Professor William R. Ginsberg Memorial Essay Contest 2022 Winners

We are pleased in this issue to feature the first-place winners of the 2022 Professor William R. Ginsberg Memorial Essay Contest. In this 35th holding of the annual competition, two essays tied for first place. In addition to the first-place essays featured here, the committee awarded a second-place prize to MacKenzie Thurman (Columbia Law School, “Agricultural Land Is a Primary Driver of Climate Change”) and Liam Fine (Columbia Law School, “Watt About It? Climate Resilience in the Electric Utility Sector”). The contest is open to all JD and LLM candidate students enrolled in a New York State law school. First place winners receive a $1,000 prize, second place winners receive a $500 prize, and third place winners receive $250.

The Clothes (Un)Make the Man: How the Textile Quota Phase-Out Wears Out the Planet

The textile sector is one of the most environmentally harmful industries in the world. From 1974-2004, trade in textiles was subject to quota-based restrictions that effectively limited global apparel production. Since these quotas were phased out in the early 2000s, global clothing production has roughly doubled. Despite this, there is currently no literature—legal or otherwise—assessing the environmental impacts of such a major change in global trade policy. This paper shows how the textile quota phase-out has enabled the ‘fast fashion’ phenomenon, and exacerbated the environmental consequences that accompany it.

In addition, it assesses the legal viability of reintroducing some form of textile quotas under current international trade law. Based on this analysis, it concludes that reforms in trade-related policy and jurisprudence are urgently needed and ultimately inevitable.

Introduction

The textile sector is both the second-largest industrial consumer1 and polluter2 of water, using some 100 trillion liters each year3 and bearing responsibility for about one-fifth of all industrial water pollution.4 It likewise accounts for one-fifth of mixed waste worldwide5 and, while uncertain, some estimates are that the textile sector is responsible for up to 10% of global greenhouse gas (GHG) emissions.6 The fashion industry has also been linked to a wide range of other environmental concerns including deforestation,7 microplastic pollution,8 deterioration of human and animal health,9 and biodiversity loss.10

Despite this, the matter has merited far less attention than other major polluting industries.11 While awareness of fast fashion’s harmful effects on the environment has increased in recent years,12 the dearth of literature discussing the environmental impacts of textile lifecycles—even cursorily—is especially notable within the realm of legal scholarship.13 As a result of this nescience, the literature that does exist often features conflicting, misleading, or inaccurate data that render problems in the sector even harder to remedy.14

Nevertheless, it is clear that the fashion industry’s activities inflict tremendous harm to the environment on a global scale. At the outset, it is important to note that the production process of apparel is complex and highly fragmented.15 Each stage—cultivating raw materials, spinning them into yarn, weaving yarn into fabric, cutting, sewing, dyeing, finishing, shipment, consumer use, and disposal—has its own environmental impact,16 making it difficult to draw broad conclusions without simplifying to some extent.17 The challenge is also compounded by the fact that different types of fibers correspond with different forms and degrees of environmental side-effects.18 On top of that, the environmental impact of each stage can vary by location,19 and the lifecycle stages of garments are geographically dispersed.20

Still, current research provides insight into the state of the industry as a whole, allowing for a big-picture overview of its environmental repercussions as well as the policies that enable them. Part I aims to introduce and assess how a major change in global trade policy—the textile quota phase-out—has contributed to the rise of ‘fast fashion’ and the environmental consequences that accompany it. Part II provides an overview of current efforts undertaken to address those consequences, while parts III and IV examine potential trade-based solutions, and the legal framework within which they must be implemented. Part V concludes.

Part I: The Role of the Quota Phase-Out

Historical Background

Textile production is distinguishable from (some) other industrial sectors in that it is very labor-intensive but requires only limited capital and technology. As a result, it tends to have few barriers to entry, making it an attractive option for countries with little capital but large unemployed populations.21 Indeed, development of the textile industry marked the onset of the industrial revolutions in Europe and the United States, and has played a similar role in other nations around the globe.22

As industrialization became more widespread throughout the world during the 20th century, more and more countries became capable of mass textile production. The resulting global overcapacity for this production led major industrialized nations to protect their local industries by restricting textile trade, most notably via the imposition of quotas on textile exports from developing countries.23

These restrictions initially applied only to cotton products, leading nations to export growing volumes of manufactured fibers not subject to them.24 In order to address this emerging reality, a multilateral agreement known as the Multi-fiber Arrangement (MFA) became effective under the 1974-version of the General Agreement on Tariffs and Trade (GATT), and was originally intended to last for a period of four-years.25 Article 1.2 of the MFA stated that its objective was “to achieve the expansion of trade . . . while at the same time ensuring the orderly and equitable development of this trade . . . in both importing and exporting countries.”26

Despite the stated purpose of achieving the “expansion of trade,” the MFA in fact limited trade in textiles, taking after its name by extending quota restrictions to encompass manufactured fibers, as well as forcing countries to push back against the policy,27 eventually leading to the Uruguay Round Agreement on Textiles and Clothing (ATC) in 1995.28

The ATC was designed to gradually phase out the MFA in several stages, ultimately ending the specialized treatment of textile trade by 2005.29 The integration of the textile sector into the general trade rules applicable to all other sectors has led to several changes in global textile trade trends. As a 2016 report points out, trade volume in clothing and textiles has increased more rapidly than the average for all world trade, with clothing trade increasing at a higher rate than general textile trade since the MFA/ATC’s expiration.30 Indeed, global clothing production has approximately doubled over the last 20 years.31

How the Quota Phase-Out Has Affected the Fashion Industry’s Environmental Repercussions

To date, little has been written about the environmental consequences of the textile quota phase-out, and there seem to be no publications dedicated to assessing this inquiry.32 Nevertheless, some literature provides insight into the general repercussions of the quota phase-out on the textile industry, and, by extension, the environment.

Labor Costs, Consumer Prices, and Globalization

The first and most obvious effect of the MFA’s expiration is the incentivization of outsourcing based on competitive labor costs, leading to accelerated globalization.33 Since apparel production is so highly dependent specifically on labor, its cost has become one of the chief considerations for sourcing in the post-MFA era.34 Without quota limitations, retailers are now free to source their apparel from whichever country provides the cheapest labor.35 And the price of labor can vary enormously: as of 2014, hourly labor costs in the textile industry averaged $0.62 USD in Bangladesh, $1.12 in India, $2.65 in China, and $17.71 in the U.S.36 Indeed, developing countries contributed only 25% of worldwide apparel exports in the mid‐1960s, but this increased to more than 70% as of 2014.37 The U.S., for instance, produced a majority of its clothing as recently as 1991, but now produces less than 3% domestically.38

The increased geographic dispersion inherent in modern apparel production contributes to a number of harmful social and environmental consequences. For one thing, it results in higher levels of transportation. While minor compared to the impact at other stages in textile lifecycles (currently, at least), increasing amounts of clothing are being shipped through air cargo, which is drastically more carbon-intensive than traditional shipping methods.39

More importantly, the globalization of apparel production following the MFA’s expiration intensified competition among clothing exporters.40 This has led to a ‘race to the bottom’41 in which producers are pressured to constantly lower their prices in order to keep up with cheap labor costs in other developing countries. As a result, the average price of clothing has fallen relative to the prices of other consumer goods.42 U.S. buyers, for example, saw the real price of apparel fall 44% between 1994 and 2017,43 while the price of clothing dropped by about 50% in the U.K. market between 1995 and 2014.44

Lower prices for consumers in wealthier countries mean lower wages for garment workers (the vast majority of whom are women) in poorer countries, obscuring labor conditions and exacerbating concerns over human rights and environmental justice.45 For instance, despite calls for improved labor conditions prompted by 2013’s Rana Plaza disaster in Bangladesh, lead retailer firms paid 13% less to Bangladeshi suppliers in the following five years.46 Similarly, a coalition of more than 180 human rights groups has raised serious concerns over forced labor in the industry (particularly amongst Xinjiang’s Uyghur population). It estimates that as many as one in five cotton products are “virtually certain” to be tainted with human rights violations.47

Indeed, globalization in the fashion industry makes it exceedingly difficult to address both environmental and social concerns48 because it is challenging for downstream manufacturers (let alone retailers and consumers) to know where the raw materials come from or how they were processed.49 In addition, the fragmented nature of globalized supply chains makes it impossible to properly assess environmental impact.50 And, since the majority of global clothing production now occurs in developing countries, those countries bear the brunt of the harmful environmental side-effects.51 In Cambodia, for example, the fashion industry has caused an estimated 60% of water pollution52—40% higher than the global average—and has exacerbated water scarcity in other major textile-producing countries.53 Similarly, instances of pesticide poisoning (pesticides are commonly used in cotton cultivation, as discussed in Part III) tend to be more common in Asia’s textile-producing countries than elsewhere.54

Lead Times and Quality

With labor costs driven ever-lower following the textile sector’s entry into a free trade regime, another increasingly important sourcing consideration is lead time, i.e., the speed with which clothing can be produced and sold.55 Demand for cheap clothing in a free apparel market has increasingly pressured producers in low-income countries to decrease both prices and lead times,56 enabling the proliferation of ‘fast fashion.’57 Indeed, companies such as Zara and H&M have managed to reduce production lead times from months to weeks.58 Among all European apparel companies, the average number of clothing collections59 more than doubled in the ATC’s wake, from two a year in 2000 to about five in 2011.60 Once the fabric is ready, the rest of the process—from sewing in Asia to final sale in Europe—can now take as little as 12-14 days.61

Unfortunately, both the accelerated pace of clothing production, as well as the concentration of that production in lower-income countries, can exacerbate the fashion industry’s environmental repercussions. Some research has suggested, rather intuitively, that the faster an article of clothing is produced, the more likely it is to result in higher levels of pollution.62 And, since environmental standards are often laxer in developing nations, apparel produced in those countries can tend to have greater environmental repercussions. Chemicals used in textile dyeing, for example, are strictly regulated in the U.S. and Europe, but still freely used within the industry of developing nations.63 Similarly, wastewater treatment is common practice in Europe, but other textile-producing countries pump wastewater directly into water bodies.64 This is often the case in terms of emissions as well; textiles produced in China, for example, have a 40% larger carbon footprint as compared to those produced in Europe.65

Along with decreased prices, the quota phase-out also spurred a decrease in the average quality of clothing, in part because exporters operating under a quota are more likely to produce pricier, higher-quality garments.66 The increase in clothing production coupled with decreased prices, lower quality, and faster lead times has also affected consumer behavior, resulting in increased purchasing but decreased use of purchased garments. According to the United Nations Environment Programme, the average consumer bought 60% more articles of clothing in 2019 as compared to 2004, but kept them only half as long.67 In the U.S., for example, the average consumer now purchases one new item of clothing every five and a half days, while the average duration of garment-use has decreased 36%.68 This is significant in light of findings that extending the active life of clothing by nine months is estimated to reduce carbon, waste, and water footprints by 20 to 30%.69

Shortened garment lifetimes and increased consumption led to a 40% increase in landfilled textile waste in the U.S. between 1999 and 2009; textiles now account for up to 22% of mixed waste worldwide.70 This is especially perplexing because much of the apparel is never sold, let alone worn. For example, there are reports that as much as a third of all imported clothing in the EU is never sold, and is instead sent straight to a landfill or incinerated.71 Even clothes that are purchased may never see actual use; the World Bank has stated that, in some countries, 40% of purchased clothing is never worn,72 and it is estimated that more than half of ‘fast fashion’ apparel is disposed of in under a year.73

On the whole, global trade in textiles has increased significantly over the last 60 years, from roughly $6 billion in 196274 to $100 billion in 2020.75 Apparel trade specifically has grown more rapidly than general textile trade; apparel exports now constitute more than half of all textile exports, nearly three-quarters of which are accounted for by developing countries.76 From a more recent perspective, global clothing sales doubled between 2000 and 2015,77 and apparel now accounts for more than 70% of global textile products.78

The precise environmental repercussions of the quota phase-out may not be possible to identify. Nevertheless, its demonstrated impact on price decreases, quality downgrading, and globalized production indicate that it has played a significant part in the growing environmental consequences of fast fashion. Indeed, while the environmental implications of the MFA’s expiration have merited virtually no attention in current literature,79 general concerns over apparel production have elicited some responses.

Part II: Current Efforts and the Need for a Global Approach

Textile sustainability—particularly in the context of fast fashion—has gained increasing attention in recent years,80 leading to a wide range of initiatives designed to alleviate environmental concerns. Such measures include recycling, use of more eco-friendly fibers and processes, private “sustainable brands,” and domestic regulation. However, the large variety of actions being taken can obscure their limited efficacy, leading to concerns over ‘greenwashing’81 within the industry.82 Indeed, while many of the current approaches can be helpful, their usefulness is often limited in context of the challenges posed by a highly globalized and fragmented industry.

Recycling

Recycling, for example, often sounds like a promising solution, but less than 12% of the material used for clothing is recycled,83 with under 1% actually converted into new clothing.84 This is due to a number of reasons, including the fact that clothing made from multiple fibers is difficult to recycle;85 a typical jacket, for example, contains 25 different components, many of which are comprised of different materials themselves.86 Further, many of the base components used can render products unsuitable for recycling altogether, as most existing methods cannot separate dyes or other contaminants from the original fibers.87 As a result, roughly 85% of all textiles in the U.S. end up in a landfill or incinerated; by weight, textiles in the municipal solid waste stream increased 78% from 2000-2017.88 Globally, the equivalent of a garbage truck full of clothes is discarded every second.89

Even when apparel is successfully reused, much of it occurs outside the country in which it was purchased90 as roughly 60% of recovered wearable clothes are exported to other nations.91 The excessive importation of second-hand garments from wealthier countries to developing countries can undermine local industries in the latter while also creating superfluous waste.92 Indeed, this concern—sometimes referred to as “textile dumping”93—has prompted some nations to impose restrictions on such imports.94

Sustainable Sourcing and Processing

As noted earlier, the environmental impact of a given textile is heavily determined by the type of fiber from which it is composed,95 and efforts are being made to make clothes from more eco-friendly materials.96 There is a broad range of such materials, including hemp, linen, bamboo, organic cotton, and mushroom-based fibers.97 Yet despite the large number of seemingly appealing alternatives to traditional fibers and processing methods, their use and viability are rather limited.

Organic cotton, for example, can reduce freshwater consumption by 91%, acidification by 70%, global warming potential by 46%, and soil eutrophication by 26% as compared to conventional cotton cultivation.98 Yet organic cotton accounts for less than 1% of all cotton produced globally,99 and it is extremely difficult to scale because it produces lower yields than conventional cotton.100 Even its touted advantages can be misleading—it’s true that organic cotton is far less freshwater-intensive per plant, but lower yields mean more plants (and land) are required to produce the same amount of fiber. It has even been suggested that significantly more water is needed to produce a t-shirt made from organic cotton than conventional cotton.101

This sort of misleading or incomplete information has become highly prevalent in fashion marketing; indeed, “green advertising” increased more than tenfold in the last few decades.102 One report shows that the number of “green” products increased 77% over a one-year period, finding that 95% of those products used greenwashing tactics in their marketing;103 another recently concluded that sustainable ‘certifications’ were “unambitious, opaque, unaccountable and compromised.”104

Zara’s use of labels certified by the Forest Stewardship Council (FSC) provides a particularly striking example of the greenwashing phenomenon. According to its website, FSC trademarks “provide a guarantee to consumers that the products they buy come from responsible sources[.]”105 Yet the FSC label on Zara’s clothes often refers to the materials from which the tag is made of, not the clothing article itself.106 Even brands (such as Patagonia and People Tree) that make a genuine effort to implement sustainable practices offer limited solutions on a global scale because the more sustainable they are, the less scalable they tend to be.107

Domestic Regulation

The proliferation of greenwashing has prompted some European countries to begin addressing the pervasive lack of transparency in apparel supply chains. The Norwegian Consumer Authority, for instance, has demanded greater clarity from some major brands concerning their environmental footprint.108 Similarly, France has developed a scoring system to grade the environmental impact of fashion items, and may soon require brands to display their ratings on products sold in the country.109 Work on a social score is underway as well, and its “Devoir de Vigilance” law compels business with more than 5,000 employees to state the social and environmental risks linked to their activities.110 And in 2015, the U.K.’s Modern Slavery Act made it the first country in the world to require businesses to report their progress in identifying and addressing modern slavery risks in their operations and supply chains.111

Efforts outside of Europe are beginning to emerge as well. In the U.S., New York recently unveiled the Fashion Sustainability and Social Accountability Act. If passed, the bill will require major companies selling apparel in New York to map at least 50% of their supply chain while disclosing their greatest social and environmental impact, or else be fined 2% of their annual revenue.112 In 2018, private fashion companies and NGOs collaborated to form the global Fashion Charter for Climate Action.113 Similar to the Paris Agreement, signatories must commit to reducing GHG emissions 30% by 2030 and achieving carbon-neutrality by 2050 (and also like the Paris Agreement, these commitments are non-binding).114

While such actions are promising, they focus on transparency and disclosure; there is far less progress in terms of actual regulatory constraints on companies’ operations.115 Even when such regulations are imposed, their domestic nature allows companies to avoid responsibility by outsourcing production to suppliers that are not subject to them.116 Taking the actions necessary to fully address both the social and environmental costs of textile operations will therefore require global collaboration.117 In addition, because sustainable apparel production is currently impossible to scale,118 the solution must necessarily involve a decrease in overall production; the aim cannot be to buy better, but to buy (and produce) less.119

Given the ATC’s role in exacerbating environmental concerns in the fashion industry, a (gradual) reintroduction of textile trade quotas stands to facilitate this aim by reducing overall global apparel production. Less raw material would be cultivated, resulting in lower consumption of water and energy, as well as decreased pesticide usage and GHG emissions. Less fabric would need to be dyed as well, reducing wastewater pollution and chemical usage. And, since the average price and quality of clothing would increase, garments might be kept longer and used more frequently, leading to lower levels of waste and overall environmental impact. While the practical implications of a textile quota ‘phase-in’ are considered in Part IV, it is first worth examining the legal viability of such a measure under international trade law.

Part III: The Legality of Textile Trade Quotas

Perhaps the most straightforward means of reducing global apparel production is to reverse the effects of the quota phase-out by gradually reintroducing quotas. The two largest apparel importers120—the U.S. and EU—for example, could set specific quotas for different garments, depending on the type of fiber used to produce them. Alternatively, export quotas could be negotiated with textile-producing countries in a manner somewhat similar to the MFA. In either case, clothing made from more harmful materials could be subject to stricter quotas while apparel made from more eco-friendly fibers, such as linen or hemp, could be subject to higher quotas or exempted altogether. Additional considerations, such as compliance with environmental standards,121 might also be considered when setting quota levels.

The GATT’s Environmental Exceptions

As its name implies, Article XI of the GATT-1994—entitled General Elimination of Quantitative Restrictions—prohibits the use of quotas in international trade. Specifically, it provides that “no prohibitions or restrictions other than duties, taxes or other charges . . . made effective through quotas . . . shall be instituted or maintained by any contracting party[.]”122 Reintroducing some form of the MFA’s textile trade quotas would therefore seem presumptively invalid. Nevertheless, it may be permissible under one of the GATT’s exceptions.

Article XX (General Exceptions) of the GATT lists several trade-restrictive measures that are exempt from the GATT’s general provisions, including Article XI’s elimination of quantitative restrictions.123 Two of these—Article XX(b) and XX(g)—are known as the “environmental exceptions,” and apply to measures undertaken to either protect health or to conserve exhaustible natural resources, respectively. While each of Article XX’s listed exceptions specify the conditions under which they may apply, their invocation must also meet the requirements set forth in its introductory clause, known as the Chapeau (discussed below).124 Although trade-restrictive measures may potentially be justified under other provisions in the GATT, this article will only examine justifications based on one of Article XX’s environmental exceptions.

Textile Trade Quotas Under Article XX(b)

Article XX(b) of the GATT allows for quantitative restrictions “necessary to protect human, animal or plant life or health.”125 Important decisions issued by the WTO’s Dispute Settlement Body (DSB)126 have led to the development of a ‘necessity test’ used to determine whether a measure meets the definition of “necessary” as used in Article XX(b). The test consists of two parts. First, the policy objective pursued by the GATT-inconsistent measure must be the protection of life or health of humans, animals or plants; second, the measure must be necessary to fulfill those policy objectives.127

As explained below, trade quotas might be viably justified based on a number of health risks—to humans, plants, and animals alike—associated with the textile industry. These risks are inherent in both synthetic and natural fibers, manifesting in various ways and at different stages in textile lifecycles. Two fibers in particular, one synthetic and the other natural, dominate the textile and apparel landscape: polyester and cotton. Together, they constitute roughly 80-85% of all textile materials worldwide.128

Polyester accounts for 60% of global clothing fiber produced129 and, like water bottles, is made predominantly from polyethylene terephthalate (PET).130 Its production begins with the extraction of crude oil, a process known to harm wildlife and human health.131 Later stages in its development involve the use of up to 15,000 chemicals, many of which are hazardous to human health, and some of which can cause cancer.132 Furthermore, enormous quantities of microfibers are released during both production and consumer use; indeed, the textile sector is the largest contributor of microplastics to Earth’s oceans, accounting for an estimated 35% of all oceanic primary microplastic pollution133 (roughly equivalent to 50 billion plastic bottles annually).134

Microplastics are now virtually omnipresent; they have been detected in rivers, deep-sea sediments, arctic ice, soil, air, beer, table salt, birds, fish, plants, animals, and humans.135 While much is still unknown about this phenomenon, it is clear that synthetic microfibers pose a health risk on both an individual and ecosystemic level.136 Frequent, long-term exposure (as is the case for many garment workers) has been associated with increased risks of asthma, lung disease, and lung cancer.137 Research regarding the impacts of microfibers has also raised a number of ecological concerns due to their effects on plant and animal health. For example, they were found to cause an increase in mortality for certain marine organisms, which are often critical in food chains.138 They can also inhibit the growth of some plants, and there is research suggesting that sufficient concentrations of microfibers could threaten food security and biodiversity.139

While microplastics are a feature unique to synthetic textiles, apparel made from natural fibers comes with its own set of health concerns. Cotton, the most common natural fiber, accounts for over two-thirds of natural fibers worldwide and about one-third of textiles overall.140 Though it occupies only 2.4% of Earth’s cultivated land, cotton is allocated 16% of global insecticide usage—more than any other crop141—a significant proportion of which are not absorbed by the plant.142 Instead, these pesticides wash out into rivers and groundwater bodies, creating severe health risks;143 one review of data provided by the World Health Organization found that nearly half (44%) of all farmers worldwide suffer pesticide poisoning every year.144

According to the DSB, health risks are encompassed by Article XX(b) if the evidence “tends to show . . . a risk to health rather than the opposite,”145 an evidentiary standard requiring even less than a “‘preponderant’ weight of the evidence,” nor a quantification of the risks to life or health.146 This standard could almost certainly be met by the plethora of evidence indicating that there are significant health risks associated with the production, use, and accumulation of natural and synthetic textile fibers. Indeed, some of these risks—such as lung cancer147 and air pollution148—have already been recognized as health risks under Article XX(b).

Like the trade measures used to combat these risks (and others), a measure designed to address the health risks implicated by excessive textile trading would almost certainly meet the first element of the “necessity test.”149 This is especially true when considering that “it is within the authority of a WTO member to set the public health or environmental objective it seeks to achieve, as well as the level of protection that it wants to obtain.”150

In terms of the second element (i.e., whether a measure is “necessary” to fulfill a policy objective), “the question . . . is whether there is an alternative measure that would achieve the same end and that is less restrictive of trade[.]”151 Since “necessity” is more likely to exist where a measure’s pursued societal value—and its contribution to that value—is high,152 the scale of the textile industry’s health risks weighs in favor of finding necessity. Similarly, a quota-based measure is “apt to produce a material contribution to the achievement of its objective,”153 as trade quotas would lead to decreased textile production overall, thus mitigating health risks at every stage of textile lifecycles.154

Indeed, the Appellate Body has explicitly recognized that “certain complex public health or environmental problems may be tackled only with a comprehensive policy comprising a multiplicity of interacting measures.”155 It has also noted that “the results obtained from certain actions—for instance, measures adopted in order to attenuate global warming and climate change . . . can only be evaluated with the benefit of time.”156 These considerations are acutely pertinent to the current and future health risks inherent in the modern fashion industry.

Further, while textile quotas would undoubtedly produce restrictive effects on international commerce—thus weighing against their legitimacy157—more restrictive measures, such as categorical bans, have passed the necessity test.158 In Brazil—Retreaded Tyres, a categorical import ban was held to be necessary under Article XX(b) because no other measure would accomplish its goals to the same extent.159 Given that this extent is determined solely by the country implementing a measure,160 there is arguably no less-restrictive alternative that could mitigate the health risks posed by the textile industry to the same extent as a quota. This is especially true considering the current lack of progress in mitigating these risks,161 and that quota levels could be calibrated to correspond with the level of risk posed by a particular fiber. Thus, the imposition of trade quotas (being less restrictive than a ban) on health-threatening fibers could satisfy both elements of Article XX(b)’s necessity test.

Textile Trade Quotas Under Article XX(g)

Article XX(g)’s exception for resource conservation may be suited to justify trade quotas as well, as it permits GATT-inconsistent trade restrictions “relating to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption.”162 There are therefore three conditions that must be met when invoking this exception: (1) the measure involves something that is considered an exhaustible natural resource; (2) the measure relates to the conservation of that resource; (3) the measure is made effective in conjunction with domestic restrictions. Given that the phrase “exhaustible natural resource” has been interpreted broadly to include both living and non-living resources,163 many of the resources threatened by excessive textile trade are encompassed by this definition. Major examples could include water, clean air, forests, and wildlife.

The textile industry is the second most water-intensive industry in the world (after agriculture),164 with the production stage alone consuming about 80 trillion liters of water each year (at an estimated ratio of 200 tons of water for each ton of textiles).165 An additional five trillion liters of water are used in the global dying process,166 while laundry usage consumes another 20 trillion liters,167 for a total of approximately 100 trillion liters consumed annually. Clothing specifically accounts for more than two-thirds of this consumption.168

Clean air—an already-recognized exhaustible natural resource169—is also compromised by textile production. Textiles are among the most GHG-intensive products per unit of material produced (emitting roughly 25 tons of GHGs per ton of fabric),170 and the industry is estimated by the World Bank to account for 10% of global GHG emissions.171 In more fathomable terms, producing the cotton for a single pair of jeans requires roughly 8,000 liters of water,172 and the industry emits more GHGs than all international flights and maritime shipping combined.173

Aside from exhausting the fundamental natural resources of air and water, the textile sector has also been linked to the depletion of a wide range of other resources including forests, soil, and biodiversity.174 The conservation of each of these resources serves not only to preserve the resource itself, but a range of other resources that both support and are supported by it.

The effect of pesticides on honeybees provides a useful illustration of the close linkage between different natural resources. According to the U.S. Department of Agriculture, bees are responsible for about one-third of the U.S.’ food supply;175 they also contribute significantly to medicine, animal agriculture, biodiversity preservation, and—of course—textiles.176 In recent years, however, strong evidence has surfaced indicating an alarming decline in global bee populations,177 and populations of managed bees in the U.S. decreased by 40.7% in 2019 alone.178 These losses have been strongly linked to pesticide usage179—particularly those applied to citrus and cotton.180

Thus, limiting apparel production stands to aid in the conservation of not only clean air and water, but rather an entire ecosystemic structure upon which modern society has come to depend. The importance of this conservation is emphasized even further by the projection that, if consumption continues at its current rate, the textile industry is expected to exhaust three times as many natural resources by 2050 compared to 2000.181

However, in order for a measure to be justified under Article XX(g), it must “relate” to resource conservation. Though criticized by scholars as an “unwarranted amendment” of Article XX,182 this has been interpreted to mean that the measure must be “primarily aimed” at the conservation of the targeted resource183 such that the relationship between it and the measure at hand is “observably a close and real one.”184 Thus, it may be easier to justify quotas on the basis of conserving resources that are more directly compromised by the textile industry, such as water. Conversely, while reducing textile production would facilitate the conservation of a wide range of other resources, their less-direct relationship with the industry may weaken their contribution to a legal justification. The analysis is also complicated by the lack of clarity regarding the precise meaning and significance of “primarily aimed.”185

On the whole, though, Article XX(g)’s ‘primary aim’ requirement would seem satisfied by trade quotas intended to reduce overall consumption, and specifically tailored to the conservational impact of the targeted fibers. This is especially true because Article XX(g) measures can apply to the conservation of global resources,186 even if they require “a substantial period of time, perhaps years,” to take effect.187

Finally, in terms of Article XX(g)’s requirement that a measure be made in conjunction with domestic efforts, a range of options is available to countries seeking to implement a trade quota. The DSB has determined that compliance with Article XX(g) does not require “identical treatment of domestic and imported products,” and instead amounts to “a requirement of even-handedness in the imposition of restrictions[.]”188 Accordingly, this requirement could be satisfied so long as countries imposing an import quota also took some action on the domestic front, such as limiting polyester-based clothing production while incentivizing the production of clothes made with more sustainable fibers.

Textile Trade Quotas Under the Chapeau

In addition to meeting the conditions specific to each provision of Article XX, a GATT-inconsistent measure must also conform to the requirements of Article XX’s introductory clause, the Chapeau. The Chapeau states that all of the listed exceptions are “[s]ubject to the requirement that such measures are not applied189 in a manner which would constitute a means of arbitrary or unjustifiable discrimination,190 between countries where the same conditions prevail, or a disguised restriction on international trade.”191 The Chapeau has proven to be a major obstacle to the successful invocation Article XX, as the DSB has interpreted its requirements quite strictly.

For example, while Brazil’s ban on retreaded tires was held to satisfy the requirements of Article XX(b), the Appellate Body held that it violated the Chapeau because Brazil did not apply the ban to imports from a trade-bloc of South American countries (known as Mercosur).192 This was found to constitute “arbitrary discrimination” under the Chapeau, despite the fact that Brazil exempted Mercosur imports from the ban only in order to comply with an unrelated tribunal ruling.193 The Appellate Body determined that complying with the tribunal’s ruling “is not an acceptable rationale for the discrimination, because it bears no relationship to the legitimate objective pursued by the Import Ban[.]”194

Retreaded Tyre’s conclusion that rationales unrelated to a measure’s objective cannot justify discrimination as non-arbitrary has been criticized as “excessively rigid and rather senseless.”195 Furthermore, while the Chapeau prohibits discrimination “between countries where the same conditions prevail,” the famous Shrimp-Turtle decision held that it is also prohibited between countries where different conditions prevail if those conditions are not taken into account when applying a uniform regulatory program.196 In addition, discrimination may be found even if it does not undermine the objective of a trade-restrictive measure,197 and unilateral measures are deemed a “disguised restriction on international trade” if they are imposed without prior attempts to negotiate with affected parties.198

Given that arbitrary discrimination can exist even when it does not undermine a measure’s objective,199 and that differing conditions between countries must be taken into account,200 a uniform quota allocation applicable to all countries seems invalid. Perhaps this obstacle could be surmounted by allocating different quotas to countries based on their prior export share vis-à-vis a given apparel fiber. As a hypothetical example, if 20% of prior U.S. polyester imports came from China and 10% from Bangladesh, the quota limit on the former would be twice as high as the latter. Implementing a quota restriction in this way would preserve the percentage of textile trading between countries while decreasing its overall volume. This would therefore contribute to the goal of health protection or resource conservation while arguably avoiding arbitrary discrimination under the Chapeau.

However, because WTO jurisprudence has not yet addressed quota allocation under Article XX, the legal viability of such an approach is uncertain. This uncertainty is compounded by the fact that it is unclear whether and to what extent rationales unrelated to a measure’s objective can be used to justify asymmetrical trade-restrictions.201 In Brazil—Retreaded Tyres, the Appellate Body effectively held that arbitrary discrimination can be justified only by a rationale that relates to the objective of the GATT-inconsistent trade measure.202 In contrast, a later decision issued in EC –Seal Products explicitly stated that “the relationship of the discrimination to the [purported] objective of a measure is one of the most important factors, but not the sole test that is relevant to the assessment of arbitrary or unjustifiable discrimination.”203

If the approach taken in Retreaded Tyres is followed, allocating different quota levels to different countries would not be justified because the rationale for doing so does not relate to the measure’s general objective of protecting health or conserving resources. This is ironic in light of the fact that uneven quota allocation would be implemented specifically in order to avoid discrimination under US—Shrimp.204 Indeed, this illustrates the tension between being unable to justify discrimination through rationales unrelated to a measure’s objective while also complying with US—Shrimp’s requirement of taking differing conditions into account when applying a uniform regulatory program.205 On the other hand, such justification may be possible under Seal Product’s suggestion that rationales unrelated to a measure’s objective might sometimes justify discrimination as non-arbitrary.206 This possibility is tentative at best, however, since the scope of rationales capable of justifying discrimination remains unclear.207

A more promising approach may be to read US Shrimp’s holding more narrowly to require consideration of differing conditions only when a “regulatory program” is being applied.208 In that case, the U.S. had not only required imported shrimp to have been caught using specific technology, but also effectively required countries to employ a regulatory program comparable to its own before accepting their shrimp exports.209 In contrast, imposing textile trade quotas would not require other governments to apply any regulatory programs. Read in this way, the US—Shrimp decision would not require consideration of differing conditions when allocating quotas (thus allowing for the possibility of uniform quota allocation), though countries would still need to first exhaust the possibility of entering into cooperative agreements with exporting countries in order to comply with US—Gasoline.210

However, while this reasoning is sound, the WTO’s jurisprudential history suggests that it may not be welcomed. Indeed, the DSB’s constricting interpretations of “necessary”211 in Article XX(b), “related to”212 in Article XX(g), and “discrimination”213 as well as “disguised restriction”214 in the Chapeau have rendered Article XX’s exceptions all but moot: of the 45 attempts to justify a measure under Article XX, only two215 have succeeded.216

Part IV: Reintroducing Textile Trade Quotas?

The DSB’s approach to Article XX suggests that the WTO favors the short-term economic advantages of free trade over its long-term environmental consequences. To be sure, there is good cause for apprehension at the prospect of limiting free trade in the textile sector. For one thing, it would cause apparel prices to rise, frustrating consumers in wealthier countries who have become accustomed to buying clothes that do not reflect their social and environmental costs. More profoundly, low-income populations in wealthier and poorer countries alike currently benefit from the availability of cheap clothing, allowing them to allocate more of their income to the purchase of other necessities, such as food and shelter.217 And, limiting apparel production could have severe consequences for developing countries whose economies rely heavily on this industry.

Bangladesh’s textile sector, for example, generates up to 20% of the country’s gross domestic product (GDP), accounts for ~80% of its export earnings, and employs some 4.5 million people.218 Estimates vary widely, but it is clear that the industry also contributes significantly to the GDP of several other developing nations, including Cambodia (~16%),219 Vietnam (~16%),220 China (~7%),221and India (~3%).222 By comparison, textile manufacturing accounts for only 0.12% of the U.S.’ GDP while employing some quarter-million workers.223 Any reasonable approach to solving the problems precipitated by the ATC’s quota phase-out must acknowledge the impact it would have on such vulnerable economies.

However, that impact must also be weighed against the environmental and economic consequences—both global and national—of continuing with current trade practices. Globally, the gap between water supply and demand is projected to reach 40% by 2030 if current practices continue, and water insecurity risks triggering a global food crisis.224 This gap is far from evenly distributed, however; while some countries are less vulnerable to water scarcity, the threat looms especially large over some of the major players in the textile industry.

Water demand in India, for example, is expected to reach twice the available supply within the next decade, leading to a ~6% loss in its GDP.225 Similarly, China faces growing water scarcity,226 while Bangladesh suffers from widespread pollution throughout its rivers, driven in large part by textile dyeing.227 The world is also set to face a global waste crisis in the coming years, as countries (such as China, Thailand, Vietnam, and Malaysia) that previously accepted garbage from other nations are beginning to ban the importation of solid waste.228

In terms of emissions, the fashion industry’s impact on climate change increased by 35% between 2005 and 2016 (during the latter of which it emitted roughly four billion tons of CO2 equivalents),229 and is projected to increase by 49% between 2016 and 2030 if business-as-usual prevails.230 Globally, the world can afford to emit roughly 300 billion more tons of carbon without causing a temperature rise of at least one and a half degrees Celsius.231 We currently emit about 35 billion tons annually, leaving us with less than a decade to drastically reduce our emissions.232 In order to meet the Paris Agreement’s goal of keeping global warming “well below” two degrees Celsius, humanity will need to halve its annual GHG emissions within the next eight years.233

In the short-term, at least, these reductions will necessarily come at the expense of our consumptive habits,234 demanding that we use less conventional energy,235 fuel, meat236—and clothing. Though far from easy, limiting the consumption of apparel is arguably the least daunting of these prospects.237

Likewise, the economic costs of limiting textile trade, while considerable, must also be weighed against the costs of the climate crisis. Climate change is projected to reduce global economic output 11-14% by 2050, with a disproportionately high impact on some of the major textile producers in Asia238 (trade in textiles represents about 4.62% of global trade).239 Conversely, a 2017 report found that improving social and environmental practices throughout the fashion industry stands to benefit the world economy by over $150 billion annually.240

Thus, even if economics are to remain prioritized over sustainability, such statistics demonstrate a need to take the economic ramifications of environmental degradation into account when contemplating costs and benefits. Currently, however, economic and environmental analyses remain largely insulated from one another, and “there is an urgent need for more rapid integration of . . . sustainable development into the core of economics.”241 Indeed, though the economic consequences of the quota phase-out have been examined since the ATC’s expiration,242 its environmental impact has scarcely been considered, and never assessed.243

As some scholars have noted, failure to integrate environmental considerations into global trade policy raises four major concerns. First, “trade may cause environmental harm by promoting economic growth that results in the unsustainable consumption of natural resources.” Second, “trade liberalization often entail[s] market access agreements that can be used to override environmental regulations[.]” Third, trade “restrictions should be available as leverage to promote worldwide environmental protection, particularly to address global or transboundary environmental problems[.]” Fourth, “countries with lax environmental standards have a competitive advantage in the global marketplace and put pressure on countries with high environmental standards to reduce the rigor of their environmental requirements.”244

This formulation of these concerns was articulated in 1994, just as the ATC was being introduced; each has proven exceedingly prescient, particularly with respect to the fashion industry. While municipal, state, federal, and private efforts certainly have a role to play, they are ill-suited to comprehensively address the problems posed by a highly globalized, fragmented, and obscure industry. Conversely, international trade policy is better equipped to do so and may indeed have potential to serve as an invaluable tool in the fight against climate change. So long as trade policy and jurisprudence fail to accord environmental risks the gravity they warrant, however, that potential will remain dormant while the risks become actualized in its place.

Part V: Conclusion

As this paper hopes to show, the damaging environmental consequences of a globalized textile industry have been significantly exacerbated by the sector’s entry into a free trade regime, manifesting primarily in the form of fast fashion. While this makes regulation of textile trade a prime candidate for reversing some of these consequences, the current state of WTO jurisprudence renders such efforts uncertain to survive legal challenge, particularly under its interpretation of Article XX’s Chapeau.

Though limiting global apparel production may come at great cost, trade policy must wrestle with the prospect that it is too far along in the climate crisis to do otherwise. The current status quo—artificially low prices effected by a free trade regime—is, quite literally, unsustainable. Aside from driving an economic race to the bottom among developing nations, Earth simply lacks the resources and stability to keep up with consumer demand for much longer.

As our resources dwindle and the planet warms, humanity will face increasingly perilous threats from the degradation of our shared environment. These threats will undoubtedly beget agonizing choices; trading fashion for sustainability may be least among them.

This is an original manuscript of an article previously published by Taylor & Francis in Environmental Claims Journal (vol. 35, 2023).

Tal Avrhami is a recent 2023 graduate of Columbia Law School and holds an LL.M. in International Criminal Law from the University of Amsterdam. As a Bernstein Litowitz Berger & Grossman Fellow, he is spending the next two years working with ClientEarth to hold companies accountable for their climate impacts.