutions, and American Trade Policy, 42 Int’l Org. 179, 197–98 (1988) (describing the rise of “fair trade” laws in the United States). These laws are known as “trade remedies.” Although trade remedies are imposed under national laws and do not require the WTO’s permission, the WTO has rules on their use. 116 See Request for Consultations by the United States, China—Measures Concerning Wind Power Equipment, supra note 113 (alleging procedural violations of the GATT and SCM Agreement for, inter alia, failing to properly notice various trade measures). Hence, countries can challenge an­other nation’s imposition of trade remedies before the WTO.

The two most relevant kinds of trade remedies are (1) countervail­ing duties and (2) antidumping duties. Countervailing duties offset the effects of subsidies by another government. 117 GATT, supra note 36, art. VI.3. (“No countervailing duty shall be levied on any product of the territory of any contracting party imported into the territory of another contracting party in excess of an amount equal to the estimated bounty or subsidy determined to have been granted . . . .”). Countervailing duties thus attack the same problem as the SCM Agreement (and, in fact, the SCM Agreement contains rules on countervailing duties). 118 See SCM Agreement, supra note 10, pt. V (discussing countervailing measures). A government seeking to respond to a subsidy can thus either bring a WTO case directly under the SCM Agreement, or impose countervailing duties. Imposing countervailing duties requires findings similar to those necessary to make out an SCM claim, including the existence of a subsidy within the mean­ing of WTO rules. 119 Specifically, the domestic investigating authority must find (1) a subsidy, (2) injury to domestic industry, and (3) causation. Id. art. 11.2. Significantly, however, those find­ings are made by the national government imposing the duties, rather than by a neutral international tribunal. 120 See id. pt. V (describing the rules for national investigating authorities assessing whether to impose countervailing duties).

Antidumping duties are more flexible than countervailing duties and can also respond to government subsidization. Dumping—the trig­ger, unsurprisingly, for the imposition of antidumping duties—involves a producer’s selling its good in the importing country at less than what the importing government considers “normal” value. 121 In order to impose antidumping duties, the domestic authority must find (1) dumping, (2) injury to a domestic industry, and (3) causation. General Agreement on Tariffs and Trade art. VI.1, July 1986, [] [hereinafter GATT 1986]; Protocol Modifying Part II and Article XXVI of the General Agreement on Tariffs and Trade art VI.1, Sept. 14, 1948, 62 U.N.T.S. 80 [hereinafter GATT 1948 Amendments]. Anti­dumping duties therefore target private conduct: the pricing decisions of firms. 122 A government cannot bring a WTO action directly challenging dumping, since WTO rules only discipline government action. Governments have a great deal of flexibility, however, in how they calcu­late normal value. 123 Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 art. 2.2, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1A, 1868 U.N.T.S. 201 (explaining that, generally, “the margin of dumping shall be determined by comparison with a comparable price of the like product when exported to an appropriate third country”). This flexibility allows governments to use antidump­ing duties to respond to prices that are artificially low due to another government’s financial support. Indeed, the original GATT pro­visions on antidumping and countervailing duties recognized that either could be imposed in response to the same underlying set of facts. 124 GATT 1948 Amendments, supra note 121, art. VI.5 (“No product of the territory of any contracting party imported into the territory of any other contracting party shall be subject to both anti-dumping and countervailing duties to compensate for the same situation of dumping or export subsidization.”).

Yet a third kind of trade remedy, rarely used, is that of safeguards. In January 2018, President Trump imposed thirty percent tariffs on solar cells imported into the United States pursuant to a safeguards investiga­tion. Unlike antidumping and countervailing duties, safeguards do not require a finding that another country (or its producers) has behaved unfairly. 125 See GATT, supra note 36, art. XIX. Instead, safeguards focus entirely on the degree of injury to the domestic producer. 126 Id. In February 2018, China and the EU chal­lenged the United States’ imposition of safeguards before the WTO. As this Article goes to print, this case remains in its infancy. 127 Bradley, supra note 101.

Returning to the disputes described in Table 1, eighteen of the re­maining nineteen disputes involve the national application of trade remedies. 128 The eighteenth involved a Chinese domestic investigation of U.S. support for renewable energy in violation of the national treatment rule. See supra Table 1. These disputes target solar, wind, and biofuel products. 129 See supra Table 1. This wide range of products, spanning the renewable energy sector, demonstrates countries’ willingness to use trade law to challenge gov­ernment support for different sources of renewable energy with roots in different areas of the economy. Solar and wind energy, for instance, tend to be manufacturing industries, while the biofuel industry is grounded in agriculture. Only three of these disputes have made it to the WTO. In one of these cases, China prevailed in a challenge to the United States’ imposition of countervailing duties on a number of products, including solar panels and wind turbines. 130 Appellate Body Report, United States—Countervailing Duty Measures on Certain Products from China, para. 5, WTO Doc. WT/DS437/AB/R (adopted Dec. 18, 2014). The essence of China’s claim was that the United States impermissibly counted subsidies from Chinese state-owned enterprises (SOEs). Id. The Appellate Body accepted that the United States had impermissibly used the subsidies provided by Chinese SOEs but declined to rule that WTO members could never rely on such subsidies in a CVD analysis. Id. In the other two disputes, Argentina and Indonesia each challenged the EU’s imposition of antidumping du­ties on biodiesel fuels, which are fuels made from plant and animal fats that emit fewer greenhouse gases than fossil fuels. 131 Appellate Body Report, European Union—Anti-Dumping Measures on Biodiesel from Argentina, paras. 1.3–1.4, WTO Doc. WT/DS473/AB/R (adopted Oct. 6, 2016); Request for Consultations by Indonesia, European Union—Anti-Dumping Measures on Biodiesel from Indonesia, supra note 95, paras. II.1–II.15. These disputes are at the heart of a broader trade war over biofuels that includes not only these three countries, but also Australia, China, Peru, and the United States. 132 See supra Table 1.

In contrast to this robust history of challenges to renewable energy, not a single case has ever been brought before the GATT or WTO directly challenging government support for fossil fuels on either SCM or NT grounds. 133 De Bièvre, Espa & Poletti, supra note 16, at 412–13; Meyer, Explaining Energy Disputes, supra note 16, at 392–93. In terms of trade remedies, the EU imposed antidumping duties on a range of Russian products, such as steel and ammonium ni­trate, partly on the grounds that Russia subsidizes energy consumption. 134 Request for Consultations by the Russian Federation, European Union—Cost Adjustment Methodologies and Certain Anti-Dumping Measures on Imports from Russia (Second Complaint), paras. 3, 5, WTO Doc. WT/DS494/1 (May 19, 2015). Notably, though, these antidumping duties were not imposed directly in response to energy subsidies. Rather, they were imposed on energy-intensive downstream products that presumably benefit from such subsi­dies. 135 See id. (describing subsidies on ammonium nitrate as well as several types of tubing that would conceivably benefit from subsidized energy prices). Russia currently has a challenge to the use of antidumping duties in this way pending at the WTO. 136 Id. paras. 5.1–5.2. Although energy consumption subsidies may have played a role in the imposition of other antidumping duties on energy-intensive products in other instances, my research has not turned up other such instances. Barring a WTO challenge, such instances are difficult to identify precisely because the energy subsidies and energy-related products are not themselves the target of the actions.

Nor does this absence of disputes against fossil fuels stem from any systematic differences between fossil fuel subsidies and renewable energy subsidies. 137 See Meyer, Explaining Energy Disputes, supra note 16, at 399. To be sure, some have alleged that fossil fuel subsidies are systematically different from renewable energy subsidies. This claim rests in part on the existence of local content requirements in renewable energy subsidies and in part on the claim that fossil fuel subsidies are structured in ways that are less amenable to challenge under the SCM Agreement. 138 See De Bièvre, Espa & Poletti, supra note 16, at 417–18. In fact, however, local content requirements are rampant in the fossil fuel sector as well. A 2013 World Bank study identified local content policies supporting the fossil fuel sector in forty-eight nations. 139 Silvana Tordo, Michael Warner, Osmel E. Manzano & Yahya Anouti, World Bank, Local Content Policies in the Oil and Gas Sector xiii, 93–155 (2013) (discussing the overall findings of the study on local content policies); see also Meyer, Explaining Energy Disputes, supra note 16, at 399 (discussing research on local content policies in the fossil fuel sector). Moreover, if renewable energy subsidies were systematically more suscep­tible to challenge under the SCM Agreement, we would expect to see nations regularly relying on the SCM Agreement to bring their renewa­ble energy challenges. Yet as Table 1 attests, the SCM Agreement is used only rarely to challenge renewable energy subsidies. This strongly suggests that the structures of the subsidies are not driving the rate of challenge.

Disputes over government support for the energy sector thus show a clear trend. Governments are willing to use WTO rules to challenge other governments’ financial support for renewable energy, but not for fossil fuels, despite the fact that support for fossil fuels is many times that for renewable energy. 140 A greater amount of the world’s energy still comes from fossil fuels as opposed to renewables. See Energy Info. Admin., Monthly Energy Review: October 2017, at 4 fig.1.2 (2017), [] (showing that coal, crude oil, and natural gas have produced more energy than renewable sources in every year since 1949). Thus, on a per-unit basis, the discrepancy between the amounts of subsidies likely would not be as great. I thank Joe Margolies for this point. Even a smaller discrepancy between the amounts of subsidies, though, cannot account for the complete lack of challenges to fossil fuel subsidies. As Parts III and IV explain, this selective enforce­ment magnifies the discrepancy in subsidization between fossil fuels and renewable energy. In doing so, selective enforcement reinforces the market dominance of increasingly scarce and environmentally harmful fossil fuels.

B. Selective Enforcement in the Fisheries Sector

A remarkably similar story plays out in the world of fish. Seafood is the highest traded food commodity by value in the world. 141 Emiko Terazono, Salmon Leaps Past Shrimp in Global Fish Market, Fin. Times (Jan. 18, 2016), (on file with the Columbia Law Review) [hereinafter Terazono, Salmon Leaps Past Shrimp]. It is central to the livelihood and food security of billions of people; indeed, over three billion people rely on fish as their primary source of protein. 142 Sustainable Seafood: Overview, World Wildlife Fund,
industries/sustainable-seafood [] (last visited Jan. 23, 2018).
Like energy, fish can be divided into natural resources that must be cap­tured (wild fish) and substitute resources that are largely “manufac­tured” (fish produced through aquaculture, also called farmed fish). Fish­ing nations have for years granted huge subsidies to their fishing fleets to capture wild fish. 143 See infra section IV.B. The result has been widespread overfishing, leading to dangerously low stocks of certain breeds of fish. 144 See Margaret A. Young, Trading Fish, Saving Fish: The Interaction Between Regimes in International Law 6 (2011) (discussing the “global fisheries crisis”). Aquaculture seeks to provide an alternative to wild stocks—allowing wild stocks to recover—while also achieving greater efficiency than fishing fleets. 145 To be sure, in practice aquaculture has a mixed environmental record. Investment in aquaculture is ultimately necessary to make it environmentally sustainable. This Article explores this question in more detail in Part IV. Yet just as nations do not enforce trade laws against subsidies for traditional fossil energy, nations do not invoke trade rules to challenge subsidies for wild fishing. Instead, nations—most notably the United States—regularly invoke trade rules to challenge government support for aquaculture.

Just as in the energy sector, no WTO member to date has ever di­rectly challenged another WTO member’s financial support for fisheries before the WTO under either the SCM Agreement or the GATT’s gener­ally applicable measures. 146 See infra Table 2. But while governments are reluctant to bring fisheries support cases directly to the WTO, they have few qualms about challenging such support through national trade remedies. The UN Food and Agriculture Organization (FAO) notes that “[t]he only cases so far in international trade related to subsidies and fish exports stem from aquaculture.” 147 Subsidies and Trade Distortion, UN Food & Agric. Org.,
fishery/topic/12358/en [] (last visited Oct. 10, 2017); see also Frank Asche, Aquaculture: Opportunities and Challenges 12 (2015), http:// [http://] (noting that the “dumping of seafood has been a WTO concern primarily in relation to aquaculture”).
In other words, trade rules on govern­ment support are enforced exclusively against farmed fish.

Nine WTO fisheries disputes challenged the imposition of anti­dumping and countervailing duties. 148 See infra Table 2. A number of other WTO disputes have challenged a range of market-access restrictions, ranging from outright bans to labeling requirements. WTO, Dispute Settlement Gateway, supra note 72. Seven of these disputes challenged the United States’ imposition of trade remedies on shrimp from Asian and South American countries. 149 See infra Table 2. An eighth case challenged the United States’ imposition of duties on Chilean salmon. 150 See infra Table 2. In the lone WTO dispute in which the United States is not the respondent, Norway challenged the European Union’s imposition of duties on salmon. 151 See infra Table 2. At least seven other domestic investigations in the United States and Europe have resulted in the imposition of domestic trade remedies that have not been challenged before the WTO. 152 See infra Table 2.

Table 2: Fisheries Trade Disputes

Date 153 For WTO cases, the date listed refers to the date of the request for consultations. Of course, a WTO case is premised on an underlying domestic investigation, but Table 2 does not list the domestic investigations separately. Thus, for WTO cases, the initial imposition of trade remedies will have been earlier. The exception is for the 2005 imposition of duties by the United States on shrimp from a number of countries. Since not all subject countries brought WTO disputes, Table 2 lists that dispute separately. For domestic cases, the date listed is the date on which the government initiated an investigation. Table 2 lists all WTO complaints but lists only the first domestic investigation, rather than including each subsequent review and reimposition of duties. Type of Dispute (Legal Claim) 154 “AD” refers to an antidumping investigation or claim; “CVD” refers to a countervailing-duty investigation or claim. Complainant 155 “Complainant” and “Respondent” refer to the countries involved. In trade remedy investigations, however, the actual petitioner and respondent will be private parties from the named countries. Note that this means that, for WTO cases, the respondent is the party imposing duties. Respondent 156 See supra note 155. Industry/ Program Targeted
June 1990 157 Fresh and Chilled Atl. Salmon from Nor., Inv. No. 701-TA-302, USITC Pub. 2371 (Apr. 1991) (Final). Domestic (AD/CVD) United States Norway Salmon
September 1996 158 Crawfish Tail Meat from China, Inv. No. 731-TA-752, USITC Pub. 3002 (Nov. 1996) (Preliminary). Domestic (AD/CVD) United States China Crawfish
August 1997 159 Request for Consultations by Chile, United States—Countervailing Duty Investigation of Imports of Salmon from Chile, WTO Doc. WT/DS97/1 (Aug. 12, 1997). WTO


Chile United States Salmon
June 2002 160 Certain Frozen Fish Fillets from Viet., Inv. No. 731-TA-1012, USITC Pub. 3617 (Aug. 2003) (Final). Domestic (AD/CVD) United States Vietnam Frozen Fish Fillets
December 2002 161 Notice of Initiation of an Anti-Dumping Proceeding Concerning Imports of Large Rainbow Trout Originating in Norway and the Faeroe Islands, 2002 O.J. (C 318) 2. Domestic (AD/CVD) European Union Norway/Faroe Islands Trout
December 2003 162 Certain Frozen or Canned Warmwater Shrimp and Pawns [sic] from Braz., China, Equador [sic], India, Thai. & Viet., Inv. Nos. 701-TA-1063-1068, USITC Pub. 3748 (Jan. 2005) (Final) [hereinafter Warmwater Shrimp Investigation]. Domestic (AD/CVD) United States Brazil, China, Ecuador, India, Thailand, Vietnam Shrimp
December 2004 163 Request for Consultations by Thailand, United States—Provisional Anti-Dumping Measures on Shrimp from Thailand, WTO Doc. WT/DS324/1 (Dec. 14, 2004). WTO


Thailand United States Shrimp
November 2005 164 Request for Consultations by Ecuador, United States—Anti-Dumping Measure on Shrimp from Ecuador, WTO Doc. WT/DS335/1 (Nov. 21, 2005). WTO


Ecuador United States Shrimp
March 2006 165 Request for Consultations by Norway, European Communities—Anti-Dumping Measure on Farmed Salmon from Norway, WTO Doc. WT/DS337/1 (Mar. 22, 2006). WTO


Norway European Union Salmon
April 2006 166 Request for Consultations by Thailand, United States—Anti-Dumping Measures on Shrimp from Thailand, WTO Doc. WT/DS343/1 (Apr. 27, 2006). WTO


Thailand United States Shrimp
June 2006 167 Request for Consultations by India, United States—Customs Bond Directive for Merchandise Subject to Anti-Dumping/Countervailing Duties, WTO Doc. WT/DS345/1 (June 12, 2006). WTO


India United States Shrimp
February 2010 168 Request for Consultations by Viet Nam, United States—Anti-Dumping Measures on Certain Shrimp from Viet Nam, WTO Doc. WT/DS404/1 (Feb. 4, 2010). WTO


Vietnam United States Shrimp
February 2011 169 Request for Consultations by China, United States—Anti-Dumping Measures on Certain Frozen Warmwater Shrimp from China, WTO Doc. WT/DS422/1 (Mar. 2, 2011). WTO


China United States Shrimp/ Diamond Sawblades
February 2012 170 Request for Consultations by Viet Nam, United States—Anti-Dumping Measures on Certain Shrimp from Viet Nam, WTO Doc. WT/DS429/1 (Feb. 27, 2012). WTO


Vietnam United States Shrimp
February 2014 171 Notice of Initiation of an Anti-Subsidy Proceeding Concerning Imports of Certain Rainbow Trout Originating in Turkey, 2014 O.J. (C 44) 9. Domestic (AD/CVD) European Union Turkey Trout
August 2015 172 Notice of Initiation of an Anti-Subsidy Proceeding Concerning Imports of European Sea Bass and Gilthead Sea Bream Originating in Turkey, 2015 O.J. (C 266) 4 [hereinafter Turkish Bass and Bream Notice]. Domestic (AD/CVD) European Union Turkey Sea Bass/ Sea Bream


Sometimes, trading partners employ trade remedies explicitly against farmed fish. In 2006, the EU imposed antidumping duties on farmed salmon from Norway. 173 Panel Report, European Communities—Anti-Dumping Measure on Farmed Salmon from Norway, para. 2.4, WTO Doc. WT/DS337/R (adopted Nov. 17, 2007) [hereinafter EC–Salmon—Norway]. In announcing the antidumping duties, the European Commission noted that “Norway decided in the early 1990s that, like oil, farmed fish is of strategic economic importance and this sector received considerable financial, organisational, political and research support from the Norwegian state.” 174 Press Release, European Comm’n, Norwegian Salmon (Feb. 21, 2006), http:// [] (emphasis added).
As a result of this subsidization, both the United States and the EU imposed trade reme­dies in the 1990s against Norwegian salmon. 175 See Mark T. Peterson, U.S. Trade Law and Imported Farmed Atlantic Salmon: Protectionism or Protection of Free Trade, 2 Ocean & Coastal L.J. 33, 35–37 (1996) (describing the case leading to the imposition of antidumping and countervailing duties against Norwegian salmon); Press Release, European Comm’n,  Commission Agrees Measures to Curb Norwegian Salmon Dumping (June 2, 1997),
rapid/press-release_IP-97-468_en.htm [] (describing the European Commission’s finding that from August 1995 to July 1996 “Norwegian salmon was being exported to the EU at dumped prices and was also enjoying subsidies”).
In its WTO case challenging the EU’s imposition of trade remedies, Norway noted that the 2006 antidumping measures were a de facto continuation of the anti­dumping and countervailing duties the EU had imposed since the 1990s. 176 EC–Salmon—Norway, supra note 173, paras. 4.5–4.7 (“For the past 17 years, hardly a year has gone by without some kind of restriction in place.”). Indeed, the European Commission itself suggested this connec­tion in its order imposing the duties. 177 Commission Regulation 628/2005 of Apr. 22, 2005, Imposing a Provisional Anti-Dumping Duty on Imports of Farmed Salmon Originating in Norway, 2005 O.J. (L 104) 5 (EC) [hereinafter Commission Regulation 628/2005]. In most trade remedies cases brought to the WTO the complainant prevails, and this case was no different. Norway won and the EU removed its antidumping measures in 2009. EU Repeals Anti-Dumping Measures Against Norwegian Salmon, Nor.: Mission to the EU, (June 8, 2009),
anti_dumping/#.WeWB01KZNE4 []. In 2012, the United States lifted its antidumping duties on Norwegian salmon, marking the first time since 1991 that neither the United States nor the EU applied trade remedies against Norwegian salmon. See Org. for Econ. Cooperation & Dev., OECD Review of Fisheries: Policies and Summary Statistics 37 (2013).
The Commission found that viola­tions of the earlier antidumping and antisubsidy measures meant that EU producers had never been able to compete on cost with Norwegian producers, leading in part to the difficulties European pro­ducers faced in the mid-2000s. 178 Commission Regulation 628/2005, supra note 177.

Most other trade remedies cases do not formally distinguish between farmed fish and wild fish. Farmed salmon and wild salmon, for example, might be considered “like” products and thus fall within the same trade remedies investigation. 179 This distinguishes fish from energy. Farmed fish and wild fish are much more similar, and hence more likely to be “like” products within the meaning of trade law, than are fossil fuels and renewable energy products such as solar panels. Nevertheless, as Professor Frank Asche, a noted marine economist, has pointed out, the “dumping of seafood has been a WTO concern primarily in relation to aquaculture.” 180 See Asche, supra note 147, at 12. The reason is that in the WTO era, the extraordinary growth in global fish production has been primarily the result of aquaculture. 181 See, e.g., UN Env’t Programme, Green Economy and Trade: Trends, Challenges, and Opportunities 93 (2013), [] (“[T]otal catch from wild fisheries has remained relatively stable throughout the last three decades. The source of significant growth in the global production of fish since the late 1980s has been aquaculture . . . .”); Walter R. Keithly, Jr. & Pawan Poudel, The Southeast U.S.A. Shrimp Industry: Issues Related to Trade and Antidumping Duties, 23 Marine Resource Econ. 459, 462 (2008) (“Much of the growth in world warm-water shrimp production since the 1980s has been the result of successful farming activities . . . . [G]rowth [in farmed shrimp] has been particularly pronounced since the mid- to late-1990s.”).
Indeed, the World Wildlife Fund reports that “[s]almon aquaculture is the fastest growing food pro­duction system in the world—accounting for 70 percent . . . of the market.” 182 Farmed Salmon, World Wildlife Fund,
farmed-salmon [] (last visited Oct. 11, 2017).
This increase in production from aquaculture puts downward pressure on “dockside” prices—prices paid to fishermen 183 Keithly & Poudel, supra note 181, at 459. —for wild caught fish, especially in developed countries like the United States and the members of the EU. 184 See id. at 463–64 (explaining that increased farmed-shrimp production coincides with a decline in shrimp prices and suggesting that aquaculture production is responsible for that decline).

U.S. antidumping investigations into shrimp illustrate why aquacul­ture has been the primary target of antidumping duties. World exports of shrimp more than quadrupled between 1980 and 2005, but the inflation-adjusted value of such shrimp only slightly more than dou­bled. 185 Id. at 463. The result was a more than 50% decrease in the real price of shrimp in those twenty-five years, owing primarily to increased shrimp farming. 186 Id. More­over, estimates put almost 90% of fish farming in Asia. 187 UN Food & Agric. Org., The State of World Fisheries and Aquaculture 23 tbl.6 (2016), [] [hereinafter State of World Fisheries 2016] (indicating that, of the approximately 74 million metric tons of global aquaculture production, Asia is responsible for approximately 66 million metric tons, or roughly 89%). This dramatic growth in supply created significant hardship for American shrimpers, especially along the Gulf of Mexico. 188 See Simon Worrall, Q&A: Explaining the Decline of American Seafood, Nat’l Geographic (July 12, 2014), [] (“90 percent of [American] shrimp is imported, the majority of it from Asia.”). Indeed, some have compared the outsourcing of the American shrimp industry to outsourcing in other sectors such as textiles or manufacturing. 189 See, e.g., Keithly & Poudel, supra note 181, at 468.

In 1985, the U.S. International Trade Commission (ITC) evaluated the shrimp sector for possible action. 190 Id. at 468. Increases in overseas shrimp farm­ing prompted a complaint from southeastern U.S. shrimp harvesters alleging that foreign governments provided financial assistance to shrimp farmers, hurting the domestic shrimp harvesters. 191 Id. The ITC declined to impose antidumping duties at that time. But in 2003 the Southern Shrimp Alliance filed another petition seeking antidumping duties against shrimp from six countries: Brazil, China, Ecuador, India, Thailand, and Vietnam. 192 Id. at 469.

In 2005, following an investigation, the United States imposed anti­dumping duties against the six countries. In its finding, the ITC empha­sized two points about the imports. First, the ITC noted that “[i]mports from subject countries include both farmed and wild-caught warmwater shrimp. However, production of farmed warmwater shrimp plays a much more important role in subject country production than in U.S. production.” 193 Warmwater Shrimp Investigation, supra note 162, at II-4.

Second, the ITC found that foreign shrimp producers benefit from “substantial government assistance.” 194 Id. (“[S]ubject country governments have been active in assisting the growth of their warmwater shrimp industries, using subsidies, loans, prohibitively high tariffs on imports . . . , government efforts in research and development and in developing a seed stock . . . for farms, government aid in response to [shrimp] epidemics . . . , and training.”). Offsetting this government sup­port seems to have been at least part of the motive for the 2005 U.S. anti­dumping duties on shrimp. These antidumping duties alone pro­duced four WTO disputes. 195 In two of these cases, Thailand challenged the provisional application of duties in 2004 and the final application in 2005, ultimately pursuing only the latter (in a case not initiated until 2006). See supra notes 163, 166. The core challenge in most of these cases—won by the complainants—was to the U.S. method of “zeroing.” See Index of Disputes Issues, WTO, [] (last visited Oct. 30, 2017) (listing many of these fisheries disputes as disputes about zeroing). Zeroing is a controversial method for calculating dumping margins. It involves setting negative dumping margins on products subject to investigation (that is, those products not being dumped) to zero, and then averaging them with products subject to investigation that yield positive dumping margins. For instance, if farmed shrimp is being dumped at a margin of $2 per pound, that means it is being sold at $2 below “normal value.” If captured shrimp is sold at a dumping margin of -$2 per pound, that means it is being sold at $2 more than normal value, and hence is not being dumped. Averaging these two products, since they compete, would yield a dumping margin of zero (assuming they are sold in the same quantity). A zeroing methodology, however, requires you to treat the dumping margin on captured shrimp as $0, rather than -$2. The effect of this technique is to increase dumping margins in investigations that involve “like” products, some of which are being dumped and some of which are not. Another of these cases involved a challenge to a U.S. order imposing enhanced bonding requirements on “agriculture/aquaculture merchandise” subject to antidumping duties (in practice, only the imported shrimp). See Panel Report, United States—Customs Bond Directive for Merchandise Subject to Anti-Dumping/Countervailing Duties, para. 2.2, WTO Doc. WT/DS345/R (adopted Feb. 29, 2008).

The growth, and government support, of aquaculture have thus driven trade remedies cases and associated WTO disputes. Beyond shrimp and salmon, the United States and the European Union have im­posed trade remedies on Vietnamese catfish, 196 Antidumping Duty Order: Certain Frozen Fish Fillets from the Socialist Republic of Vietnam, 68 Fed. Reg. 47,909, 47,909 (Aug. 12, 2003); see also Certain Frozen Fish Fillets from the Socialist Republic of Vietnam: Final Results of Antidumping Duty New Shipper Review; 2014–2015, 81 Fed. Reg. 44,272, 44,273 (July 7, 2016) (confirming antidumping duties on Vietnamese frozen fish fillets after a review). Chinese crawfish tails, 197 Notice of Amendment to Final Determination of Sales at Less than Fair Value and Antidumping Duty Order: Freshwater Crawfish Tail Meat from the People’s Republic of China, 62 Fed. Reg. 48,218, 48,219 (Sept. 10, 1997). Norwegian trout, 198 Council Regulation 437/2004 of Mar. 8, 2004, Imposing Definitive Anti-Dumping Duty and Collecting Definitively the Provisional Duty Imposed on Imports of Large Rainbow Trout Originating in Norway and the Faeroe Islands, 2004 O.J. (L 72) 23, 38 (EC). and Turkish trout, 199 Commission Implementing Regulation 1195/2014 of Oct. 29, 2014, Imposing a Provisional Countervailing Duty on Imports of Certain Rainbow Trout Originating in Turkey, 2014 O.J. (L 319) 1, 29 (EU); see also Eva Tallaksen, Turkish Trout Exporters Could Face 7–10% Anti-Dumping Duties to EU, Undercurrent News (Oct. 15, 2014), (on file with the Columbia Law Review) (noting that Turkish trout producers would be assessed countervailing duties on their rainbow trout after a finding that Turkey had subsidized those producers). among others. Trade disputes centered on wild-caught fish are conspicuously absent.

Despite all this activity surrounding aquaculture, governments have not formally challenged one another’s considerably more extensive sup­port for wild fisheries. To be sure, the WTO has been the site of more comprehensive efforts to deal with government support for wild fisheries. Indeed, the regulation of fisheries subsidies has been a hot topic in trade negotiations since the 1980s. 200 A number of international organizations have worked on fisheries subsidies issues, including not only the WTO but also the Food and Agriculture Organization of the United Nations, the Organization for Economic Co-operation and Development, and Asia-Pacific Economic Cooperation. For a comprehensive discussion of the interaction among these various efforts at reform, see Young, supra note 144, at 85–133. During the Uruguay Round, which led to the creation of the WTO, countries debated how best to regulate fisher­ies subsidies. 201 See id. at 70–73 (discussing the structure of the WTO agreements and the decision to constrain the use of subsidies). Ultimately, they elected to exclude fisheries from the Agreement on Agriculture. 202 Id. at 52, 73. Fisheries subsidies are governed, though, by the general subsidy rules in the SCM Agreement. 203 Id. at 73–74, 91–93. This fact alone makes the absence of wild-fisheries challenges puzzling. Nations wish to reform fisheries subsidies, but are unwilling to use existing rules and the dispute-settlement process to do so. 204 The Doha Round of negotiations, which began in 2000 and continues slowly today, includes a mandate to “clarify and improve WTO disciplines on fisheries subsidies.” WTO, Ministerial Declaration of 14 November 2001, para. 28, WTO Doc. WT/MIN(01)/DEC/1, 41 ILM 746, 750 (2002). Despite this mandate, a broad consensus that some form of additional regulation is necessary, a number of proposals from leading WTO members, and fifteen years of negotiations, an agreement has proved elusive.

*   *   *

Energy and fish are two of the most important sectors of the global economy. Not surprisingly, they both benefit from substantial govern­ment support, some of which runs afoul of global trading rules. Yet, bizarrely, those trade rules are enforced against only a subset of products within each industry. Traditional natural resources—fossil fuels and wild fish—are left alone. Natural resource substitutes—renewable energy and farmed fish—are targeted over and over again. This Article does not claim that government support for natural resource substitutes is or should be lawful. Nor does it argue that these cases are by themselves unfair. Rather, its claim is that fairness is a relative concept when prod­ucts compete in the marketplace. Enforcing trade laws against products with the potential to reduce consumption-imposed environmental bur­dens, but not against competing products whose consumption harms the environment, ultimately hurts efforts to develop and promote envi­ron­mentally friendly products. The following Part takes up this claim in detail.

III. How Selective Enforcement of Trade Laws Hurts Environmental Products

The traditional fear about selective enforcement has been that it will confer an unfair advantage on some nations at the expense of others. In so doing, selective enforcement will undermine certain values that many would like to see the trade regime further, or at least not undermine, such as environmental protection, labor rights, or economic develop­ment. This Part unpacks how selective enforcement of trade law can inhibit, rather than promote, competition—a core economic value. In the commercial context, selective enforcement can distort the develop­ment of product markets. Even worse, when products have broader social costs and benefits, such as environmental externalities, these distortions do more than simply affect the welfare of firms and consumer choice. They can create market failures, leading products with significant social costs to capture greater market share than they would in the presence of evenhanded enforcement. Selective enforcement of trade rules thus creates a paradox. In enforcing rules designed to keep markets free and open, selective enforcement can actually create or exacerbate market failures.

Section III.A describes how selective enforcement benefits products not targeted for enforcement at the expense of those products that are targeted. Section III.B considers how competitive the relationship between two products has to be for selective enforcement to create market distortions. Section III.C explains the social costs of selective enforcement that protects natural resource consumption.

A. The Financial Costs of Selective Enforcement

Selective enforcement creates costs that some market participants, but not others who are similarly situated, must bear. These costs create a market advantage for those firms not targeted for enforcement. This advantage, the result of government action, is an implicit subsidy. And by distorting competition, selective enforcement can affect how markets function, with far-reaching effects for both market participants and society as a whole. Selective enforcement produces at least three costs: litigation costs, liability, and lost investment. These costs are generaliza­ble to selective enforcement in any commercial context, not just international trade law. This section unpacks these costs, which Part IV will discuss more concretely.

1. Litigation Costs. — First, and most obviously, defendants must bear the cost of defending themselves in litigation. These costs can be significant. Bringing or defending a WTO dispute, much less participat­ing in WTO disputes on a regular basis, can tax the resources of any country. 205 See Busch et al., supra note 60, at 560–61 (documenting the burdens of the WTO dispute settlement process); Guzman & Simmons, supra note 53, at 583, 591 (finding evidence that capacity to bear the significant costs of trade disputes affects developing countries’ litigation decisions). In a survey of WTO members, 88% of respondents said that legal capacity is the single most important factor distinguishing the most powerful WTO members from run-of-the-mill members. 206 Busch et al., supra note 60, at 567. The most power­ful countries are better able to afford the “high cost of WTO litiga­tion” and benefit from “greater private sector support.” 207 Id. (internal quotation marks omitted). Most coun­tries, on the other hand, struggle to marshal the resources necessary to protect their rights in WTO dispute settlement. For these countries, the risk of becoming a respondent in a WTO action is a significant one, to be avoided if at all possible.

The same point applies to individual producers that might be tar­geted by trade remedies investigations. Trade remedies investigations do not target nations but rather focus on individual companies. Those com­panies are entitled to contest the imposition of trade remedies against them. 208 See, e.g., Craig A. Lewis, Jonathan T. Stoel & Brian S. Janovitz, The United States Court of International Trade in 2010: Is Commerce Suffering from Adverse Decisions It Wasn’t Double-Counting On?, 43 Geo. J. Int’l L. 47, 50 (2011) (noting that the Court of International Trade’s docket “remains overwhelmingly dominated by appeals contesting the issuance and administration of trade remedies”). Doing so is more burdensome for targeted foreign producers than it is for the domestic producers that seek their government’s protection via trade remedies. Foreign producers must bear the cost and inconvenience of contesting legal proceedings in a foreign country, often in part by hiring expensive foreign counsel. These costs can be especially burdensome for relatively new producers or industries that are challeng­ing incumbent producers. Indeed, the costs of contesting trade remedies investigations, both at the national and international levels, have led countries to avoid dispute resolution for most trade remedies. 209 Chad P. Bown, Trade Remedies and World Trade Organization Dispute Settlement: Why Are So Few Challenged?, 34 J. Legal Stud. 515, 518–24 (2005) [hereinafter Bown, Trade Remedies] (explaining the costs of challenging a trade remedy at the WTO and suggesting countries may be choosing antidumping retaliation in lieu of formal dispute resolution).

2. Liability. — Second, defendants may have to bear financial liabil­ity or other penalties if a tribunal finds they have acted unlawfully or engaged in “unfair” trade practices. Defendants may also spend more on precautionary measures in the future than those market participants that do not believe they are likely targets for enforcement action. 210 See, e.g., Robert Cooter, Unity in Tort, Contract, and Property: The Model of Precaution, 73 Calif. L. Rev. 1, 2–3 (1985) (analyzing how the common law establishes incentives for actors to take precautionary measures); Charles D. Kolstad, Thomas S. Ulen & Gary V. Johnson, Ex Post Liability for Harm vs. Ex Ante Safety Regulation: Substitutes or Complements?, 80 Am. Econ. Rev. 888, 890–91 (1990) (discussing theories of optimal liability and precaution in the presence of uncertainty about how a court will view the defendant’s actions). In this way, selective enforcement is a “force multiplier” for similarly situated producers that do not expect to be challenged. Producers that might be challenged incur costs to avoid liability even if they are never challenged.

In the context of international trade rules on government support, two particular forms of liability are worth discussing in further detail: withdrawn government support and higher duties on imports.

a. Withdrawn Government Support. — Unsurprisingly, enforcement of WTO rules against government financial support can neutralize the ef­fects of that support. This can occur in several ways. First, in cases brought to the WTO’s Dispute Settlement Body, reports (that is, deci­sions) against the respondent come with a recommendation that the re­spond­ent government bring itself into compliance with the DSB’s decision. 211 See DSU, supra note 54, art. 19. Under the Dispute Settlement Understanding, an initial panel hears the dispute, makes factual determinations, and applies relevant WTO agreements to determine whether the respondent has in fact violated its WTO obligations. See id. art. 11. The losing party may appeal to the WTO’s Appellate Body, which reviews the panel’s determination. See id. art. 17. Both the panel’s and the Appellate Body’s decisions become binding on the parties unless all WTO members—including the prevailing party—agree otherwise. See id. arts. 16.4, 17.14. A losing respondent should thus remove its unlawful finan­cial support, or at least modify the measure to make it consistent with WTO rules.

Of course, the effect of such a ruling is to reduce government sup­port for the relevant domestic industry. For example, in 2014, the United States agreed to replace most of its subsidies to cotton farmers with an insurance system. 212 Notification of a Mutually Agreed Solution, United States—Subsidies on Upland Cotton, WTO Doc. WT/DS267/46 (Oct. 23, 2014). The United States did so to comply with a WTO decision finding that U.S. cotton subsidies caused adverse effects to Brazilian cotton producers and were, in some cases, prohibited. 213 See Appellate Body Report, United States—Subsidies on Upland Cotton, paras. 4–5, 763–764, WTO Doc. WT/DS267/AB/R (adopted Mar. 3, 2005) [hereinafter United States—Cotton]. Similarly, the United States, the European Union, and Canada challenged certain measures China applied to imported automobile parts as violations of, inter alia, the SCM Agreement. 214 Neither the panel nor the Appellate Body actually reached the SCM claim, as the measures were found to violate the (simpler) national treatment rule found in GATT Article III. See Appellate Body Reports, China—Measures Affecting Imports of Automobile Parts, paras. 253–254, WTO Docs. WT/DSB339/AB/R, WT/DS340/AB/R, WT/DS342/AB/R (adopted Dec. 15, 2008). After losing, China repealed the of­fend­ing measures. 215 Id.

If the respondent declines to remove the measure, the complainant can obtain authorization to retaliate. 216 See DSU, supra note 54, art. 22 (discussing “[c]ompensation and the suspension of concessions” as retaliatory measures available to complaining members). Retaliation takes the form of sus­pending concessions—in other words, imposing trade barriers that the complainant had otherwise agreed to remove—to get the respondent to remove its unlawful measure. 217 Id. The complainant has significant discre­tion in how it retaliates. For example, it may try to suspend concessions that hurt the recipients of government support in an effort to encourage those recipients to ask the government to stop supporting them. 218 See id. art. 22.3 (describing the general principles guiding a successful complainant’s decision as to which concessions to suspend); see also Decision by the Arbitrators, European Communities—Measures Concerning Meat and Meat Products (Hormones), para. 18, WTO Doc. WT/DS26/ARB (July 12, 1999) (determining that a respondent may challenge only the “level” of suspended concessions, not the “nature” of the concessions the complainant chooses to suspend).

Alternatively, the losing party in a WTO subsidy dispute might end up paying foreign producers to avoid sanctions. Brazil initially won its challenge to U.S. cotton subsidies in 2005. 219 See United States—Cotton, supra note 213, paras. 763–764. The United States agreed in 2010 to pay Brazilian cotton farmers $147.3 million per year to avoid other retaliatory penalties. 220 Editorial, U.S.–Brazil Cotton Deal Perpetuates an Unhealthy Status Quo of Subsidies, Wash. Post (Oct. 7, 2014), (on file with the Columbia Law Review). While U.S. cotton farmers continued to receive subsidies, the U.S. government’s payment to Brazilian farmers—effectively forcing U.S. tax payers to subsidize Brazilian cotton growers—reduced the advantage of U.S. subsidies.

b. Higher Duties on Natural Resource Substitutes. — Trade remedies provide the second way that governments can neutralize the effect of government support. Antidumping duties (or countervailing duties) are designed to offset unfair trade practices. They can thus be set up to the amount needed to offset the unfair practice (either the “margin of dumping” or the amount of the subsidy). 221 See GATT, supra note 36, art. VI.2 (“In order to offset or prevent dumping, a contracting party may levy on any dumped product an anti-dumping duty not greater in amount than the margin of dumping . . . .”); id. art. VI.3 (“No countervailing duty shall be levied on any product . . . in excess of an amount equal to the estimated bounty or subsidy . . . .”).

Moreover, because trade remedies are essentially higher tariffs, they are paid by the very producers that otherwise would have charged lower prices.

Those higher prices, in turn, can reduce the market share of the imported product. For example, a production subsidy might allow a Chinese producer of solar panels to sell its products for $100 less per panel in the United States than it would otherwise have to charge. The Chinese producer might therefore be able to sell more of its products. Moreover, cheaper solar panels would lead to more solar power and therefore fewer greenhouse gas emissions—a social benefit shared by the community as a whole, not only the purchaser of the solar panel. But if the United States imposes countervailing duties on Chinese solar panels, nullifying the $100 subsidy, the price goes back up. The Chinese pro­ducer now fails to capture greater market share, fewer solar panels are sold overall, and the environment does not benefit from the emissions reductions that attend cheaper solar power.

Equally importantly, a government need not seek the WTO’s per­mission before resorting to trade remedies. 222 See id. art. VI.2–.3 (granting parties the right to impose antidumping and countervailing duties). Instead, a government may impose trade remedies after determining, through a domestic investiga­tion, that such remedies are justified. 223 See id. art. VI.6 (requiring states to determine that dumping or subsidization has caused or threatened “material injury” to a domestic industry). In the United States, for exam­ple, the Department of Commerce and the International Trade Commission conduct these investigations; 224 Jay Charles Campbell, The Trade Litigant’s Gauntlet: The Hanging Judge and the Teflon Tribunal, 31 Nw. J. Int’l L. & Bus. 1, 3 (2011). in the European Union, the European Commission does. 225 See Wentong Zheng, Reforming Trade Remedies, 34 Mich. J. Int’l L. 151, 179 & n.151 (2012) (noting the European Commission’s investigative role in advance of trade remedies). If the exporting government feels that the importing government has imposed trade remedies unlawfully, the exporting government can bring a case to the WTO Dispute Settlement Body. 226 Reid M. Bolton, Anti-Dumping and Distrust: Reducing Anti-Dumping Duties Under the W.T.O. Through Heightened Scrutiny, 29 Berkeley J. Int’l L. 66, 66–67 (2011).

Trade remedies have this protectionist effect by design. They are, in effect, a safety valve to ensure that governments can protect certain domestic industries from the effects of unfair foreign competition. 227 Cf. Alan O. Sykes, Protectionism as a “Safeguard”: A Positive Analysis of the GATT “Escape Clause” with Normative Speculations, 58 U. Chi. L. Rev. 255, 273 (1991) (noting that the GATT “escape clause” “legitimates protection” for domestic industries). Governments’ responding to unfair trade practices is not in and of itself problematic. The difficulty, it bears repeating, is the selective way in which they do so. When disputes systematically target only one set of actors benefitting from the unfair trade practice, then the enforcement policies themselves have become unfair and, if the targeted set of actors produces a great social benefit, socially harmful. 228 See infra section III.C.

3. Lost Investment and Higher Costs of Capital. — Litigation costs and liability lead to the third, and most important, cost of selective enforce­ment in the trade context: decreased investment and increased costs of capital, which together further erode profitability and can slow innova­tion. To see how selective enforcement deters investment, notice that litigation and liability costs increase the minimum price a producer must charge to break even. If money coming in goes back out the door in response to enforcement efforts, businesses cannot use it to settle other debts, make further investments in their business, or simply to take a profit. If they continue to charge the same prices despite additional enforcement-related costs, they lose profitability. If they raise prices, they risk losing market share to less expensive competitors.

This reduction in profits can create a vicious cycle. Investors direct money to firms that make profits. Increased costs reduce profits and therefore reduce investors’ willingness to put money behind a firm. This investor reluctance itself translates into higher costs of capital. If profits are lower, a firm may have to offer more—for example, it may have to pay higher interest rates—in order to attract the funds it needs to sup­port its operations. These increased costs of capital decrease the firm’s profitability, making it more difficult to attract capital, and so on.

Critically, the targeted firms’ similarly situated competitors—those that engage in or benefit from the same allegedly unlawful conduct but against whom the law is not enforced—come out ahead. They do so in two ways. First, they do not bear the direct litigation and liability costs that the targeted entities bear. They do, however, continue to benefit from the allegedly unlawful conduct. In this way, selective enforcement acts like an implicit subsidy for those who are not targeted. The costs of engaging in behavior deemed advantageous from a business perspective is lower for those firms that avoid enforcement.

Competitors who avoid enforcement also may gain access to cheaper capital. Investors looking to gain in a particular sector may shift their investments from companies targeted for enforcement to those not so targeted. Greater access to capital may allow firms that predictably avoid enforcement to expand their operations, taking advantage of greater economies of scale or capturing additional market share. It may also allow them to negotiate lower interest rates than those available to their targeted competitors.

Two examples from outside the world of international trade illus­trate the general point. First, rules requiring lenders to maintain a cer­tain amount of capital on hand to cover debts gone bad are enforced against traditional banks, but not against so-called “shadow banks.” 229 See Sridhar Natarajan, The SEC’s Beef with Shadow Banks Could Be Bad for Some Businesses, Bloomberg (Oct. 26, 2016),
2016-10-26/shadow-banks-on-wall-street-turf-confront-sec-on-lending-rules (on file with the Columbia Law Review) (discussing an SEC proposal to impose capital constraints against shadow banks, as well as traditional banks).
As a result, shadow banks have captured a significant share of the market for loans to small and mid-market businesses. 230 Id. These loans are riskier than loans to large corporations and thus harder to make for traditional banks that are subject to capital constraints. Id. Selective enforcement of rules on capital adequacy thus protect and benefit shadow banks at the expense of traditional banks.

Second, consider tobacco litigation in the United States. In the 1990s, state attorneys general reached what is known as the Master Settlement Agreement (MSA) with major tobacco companies such as Philip Morris and R.J. Reynolds. 231 See W. Kip Viscusi, Tobacco: Regulation and Taxation Through Litigation, in Regulation Through Litigation 22, 46–48 (W. Kip Viscusi ed., 2002) (describing the Master Settlement Agreement); see also Tobacco Control Legal Consortium, The Master Setlement Agreement: An Overview 1 (2015),
sites/default/files/resources/tclc-fs-msa-overview-2015.pdf [].
The MSA required the major tobacco companies to make payments to the states to settle their liability for state tobacco-related health expenditures. 232 Viscusi, supra note 231, at 48–50. The MSA also imposed re­strictions on certain business practices, such as advertising tobacco prod­ucts to youth. 233 Charles King III & Michael Siegel, The Master Settlement Agreement with the Tobacco Industry and Cigarette Advertising in Magazines, 345 New Eng. J. Med. 504, 504 (2001). Notably, however, the MSA only applied to those tobacco producers targeted for enforcement by the states. Small-scale tobacco producers were exempted from enforcement, and hence the financial and advertising restrictions imposed by the settlement. 234 Martha A. Derthick, Up in Smoke: From Legislation to Litigation in Tobacco Politics 202–03 (2002). In the after­math of the MSA, a number of these small producers thrived, contributing to a declining market share for the big tobacco companies. 235 Id.

B. What Kind of Relationship Between Products Makes Selective Enforcement Pernicious?

Selective enforcement in trade law is unfair only if it disadvantages a competitor. Selective enforcement of environmental and labor laws gives a country a competitive advantage over countries that evenly and robustly enforce their laws. Selective enforcement against weak countries gives strong countries an advantage by allowing them to flout trade law to a greater degree. Likewise, the selective enforcement of trade laws is unfair only if it harms some product’s chance to compete in the market. But how similar do two products have to be for this harm to arise?

The United States appears to take the position that products must be very similar for trade enforcement to affect competitive opportunities. In 2016, the United States prevailed in a WTO case against India challeng­ing India’s National Solar Mission on the grounds that it con­tained a lo­cal content requirement. 236 India—Solar Cells and Modules, supra note 112, para. 6.4. In discussing the United States’ victory at the panel stage, U.S. Trade Representative Michael Froman stated:

[T]he Obama Administration is committed to strengthening the clean energy sector and the millions of jobs it supports here in America and all over the world. Trade enforcement is critical for ensuring that world-class U.S. clean energy goods and services can compete on an equal footing around the world . . . . 237 Press Release, Office of the U.S. Trade Representative, United States Prevails in WTO Dispute Challenging India’s Discrimination Against U.S. Solar Exports (Feb. 24, 2016), [].

When Froman talks about “competing on an equal footing,” he as­sumes that the American renewable energy sector competes primarily with the Indian renewable energy sector (and perhaps the renewable en­ergy sectors of other countries, such as China) in the Indian market. In drawing such a tight comparison, Froman is following the letter of inter­national trade law. In bringing a discrimination claim, for instance, a complainant must first demonstrate that the product receiving prefer­ential treatment and the one being discriminated against are “like.” 238 See Appellate Body Report, Japan—Taxes on Alcoholic Beverages, 18–19, WTO Doc. WT/DS8/AB/R, WT/DS10/AB/R, WT/DS411/AB/R (adopted Oct. 4, 1996) [hereinafter Japan—Alcohol] (noting the importance of determining “whether the taxed imported and domestic products are ‘like’”). Likeness is a narrower relationship than merely a competitive relation­ship in the marketplace. 239 See id. at 21. Indeed, the WTO Appellate Body has described substitutability in the marketplace as only one factor relevant to an assessment of likeness, along with factors such as a product’s physical characteristics and its tariff classifications. 240 Id. at 25. As a consequence, farmed fish and wild fish of the same species might be treated as like products in a trade case, but renewable energy products and fossil fuels—which surely compete with each other—would almost certainly not be.

But this assumption is unwarranted. The American renewable energy sector competes not only with the Indian renewable energy sec­tor but also with the fossil fuel sector. A report from the International Institute for Sustainable Development, entitled How Subsidies for Kerosene Are Holding Back Solar Power in India, makes the point clear. 241 Vibhuti Garg, Shruti Sharma & Kieran Clarke, How Subsidies for Kerosene Are Holding Back Solar Power in India, Int’l Inst. for Sustainable Dev.: Glob. Subsidies Initiative (July 11, 2016),
back-solar-power-india [].
Millions of rural households in India rely on kerosene for lighting. 242 Id. The report notes that:

Off-grid solar technologies, such as solar lanterns and solar home systems, can effectively replace kerosene use for lighting in rural areas that are unserved (or poorly served) by the electricity grid, and are likely to be so for some time. Solar can provide safer and better quality lighting, and is also friendly to the environment. 243 Id.

Yet solar has not been a successful replacement for kerosene in off-grid communities, in large part due to subsidies for kerosene. 244 Id. The study finds that under current conditions over a two-year period, house­holds lose money by shifting to solar products. 245 Id. Solar becomes the better option, however, if kerosene subsidies are zeroed out. 246 Id. One might argue that the Indian government could eliminate government support for both fossil fuels and renewables. In the short term, this is likely impractical for domestic political reasons. More importantly, though, it neglects the impact of international fossil fuel subsidies, which are beyond India’s ability to control. Subsidies for kerosene, in other words, have allowed it to gain greater market share at the expense of solar. This market impact, in turn, has significant nega­tive health consequences for kerosene users in addition to the broader environmental impacts of fossil fuel consumption. 247 Id.

This example highlights the centrality of the competitive relation­ship between products. If two products are substitutes in the market, removing subsidies for one but not the other will hamper competition. Trade lawyers have understood this point since the earliest days of the GATT. In 1950, Chile challenged Australia’s decision to remove a war­time subsidy on sodium nitrate. 248 Working Party Report, The Australian Subsidy on Ammonium Sulphate, paras. 2–6, GATT/CP.4/39 (adopted Apr. 3, 1950), GATT BISD at 188, 189–91 (2003). Chile had negotiated with Australia for duty-free entry of its sodium nitrate at a time at which Australia had wartime subsidies on both sodium nitrate and its competitor, ammonium sulphate. 249 Id. A GATT Working Party (a predecessor to the more formal dispute process in existence today) found that Australia’s action harmed Chile’s legitimate expectations. 250 Id. para. 12. By removing the subsidy on sodium nitrate and not on ammonium sulphate, Australia had unfairly dis­advantaged it against its (still subsidized) competitor.

Economists refer to the strength of the competitive relationship be­tween two products as the cross-elasticity of demand. 251 See, e.g., 62B Am. Jur. 2d Private Franchise Contracts § 71 (2018) (“[C]ross-elasticity of demand may be defined as the extent to which purchasers will accept substitute products in the event of price fluctuations and other changes.”). Technically, the cross-elasticity of demand measures how demand for one product changes in response to a change in the price of another product, holding the price of the first good constant. 252 Gregory J. Werden, The History of Antitrust Market Delineation, 76 Marq. L. Rev. 123, 130 (1992). Cross-elasticities of demand are commonly used in antitrust law to determine the “relevant market”—that is, to define legally significant competitive relationships among products. 253 See id. (describing the Supreme Court’s use of cross-elasticity in market delineation).

In economic terms, selective enforcement becomes more pernicious the higher the cross-elasticity of demand between two products. 254 Id. at 131 (“[A] high cross-elasticity indicates close substitutes.” (internal quotation marks omitted) (quoting Joe S. Bain, Price Theory 52 (1952))). The reason is straightforward. If a product has few close substitutes and the demand for a product is inelastic, producers can pass along price increases to the consumer. Hence, if a product has relatively inelastic demand, the costs discussed above—litigation, liability, and lost invest­ment—will not affect a product’s bottom line. But as a demand becomes more elastic, consumers will purchase less of the product as these costs force producers to raise prices. If the product has a high cross-elasticity of demand with another product, consumers will substitute that second product for the newly more expensive original one. In this way, one pro­ducer’s loss is another producer’s gain. In the kerosene–solar example discussed above, consumers purchase less solar (and more kerosene) as the price of kerosene falls in response to subsidies. 255 See supra notes 241–247 and accompanying text.

The idea of cross-elasticities of demand also illustrates the problem with Froman’s focus on competition exclusively among renewable energy, namely that he is looking at an incomplete set of competitive re­lationships. 256 See supra notes 237–238 and accompanying text. Consider three products: American solar panels, Indian solar panels, and fossil fuels. For simplicity, American solar panels sell in India for $30 a panel. Indian solar panels would also sell for $30 unsubsidized, but the effect of India’s subsidy is to allow the panels to sell for $20. Subsidized kerosene necessary to produce an equivalent amount of energy sells for $25, but would sell for $35 unsubsidized. The United States’ success at the WTO may succeed in removing the Indian subsidy for Indian solar panels. Doing so would make American and Indian solar panels competitive with each other. But the effect would also be to de­crease the purchase of solar panels—American or Indian—by redirecting purchases to kerosene. Subsidized kerosene, after all, remains cheaper at $25 than solar power at $30. 257 One could formalize the exact size of this effect more precisely by imagining the cross-elasticity of demand between kerosene and solar energy. The basic point, however, is made clear by this simple example.

Of course, as Froman notes, his concern is not with the overall share of renewable energy versus fossil fuels; he is concerned specifically with the share of the market that American solar producers capture. 258 See supra text accompanying note 238. In the simple example above, consumers willing to pay an extra $5 for clean energy will now be more likely to purchase American solar panels as opposed to Indian solar panels. The incidental benefit to fossil fuels may be, in Froman’s view, beside the point. But this effect matters if the notion of “fair trade” encompasses values beyond liberalized trade.

C. The Social Costs of Selective Enforcement

Critically, selective enforcement of trade law has its worst effects when targeted products have social benefits beyond those captured by producers, while similarly situated, untargeted products have social costs. In these circumstances, selective enforcement reinforces a market failure.

The theory of externalities demonstrates why this occurs. In general, free markets work well because freely determined prices carry information about how much society values a product or activity. 259 See, e.g., Donald J. Boudreaux, Information and Prices, The Concise Encyclopedia of Economics, Library of Econ. & Liberty,
InformationandPrices.html [] (last visited Oct. 30, 2017) (discussing the role of prices in conveying information about social preferences).
This assumes, however, that private market participants capture the entire social value of a transaction. If they do not, then market prices will not reflect the true social cost or benefit of a product or service. 260 See Dennis W. Carlton & Alan S. Frankel, Transaction Costs, Externalities, and “Two-Sided” Payment Markets, 2005 Colum. Bus. L. Rev. 617, 622–23 (“Externalities arise when the private cost facing a buyer or seller differs significantly from the social cost.”).

Since Ronald Coase introduced the problem of social costs in 1960, 261 R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1, 40–42 (1960). law and economics scholars have argued that a key purpose of the law is to get private actors to take into account these broader social costs and benefits of their transactions. 262 See Robert Cooter, Prices and Sanctions, 84 Colum. L. Rev. 1523, 1523, 1535 (1984) (arguing that the government should impose prices to ensure that decisionmakers internalize the complete costs of their actions). In economic terms, the law should encourage private actors to internalize the externalities created by their actions. In the context of subsidies, this has generally meant two things. First, governments should not subsidize products that have signif­icant social costs. Government support for these two sectors keeps prices artificially low, increasing consumption and thereby increasing the over­all costs on society. For exactly this reason, environmentally minded people have supported policy goals that reduce or eliminate subsidies for fossil fuels and wild fishing. 263 See, e.g., Allison Kirsch & Timmons Roberts, The Ghosts of Resolutions Past: The G20 Agreement on Phasing Out Inefficient Fossil Fuel Subsidies, Brookings Inst.: Planet Policy (Nov. 14, 2014),
11/14-g20-fossil-fuel-subsidies-kirsch-roberts [] (discussing government promises to phase out fossil fuel subsidies); see also Young, supra note 144, at 6 (discussing efforts to reform subsidies for wild fisheries).
Because of governments’ failures to elimi­nate these subsidies, the market produces and consumes more fossil fuels and wild fish than it would without subsidies. 264 Indeed, many argue that a tax further raising the cost of fossil fuels is necessary to align the price of fossil fuels with their total social good. See, e.g., Carlton & Frankel, supra note 260, at 622 (“By taxing pollution or charging highway tolls in amounts equal to the social costs imposed by these activities, regulators can theoretically correct these market failures.”).

Second, and by contrast, law and economics scholars have argued that governments should financially support products that have social benefits. 265 See, e.g., Timothy Meyer, How Local Discrimination Can Promote Global Public Goods, 95 B.U. L. Rev. 1937, 1969 (2015) [hereinafter Meyer, Local Discrimination] (discussing how discriminatory measures like local content requirements can correct the undersupply of global public goods). The producers of such products do not capture the benefits from their products. Because producers of such products do not capture the resulting benefits, they will underproduce relative to what would be socially desirable. 266 See id. (explaining why governments “undersupply public goods” when they create benefits outside of the relevant jurisdiction). Subsidies, by increasing the value of producing envi­ronmentally sustainable products, can ensure that environmentally sus­tainable (or otherwise socially desirable) products appear in the market in greater numbers. This approach to subsidization has long justified subsidies for basic research into infrastructure, medicine, and tech­nology—including renewable energy technology—for which the initial market benefits may be too small to justify private investment despite significant potential social benefits from innovation

Trade rules by and large do not reflect this economic theory of sub­sidies. The SCM Agreement initially contained provisions permitting certain kinds of beneficial subsidies, but those provisions expired in 2000. 267 See SCM Agreement, supra note 10, arts. 8, 31 (explaining that green light subsidies expire five years after the SCM Agreement enters into force). Since then, WTO rules have not distinguished among subsidies based upon their purpose. 268 See Young, supra note, 144, at 6 (noting the disputes surrounding attempts to reform the SCM Agreement to incorporate environmental objectives by distinguishing between harmful and beneficial subsidies). Indeed, as Professor Margaret Young notes, discussions over subsidies for wild fisheries have struggled in part because of the need to distinguish harmful from beneficial subsidies. Some fisheries subsidies, such as subsidies for larger fleets, have negative consequences, while others, such as support for retraining, have environmental benefits by reducing the number of fish caught. See id. Many have called for the reauthorization of so-called “green light” subsidies for environmental products. 269 See, e.g., Mark Wu, E15 Initiative, Re-Examining ‘Green Light’ Subsidies in the Wake of New Green Industrial Policies 10 (2015), [] (discussing “how Article 8 might be reconstituted for those who desire its reintroduction”). To date, though, no action has been taken on this suggestion. 270 See Agreement on Subsidies and Countervailing Measures (“SCM Agreement”), WTO, [] (last visited Oct. 11, 2017) (noting only two categories of subsidies: prohibited and actionable). WTO members are currently negotiating an Environmental Goods Agreement, but the negotiations do not cover any subsidies rules. 271 Environmental Goods Agreement (EGA), WTO,
tratop_e/envir_e/ega_e.htm [] (last visited Oct. 11, 2017) (noting that negotiations are limited to “seeking to eliminate tariffs on a number of important environment-related products”).
Moreover, while govern­ments have regularly agreed to reform environmentally harmful subsidies, including both fisheries and fossil fuels, action has been relatively slow. 272 See Oby Ezekwesili, Why We Need to End Fisheries Subsidies, World Econ. F. (Oct. 2, 2015), [] (“Nearly 60% of the WTO’s membership supports controlling fisheries subsidies . . . .”); Karl Mathiesen, G7 Nations Pledge to End Fossil Fuel Subsidies by 2025, Guardian (May 27, 2016),
environment/2016/may/27/g7-nations-pledge-to-end-fossil-fuel-subsidies-by-2025 [http://] (quoting Shelagh Whitley of the Overseas Development Institute as saying that “[w]e already see [some in] the G7 going in the wrong direction since Paris” but “[j]ust because they are saying this [about fossil fuel subsidies], it’s not a fait accompli” (alterations in original) (internal quotation marks omitted)).

The result is a selective enforcement policy that compromises not only the value of fair competition but also the public’s interest in markets that produce socially responsible products. It bears noting that selective enforcement will not always have these larger costs. If selective enforce­ment disadvantages a run-of-the-mill competing product, the policy may merely be inefficient. But natural resource scarcity has become one of the central issues of the twenty-first century. 273 See Harlan Grant Cohen, International Law in a Time of Scarcity: An Introduction, 42 Ga. J. Int’l & Comp. L. 1, 1–6 (2013) (discussing how international law responds to increasingly scarce natural resources). It is imperative to identify and develop products that can replace scarce natural resources. A trade policy that actively harms that interest by reinforcing the market position of natural resources and natural resource–intensive products is, like the consumption of the natural resources themselves, ultimately unsustainable.

*   *   *

While trade rules may not distinguish between subsidies based on their social costs and benefits, enforcement of trade rules should at least aim to do no harm. Enforcement policies could, of course, focus on those subsidies that are harmful. Doing so would mimic the effect of the SCM’s expired “green light” subsidy provisions. Instead, through litiga­tion costs, liability, and the knock-on effect of lost investment and higher capital costs, enforcement policies counteract the effect of environmen­tally sustainable subsidies while leaving in place environmen­tally harmful subsidies. In other words, selective enforcement operates as an implicit subsidy for environmentally harmful incumbent products like fossil fuels and wild fisheries. The next Part turns to the way that subsidy works.

IV. Selective Enforcement in Energy and Fisheries

Environmentally sustainable products are often new products, pred­icated on technological innovation, that seek to compete with and ultimately replace incumbent, natural resource–intensive products. En­couraging this competition through government support can have beneficial consequences. Instead, as this Part demonstrates, the selective enforcement of trade laws places innovative environmentally friendly products at a competitive disadvantage. This Part makes concrete the harms theorized in Part III, first in the energy sector (section IV.A) and then in the fisheries sector (section IV.B). Each section begins by discuss­ing evidence that government support is necessary to help sus­tainable products compete with natural resource–intensive products. This Part then turns to evaluating the effects of some of the individual cases dis­cussed in Part II, demonstrating that they resulted in litigation costs and lost government support.

A. The Effects of Selective Enforcement in Energy

Reducing the consumption of fossil fuels and meeting increasing energy demand through renewable energy are imperative if the world is to avoid the most catastrophic effects of global climate change. 274 See Intergovernmental Panel on Climate Change, supra note 68, at 208 (“Without additional efforts to reduce [greenhouse gas]  emissions beyond those in place today, global emissions growth is expected to persist driven by growth in global population and economic activities.”). In order to achieve this goal, public officials and experts have repeatedly called for greater innovation and investment in order to close the gap between renewables and fossil fuels. In May 2016, for instance, govern­ment officials in charge of energy policy convened for the Clean Energy Ministerial in San Francisco. 275 Energy Secretary Ernest Moniz Says Government Can Help Clean Energy Innovation, NPR (May 29, 2016), (on file with the Columbia Law Review). The energy ministers in attendance dis­cussed plans to achieve the goals established by the Paris Agreement on Climate Change, including by “spurring companies to develop new, cleaner technologies.” 276 Id. Former U.S. Energy Secretary Ernest Moniz em­phasized the importance of government support in this task, saying, “I think the government role is often undersold in the way it permeates across our innovation system.” 277 Id. British Secretary Ed Davey has touted his government’s investments in renewable energy in response to “an historic legacy of underinvestment and neglect that threatened to un­dermine the whole economy.” 278 Press Association, Investment in UK Renewable Energy Sector Almost £8bn in 2013, Guardian (July 17, 2014),
investment-uk-renewable-energy-sector-8bn [].

Social science studies have confirmed the important role of public policy in supporting private innovation in renewables. Nick Johnstone, Ivan Haščič, and David Popp have found that “[government] [i]nvestment in renewable energy sources—wind, solar, geothermal, ocean, biomass, and waste-to-energy—can significantly contribute to the realization of public environmental objectives.” 279 Nick Johnstone, Ivan Haščič & David Popp, Renewable Energy Policies and Technological Innovation: Evidence Based on Patent Counts, 45 Envtl. & Resource Econ. 133, 134 (2010). The need for invest­ment in renewa­bles, including from the government, stems from the uncertainties attached to developing and deploying new energy technol­ogies. Investments in new technology may not pay off, and the returns on those investments depend in critical part on fossil fuel prices. Low fossil fuel prices give electricity generators and consumers little reason to switch to renewables.

As Carolyn Fischer and Richard Newell note, a “production subsidy for renewable energy improves the competitiveness of these sources vis-à-vis fossil fuels” by reducing their production costs and therefore boosting profits. 280 Carolyn Fischer & Richard G. Newell, Environmental and Technology Policies for Climate Mitigation, 55 J. Envtl. Econ. & Mgmt. 142, 144 (2008) (emphasis omitted). Subsidies for renewable energy can “reduce[] the risk of invest­ments and offer[] a secure basis” for companies to innovate and expand capacity. 281 Ger Klaassen, Asami Miketa, Katarina Larsen & Thomas Sundqvist, The Impact of R&D on Innovation for Wind Energy in Denmark, Germany and the United Kingdom, 54 Ecological Econ. 227, 229 (2005) (emphasis omitted). Innovation and expansion are key for a sector trying to compete with the established fossil fuel industry, which has had years to invest in infrastructure and develop economies of scale.

Despite the critical role of government support for clean energy innovation, selective enforcement has only reinforced the market posi­tion of fossil fuels. Both aggregate data on energy subsidies and the outcomes of individual trade disputes demonstrate this fact. 2009 is a useful benchmark for comparing the effects of selective enforcement on global energy subsidies. 282 See supra section II.A. As discussed in Part II, the pattern of selective enforcement in the energy sector began in earnest that year, with renew­ables becoming the subject of trade cases while fossil fuels remained beyond the reach of trade authorities. 283 See supra section II.A. As a consequence, trade law has not led to any reduction in fossil fuel subsidies and, indeed, fossil fuel subsidies have continued to benefit from generous government support.

Start with the aggregate data. We would like to know counterfactu­ally if, absent selective enforcement, the ratio between fossil fuel subsi­dies and renewable energy subsidies would be more favorable to renewable energy than it is currently. That is, would renewable energy subsidies be higher or fossil fuel subsidies lower? Given their environ­mental benefits, one might hypothesize that renewable energy subsidies would grow at a faster rate over time, causing the ratio to adjust in favor of renewable en­ergy. The public good that renewable energy subsidies create, in terms of reducing greenhouse gas emissions, lowers the real cost of such subsidies. Conversely, the environmental externalities cre­ated by fossil fuels raise the real cost of fossil fuel subsidies.

In fact, however, we see the opposite pattern. Fossil fuel subsidies have maintained their ratio vis-à-vis renewable energy subsidies and, in absolute dollar terms, have grown faster. Fossil fuel subsidies can be divided into two kinds: subsidies for production and subsidies for con­sumption. 284 This distinction is a bit artificial, although it is commonly used in the literature. A subsidy for production can, of course, spur consumption by lowering prices. Likewise, if structured properly a consumption subsidy can spur production to meet increased demand. The International Energy Agency (IEA) estimates that from 2009 to 2014, fossil fuel consumption subsidies increased from $300 billion per year 285 Int’l Energy Agency, World Energy Outlook 2011, at 508 (2011),
publications/freepublications/publication/WEO2011_WEB.pdf [] [hereinafter World Energy Outlook 2011].
to $490 billion. 286 Int’l Energy Agency, World Energy Outlook 2015, at 27 (2015),
publications/freepublications/publication/WEO2015.pdf [] [hereinafter World Energy Outlook 2015].
Indeed, fossil fuel consumption subsidies peaked at approximately $550 billion in both 2012 and 2013. 287 Int’l Energy Agency, World Energy Outlook 2013, at 25 (2013),
publications/freepublications/publication/WEO2013.pdf []; Int’l Energy Agency, World Energy Outlook 2014, at 320 (2014),
publications/freepublications/publication/WEO2014.pdf [].
Declining fuel prices accounted for the drop in subsidies from 2013 to 2014, rather than a change in government policies regarding support for fossil fuel consumption. 288 See World Energy Outlook 2015, supra note 286, at 98. Moreover, the IEA estimates that consumption subsidies would have been $610 billion in 2014 without reforms enacted following the 2009 pledge by the G20 to phase out fossil fuel subsidies. Id. at 27. Given the effect of declining prices on subsidy levels, however, skeptics wonder whether the pledge has had any real effect on policy. Kirsch & Roberts, supra note 263. In terms of production subsidies, experts esti­mate that in 2014, the G20 alone provided $444 billion in fossil fuel production subsidies. 289 Elizabeth Bast, Alex Doukas, Sam Pickard, Laurie van der Burg & Shelagh Whitley, Overseas Dev. Inst. & Oilchange Int’l, Empty Promises: G20 Subsidies to Oil, Gas and Coal Production 11 (2015),
empty_promises_full_report_update.pdf [].
Taken together, then, total fossil fuel subsidies in 2014 reached, at a minimum, $934 billion. 290 See id. (noting that G20 governments provide $444 billion per year in fossil fuel production subsidies); World Energy Outlook 2015, supra note 286, at 27 (estimating $490 billion in fossil fuel consumption subsidies).

Subsidies have also increased on the renewable side of the ledger, though fossil fuel subsidies continue to dwarf renewable subsidies. In 2009, renewable energy subsidies—including subsidies for both con­sumption and production—totaled $60 billion. 291 World Energy Outlook 2011, supra note 285, at 508. In 2014, they totaled approximately $135 billion. 292 World Energy Outlook 2015, supra note 286, at 343. Despite this growth, the ratio of renewable energy subsidies to fossil fuel subsidies has remained relatively stable over these five years. In 2009, the IEA reports that fossil fuel consumption subsidies were five times renewable energy subsidies ($300 billion to $60 billion). 293 World Energy Outlook 2011, supra note 285, at 508. Confusingly, the IEA typically compares only consumption subsidies for fossil fuels to all renewable energy subsidies. Id. That ratio came down slightly, with fossil fuel consumption subsidies 3.6 times the size of all renewable energy subsidies in 2014 ($490 billion 294 World Energy Outlook 2015, supra note 286, at 27. compared to $135 billion 295 Id. at 343. ). Using the combined estimate of both consumption and production subsidies above ($934 billion), fossil fuels received approximately seven times the amount of subsidies today that renewables received in 2014. Moreover, in absolute terms, renewable energy subsidies increased by only $75 billion, while fossil fuel consumption subsidies alone (not including production subsidies) increased by $190 billion.

This evidence demonstrates a correlation between selective enforcement and a discrepancy in government support for renewable energy—a discrepancy that governments themselves have pledged to close by reducing fossil fuel subsidies. 296 Leaders’ Statement, The Pittsburgh Summit, para. 24 (Sept. 24–25, 2009), http://
leaders_statement_250909.pdf [] (committing G20 governments to “phase out and rationalize over the medium term inefficient fossil fuel subsidies” (emphasis omitted)).
To be sure, this aggregate data is not conclusive evidence of the impact of selective enforcement. Selective enforcement is surely not the only cause of this policy failure. The power of the fossil fuel lobby and downstream industries that benefit from cheap fuels plays an important role. But part of the way in which lobby­ing affects subsidies policies is through the kinds of trade cases gov­ern­ments bring. And selective enforcement does demonstrably reduce the availability of government support for renewable energy. Starting in 2009, governments have exclusively brought trade cases challenging gov­ernment support in the energy sector against renewable energy. During that same period, absolute support for fossil fuels grew approximately two and a half times as much as support for renewable energy ($190 billion compared to $75 billion). 297 See supra notes 291–295 and accompanying text.

Beyond aggregate data on subsidies, one can see the direct impact of selective enforcement by looking at the effects of some cases discussed in Part II. Consider the impact of the Canada—Renewables case. There, the European Union and Japan challenged Ontario’s Feed-in Tariff Program as violating the WTO’s subsidies rules, as well as discriminating against foreign products in violation of the national treatment rule found in the GATT and the Agreement on Trade-Related Investment Measures. 298 Canada—Renewables Appellate Body Report, supra note 14, paras. 1.6–1.7. A feed-in tariff is a program that guarantees electricity producers, in this case those generating electricity from renewable sources, a certain rate. 299 Feed-in Tariffs, Nat’l Renewable Energy Lab., [] (last visited Oct. 30, 2017). These rates are set high enough to encourage investment in a sector in which it might not otherwise be profitable to invest. 300 Id.

The EU and Japan ultimately prevailed on the discrimination claim. 301 Canada—Renewables Appellate Body Report, supra note 14. In response, Ontario canceled its feed-in tariffs for all large pro­jects. 302 Communication from Canada, Canada—Certain Measures Affecting the Renewable Energy Sector, WTO Docs. WT/DS412/19, WT/DS426/19 (June 6, 2014) [hereinafter Communication from Canada]; see also Gipe, supra note 14. In theory, Canada could have simply eliminated the local content requirements that the WTO found discriminatory and otherwise continued its program. As Canada’s communication to the WTO indicated, however, the Ontario legislature was unsuccessful in amending the legislation governing the feed-in tariff. See Communication from Canada, supra, at 1. As I have argued elsewhere, this kind of stalemate is likely, especially at the subnational level. Discriminatory conditions are often necessary to pass green subsidies. Without the discriminatory subsidy, the subsidy vanishes entirely. See Meyer, Local Discrimination, supra note 265, at 1942. Ontario did leave its feed-in tariff in place for smaller pro­grams. 303 Communication from Canada, supra note 302; Gipe, supra note 14. The province capped the program’s availability, however, and cut the rates it paid these smaller programs by 25% to 39%. 304 Christian Roselund, Ontario Reduces Feed-in Tariffs 1–5% for Solar PV, PV Mag. (Oct. 8, 2014), []. The sub­sidy for investing in renewable energy came in the form of guaranteed higher rates, so the combination of eliminating the program for large projects and cutting rates for those projects that remain amounts to a dramatic reduction in government support for renewable energy in Ontario.

The long-term effects of the sudden cancellation of the feed-in tariff remain to be seen. On the positive side, at least some renewable energy equipment producers seem to have taken advantage of the Feed-in Tariff Program while it existed to become profitable without government sup­port. 305 See Peter Kuitenbrouwer, Ontario Solar Industry Finds Place in the Sun After Green Energy Flameout, Fin. Post (Dec. 28, 2015), (on file with the Columbia Law Review) (last updated Jan. 4, 2016). Establishing such producers and then weaning them off govern­ment support as they become cost-competitive is the ideal for programs like Ontario’s Feed-in Tariff Program. On the other hand, though, these producers have become export focused, targeting foreign markets like the United States that still have incentives in place for renewable en­ergy. 306 Id. (quoting Ontario solar panel producers as having “adjusted our strategy as a result of the U.S. tax incentives” after Ontario cancelled the Feed-in Tariff Program (internal quotation marks omitted)). Those Ontario producers thus continue to rely on government incentives in order to be competitive. This fact has led critics to fear that the absence of the feed-in tariff endangers Ontario’s own long-term tran­sition to renewable energy. 307 See, e.g., Leah Stokes, Opinion, Ontario’s Backward Step on Renewable Energy, Toronto Star (July 22, 2013),
22/ontarios_backward_step_on_renewable_energy.html [].

The United States’ WTO case challenging discriminatory Chinese subsidies for wind power provides another case in point. Following the filing of the case, China agreed to remove its subsidies, which “could [have] collectively total[ed] several hundred million dollars” between 2008 and 2011. 308 Palmer & Walet, supra note 114. Just as in Ontario, however, industry experts speculated that at least several firms had used the subsidies to become sufficiently competitive to survive without the subsidies. 309 Id. (quoting an investment analyst as saying that “Chinese wind power companies have reached a stage that, regardless of subsidies, they will head out and aim for overseas markets if there are opportunities” (internal quotation marks omitted)). Indeed, Chinese subsidies and the cheap wind power they make available are in part responsible for the growth in wind energy in other countries, including the United States. 310 See, e.g., Wind Energy Globally on the Rise: Policies Drive Growth in China, the US and Germany, UN Framework Convention on Climate Change (Aug. 19, 2015), [] (highlighting the growth in wind power in the United States, Germany, and Europe more broadly).

Trade restrictions on biofuels offer a third case in point. Biofuels compete directly with fossil fuels. Ethanol and biodiesel can substitute for fossil fuels, or blend into gasoline and diesel fuels. 311 See Alternative Vehicle Fuels, EPA, [] (last updated Oct. 24, 2016). As the price of oil falls, traders consume fewer biofuels because blending becomes less cost-effective. The long-term relationship between the two also affects energy consumers’ fuel choices.

Like other forms of renewable energy, biofuels have been criticized as not cost competitive with fossil fuels. Writing in Forbes, Jude Clemente argues that “[a] major hurdle to [the] commercialization of biofuels is their cost in comparison to petroleum-based fuels.” 312 Jude Clemente, Why Biofuels Can’t Replace Oil, Forbes (June 17, 2015), []. He notes that between 2008 and 2022, biofuels are expected to receive approximately $400 billion in subsidies. 313 Id. As noted above, subsidies for fossil fuels ex­ceed this amount in a single year. 314 See supra notes 285–287 and accompanying text. This fact makes the cost-competitive­ness argument against biofuel subsidies difficult to sustain.

Yet just as subsidies for solar and wind have been under attack, so too has government support for biofuels. As discussed in Part II, Argentina and the EU have engaged in a trade war at the WTO over bio­diesel. In 2008, the EU issued rules requiring that 10% of land transport energy come from biofuels by 2020. 315 Jonathan Stearns, Argentina, Indonesia Hit with EU Tariffs on Biodiesel, Bloomberg Tech. (Nov. 19, 2013), (on file with the Columbia Law Review). This rule stimulated investment in biofuels, including biodiesel, within Europe. European producers, how­ever, found themselves losing market share to foreign producers. Decid­ing that the loss of market share stemmed from dumping and illegal subsidies, the European Union imposed antidumping and coun­tervailing duties on U.S. biodiesel in 2009. 316 Id. Filling the void left by U.S. biodiesel, Indonesian and Argentinian shares of the European market rose from 9.1% in 2009 to 19.3% in 2012. 317 Id.

The European Biodiesel Board, an industry group, responded by filing a complaint with the European Commission, which then launched both antidumping and countervailing-duty investigations. 318 Panel Report, European Union—Anti-Dumping Measures on Biodiesel from Argentina, para. 7.179, WTO Doc. WT/DS473/R (adopted Oct. 26, 2016) [hereinafter EU—Biodiesel]. In 2013, the EU imposed antidumping duties on biodiesel producers from both countries. 319 Council Implementing Regulation 1194/2013, supra note 15, at 2–26. The crux of the complaint against Argentina was that it main­tained a selective export tax. 320 EU—Biodiesel, supra note 318, para. 7.180. In short, Argentina had an export tax on soy and soybeans, used to make biodiesel, that was higher than the export tax on the biodiesel itself. 321 Id. Preferential tax treatment of this kind is often viewed as a financial contribution by a government. 322 See SCM Agreement, supra note 10, art. 1 (discussing how revenue otherwise foregone, such as tax breaks, can qualify as a financial contribution by a government). In this case, Argentina provided financial support to biodiesel producers by foregoing the higher tax rates on soybeans in favor of the lower tax rates on biodiesel. This selective tax system meant that Argentinian biodiesel producers could acquire the goods—soybeans—necessary to make bio­diesel for less than European producers could. The EU imposed anti­dumping duties in response. In March 2016, the WTO ruled that the EU had improperly calculated the “dumping margin”—the extent to which pricing deviates from “normal”—because of how it treated the selective export tax. 323 EU—Biodiesel, supra note 318, paras. 7.20, 8.1.c.

Despite this ruling, the effect of the EU’s antidumping duties on the Argentinian biodiesel sector has been profound. In 2013, Argentina exported only one third of the amount of biofuels to Europe that it had sold there in 2011 and 2012. 324 Carlos Manzoni, La Industria Local de Biodiésel Trata de Salvarse con el Mercado Interno [The Local Biodiesel Industry Tries to Save Itself with the Internal Market], La Nacion (Nov. 10, 2013), []. Some Argentine producers had stopped production at the end of 2012 and large exporters found that 60% of their production capacity lay idle. 325 Some speculate that the drought and subsequent soy scarcity contributed to, if not caused, the biofuel production halt in 2012 and 2013. Biodiesel Argentina Noticias, supra note 15. The Argentine Biofuels and Hydrogen Association estimates that Argentinian biofuel production fell by more than 30% from 2014 to 2015, while exports fell by an astonishing 55%. 326 Charles Newbery, Argentina Raises Biodiesel Prices, Export Taxes, Platts (Feb. 16, 2016), [].

Exports recovered in 2016 with the opening of the U.S. market to Argentinian biofuels. 327 Maximilian Heath, Argentina Biodiesel Exports Surge as Sales to U.S. Grow, Reuters (Sept. 6, 2016), []. The United States was estimated to have received 75% to 80% of Argentina’s 2016 exports. Id. But two new trade conflicts now threaten the Argentine biofuel sector’s resurgence. In late 2016, Peru—Argentina’s second largest export market—slapped antidumping duties on Argentinian biofuels, claiming that the fuel was unlawfully subsidized by the Argentinian government. 328 Meghan Sapp, Peru Accuses Argentina of Dumping Biodiesel, Biofuels Dig. (Apr. 21, 2016), []. In August 2017, the United States—Argentina’s largest export market—imposed significant preliminary countervailing duties on Argentinian biofuels. 329 Press Release, U.S. Dep’t of Commerce, U.S. Department of Commerce Issues Affirmative Preliminary Countervailing Duty Determinations on Biodiesel from Argentina and Indonesia (Aug. 22, 2017), [
Prior to the imposition of these duties, the head of Argentina’s biodiesel trade association had worried that “[i]f a sanction is applied against Argentina in the U.S. market, our exports will no longer be viable. At this point, there is no alternative market.” 330 Maximiliano Rizzi, Dumping Complaint Could Kill Argentine Biodiesel Exports, Groups Says [sic], Reuters (Mar. 31, 2017), []. Following the United States’s imposition of duties, Argentinian producers halted exports to the United States. 331 Maximilian Heath, Argentina Bets on European Biodiesel Market After U.S. Imposes Duties, Reuters (Aug. 23, 2017), [].

Taken together, these cases demonstrate the powerful impact that selective enforcement can have. In Canada, China, and Argentina, gov­ernments and industry groups have incurred litigation costs defending themselves against trade claims. They have also incurred significant liabil­ity. The governments of Ontario and China have been forced to withdraw green subsidies, a significant loss for the renewable sector. Argentinian bio­fuel producers have been forced to pay higher duties in all of their major export markets, cutting into their profits and poten­tially rendering the entire industry nonviable.

Yet the kind of subsidies for renewable energy challenged here are necessary simply to create a level playing field with the heavily subsidized fossil fuel industry. Fossil fuel companies benefit from substantial gov­ernment support without suffering any consequences under the interna­tional trade regime. Fossil fuel prices can be lower and company profits higher because of these subsidies. Renewable subsidies continue to lag behind fossil fuel subsidies, despite the beneficial environmental impacts of such subsidies. Moreover, renewable subsidies have been reduced, and their effectiveness undercut, by trade enforcement. While this enforce­ment might well improve competitiveness among renewable energy sources, it severely disadvantages renewable energy against its primary competition, fossil fuels.

B. The Effects of Selective Enforcement in Fisheries

Investment in innovation is as important for fisheries as it is for en­ergy. Seafood is the most highly traded food commodity in the world. 332 Terazono, Salmon Leaps Past Shrimp, supra note 141. Wild fisheries cannot continue to produce sufficient fish to feed the world’s population. 333 See State of World Fisheries 2016, supra note 187, at 5–6. Preserving fish stocks thus means either reducing the worldwide consumption of fish (unlikely) or developing alternative means of satisfying the world’s need for fish.

The FAO reports that as of 2013, 31.4% of the world’s fisheries are overfished, meaning that the level of fishing is biologically unsustaina­ble. 334 Id. at 38. Another 58.1% are fully fished, meaning that they “have no poten­tial for increases in production.” 335 Id. at 6, 38. One of the most dire predic­tions, published in the journal Science, is that stocks of all species cur­rently fished for food will collapse by 2048 if fishing practices do not change. 336 Boris Worm et al., Impacts of Biodiversity Loss on Ocean Ecosystem Services, 314 Science 787, 790 (2006). Consequently, capture fisheries are unlikely to be able to satisfy the global demand for fish.

This limitation has spurred the dramatic growth in aquaculture. Global aquaculture production grew at an average annual rate of 8.6% from 1983 to 2012. 337 Org. for Econ. Cooperation & Dev., OECD Review of Fisheries: Policies and Summary Statistics 2015, at 16 (2015) [hereinafter OECD 2015]. By contrast, global harvesting of wild fish peaked in 1996 and has been flat or declining since then. 338 Id. at 14. Wild harvest in countries that are members to the Organization for Economic Cooperation and Development (OECD) declined 15% between 2005 and 2015 and 39% since its peak in 1988. 339 Id. In 2015, 80% of global aquacul­ture production was located in just five countries: China, India, Vietnam, Indonesia, and Bangladesh. 340 Id. at 10. By contrast, the top five OECD produc­ers—Norway, Chile, Japan, South Korea, and the United States—accounted for only 6% of global produc­tion. 341 Id. In 2014, humans con­sumed more farmed fish than wild-caught fish for the first time ever. 342 Emiko Terazono, Farmed Fish on Course to Overtake Wild Catch in 2019, Fin. Times (May 28, 2017), (on file with the Columbia Law Review) (“Farmed fish became the chief source of fish consumed by humans in 2014.”).

Although some have hailed aquaculture as a possible way to relieve the strain on wild fish stocks, 343 See, e.g., Farmed in the EU, European Comm’n,
inseparable/en/farmed-eu [] (last updated Jan. 25, 2018) (“Fish farming, or aquaculture, can help to relieve this burden on wild fishes when it comes to satisfying our ever growing demand for fish . . . .”).
the environmental benefits of contempo­rary aquaculture are mixed in practice. First, fish production from the wild continues to exceed fish production from aquaculture due to wild fish that are used as feed for farmed fish. 344 OECD 2015, supra note 337, at 16. Hence, current aquaculture practices—and consumers’ taste for carnivorous fish such as shrimp and salmon—actually create demand for certain wild-caught fish, even as aquaculture satisfies demand for other wild-caught fish. Second, natural habitats are often destroyed to make way for fish farms. 345 See J.H. Primavera, Overcoming the Impacts of Aquaculture on the Coastal Zone, 49 Ocean & Coastal Mgmt. 531, 533 (2006) (describing the destruction of mangrove forests, estuaries, and tidal creeks to make way for fish ponds and cages). The loss of mangrove forests, which have suffered severe declines in the Asian countries where they are predominantly located, is perhaps the most commonly cited environmental harm arising from aquaculture. 346 See id. at 534 (“Mangrove conversion to shrimp ponds is the single major factor that has contributed to the negative press received by aquaculture.”). Other risks include the release of chemicals or antibiotics used to clean fish farms into the gen­eral water supply, the release of waste, and the introduction of farmed fish into the wild in ways that disrupt local ecosystems (if, for example, they are non-native). 347 See id. at 535–37 (listing and discussing the environmental risks of aquaculture).

The long-term viability of fish as a source of human food, as well as the potential for aquaculture to relieve the strain on capture fisheries without causing additional environmental damage, thus depends on in­vestment and innovation. As the Economist put it, “By boosting basic research and infrastructure for aquaculture, governments could hasten a welcome trend.” 348 Net Positive: How to Stop Overfishing on the High Seas, Economist (July 16, 2016), (on file with the Columbia Law Review) (“So in parallel with efforts to protect wild stocks, another push is needed: to encourage the development of aquaculture . . . .”).

Governments and international organizations have recognized this need. In 2015, the OECD launched its “Fisheries and Aquaculture Innovation Platform” (FAIP). 349 See About the Platform, Fisheries and Aquaculture Innovation Platform (FAIP), Org. for Econ. Cooperation & Dev., [] (last visited Oct. 11, 2017). FAIP aims to collect data on nations’ efforts to support innovation for both wild catch fisheries and aquacul­ture. 350 See Towards Innovation, Fisheries and Aquaculture Innovation Platform (FAIP), Org. for Econ. Cooperation & Dev., [] (last visited Oct. 11, 2017). More specifically, FAIP attempts to estimate both the percent of government fisheries’ spending devoted to research and development, as well as the level of patenting activity in different countries. See id. Patenting activity is broken down into three categories: (1) harvesting or fisheries-technology patents (also called wild-catch patents), (2) aquaculture-technology patents, and (3) patents in new products and markets. See Data, Fisheries and Aquaculture Innovation Platform (FAIP), Org. for Econ. Cooperation & Dev.,
fisheries-innovation/data/ [] (last visited Oct. 11, 2017). Interestingly, large developed economies like the United States, Canada, and the United Kingdom produce many more harvesting patents, while countries like Norway and China produce more aquaculture patents (although the overall rate of patent activity in the latter countries is much lower). Id.
Similarly, the FAO has argued that “a lack of technical innovation” constrains aquaculture development, especially among the small-scale producers that are prevalent in Asia and Africa. 351 State of World Fisheries 2016, supra note 187, at 26.

Individual nations have placed a similar emphasis on investment and innovation. For example, in June 2016 the Canadian Council of Fisheries and Aquaculture Ministries (a group composed of the Canadian federal and provincial fisheries and aquaculture ministers) released an “Aquaculture Development Strategy” designed to capitalize on “aquacul­ture’s potential to create jobs, economic growth and prosperity in remote, rural, coastal and Indigenous communities.” 352 Ministers Restate Aquaculture Importance and Canadian Market Expansion, Fish Info. & Servs. (June 23, 2016), []. The Strategy noted the ministers’ view that “[i]nvestment in innovation and scientific research is required to support further development of this important industry. Investment in these areas will foster continual improvement in environmental sustainability and economic viability while increasing public confidence.” 353 Canadian Council of Fisheries & Aquaculture Ministers, Aquaculture Development Strategy 2016–2019, at 3 (2016), [].

Bangladesh provides another case in point. With the assistance of the FAO, Bangladesh released a “National Aquaculture Development Strategy and Action Plan” for 2013 to 2020. 354 Ministry of Fisheries & Livestock, Gov’t of the People’s Republic of Bangl. & UN Food & Agric. Org., National Aquaculture Development Strategy and Action Plan of Bangladesh 2013–2020 (2014), []. The report notes that Bangladesh has become the fifth-largest producer of aquaculture prod­ucts in the world, with its farmed fish supply increasing by more than 200% during the first decade of the twenty-first century. 355 Id. at vii. In the report, Bangladesh argues that aquaculture can fulfill a number of important strategic objectives, including assuring food and nutrition security and creating employment opportunities, especially for small-scale fish farmers. 356 Id. at 2. Critically, the report recognizes that while the growth of aqua­culture presents opportunities, it also presents challenges if it is not managed and pushed to develop in environmentally and socially sustain­able ways. 357 See id. at 1.

Innovation plays a key role in how Bangladesh sees aquaculture developing in a beneficial way. For example, Bangladesh intends to “de­velop[] low-cost aquaculture technologies,” “improv[e] hatchery man­age­ment practices and genetic quality of culture fish spe­cies,”  “strengthen[] research and development,” and “enhanc[e] commer­cial aquaculture productivity under a public–private partner­ship.” 358 Id. at 5. The explicit purpose of the plan is to guide “public investments in develop­ment and support services [for aquaculture] and encourag[e] private investment in aqua­culture enterprises.” 359 Id. at 2.

Despite these ambitions, government support for aquaculture is still dwarfed by government support for wild fisheries. Like fossil fuel subsidies, subsidies for capture fisheries are about a century old. Major fishing nations such as Norway introduced fisheries subsidies as early as the 1920s as part of their industrialization policies. 360 UN Env’t Programme, Fisheries Subsidies, Sustainable Development and the WTO 115–16 (Anja von Moltke ed., 2011). Subsidies became more widespread in the 1930s and 1940s when they aimed to encourage investment and modernization in the fishing industry. 361 See William E. Schrank, UN Food & Agric. Org., Introducing Fisheries Subsidies 15, 18, 20–22 (2003), [
H2HH-8RMX] (discussing fisheries subsidies introduced by the United States, Canada, Norway, and Iceland).
Over the next several decades, fisheries subsidies spurred improvements in vessel and gear design, increased vessel capacity, and led to improvements in preservation methods, thereby allowing vessels to range farther out to sea to catch fish. 362 See UN Food & Agric. Org., Marine Fisheries and the Law of the Sea: A Decade of Change 1, 4–5, 21–24 (1992), [].

By 1992, however, the FAO sounded the alarm: Fisheries subsidies were leading to dangerous levels of overfishing, putting the world’s fish stocks at risk of extinction. 363 See id. at 6–7. The FAO initially estimated global fisheries subsidies at approximately $54 billion per year, a number that has proven to be too high, even adjusting for inflation. 364 Id.; U. Rashid Sumaila et al., Directorate-Gen. for Internal Policies, Eur. Parliament, Global Fisheries Subsidies 11 (2013),
join/2013/513978/IPOL-PECH_NT(2013)513978_EN.pdf [].
A more recent estimate from the European Parliament in 2013 put global fisheries subsidies at $35 billion annually in 2009. 365 Sumaila et al., supra note 364, at 27. Of that, $20 billion are “capacity-enhancing subsidies”—subsidies that directly increase the global fish catch. 366 Id. Other subsidies are either ambiguous in their effects on fishing practices—such as social security programs for fishermen—or may lead to more sustainable fishing practices—such as subsidies for research into best management practices or vessel decommissioning. 367 Id. at 22, 25, 30. Interestingly, the European Parliament estimates that 22% of fisheries subsidies are also energy subsidies in the form of fuel subsidies for vessels. Id. at 27. The largest subsidizing countries in the world are, in descending order, Japan, China, the United States, Russia, and Micronesia. 368 Id. at 31–32. Except for those in the United States, the great majority of each country’s subsidies are capacity enhancing. 369 Id.

Just as with fossil fuels, trade cases—in particular trade remedies that target government support without the need to bring a WTO case first—have resulted in withdrawn support for aquaculture. Most recently, Turkey agreed to withdraw subsidies for farmed sea bass and sea bream. 370 See Commission Implementing Decision 2016/1360 of Aug. 8, 2016, Terminating the Anti-Subsidy Proceeding Concerning Imports of European Sea Bass and Gilthead Sea Bream Originating in Turkey, 2016 O.J. (L 215) 31, 32 (EU) [hereinafter Turkey Bass and Bream Decision]; see also EC Puts End to Turkish Seabream and Seabass Import Dispute, Fish Info. & Servs. (Aug. 11, 2016),
monthyear=8-.2016&day=11&id=86265&l=e&country=0&special=&ndb=1&df=0 [http://] (noting Turkey’s decision to terminate its main subsidy program).
The European Commission initiated an antisubsidies investiga­tion against Turkish sea bass and sea bream on August 14, 2015. 371 Turkey Bass and Bream Decision, supra note 370, para. 1. Following consultations between the Commission and the Turkish gov­ernment, Turkey announced in May 2016 that it was canceling its subsidies retroactive to January 1, 2016. 372 Id. para. 8. In response, the Spanish trade organization that initially sought the investigation withdrew its complaint and the Commission dropped the investigation. 373 Id. paras. 9–10. Notably, the revoked subsidies include direct subsidies for the organic farming of sea bass, the development of which would enhance the health benefits flowing from the fish trade. 374 See Turkish Bass and Bream Notice, supra note 172, at 4 (describing the subsidies as “[d]irect transfer[s] of funds . . . for the organic production of sea bass and sea bream”).

As another example, Vietnam’s catfish farms showcase all three effects: litigation costs, liability, and lost investment. As part of normaliz­ing its relations after the Vietnam War, the United States lifted an embargo on Vietnamese products in 1994. 375 See Irene Brambilla, Guido Porto & Alessandro Tarozzi, Adjusting to Trade Policy: Evidence from U.S. Antidumping Duties on Vietnamese Catfish, 94 Rev. Econ. & Stat. 304, 304–06 (2012) (describing the exportation of Vietnamese catfish to the United States after the termination of the latter’s embargo). Catfish farming quickly became a major source of economic livelihood, especially in the Mekong Delta. 376 See id. (describing the impact of trade with the United States on Mekong Delta producers). By 2002, Vietnamese producers exported 50% of their catfish to the United States and had almost 20% of the U.S. catfish market. 377 Id. at 305. The Association of Catfish Farmers of America sought protection from the U.S. government, first in the form of labeling measures and then in the form of antidumping duties. 378 See id. at 306 (describing American catfish producers’ efforts to restrict the use of the label “catfish” to U.S. subspecies and a subsequent lawsuit against Vietnam). The U.S. government imposed antidump­ing duties in the summer of 2003, justified in part on the grounds that Vietnamese catfish sold at artificially low prices due to Vietnamese gov­ernment subsidies. 379 See David Lamb, U.S., Vietnam in Dispute over Catfish Exports, L.A. Times (Dec. 8, 2002), [] (“The ruling appears to support the U.S. catfish farmers’ contention that Mekong Delta farmers sell their catfish at artificially low prices because of government subsidies.”).

By the end of 2003, Vietnamese catfish exports to the United States had dropped by 85% from 2002. 380 See Brambilla et al., supra note 375, at 306 (showing that monthly Vietnamese catfish imports dropped from nearly 380 tons in 2002 to 56 tons in late 2003). The loss of market share owing to higher duties clearly illustrates the liability effects flowing from the appli­cation of trade remedies. But Vietnam’s experience also illustrates the effects of litigation costs and lost investment as well. On the one hand, Vietnamese producers struggled to provide the information required in an antidumping investigation, including information about labor costs and sales contracts, which must be provided on English-language forms. 381 See Do Thanh Cong, Catfish, Shrimp, and the WTO: Vietnam Loses Its Innocence, 43 Vand. J. Transnat’l L. 1235, 1250 (2010) (“Many Vietnamese exporters found it difficult to understand the [antidumping] questionnaires because they are com­plicated and in English . . . .”). Equally importantly, the U.S. antidumping duties caused signifi­cant diversion of investment away from catfish production in Vietnam. A recent study estimated the declines in investment in catfish aquaculture to be between 28% and 62%, depending on the producer’s exposure to the U.S. market. 382 Brambilla et al., supra note 375, at 317. Such cuts in investment are crippling to the development of a sustainable aquaculture industry that can meet the needs of global food production. Little reason exists to make long-term investments in improving aquaculture practices when market access hinges on the decisions of U.S. antidumping authorities. Meanwhile, with no chal­lenges to Vietnam’s substantial subsidies for its fishing fleets, Vietnam and its fish exporters might reasonably divert their investment to captur­ing wild fish. U.S. antidumping actions, in other words, reinforce the place of Vietnamese capture fisheries at the expense of its aquaculture industry.

*   *   *

Consumption of energy and fish are central to the global economy, among the most highly traded commodities in the world. In both sectors, existing consumption relies too heavily on natural resources. Providing alternatives that meet the demand for energy and fish in a sustainable way is imperative. In both sectors, governments and researchers have confirmed over and over again the central role of investment in innova­tion. Yet as the evidence here shows, selective enforcement has consist­ently undermined natural resource substitutes. Producers in developing countries have struggled with the burden of defending themselves in trade actions in the United States and Europe. The imposition of trade remedies has closed markets and threatened to shutter entire industries, such as the Argentinian biofuel or Vietnamese catfish farming sectors. And these trade remedies have resulted in lost investment in almost every case. With no similar action nor costs imposed on them, the fossil fuel and wild fisheries industries win big.

V. Reforming Trade Law Enforcement

Fixing selective enforcement is imperative if we are to have a fair trade policy, in the sense of a trade policy that does not undermine core noncommercial values. Yet selective enforcement arises in part because of the way in which the WTO is structured. The WTO enforcement pro­cess is akin to a system of private attorneys general. Each nation brings claims when it believes its industries have been hurt, but no centralized enforcement system exists. Industries affected by a foreign government’s laws can lobby their government to impose trade remedies. Only some industries will lobby successfully. Due to resource and time constraints, governments cannot bring a trade case every time an industry is hurt. 383 Moreover, many trade law violations or unfair practices will create both winners and losers in other countries. For example, energy-intensive industries might be opposed to cases challenging fossil fuel subsidies, while renewable energy companies might favor such a challenge. Whether the government chooses to bring a case depends on which side the government intends to take. Together, differences in domestic political power and resource con­straints contribute to selective enforcement.

Selective enforcement is not necessarily problematic. Selective enforcement becomes unfair only when it stacks the deck against prod­ucts that create broader social benefits. This Part presents two proposals for reforming the WTO enforcement process to reduce this effect: (1) creating a centralized enforcement procedure for environmental prod­ucts, and (2) reforming trade remedies investigations. In both cases, the goal of reform is to address selective enforcement of trade obligations by increasing enforcement. In other words, member states should look for ways to ratchet enforcement up, rather than down.

A. Reforming the WTO Enforcement Process

The WTO should consider creating a centralized enforcement pro­cess. The cornerstone of this process would involve the WTO Secretariat identifying products that are similarly situated to products targeted by WTO dispute settlement or trade remedies. In a strong version, a WTO prosecutor could be empowered to bring cases to remove the pernicious effects of selective enforcement. In a weaker version, the list of similarly situated products would be circulated to WTO members. Significantly, in identifying similarly situated products, officials should use a test that focuses on competition in the marketplace, rather than the narrower “like products” test common in trade law.

The development of the list of similarly situated products would follow a two-step process. First, when a WTO member prevails in a dis­pute before the DSB or notifies the WTO that it has levied antidumping or countervailing duties, the WTO Secretariat would identify products that compete with the product at issue in the dispute. 384 Antidumping and countervailing-duty investigations are not, of course, WTO disputes unless a member state challenges their imposition. However, because WTO members use antidumping and countervailing-duty investigations to create protectionism and their imposition can create the same kind of harmful effects for competitive products as a WTO dispute, the centralized enforcement process should include both. Antidump­ing and countervailing-duty investigations are already subject to WTO oversight through both notification requirements and the DSU, so including them within the scope of the centralized enforcement process does not expand the WTO’s reach. For example, if this process had been in place following the CanadaRenewables case, officials would have looked for products that compete with solar panels, such as wind energy or fossil fuels. In theory, this exercise could be done for all disputes. A more efficient system, however, would require the Secretariat to undertake this task upon the request of a single member. In this way, the request to develop a list of similarly situated products would be akin to the request to form a panel: A single nation could initiate it.

Significantly, officials should not apply the ordinary WTO “like-products” test. Such a test unduly privileges a variety of extraneous fac­tors, such as the physical composition of two products or their tariff classifications. 385 See, e.g., Japan—Alcohol, supra note 238 (discussing the factors used in a like-products analysis). Instead, officials should focus on whether two products are substitutes within the marketplace. In conducting this analysis, offi­cials should employ a test like the “relevant market” analysis used in antitrust law. 386 See, e.g., United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 395 (1956) (setting out the test for relevant market analysis). The relevant market includes all products that are “reason­ably interchangeable by consumers for the same purposes.” 387 Id. Cross-elasticities of demand are used to mark out the boundaries of rele­vant markets. The EU uses a similar approach to market definition: “identify[ing] the competitors of the undertakings concerned by a par­ticular case that are capable of constraining their behavior.” 388 Mario Monti, European Comm’r for Competition Policy, Address at the Workshop on Market Definition at the Helsinki Fair Centre: Market Definition as a Cornerstone of EU Competition Policy (Oct. 5, 2001), []. Indeed, the EU applies this methodology even in the context of EU antisubsidy law, known as “state aid.” 389 See generally Leo Sleuwaegen, European Comm’n, Competitive Distortions and State Aid to Firms. How to Define the Relevant Market? (1999),
en (on file with the Columbia Law Review) (discussing the EU’s use of antitrust market definition methods in the context of state aid).
Competition agencies throughout the world—such as the Federal Trade Commission or Department of Justice in the United States, or the European Commission in the EU—are familiar with this basic inquiry, suggesting that the WTO Secretariat could credibly perform this task without significant difficulty.

Second, after defining the universe of relevant products, the Secretariat would identify which products in the relevant market benefit from the same conduct that either the WTO found unlawful or a trade remedy investigation targeted. Similarly situated products are not, after all, any products that benefit from an enforcement action. An unsubsi­dized wind farm, for instance, should benefit from the removal of subsi­dies on solar panels in response to a DSB ruling. Returning to the Canada—Renewables example, the Secretariat would ask which of the products that compete with solar panels, identified at step one, also benefit from local content requirements (the measure found unlawful).

The result would be a list of identified measures that the Secretariat has preliminarily determined both violate WTO rules and affect the competitive conditions for the product subject to the original dispute. Having compiled the list, what should the Secretariat do with it? Here, one can imagine a weak version and a strong version of the enforcement process. In the weak version, the Secretariat would circulate the list of identified measures to all WTO members. 390 Others have called for similar kinds of transparency-based measures to compensate for underenforcement at the WTO. See Chad P. Bown & Bernard M. Hoekman, Developing Countries and Enforcement of Trade Agreements: Why Dispute Settlement Is Not Enough (World Bank Dev. Research Grp., Policy Research Working Paper No. 4450, 2007), (on file with the Columbia Law Review). At that point, its involve­ment would end. Individual member states could then consider whether to initiate further proceedings challenging any of the identified measures. 391 This process is “weak” precisely because it leaves open the possibility that states would do nothing at this stage. States may be captured by, for instance, their fossil fuel industries in a way that drives them to use enforcement as an implicit subsidy for fossil fuels (and a weapon against renewable energy). The weak enforcement does not prevent this outcome. It merely creates public information that can encourage nations to take action. Such action could result simply from possessing knowledge about the effects of enforcement policies or from lobbying by other organizations responding to the list of similarly situated products.

These proceedings could, of course, include a formal complaint through the DSB. But the proceedings could also take less adversarial forms. WTO member states organize themselves into committees that correspond to each of the WTO agreements, such as the Committee on Subsidies and Countervailing Measures. These committees serve as party-led forums for informally resolving disputes. In particular, members can question each other’s practices through the committees, and the chal­lenged member has the opportunity to respond to those questions. India, for exam­ple, posed questions to the United States about certain state and local renewable energy subsidies in the Committee on Subsidies and Counter­vailing Measures and the Committee on Trade-Related Investment Measures. 392 Committee on Subsidies and Countervailing Measures, Minutes of the Regular Meeting Held on 22 April 2013, paras. 116–125, WTO Doc. G/SCM/M/85 (Aug. 5, 2013) (stating questions posed by India to the United States under Article 25.8 of the Agreement on Subsidies and Countervailing Measures); Committee on Trade-Related Investment Measures, Certain Local Content Requirements in Some of the Renewable Energy Sector Programs, Questions by India to the United States, para. 1, WTO Doc. G/TRIMS/W/117 (Apr. 17, 2013) (stating that some renewable energy programs in the United States “make the availability of incentives contingent upon the use of domestic or state specific products, which raises concerns about their compatibility with . . . Article 2 of the TRIMs Agreement read with Article III:4 of GATT 1994”).

This weak version of the enforcement process preserves the funda­mentally party-driven nature of WTO dispute settlement. The Secretariat would provide a public good to members in the form of information. This service would be especially valuable to the majority of smaller mem­bers that lack the resources to conduct this kind of inquiry themselves but might nevertheless be interested in the result. In this sense, the weak version is very similar to the Advisory Centre on WTO Law created by members to address selective enforcement against weak or capacity-constrained states. 393 See supra Part I. It tries to level up enforcement and participation in WTO dispute settlement by easing the burdens of participation.

Members themselves would still have to choose to act on this information. As a consequence, WTO dispute settlement would retain a diplomatic character, with both politics and relationships playing a large role in determining whether and how states address identified measures. This feature makes the weak form of centralization more polit­ically feasible. It is also more consistent with a view of the WTO in which the agreements are contracts among parties. Whether the parties to a contract enforce their rights is, at the end of the day, up to them.

The strong version of this enforcement process would call for a prosecutor’s office within the Secretariat with the authority to initiate WTO disputes challenging identified measures. 394 See Claus D. Zimmermann, Rethinking the Right to Initiate WTO Dispute Settlement Proceedings, 45 J. World Trade 1057, 1068–69 (2011). Such a mechanism would remove the political filter that prevents some meritorious cases from being brought. As lawyer and economist Claus Zimmermann has argued, leaving dispute initiation in the hands of governments sits uneasily with the theory underlying free trade. 395 Id. at 1062–66. Trade agreements aim to tie domestic political actors’ hands by giving them a long-run incentive to reduce trade barri­ers. Putting enforcement in the hands of govern­ments effectively “untie[s]” governments’ hands by allowing violations to go unpunished, resulting in trade barriers for foreign exports and higher prices for do­mestic consumers. 396 Id. at 1065 (internal quotation marks omitted).

The prosecutor’s mandate would be limited to addressing identified similarly situated measures. Hence, the prosecutor would not have the authority to investigate just any potential violation of WTO law. The of­fice’s focus would be on ensuring that the normal operation of the DSU does not distort the development of socially beneficial products. In this sense, the prosecutor would supplement the state-driven dispute settle­ment system, rather than replace it.

Several additional safeguards could also limit the possibility of over­reach by the prosecutor. First, as part of bringing a case the prosecutor could be required to demonstrate harm to the development of socially beneficial products in the absence of the case. For instance, before bring­ing a case against a local content requirement for fossil fuels, the prose­cutor would have to show that failing to do so would harm renewable energy products benefitting from local content requirements. This requirement that the prosecutor demonstrate harm is a departure from the norm in most WTO cases, in which “nullification and impair­ment” (trade speak for injury) is presumed. 397 DSU, supra note 54, art. 3.8.

Consider again the Canada—Renewables example. 398 See supra notes 109–111, 298–304. Recall that there the DSB adopted reports finding that Ontario’s Feed-in Tariff Program discrimi­nated against foreign producers of renewable energy equipment because, to qualify for the program, electricity generators had to purchase locally produced renewable energy equipment, a local content requirement. 399 See Canada—Renewables Appellate Body Report, supra note 14, para. 6. The Secretariat would therefore look to see whether fossil fuels or other forms of renewable energy available in Ontario also benefit from local content requirements. In each example it identified, the Secretariat would make a preliminary determination as to whether the allegedly unlawful measure is sufficiently similar to the measure found illegal by the WTO, and hence illegal itself. This determination would be based upon an unbiased analysis of both the similarities and the differences between the two measures. Moreover, the Secretariat would investigate whether those benefits harmed the competitive opportunities in Ontario for solar panels, the primary type of renewable energy equipment at issue in Canada—Renewables. The goal, of course, is to ensure that solar panels are not disadvantaged in the marketplace because they happened to be the target of a dispute, while other products with which solar panels compete continue to receive and benefit from unlawful measures.

Second, a pretrial chamber could be created to evaluate the merits of disputes that the prosecutor wishes to initiate. The pretrial chamber would authorize the prosecutor to initiate a dispute upon a preliminary showing that the identified measure is both inconsistent with WTO rules and affects the competitive conditions for similarly situated products involved in other disputes. Absent such authorization, the prosecutor could not initiate a dispute. The initiation of a dispute would also be sub­ject to the DSU’s ordinary reverse consensus rule. In other words, the prosecutor would notify member states of her intention to initiate a dis­pute. Member states could block the prosecutor from proceeding by consensus, just as they can block the ordinary operation of the DSU only through consensus. 400 See, e.g., DSU, supra note 54, art. 17.14 (“An Appellate Body report shall be adopted by the DSB and unconditionally accepted by the parties to the dispute unless the DSB decides by consensus not to adopt the Appellate Body report . . . .”).

Finally, the WTO’s system of retaliation would also operate as a check on the prosecutor. Under ordinary WTO rules, sanctions for viola­tions are not imposed by the WTO. They are, rather, imposed by individ­ual member states after receiving WTO authorization. 401 See supra Part I. The creation of a prosecutor would not change that. The prosecutor would have no independent ability to seek or impose sanctions against a respondent state that refused to comply with an adverse decision of the DSB. If the respondent did not remove the unlawful measure, a member state would have to come forward and seek authorization to retaliate on the basis of the decision. The state seeking authorization to retaliate would not have to relitigate the legality of the measure, though. In this way, the prosecu­tor provides a public good to members by establishing that a nation’s measures violate a WTO obligation. That finding alone would subject the member state to reputational sanctions. 402 See generally Andrew T. Guzman, How International Law Works: A Rational Choice Theory (2008) (setting forth a reputational theory of international law). But the reciprocal suspension of trade concessions—the backbone of the WTO’s relatively effective enforcement procedures—would remain subject to states’ political and diplomatic calculations.

To be sure, the prosecutor’s office would still face the capacity con­straints common to prosecutors in domestic systems. There would thus be an element of prosecutorial discretion in selecting investigations to pursue. But even with these capacity constraints, selective enforcement would still decline. A prosecutor’s office would also offset the capacity constraints that limit the ability of developing countries to pursue potentially meritorious claims. 403 See generally Guzman & Simmons, supra note 53, at 583–88, 591–92 (showing that capacity constraints explain the pattern of WTO disputes better than power-based arguments).

Recent criticism of the DSB, especially from the United States, has focused on claims that the DSB has overstepped its mandate. 404 Statement by the United States at the WTO Dispute Settlement Body’s First “Dedicated Session” on the Issue of Reappointments of Appellate Body Members (Sept. 26, 2016),
Dedicated.Session.26Sep16.pdf [].
This criticism, although not necessarily shared widely by WTO members, likely means that the creation of a prosecutor’s office within the WTO is not politically viable. Still, other international dispute resolution bodies, most notably the International Criminal Court, have prosecutors that are em­powered to initiate investigations and, subject to limitations, disputes. 405 Rome Statute of the International Criminal Court art. 15, opened for signature July 17, 1998, 2187 U.N.T.S. 90 (entered into force July 1, 2002). The existence of prosecutors in international law in general suggests that even if the creation of one is ill-advised or unlikely now, members might come to view one more favorably in the future.

B. Reforming Trade Remedy Investigations

Reforming antidumping and countervailing-duties law would also reduce the pernicious effects of selective enforcement. As Part II demon­strates, selective enforcement against environmentally beneficial prod­ucts relies primarily on trade remedy investigations. 406 See supra Part II. Consequently, limiting the role of trade remedies investigations would curtail the most prevalent tool of selective enforcement. 407 See Wu, supra note 269, at 2 (arguing that “[d]eclaring certain types of subsidies to be nonactionable would be one mechanism to safeguard against the potential of the negative impact of CVDs”). However, pro­posed trade rem­edy reforms—which tend to focus on reducing trade remedies actions—are both infeasible politically and could eliminate the valuable role trade remedies play in the trade system. Consequently, any trade remedy re­forms aimed at reducing selective enforcement should focus on ensuring evenhanded enforcement.

Over the years, many commentators have called for trade remedies reform. 408 See, e.g., Jean-Marc Leclerc, Reforming Anti-Dumping Law: Balancing the Interests of Consumers and Domestic Industries, 44 McGill L.J. 111, 122–25, 138–40 (1999) (“Through incremental changes in regional trade agreements, countries like Canada will increasingly realize that national economies (and consumers in particular) benefit from dumping.”); Zheng, supra note 225, at 181–91 (proposing to “replace antidumping with a country-specific safeguard equipped with a heightened injury standard and a mandatory public interest clause”); see also André Sapir, Some Ideas for Reforming the Community Anti-Dumping Instrument 2–6 (2006),
files/213/5081012.pdf [] (detailing various trade remedy reforms, including reforming the Community interest clause, increasing the transparency of antidumping measures and proceedings, and shifting the focus of antidumping cases from protecting consumers to fostering competition).
The general argument for reform is two-fold. As Nobel Prize–winning economist Paul Krugman and Maurice Obstfeld have noted, the economic theory underlying trade liberalization does not support anti­dumping laws, the far more commonly used trade remedy. 409 Paul R. Krugman & Maurice Obstfeld, International Economics: Theory and Policy 131–36 (7th ed. 2006) (“Economists have never been very happy with the idea of singling dumping out as a prohibited practice. For one thing, price discrimination between markets may be a perfectly legitimate business strategy . . . . Also, the legal definition of dumping deviates substantially from the economic definition.”). Price discrimination—the essence of what antidumping laws forbid—“may be a perfectly legitimate business strategy . . . . A firm may well be willing to sell a product for a loss while it is lowering its costs through experience or breaking into a new market.” 410 Id. at 134. Moreover, as Professor Maurizio Zanardi writes, “nowadays [antidumping laws are] widely recognised as a successful form of protectionism that basically lost any connection with dumping.” 411 Maurizio Zanardi, Anti-Dumping: What Are the Numbers to Discuss at Doha?, 27 World Econ. 403, 403–04 (2004). Most commonly, governments accomplish this protectionist objective by massaging the data used to calculate dumping margins. Recall that dumping occurs when the price charged in the importing country is less than “normal value,” which is usually the price charged in the exporting country. 412 GATT, supra note 36, art. VI. As mentioned above, governments have figured out a variety of techniques to increase “normal value” so that dumping is easier to find. 413 Zeroing involves dividing a product into subproducts and calculating the dumping margins on the subproducts. The dumping margins on the subproducts are then added together to get an overall dumping margin. The trick is that subproducts with negative dumping margins (that is, those that are not being dumped) are rounded to zero, an act which pushes the overall dumping margin higher. See supra note 195. One common technique is to use a “fair” price, rather than the price a producer actually charges, in order to figure out whether a pro­ducer dumps its goods. 414 Krugman & Obstfeld, supra note 409, at 134. That price is usually higher than what the producer actually charges in its own country, making it easier to demon­strate dumping.

Some have also worried that the language of fairness in antidumping disputes increases pressure on governments to bring cases. Trade policy analyst Simon Lester, for example, has argued that “[t]he constant accusations of ‘unfair trade’ have convinced domestic groups around the world that foreigners are cheating them in one nefarious way or another.” 415 Simon Lester, Trade Remedies vs. Trade Liberalization, Int’l Econ. Law & Policy Blog (June 20, 2016), []; see also Bruce A. Blonigen & Thomas J. Prusa, Antidumping, in 1 Handbook of International Trade 251 (E. Kwan Choi & James Harrigan eds., 2003) (discussing the disparity between “the fair trade rhetoric stressed by [antidumping’s] supporters” and the practical effects of antidumping policies). This perception of unfairness leads to a breakdown in trust. Indeed, as economist Chad Bown has argued, antidumping investigations have the flavor of “vigilante justice,” with countries increasingly electing to retaliate against each other using antidumping investigations rather than the WTO’s multilateral system. 416 Bown, Trade Remedies, supra note 209, at 524.

The primary argument in favor of trade remedies holds that they are a necessary safety valve for the trading system. As Michael Punke and Timothy Reif—the U.S. Ambassador to the WTO and the U.S. Trade Representative General Counsel, respectively—put it, “[w]ithout [anti­dumping and countervailing-duty laws], support for trade liberalisation would disappear altogether.” 417 Michael Punke & Tim Reif, Letter to the Editor, Problem of Over-Reach by Appellate Body Decisions Has Become More Serious, Fin. Times (June 5, 2016), (on file with the Columbia Law Review). The idea is that governments face pres­sure from well-organized domestic groups to impose protectionism. Government officials will therefore not agree to trade agreements, know­ing they will be punished politically for liberalizing trade, unless they have the ability to impose protectionist barriers to trade in response to specific lobbying demands. 418 See Sykes, supra note 227, at 275 (“Ultimately, well-organized groups—those most adept at lobbying and most capable of ‘paying’ for policy initiatives—will have their interests vindicated [through elected officials], while diffuse, poorly organized interest groups may suffer.”). On this theory, protectionist rules are a second best, but they actually increase the gains from liberalizing trade by making trade agreements possible in the first place. 419 See id. at 276–77 (“[A]n industry can gain from protection in good times as well as in bad times, as the exclusion of imports under any circumstances allows profits and employment in the protected industry to rise.”).

Given their role as a safety valve, doing away with antidumping and countervailing-duty investigations is neither possible nor, indeed, neces­sarily advisable. Instead, proposals should focus on reforming the requirements for the findings governments must make before imposing trade remedies. For example, trade remedy investigations might be required to focus on the collateral consequences of imposing trade rem­edies, rather than on only whether the “unfair” trade practice injures a domestic entity. Consumers benefit, after all, from dumped or subsidized goods. The only time they do not benefit is when the dumped or subsi­dized goods force competitors out of business and lead to monopoly prices. 420 See Sapir, supra note 408, at 2 (“[D]umping can be viewed as detrimental to the importing country . . . [including the] practice [of] ‘predatory or strategic dumping,’ setting low export prices in order to drive out competitors and then imposing high monopoly prices in the importing country.”). This, in turn, can happen only when barriers to entry prevent competition from reemerging. A focus on competition, rather than dis­criminatory pricing or subsidies, would thus allow the continuation of antidumping and countervailing-duty investigations, but would target them at behavior that actually reduces the welfare gained from liberal­ized trade. 421 See Bernard M. Hoekman & Petros C. Mavroidis, Dumping, Antidumping, and Antitrust, 30 J. World Trade 27, 27–29 (1996) (detailing how a competition-focused policy “might help to defuse both market-access-related disputes and limit the use of antidumping actions”). Transparency measures, such as the weak version of centralized enforcement proposed above, would also deter some of the most egregious uses of trade remedies.


We find ourselves at a unique moment in the history of international relations. The neoliberal consensus that has, since the end of World War II, pushed countries to reduce trade barriers ever further has cracked and may be crumbling. The question of what a fair trade policy should look like—what noncommercial values trade law should protect and how it should do so—is very much on the table.

Addressing the market distortions that selective enforcement creates must be part of the answer. Free trade agreements are supposed to open markets to competition. The great irony is that the process through which these rules are enforced can have the opposite effect. While sub­stantive trade rules promote competition, government enforcement poli­cies curtail it. Worse, they do so precisely for those products that have the largest social benefits—environmentally sustainable products that could reduce the burden imposed by natural resource–intensive industries.

This Article has set the table for further research into the effects of selective enforcement. It has focused primarily on the effects of selective enforcement on markets. Future research should focus on identifying selective enforcement’s causes. Natural resource–intensive industries predate trade liberalization in most countries. Fishing and drilling for oil, for instance, are practices that long predate the GATT. It is possible that government subsidies for these sectors are thus baked into investment-backed expectations. Government support for these sectors thus does not spur anyone to lobby for trade enforcement because no one in the industry ever expected to rely on trade rules. Renewable energy and fish farming, however, came along after trade law was estab­lished. Government support in these sectors may well disturb expectations established through the GATT and WTO.

Similarly, future research should look for other sectors of the economy where we observe similar patterns of selective enforcement, and for other laws that are selectively enforced. Monitoring selective enforcement will be especially important as trade litigation challenging health and safety measures, as well as consumer protection measures, continues to increase. These measures are frequently challenged under complicated agreements like the WTO’s Agreement on Technical Barriers to Trade or the Agreement on Sanitary and Phytosanitary Measures. Like rules on government support, the technical nature of rules on health, safety, and consumer protection measures could make them ripe for exploitation by the forces of selective enforcement.

Most importantly, though, governments should look at ways to make enforcement more evenhanded across products that compete with each other. President Trump has repeatedly called for greater enforcement of trade laws, 422 See supra note 1 and accompanying text. a call that has been echoed, if in less bellicose terms, from quarters as diverse as the European Union and China. 423 China Says U.S. Trade Orders Should Respect International Rules, Reuters (Apr. 12, 2017), []; Commission Proposes Improved Rules to Enforce EU Rights Under International Trade Agreements, European Comm’n (Dec. 18, 2012), [http://].
This consensus on the need for more enforcement provides the political will to address the distortions caused by enforcement policies. A global trade prosecutor may not be in governments’ interests today, but a multilateral process to coordinate enforcement actions would go a long way toward rebuilding the trade regime’s legitimacy. As Hunter S. Thompson once said, “We cannot expect people to have respect for law and order until we teach respect to those we have entrusted to enforce those laws.” 424 Hunter S. Thompson: In His Own Words, Guardian (Feb. 21, 2005), [