Andrew Lang, Heterodox markets and ‘market distortions’ in the global trading system, Journal of International Economic Law, Volume 22, Issue 4, December 2019, Pages 677–719, https://doi.org/10.1093/jiel/jgz042
Navbar Search Filter Mobile Enter search term Search Navbar Search Filter Enter search term SearchAn important context for contemporary trade frictions is the emergence, since the 1990s, of a wide range of new forms of market capitalism, of which China’s hybrid market economy is the most significant. Institutional diversity of this kind is a source of strength and dynamism for the global trading system, but it is also the cause of very serious friction. The General Agreement on Tariffs and Trade/World Trade Organization system has dealt with this problem before, but the existing settlement regarding the legitimate boundaries of institutional diversity is under pressure and needs to be revisited. One concept that has been incorporated into World Trade Organization trade defence law (and elsewhere) to help draw these boundaries is the concept of the ‘market distortion’. The concept can be a useful one, but it has so far been interpreted and applied with an inadequate appreciation of its serious conceptual and practical difficulties. The potential result is a system of trade defences targeted in a discriminatory and even punitive manner against heterodox institutional forms, in ways that may excessively disincentivize institutional experimentation. In response, this paper argues for an approach to the interpretation and application of this concept, which proceeds from an understanding of the institutionally embedded character of markets. This does not take the form of a readily available ‘solution’, but rather a messy and evolving set of legal techniques that, in the best case, can form the legal basis of a practical and justifiable approach to the tensions caused by institutional diversity. A toolkit of legal techniques of this kind clearly cannot take the place of a more foundational political settlement of some sort, but it is a necessary accompaniment to it, if we are to preserve the aspiration towards a genuinely non-discriminatory and rules-based global economic order.
There is no single root cause of the present instability of the international economic order. For some, the core problem is the radically unequal distribution of the gains and losses associated with the contemporary period of globalization. Existing economic institutions and structures have been challenged as those who have lost or gained little from globalization withdraw their support from a system, which appears not to work to their advantage. For others, the primary explanation is the relative erosion of US global economic hegemony, which has both left the USA less willing to act as the guarantor of the system in its present form, and also given rise to increasingly urgent efforts to reshape the system in ways that may more reliably sustain existing distributions of economic power. And for others still, the present system is under attack because it has failed to deliver on its promise of economic self-determination at the national level, as political communities feel their futures constrained and directed by global rules, institutions and logics that are beyond any sense of their immediate control.
This paper starts with the claim that the current period of instability is also the result, in part, of far-reaching changes to the institutional underpinnings of the global economy that have occurred over the last quarter century or so since the end of the Cold War, as a range of former socialist and centrally planned economies have transformed into new and heterodox market forms. Like many others, I see the present instability as driven at a fundamental level by frictions in the US-China relationship. I also take my cue from those who see these frictions as, in turn, being driven by the unexpected emergence in China of an unfamiliar form of market capitalism, which has both radically upended existing patterns of global comparative advantage, and confounded many of those who expected (reasonably or not) that China’s economic system would converge towards Western models of market capitalism after its accession to the World Trade Organization (WTO) in 2001.
The argument I seek to make in this paper is in five moves. First, in Section I, I seek to reframe the problem of ‘Chinese state capitalism’ by locating it as just one illustration of a larger and more general issue, namely, the emergence of a range of new forms of market capitalism over the last quarter century, coupled with a much greater degree of integration between them in conditions of economic globalization. This institutional diversity is a strength of the global economic order, but it also leads to competitive tensions. In Section II, I examine the ways in which the General Agreement on Tariffs and Trade (GATT)/WTO regime has dealt with similar challenges in the past, and argue that while convergence around liberal market principles has been an important objective of the GATT/WTO membership, so too has the preservation of institutional diversity and the capacity for institutional innovation. I then observe, in Section III, that the notion of the ‘market distortion’ has emerged, in both public and policy discourse, as well as jurisprudentially, as one of the key concepts used to delineate legitimate from illegitimate institutional experimentation. This is not unique to trade law—in fact it is a far more general trend occurring across a range of regimes of international economic governance. I argue in Section IV that this notion is far more difficult to operationalize than is typically understood. A coherent and defensible application of the concept needs to start with an appreciation of the institutionally embedded character of markets. Accordingly, a distinction logically needs to be drawn between governmental actions, which constitute ‘market distortions’ and those that help to establish the institutional conditions in which markets operate. In Section V, I show that the application of the concept of ‘market distortion’ in WTO law so far has not adequately appreciated this, and as a result, the jurisprudence is at risk of developing in problematic directions. Finally, in Section VI, I assess the promises and pitfalls of a range of specifically legal techniques 1 for addressing these risks, and developing a more institutionally sensitive way of applying the concept of ‘market distortion’ in a legal context. None of these techniques provide clean solutions to the conceptual problems identified—this is in principle probably impossible—but, when coupled with a serious and sophisticated appreciation of relations between markets and the institutions that constitute them, they may provide a pragmatic toolbox of sorts for adjudicators seeking to avoid some of the worst of the dangers identified.
When China joined the WTO in 2001, the expectation of its most important trading partners was evidently that its economic system would continue to evolve in the direction of marketization, along the path it had started as early as the 1970s. Indeed, China has in fact liberalized its markets in significant ways since joining the WTO, and its program of economic reform continues. 2 However, the process of marketization and liberalization has taken place experimentally and incrementally, and always in a manner, which is perceived to be consistent with its larger strategic goals of economic development, rapid technological advancement, and political stability. As a result, the emergent form of market capitalism in China appears to Western eyes as an unfamiliar hybrid of intense market competition in some sectors, managed competition in others, and high degrees of state direction on others—most often termed ‘state capitalism’. 3
As a consequence, a perception has emerged within influential trade policy circles that China has reneged on the promise it was understood to have made upon its accession to the WTO that it would gradually converge towards a Western model of state-market relations. In a statement to the WTO General Council on 26 July 2018, for example, entitled ‘China’s Disruptive Economic Model’, US Ambassador Dennis Shea made the claim that
[c]ontrary to Members’ expectations, China has not been moving towards a fuller embrace of market-based policies and practices since it joined the WTO in 2001. In fact, the opposite is true. The state’s role in China’ economy has been increasing. 4
The Chinese Ambassador’s response made clear China’s view that there is ‘more than one model of market economy’, and that it is perfectly legitimate for China to pursue the model that suits its ‘own national situation and circumstances’. 5
As this exchange of views helps to make clear, however, contemporary trade tensions between the USA and China can be seen as symptomatic of a much broader problem. They are not—or not just—the result of China’s path of economic reform deviating from the expectations of its most important trading partners. They are, in a deeper sense, also the result of precisely the global spread of market capitalism since the last decades of the twentieth century, and the consequent emergence of variegated forms of market capitalism in and across different parts of the globe. In the years following the collapse of communism, as states throughout the former ‘second’ and ‘third’ worlds embarked upon transitions to market-based economic systems, it appeared to many that the world was heading in the direction of institutional convergence, to one degree or another. This was, it turns out, a mistake: markets are highly variegated and socially embedded institutional orders, and marketization is best understood as an open-ended process of competition-oriented institutional change rather than a defined institutional endpoint. 6 As we now know, national projects of marketization initiated during this period have each evolved according to different dynamics, with the result that over time a variety of new and heterodox market forms have emerged in different countries and regions of the world. ‘Every transition to capitalism’, it has been observed, has ‘produced a new variety of capitalism’. 7
Institutional diversity, it should be noted, is a systemic strength of the international economic order. It is, as the literature on comparative capitalisms has taught us, one of the key determinants of comparative advantage, and a foundation for international specialization. 8 One of the lessons of the history and development of the postwar trading system is that global institutional innovation contributes to the long-term strength, resilience, and dynamism of the global economy. Some of the major periods of structural transformation in the global economy, including periods of major gains in productive efficiency and the integration of marginalized populations, have been associated with periods of profound institutional experimentation. The presence of multiple different market forms is efficiency-enhancing at the systemic level, even if it also creates inefficiencies in the form of specific transaction costs on individual firms conducting business across borders. This is an important point: any attempt to address the issue of institutional diversity must pay careful attention to this fundamental trade-off.
At the same time, however, institutional difference is also certainly a genuine source of friction within the global trading order—that is to say, of deeply felt perceptions of unfair competition. Institutional arrangements in one state may, for one reason or another, be perceived to give firms based in that state unfair competitive advantages in global markets. In relation to China, such claims are very familiar. Chinese firms are claimed to be unfairly advantaged by virtue of a range of measures associated with the nature of the Chinese government’s involvement in its economy: the soft budget constraints associated with government ownership or control, the receipt of inputs at below-market rates by state-controlled enterprise, direct subsidies from the Chinese government, as well as more indirectly by its export and taxation regimes. ‘Level playing field’ concerns of this kind may or may not have a good economic rationale in particular cases, but they play a huge role in the political economy of trade policy, as perceptions of the fairness of global competition is crucial to maintaining support for liberal trade over the longer term. 9
In the decades since the 1990s, as more heterodox market forms have become more deeply integrated into the global economy, these frictions have intensified. The result has been to reopen one of the fundamental issues of global economic order, which concerns the legitimate or ‘fair’ range of institutional diversity permitted in conditions of global competition. At what point do new or heterodox market forms cease to constitute legitimate experimentation, or the legitimate expression of local values and choices, and become a form of ‘cheating’ on the terms of fair competition in international trade? Finding an adequate response to this question is one of the core challenges for the contemporary trading system—and, crucially, finding an adequate way of expressing that response in a legal framework is one of the core challenges for those who wish to maintain a multilateral ‘rules-based’ trading order.
There is an important distinction to be made here. What is at stake is not, or not primarily, the degree of institutional diversity permitted by international trade rules. The continued diversity of market forms is a fundamental feature of the contemporary global economic order, and it is not realistic to imagine international trade rules limiting the extent of that diversity in any simple way. 10 What is more importantly at stake is the existence of the capacity and opportunity to innovate institutionally, its direction, and its distribution between different states within the global economic system. How far will certain institutional choices on the part of one state enable other states to respond by reducing their access to foreign markets? What rules, in other words, will govern the conditions of competition between differently-instituted market orders—including the ways in which certain institutional choices are matched up with positions of advantage and disadvantage in global market competition? It is important to recall that it is not only conditions of competition between firms in a given market order, which must be fair, but also the distribution of the capacity to innovate institutionally between states. This is the second core trade-off that must be kept in mind when addressing the challenge of institutional diversity.
In thinking about this challenge, a clear contrast that has emerged between two camps. 11 There are on one side those who believe that a condition of full participation in the multilateral trading system ought to be adherence to a set of basic liberal market principles, of one or another form. On this view, the GATT/WTO system has worked well in the past precisely because it has for the most part operated as a club of like-minded market-oriented economies. Tensions caused by institutional diversity have, for the most part, been solved by gradual institutional convergence around Western market models. On the other side are those who think that the GATT/WTO system ought to be able to accommodate a broad diversity of market arrangements, and even economic systems. For those who take this view, the history of the GATT/WTO regime discloses a careful balancing between the twin values of institutional diversity and the promotion of liberal market principles, and this careful balancing has been an important source of its strength and durability.
It is common ground between the two camps that, in many respects, the original GATT 1947 was remarkably flexible in its toleration of a variety of different economic systems. Institutional diversity has always been at least a factual reality among the GATT membership, and the GATT from its beginnings has always displayed a pragmatic openness to economic systems of various kinds. This is most vividly illustrated, for example, by the treatment given to the former socialist bloc countries during its first decades. 12 While the original GATT contracting parties ended up being almost exclusively market economies of one form or another, it is also true that an invitation had been extended to the USSR to become part the initial GATT negotiations very early on. This invitation was ultimately declined—and indeed the socialist bloc as a whole initially avoided the GATT and viewed it with some suspicion as an instrument of capitalist expansion—but it represented an important signal that the key founders of the system saw it as potentially compatible with domestic economic systems of many kinds. Some years later, when a number of Eastern European countries changed course and expressed an interest in participating in the GATT, a way was ultimately found for them to do so, albeit gradually. In addition, aside from these centrally planned economies, the GATT showed itself able to accommodate within its membership a range of mixed economies, in which the state played a very central role. This included a variety of Central and South American countries, as well as India, and—perhaps most importantly for present purposes—Japan, which acceded to the GATT in 1955.
This factual reality was reflected in the formal content of the rules themselves. There is nothing in the GATT 1947, which requires adherence to a particular economic system as a condition of membership. On the contrary, the drafters of the GATT 1947 designed a system of rules, which clearly reflected, and accommodated, the diverse institutional choices made by its original contracting parties. It was always understood that there was a real limit to how far the international trade regime could, and indeed should, seek to discipline deviance from a particular idea of state-market relationships. As a consequence, as is well-known, few if any legal requirements were included to reduce state ownership of commercial enterprises, state-trading enterprises and other state monopolies were permitted and subject to only the relatively relaxed disciplines of GATT Article XVII, large-scale governmental stabilization programs in the agricultural sector were explicitly permitted, and the original GATT agreement contained no serious disciplines on either domestic or export subsidies.
Disagreements emerge, however, when it comes to attaching significance to this factual and formal openness. 13 A number of recent contributions to the literature have taken the view that the GATT’s flexibility and pluralism regarding its members’ domestic economic systems is best seen as exceptional, or as a concession to political realities, and thus not reflective of the underlying ‘liberal understanding of law and economy’ shared by the GATT membership as a whole. 14 The USSR, it is noted, never did join the GATT negotiations. Yugoslavia, Romania, Poland and Hungary did not trade with GATT contracting parties in large enough volumes to cause serious concern, and their presence in the GATT was so unusual that special rules had to be incorporated into their accession protocols to make it workable. Japan’s accession was in significant part political—a function of strong backing by its geopolitical ally, the USA—and indeed subject to non-application under GATT Article XXXV for many years by a large number of GATT contracting parties. In any case, the unusual position of Japan, it is noted in such accounts, was resolved in due course by its transition to a more recognizably Western model of economic organization. The same has also been said of other large and important mixed economies, such as India and Brazil.
But the most telling examples, for such accounts, come after the creation of the WTO, when a wide variety of former socialist economies negotiated their accession. In such cases, the process of WTO accession has routinely been used as a mechanism to encourage, or at least recognize and lock in, greater adherence to free market principles. Accession negotiations have paid particular attention, for example, to the progress of ongoing privatization programs, the administration of government pricing policies (for example in the energy sector), and the extent of administrative discretion in the regulation of economic life, and certain obligations on these and other matters frequently found their way into these states’ accession protocols. 15 The process of WTO accession for China was similar to the extent that the carrot of WTO membership was evidently deployed by major trading powers in the WTO as an instrument to encourage liberal economic reform in China. Indeed, China’s Accession Protocol contains a series of China-specific obligations requiring China to establish a domestic market-based order—including obligations regarding a general right to import and export goods, the conduct of state-owned enterprises, the elimination of technology transfer obligations and other performance requirements, as well as a broad obligations to ‘allow prices for traded goods and services in every sector to be determined by market forces’. 16
This historical account has a great deal to commend it, but at the same time is open to a number of important criticisms. Two are particularly relevant here. One is that it almost certainly overestimates the degree of institutional convergence among the majority of GATT contracting parties with market-oriented economies. The fundamental character of global capitalisms as highly variegated social and economic forms is well established. 17 Indeed, the diverse institutional features of many of the most well-known historical market forms—American liberal market capitalism, the German social market economy, French dirigisme, Scandinavian social democracy, the British postwar mixed economy, and so on—were already established in the years either side of World War 2, and clearly present in the GATT membership from the beginning. Furthermore, the perception that economies such as Japan, India and Brazil have largely converged towards a Western model is not borne out by those who have engaged in close study of those economies. 18 Even such convergence as has occurred has by and large been the result of broader historical and political developments, rather than the direct result of GATT/WTO membership. Intra-capitalist institutional diversity has always been a central feature of the global trading system, and of the GATT regime itself. 19
The second, which follows directly from the first, is that it appears to underestimate the importance and centrality to the GATT/WTO system of mechanisms to manage and minimize the frictions caused by trade between different economic systems. 20 One of the most important of these mechanisms was the right to take defensive measures in the form of increased duties against ‘dumped’ or subsidized products. It was Jackson who first perceptively noted that these trade defence mechanisms could act in practice as ‘an “interface” or buffer mechanism to ameliorate difficulties … caused by interdependence among different economic systems.’ 21 As he explained, in the context of institutional diversity, establishing a consensus view of what constitutes ‘fair’ trade—in the sense of establishing a single baseline of universal market institutions—was not only impossible, but also in tension with the pluralist institutional ethos of the GATT just described. The approach taken, then, was to give each state relatively broad freedom to take unilateral defensive action on the basis of its own vision of what constituted fair and unfair trade. Such defensive measures, as long as they were kept within certain bounds, could act ‘as a crude or blunt instrument to cause different economic systems to more equitably share the burdens of adjusting to shifts of world trade flow’. 22 These mechanisms were only relatively rarely used during the first decades of the GATT, but were one of the main instruments used to address the frictions caused by the emergence of East Asian capitalist forms—and in particular the emergence of Japan as a global trading power—over the course of the 1970s and 1980s. 23 And, as is well known, the application of a special set of rules on trade defences, applicable only to China, was a central pillar of the arrangements under which its accession to the WTO was agreed. 24
There were many other mechanisms that allowed countries to respond by lawfully limiting trade with selected countries, where institutional differences were perceived to amount to ‘unfairness’ of one form or another. Others, not typically triggered by allegations of ‘unfair’ trade practices but sometimes performing a similar function, included: safeguards under GATT Article XIX; the careful crafting tariff concessions to favour products that did not give rise to such frictions; as well as all of the ad hoc and idiosyncratic mechanisms familiar to historians of the GATT, from voluntary export restraints, to non-application procedures under GATT Article XXXV, to country-specific rules in accession protocols, and so on. Such mechanisms are not marginal and exceptional, but built into the architecture of the regime, and central to its smooth functioning, as well as its durability over time.
The better view, then, is to see the GATT/WTO’s approach to the problem of institutional diversity as a messy and evolving—but, as it turns out, mostly quite workable—mix of techniques for encouraging convergence around basic liberal market principles, and mechanisms for ensuring managing the tensions caused by institutional diversity. Institutional convergence historically had a role to play in addressing trade frictions, but it was hardly the regime’s only effective response, and perhaps not even the primary one. Indeed, it is probably fair to say that the push towards institutional convergence around liberal market principles provided by the GATT/WTO system has been somewhat weaker than is often stated—and, furthermore, that the precise content and boundaries of such principles have been more ambiguous and ill-defined than is usually acknowledged. Instead, a key lesson is that mechanisms to manage frictions caused by the coexistence of different systems—especially trade defences—have been much more important than is usually recognized in maintaining the stability of the system. That said, it is also true that existing trade defence regimes have proved themselves at times to be a particularly blunt method of fulfilling that function, and are constantly in danger of going beyond it. Even in the 1980s, the US use of such measures against imports from heterodox economies was subject to criticism for being inefficient, politically captured, and ideologically driven. 25
Why is this history important now and why is it central to my argument? In Section VI, I shall argue that a workable legal solution to the problem of institutional diversity is likely to take the form of a set of general principles, which must be interpreted and applied in a way that is deeply sensitive to the intentions and values of its drafters, to the strengths and weaknesses of the WTO as an institution, and to the limitations of what can legitimately be achieved through international adjudication. In that context, this history is crucial. It reminds us that the promotion of liberal market values has been an important objective of the GATT/WTO Membership—but that so too has the preservation of institutional diversity and the capacity for institutional innovation, against which it has always been balanced. It further suggests that expecting to achieve significant domestic institutional change through the application of WTO law is probably unrealistic, and is certainly inconsistent with our historical experience of how the regime works, when it is working well. And it makes clear that having a functional ‘buffer’ mechanism, as Jackson called it, is likely to be crucial in resolving the current tensions. 26 I shall return to each of these propositions in due course.
I said above the current US-China trade war, and the emergence of ‘Sino-capitalism’, raises the fundamental question of how to tell the difference between ‘fair’ and ‘unfair’ institutional conditions of market capitalism, and how to approach irresolvable differences of view as to what ‘fairness’ entails? My aim in the following sections is not to provide a definitive answer to this question—this is almost certainly a fool’s errand, and in any case probably unnecessary as a practical matter—but rather to assess the adequacy of one of the main legal tools, which has emerged in response to it. This legal tool is the concept of the ‘market distortion’, or, equivalently, the notion of the ‘distorted market’. The core idea here is a relatively simple one: where the institutional choice in question ‘distorts’ conditions of competition in the relevant market, it is more likely to be considered ‘unfair’, and to trigger legitimate responsive action. The attraction of such an approach is clear: market distortions are presumptively both inefficient and inconsistent with accepted norms of free and fair competition, so their presence appears to offer a relatively objective, and widely accepted, indication of an illegitimate intervention. The argument I seek to make in the remainder of this paper, however, is that, if the concept of a ‘market distortion’ is to be used as a normative yardstick in this way, it must be accompanied by a sophisticated understanding of the institutionally embedded character of markets—without this, it risks incoherence, and indeed chauvinism. I suggest that such an understanding has so far mostly been absent, I show the problems then that has led to in the WTO context, and then offer some thoughts about how this might be corrected.
The main reason for focussing on this concept here is that it has clearly emerged as a key concept in WTO trade defences law to address precisely the problems relating to Chinese state capitalism noted above. But it is also worth noting that its use is far more general than just the WTO. It has, for example, become very common across a variety of international spaces to speak of certain aspects of China’s capitalist model as ‘distorting’ conditions of competition in global markets, and on that basis to call them into question. 27 Moreover, it has become standard across many different regimes of supranational market governance to judge the legitimacy of state action in the economy in part by reference to an assessment of the degree to which they ‘distort’ market outcomes, or by reference to the cognate concept of ‘competitive neutrality’. 28 The conceptual criticisms that I offer below are quite general, and are just as relevant to these domains of governance as to WTO trade remedies law.
The first context in which WTO law has begun to incorporate the notion of a ‘market distortion’ is in relation to the application of countervailing duties under the Agreement on Subsidies and Countervailing Measures (‘Subsidies Agreement’). As is well known, in certain circumstances, WTO law permits WTO Members to impose countervailing duties to offset the adverse effects caused by ‘subsidies’ granted to imported products. Under the Subsidies Agreement, a subsidy only exists if it can be shown that a government or public body has made a financial contribution, and that a ‘benefit’ has thereby been conferred. 29 The test for the existence of a benefit is in most cases a market-benchmarking test; that is to say, whether a benefit has been conferred should normally be determined by assessing whether the recipient has received the financial contribution on terms more favourable than those available to the recipient in the market. 30
Importantly, the ‘market’ in this test is that which exists within the territory of the Member alleged to be granting the subsidy, in accordance with the market conditions prevailing in that country. 31 It is worth highlighting that this is therefore an internal, not external, market benchmark test. By ‘internal’, I simply mean that the existence of a subsidy is determined not by reference to some universal notion of a ‘true’ market price. Instead, the relevant benchmark price is identified by reference to the actually existing markets of the subsidizing member—the ‘prevailing market conditions’ as they are found ‘in the territory of [the exporting] Member’. 32 It is accepted that such conditions will legitimately vary from country to country, and that therefore so too will the appropriate price benchmark also vary from country to country.
It follows from this that, as an evidentiary matter, one would typically first look for private prices in the country of provision to establish a benchmark price to be used for market comparison. However, a difficulty arises where those domestic market prices are for some reason or another unreliable, as evidence of genuine market prices. There is a narrow and a broad version of this difficulty. The narrow version is focussed on the problem of circularity that arises where a subsidy program in question is so economically significant that it impacts the market as a whole, affecting even the prices, which private actors offer to one another in the ordinary course of business. In some cases, domestic private prices have been disregarded as appropriate evidence of the benchmark market price, because the government has been found to be so predominant a supplier of the good in question that its actions have distorted market prices generally. 33
The broader version of the difficulty is more generally focussed on the problem that the domestic market might be distorted by government action of whatever form, not just the subsidy in question. In a number of relatively recent cases, 34 it has been accepted that private prices in the domestic market of the exporting members may be disregarded for the purposes of the construction of a market benchmark, wherever government intervention causes a price distortion:
the central inquiry under Article 14(d) in choosing an appropriate benefit benchmark is whether government intervention results in price distortion such that recourse to out-of-country prices is warranted, or whether instead in-country prices of private enterprises and/or government-related entities are market-determined and can therefore serve as a basis for determining the existence of benefit. Thus, what would allow an investigating authority to reject in-country prices is a finding of price distortion resulting from government intervention in the market. 35
Where domestic markets are so distorted, then evidence of prices in other markets may be used to establish a market benchmark. 36 These may include world prices, or prices from markets in other countries or regions that are comparable. Importantly, this is an evidentiary matter only: it does not change the fact that the benchmark test is still a fundamentally internal one. The relevant market to be used for the benefit test, that is to say, is still the domestic market of the subsidizing Member—as it would be in the absence of the distorting interventions of that Member. 37 Prices from other markets are used solely as an aid to determine what prices the domestic market might produce, in the absence of that distortion. It follows that an adjustment ought to be made to take into account the differences in relevant market conditions between the foreign or world markets, on one hand, and the domestic market in question (absent the relevant distortions), on the other. 38
This line of jurisprudence is quite general, and in principle not confined to any particular subset of WTO Members. However, it is important to note that this special method for calculating the appropriate market benchmark may be more readily applicable to China, by virtue of Section 15(b) of its Accession Protocol. 39
The second context in which the concept of market distortions is important is in the application of the Antidumping Agreement, and the story behind its emergence here has been told in detail elsewhere. 40 The legal framework is different from that applicable to countervailing duties, but it involves comparable moves. WTO Members are permitted to impose antidumping to counteract the injurious effects of ‘dumping’, and ‘dumping’ occurs when the export price of a product is below its normal value, where normal value is understood to mean the price of the like product when sold in the domestic market in the ordinary course of trade. 41 The calculation of a dumping margin for a product, then, involves a comparison between the export price of a good, and its market value in the domestic market of the export country.
Unlike the subsidies context, where it is a relatively recent development, it has been recognized for a very long time in the antidumping context that there are some circumstances in which actual domestic transactions may be treated as an unreliable indication of the genuine domestic market price (‘normal value’). One of these circumstances, which arose historically in the context of trade with Eastern bloc countries, is where the exporting country is a so-called ‘non-market economy’ (NME). In such cases, it was recognized, as early as the mid-1950s, that a special methodology for calculating normal value may be appropriate, and the differential treatment of NMEs has been an accepted feature of antidumping practice from that time until very recently. 42 As is well known, a condition of China’s accession to the WTO was its agreement that other WTO Members may—in certain circumstances, and for a specified period—treat it presumptively as a NME for the purposes of antidumping proceedings, and apply this special methodology. 43 Though at the time of writing there is some doubt, 44 it was widely expected for some time that this special antidumping regime for China would cease to apply from December 2016.
In part as a consequence, an interpretation of the Antidumping Agreement has emerged, which, it is argued, may permit the use of a similar special methodology for calculating normal value in more general circumstances. The claim is simply that domestic transactions are unreliable indications of true (market-based) ‘normal value’ wherever the relevant domestic market is heavily distorted. In such circumstances, it is argued, the Antidumping Agreement permits WTO Members to calculate normal value using an alternative methodology, based either on export prices to third countries, or to costs of production in the country of origin. 45 Furthermore, where this alternative method uses costs of production in the country of origin, there has also been an effort to introduce the notion of distortions a second time. That is to say, it has been argued that, where domestic costs of production are ‘distorted’—because the domestic input markets are themselves distorted by state intervention—then WTO Members are permitted to use corrected prices in their dumping calculations. The legality of this latter practice, however, is doubtful in light of recent jurisprudence. 46
The upshot is that, under the Subsidies Agreement, and potentially also the Antidumping Agreement, the existence of a ‘market distortion’ in the exporting country can indirectly trigger a right to impose duties on products imported from that country, and/or fundamentally affect the way that they are calculated. Broadly speaking, then, we can say that WTO law distinguishes between at least three different situations. First, there are certain forms of state action that are expressly prohibited by the operation of various WTO agreements (not discussed above). Second, partially overlapping with the first category but mostly quite distinct from it, there are certain forms of state action that count as ‘distortions’ of domestic markets and that, therefore, justify the imposition of duties on importation. The underlying idea here is that these actions creates competitive advantages for exports that can permissibly be treated as ‘unfair’ and that, therefore, can be offset if desired. 47 Third, there are certain kinds of state actions that are neither prohibited, nor treated as justifying corrective duties of any kind but represent simply the background institutional conditions of market activity that states and their populations are free to establish as they see fit. It follows that the precise way in which ‘distortions’ are defined, identified, and quantified for the purposes of WTO law has an important bearing on the content of each of these three groups—in particular on the boundary between the second and third. As a consequence, it is one of the key legal concepts on which hangs the approach of the international trade regime to the question of the appropriate limits of institutional diversity and experimentation more generally.
Given this significance, it is important to ask whether this concept can actually adequately perform the function that is being asked of it. For this purpose, in the next section, I step back to consider (or indeed reconsider) some of the more conceptual difficulties associated with the application of the concept of ‘market distortion’ in a legal context. I shall return in the subsequent section to offer some more detailed thoughts on the case law briefly summarized above, where it will be my claim that the problems in the jurisprudence stem in part from inadequately recognizing these conceptual difficulties.
One of the debates familiar to scholars working in the field of US countervailing duty law, concerns whether or not it is possible to identify objectionable foreign subsidies by singling out those that ‘distort’ the efficient allocation of resources. As long ago as 1972, Schwartz and Harper argued that such a ‘distortion’ test was inoperable and incoherent, for three primary reasons. First, subsidies will not be distortive where they internalize externalities, and externalities are so pervasive as to be incalculable. Second, subsidies may represent the efficient pursuit of what may now be called ‘non-economic’ public preferences, which preferences cannot adequately be taken into account in a distortion analysis. Third, subsidies may offset other governmental burdens, and the practical difficulties of conducting a comprehensive ‘netting’ of all governmental actions are insurmountable. 48 As a result of their analysis, the idea that ‘distortion’ analysis is fundamentally ‘incomplete’ is now well accepted, and a variety of attempts have been made to reground countervailing duty law on different theoretical grounds. 49 In similar but not identical terms, Tarullo has argued against what he has called the ‘market correction’ norm in US countervailing duty law, according to which the role of such law is to correct for market distortions introduced by foreign government subsidies. The application of this norm, he has argued, not only entails impossible calculations if it is to be carried out rigorously, but also (as a consequence) leads in practice to investigating authorities imposing countervailing duties in ways that discipline deviation from an (implicit) understanding of ‘normal’ market relations. 50 Writing in the same year, Alford provided a similarly incisive criticism of the ‘market/non-market’ distinction as the justification for differential treatment in the context of antidumping and countervailing duty law. 51
It will be clear that the arguments I make in this section and the next have been deeply influenced by this literature, and overlap considerably with some of it. In my view, the lessons of this literature have been only very inadequately internalized into thinking about WTO trade remedies law, particularly in the context of US-China relations. 52 That said, the argument I make is articulated quite differently, and that is so for a very specific reason. The question at issue here is not whether a particular subsidy can properly be characterized as ‘distortive’, but rather whether a foreign country’s economy can properly be regarded as ‘distorted’. In that context, I shall be arguing that any coherent answer to this question must involve making a distinction between ‘market distortions’ and ‘background institutional conditions’. This distinction is not central to the above literature, certainly not explicitly. Furthermore, my way of drawing that distinction crucially incorporates an understanding of the role of institutions in economic life that has only clearly emerged since the late 1990s, 53 and which may or may not be shared by the earlier writers on which I drawing. In addition, I shall be arguing that this distinction is not just incomplete but deeply contested, and that as a consequence it is not the sort of distinction, which is amenable to clear and final demarcation by international judicial tribunals by reference to definitive normative or logical principles. This colours what I will have to say in Section VI on ‘solutions’. Again, this claim is present in some of the literature referred to above, but is often left unsaid—primarily because the purpose of that literature, once ‘distortion’ analysis has been discarded as incoherent, is often to find an alternative theoretical basis for trade remedies.
What, then, do we mean when we talk about ‘market distortions’? Broadly speaking, we say that a particular governmental action is a ‘market distortion’ when it affects market behaviours, prices or outcomes in a way which makes them different from what they would have been if they were ‘market-determined’. 54 Importantly, however, this does not include all governmental actions that affect market outcomes in some way, at least not in its ordinary usage. For example, most of us would distinguish between governmental acts that ‘distort’ markets, and those that ‘regulate’ markets. 55 Of course, it may be that, if they are poorly designed in one way or another, we may begin to talk about regulatory actions as distortions, but in the absence of that sort of problem we are more likely to think of this sort of governmental act as either correcting market failures, or simply ‘regulating’ markets in the public interest.
In addition, and more importantly for the purposes of the present argument, we also tend to distinguish between governmental acts that distort markets and governmental acts that ‘enable’ markets to function. 56 For example, where a market is subject to rules that prohibit cartels, or grant legal personality and limited liability to corporations, or define the length of patent terms, or establish rights of minority shareholders, we would not for that reason only speak of these markets as ‘distorted’. Although we might disagree on the appropriate content of the rules necessary to enable markets to function properly, very few of us would suggest that the mere presence of these types of rules renders the market they govern ‘distorted’. Finally, most of us also tend implicitly to hold another category of governmental actions in our heads, namely those that merely exist in the ‘background’ of market relations. For example, in relation to the market for automobiles in a particular country, most of us would not think of rules relating to succession, adoption, water quality, the sale of health insurance, or zoning as ‘distortions’ of that market, but rather as simply the background to it, or the context in which it operates. This is so even though these rules may well, and in fact often do, have profound economy-wide impacts on market behaviour and market prices, in some diffuse and hard to track, but still very real, way.
It is not only that we do make such distinctions in our ordinary language and ways of thinking, it is also that we have to make distinctions of this sort, even if only implicitly, if we want to identify market distortions in a coherent manner. This is true in at least three senses. First, it is necessary as a practical matter, wherever a situated claim is made that a particular measure actually distorts a real-world market. However complicated or involved that claim is, there will always be in practice some rules that are taken for granted for the purposes of analysis. Second, it is necessary as a logical matter, because otherwise one very quickly runs into absurdity. The alternative to making distinctions of this type is to treat every governmental action (contrary to our ordinary ways of speaking) as in principle a distortion, or at least potentially distortive. There is no logical limit. 57 Third, it is necessary as a normative matter, wherever the claim that a governmental measure is distortive is made (as it almost always is) as part of an argument for remedial action of some type—such as removing it, modifying it, compensating for it, and so on. This is because it is a basic assumption of all our real-world political discourse about market discourse that states can legitimately set the basic framework and background conditions for markets, at least to some degree. We always agree that there are some rules that are compatible with the principle of ‘free and fair’ competition in a particular market, even if we might disagree which ones they are. The distinction between rules that ‘distort’ markets, and those that are ‘enabling’ or in the ‘background’, expresses this common assumption.
In the remainder of this paper, I will lump together the categories of ‘enabling’ and ‘background’ rules, and speak of them generally as part of the ‘institutional conditions’ of a market (nothing turns on the distinction between these two categories for the purposes of my argument). The first step in my argument is that when we try to ascertain whether a particular measure constitutes a market distortion—or whether a market is distorted—we need in principle to ask whether or not the measure in question is better understood as part of the institutional conditions of the market. If the measure does more properly fall into that category, it ought not to be described as a distortion of that market at all. This process of characterization is in principle distinct from (even if it is related to) the question of to what extent and in what ways the measure makes market outcomes different from what they would otherwise have been.
Another way of putting this point is that the existence of a market ‘distortion’ can only ever be determined by way of a comparison with a baseline ‘undistorted’ (or less distorted) market, and that this baseline market is always and necessarily a product of a particular set of institutions. 58 The trick is working out whether the state action in question should be treated as part of the institutional baseline, or as an addition to it. So stated, it is clear that this is one version of the broader ‘baseline’ problem very familiar to other fields of legal enquiry. 59
In the remainder of this paper, I shall take it as entirely uncontroversial, at least as an abstract proposition and a matter of principle, that governments can perfectly legitimately help to establish a market’s institutional conditions, and indeed always do so, regardless of the position we take on the justifiability of market distortions. Institutional diversity that results from such choices is a normal, pervasive and broadly positive feature of the global economic order. The proposition that governments can and do ‘set the institutional conditions’ of markets is equivalent to the idea, expressed by others, 60 that governments can and do set the ‘parameters’ within which market competition takes place. I prefer the language of setting ‘background conditions’, however, because ‘setting parameters’ can lead to an important confusion. It can seem to imply that the institutional context merely sets limits and boundaries to competition, or exogenous constraints within which market forces act. This would be a misunderstanding. In fact, we should imagine the work of the legal rules that form part of the institutional context in a different way. They are endogenous to market relations: such rules define the objects of competition, they establish the meaning and methods of fair competition, they make available certain opportunities and resources to economic actors, they facilitate certain business strategies and strategies of interfirm cooperation, and so on. The work they do is already presumed by, and contained in, the notion of ‘market competition’, rather than a set of limits on it. This is what is meant here by the claim that markets are institutionally embedded. 61
This is an important point to clarify because it helps to forestall another potential misunderstanding. It is very common, particularly in discussions about diversity of national economic system in global trade, to categorize different forms of market capitalism along a continuum, which is either ‘closer to’ or ‘more distant from’ an imagined form of ‘pure’ free market capitalism. From this perspective, the question of what degree of institutional diversity should be permitted by the global trade regime often is heard as the question of how far from ‘pure’ market capitalism should its Members be permitted to depart? But it follows from what has just been said that this way of categorizing economic systems can only go so far, and is an inadequate lens through which to understand certain differences between economic systems. Some differences at the level of institutional conditions (in the sense defined above) can affect the nature of market competition, but not its degree. That is to say, some institutional differences have a crucial constitutive 62 influence over the nature, dynamics and methods of competition in the markets they govern—and indeed therefore over the prices and outcomes produced by these markets—but they do so, or can in principle do so, in a manner consistent with perfectly free competition. 63 They produce very different markets, but the differences between them are not best understood in the language of ‘more or less competition’ or ‘more or less freedom’. If the claim that governments are free to set the institutional conditions for markets is to mean anything, it must mean—at a minimum—that these sorts of institutional choices remain entirely for each Member to determine for itself.
The next question is how—by reference to what criteria and principles, or by what alternative method—can any particular governmental measure be characterized as either a ‘distortion’ or part of the ‘background institutional conditions’ of a market? One very common method is what we might call the ‘but-for’ effects test. According to this method, a governmental act counts as a distortion if one can show that it affects market prices, making them different from what they would otherwise be. It will be clear already, from what has been said above, why this is conceptually flawed, at least when stated so baldly. It is a method that can be applied to any governmental act, without logical limit, as long as it has some price impact, no matter how diffuse, general or unpredictable. 64 It provides no real way of isolating distortions from background institutional conditions, because even the latter will often deeply affect market behaviours. Think, for example, what difference it would make to the price of credit if limited liability corporations did not exist, or to patterns of corporate ownership if minority shareholder protections were strengthened.
In truth, however, application the ‘but for’ effects test is always accompanied by a choice, and an assumption, which do a lot of the analytical work. The choice is the decision concerning which measure to analyse as a potential distortion. The assumption is that the rules governing the market used as the comparison in the ‘but-for’ test are all properly characterized as background ‘institutional conditions’. As Tarullo has argued persuasively, when such choices and assumptions are made implicitly, they tend in practice to rely on an unspecified set of ideas of what constitute ‘normal’ market relations. 65 ‘Normal’ here might mean ‘what governments generally do’, or ‘what governments tend to do in the markets with which I am familiar’, or ‘what governments do in what I think of as a perfectly free market’, or something else again. Such ideas are sometimes conscious, sometimes not, but of course they are always ideological, in the most general sense of that word. In some contexts, this does not really matter. 66 But it is self-evidently deficient as an approach where precisely what is at stake is a difference of opinion as to what constitutes ‘normality’—or where the fairness of conditions of competition in the market as a whole are precisely what is at issue. In such cases, the precise ways in which one characterizes governmental acts—as enabling, corrective, background, distortive, and so on—will inevitably be made visible and contested, and as a consequence must be defended as a matter of principle. This is of course precisely the case in the context of international trade governance, and it is the reason why a bare bones application of a ‘but for’ effects test is inadequate in that context.
There are many other ways in which we might instinctively seek to justify the distinctions we draw between different types of governmental measures, not by reference to their effects, but as a matter of principle. Some might, for example, be more likely to characterize a measure as part of the background institutional conditions of a market when its effects on market behaviour are general, or diffuse, or indirect. Others might draw the distinction based in part on the degree of coerciveness of the measure. Others still may see as relevant the intention of the measure: was it intended to favour some market participants over others, or was this outcome reasonably foreseeable? Or we might have recourse to the perceptions of participants in the market in question: do they tend to see it as part of the institutional background, or as an interference with their choices? Or we might treat measures differently depending on whether they affect consumers’ real (authentic) preference rankings. 67 More likely, we may choose some combination of these, or indeed others. Each of them is plausible, each has its difficulties, and each is associated one way or another with different philosophical traditions of thinking about what ‘market freedom’ and ‘fair competition’ can and should mean.
The simple point I want to make for the purposes of my argument here is that once we start down the path of making and justifying these distinctions, then certain serious difficulties arise, which are of particular concern in the context of international trade governance in general, and WTO dispute settlement in particular. I will mention just three of them.
First, while each of the above-stated criteria is plausible in the abstract, they face considerable conceptual difficulties in their application, which will not be able to be ignored in the rigorous context of judicial dispute settlement between states. Take, for example, the Hayekian distinction between general and specific measures, which is often used to help distinguish between enabling rules and distorting interventions. On one hand, the generality of a measure is often an unreliable guide to its competitive impacts. As has long been noted, facially general measures can have very specific competitive effects—and indeed are often implemented in practice in full knowledge of them. A tax law providing for accelerated depreciation for machinery in use for more than 12 hours a day may in practice benefit very specific sectors or companies. 68 It is also worth noting that many of the rules and institutions in place in the Chinese economy, and challenged as distortive, are general in nature. 69 Conversely, specific measures are sometimes best understood as instantiations of broader unstated principles. Is a government-sponsored worker retraining program servicing a local aeronautical company a specific favour to that company or best seen as part of a broader collection of social protections for displaced workers? For Tarullo, this difficulty is so pervasive that ‘nearly any government program may be plausibly characterized as either “generally available” or specifically targeted.’ 70 Similar criticisms have also been made of the distinction between coercive and non-coercive measures. 71
Second, and perhaps more fundamentally, another difficulty is not so much that it is impossible to draw a distinction between different kinds of governmental acts in a principled manner. The deeper point, rather, is that this is self-evidently the stuff of politics, and as such hardly the sort of thing that amenable to judicial pronouncement—particularly at the international level, and particularly in the fraught context of trade. The choice of which principle to use is ultimately a choice between different understandings of what markets are and what kinds of laws are compatible with their existence. The more that international trade dispute settlement panels are asked to draw boundaries around the concept of ‘market distortions’, the more they will necessarily find themselves forced to take a view on heavily (and essentially) contested understandings of what sorts of law are compatible with market relations. Determining whether or not a certain measure counts as a market ‘distortion’ sometimes turns out not only to be a basis on which to determine its fairness, but just as much way of expressing a judgment about its fairness, based on a particular understanding of market relations and fair market competition.
Third, the proper characterization of a measure depends in part on where you stand. This is so in at least two important senses. First, it depends on your analytical frame of reference. Take, for example, a government measure providing subsidies for certain consumers for the purchase of health insurance, or providing for government-funded alternatives to employer-funded health care. If one takes the market for health insurance as one’s frame of reference this will be characterized by most as an ‘intervention’ or ‘distortion’ (whether justified or not, views will differ). But if one’s frame of reference is, say, the market for automobiles, then most would, I suspect, see this as simply part of the background in which the market operates—notwithstanding the fact that, in principle at least, such a measure might have a profound effect on the cost and availability of a skilled workforce for the automobile sector over the long-term. The point is a general one: what counts as a distortion of one market may not count as a distortion of another market (even where it has clear effects in both markets), such that it is often not possible to characterize any particular measure as one or the other in any decisive or final way. Second, what counts as a distortion can also depend on the institutional context in which the question is posed. For example, it is not uncommon in tax policy debates to treat the entire tax system as distortive in some manner or another, and to ask the question how its basic principles might be designed to minimize its distortive impact. In the context of international trade governance, however, it has been firmly established that it is each Member’s sovereign right to choose the basic principles on which its tax policy is established, and that (for the purposes of WTO law), these basic principles cannot be treated as distortive subsidies, but rather as the background conditions against which global trade takes place. 72 Again, the point is general: in reality, what properly counts as a market distortion for the purposes of international trade governance, is inseparably and inevitably linked to the question of what is properly inside and outside the institutional mandate and competence of the WTO.
The point of setting out these difficulties is not to encourage us to simply throw up our hands. It may be that they are insoluble as a conceptual matter, though as I shall make clearer below, they may yet be amenable to a more modest ambition of ‘good enough’ management. Nevertheless, it seems to me tremendously important to set such conceptual and normative problems out clearly and explicitly, because without a keen awareness of them, the application of the concept of ‘distortion’ is almost certain to lead to real problems—for the law, and consequently for the trade regime more generally.
One risk is that of incoherence and arbitrariness. Without clear and explicit reasoning on some of the above questions, the choices made about what does and does not constitute a ‘distortion’ are almost certain to be impenetrable to those whose actions the law seeks to guide and constrain. Another risk, and perhaps the more important one, is that of chauvinism. By this I mean that, without a clear understanding of the problems set out above, and without the careful reflexivity they demand, judicial decision-makers are likely to be guided in their interpretation of the notion of ‘distortion’ either by their own implicit assumptions about what ‘normal’ markets look like, or (more likely) by their perceptions of what WTO Members as a whole consider to be ‘normal’ market relations. The risk, of course, is of a system that treats Members asymmetrically, depending on their distance from what are perceived to be orthodox market institutions at any particular time. The great danger to avoid here is that of normalization. 73 The risk, that is to say, is to approach the question with an implicit vision of ‘normal’ or ‘ideal’ market relations—and to construct a set of rules that operates in practice to discipline deviations from this normal or ideal model. This would be problematic for many reasons, but mostly because it would normatively privilege certain kinds of market models over others, without a clear, or indeed any explicit, justification in either economic theory or shared political values.
In this section, I argue that these risks are already beginning to manifest in the law of the WTO. Both WTO and domestic jurisprudence applying the concept of market distortions is in its infancy, but already it is beset by serious difficulties. I will briefly mention four by way of illustration. All of them, I would argue, stem from an inadequate theorized, or at the very least inadequately articulated, understanding of the institutionally embedded character of markets.
One problem has been the use of what I shall call ‘kitchen-sink-ism’. This is the approach according to which, in order to characterize a market as distorted, one simply catalogues all the governmental actions that can plausibly be said to have an impact on prices and competitive in that market, relying on their sheer number and aggregated weight to make a determination that a market is relevantly distorted. It involves no explicit attempt to distinguish between ‘distortions’ and background ‘institutional conditions’, instead lumping them all together in a single category.
A few examples are worth briefly citing, to give a flavour. One is the EU’s voluminous country report on China, ‘On Significant Distortions in the Economy of the People’s Republic of China for the Purposes of Trade Defence Investigations’. 74 As noted above, the EU’s new antidumping regime permits the disregard of domestic prices or costs in the calculation of normal value where there are ‘significant distortions’ in the market of the exporting country. 75 In determining whether such distortions exist, the Commission must consider a range of factors, including: the significant presence of state-owned or -controlled enterprises; state interference in firm pricing decision or influence over costs; public policies that discriminate or ‘otherwise influenc[e] free market forces’; inadequate bankruptcy, corporate or property laws, distorted wages; and the presence of credit institutions that implement public policy objectives or otherwise are not independent of the state. Accordingly, the European Commission’s country report on China, issued in accordance with its new rules identifies a huge array of sources of distortion in the Chinese economy as a whole, ranging from state and party involvement in corporate management, the basic legal structure of ‘socialist market economy’, the risk assessment practices of financial firms, preferential government procurement practices, mechanisms for allocating land, investment screening systems, divergence of Chinese labour laws and practices from fundamental international labour standards, as well as a wide range of sector-specific policies, from research and development subsidies, preferential loans to favoured enterprises, export restrictions and incentives, tax incentives, land-use cost relief, employment stabilization plans, and much more.
Other examples can readily be found in other jurisdictions. For example, in a number of proceedings, the United States Department of Commerce (USDOC) concluded that Chinese interest rates were distorted, and could not be used as reliable indications of the interest rates that would be established by a functioning market. 76 This determination rested on a large catalogue of factor including that the primary shareholder in the largest banks continued to be the government of China and foreign ownership remained restricted; privately owned Chinese banks had a very small market share; corporate governance reform in the banking sector had not been fully implemented; Chinese banks were required to take into consideration government macroeconomic policies in their lending decisions; they lacked adequate independent risk management capacities; lending rates by Chinese banks were largely undifferentiated; and that the Chinese government directly regulated deposit and lending rates. 77 Similarly, in other proceedings, the USDOC rejected domestic Chinese land prices as an appropriate benchmark on the basis that Chinese markets for industrial law were distorted, citing a number of factors: that the Chinese government was the ultimate owner of all land; that government authorities ‘controlled the supply and allocation of land that could be used by non-[SOEs] for non-agricultural purposes’; that land-use rights were in general poorly defined and weakly enforced; that local governments ‘had often exercised broad and unrestricted powers to expropriate land’, that in practice land-use rights were ‘transferred via “closed-door” negotiations and not via public auctions, tenders or listings as required by law’. 78
For present purposes, the problem with these assessments is not their conclusions, but the manner in which they are carried out, and in particular their failure to distinguish between those measures that are properly characterized as ‘distortions’ and those that are merely baseline institutional conditions. Some of the matters covered by these determinations—the basic structure of property regime, corporate governance, the financial system, labour law, and so on—are precisely those that institutional economists regard as the most significant institutional determinants of different varieties of market capitalism. This does not mean that WTO law needs to take them entirely as given, but there must be some parts of a Member’s legal infrastructure, which in principle must be taken as given for the purposes of benchmark analysis. While we might disagree on precisely where to draw the line, a line simply must be drawn somewhere, if the international trade governance is to adhere even to a minimal toleration of institutional diversity. This, as I understand it, is part of Qin’s point in her discussion of the treatment of land-use rights in US—AD/CVD just described. 79 The problem with ‘kitchen-sink-ism’, then, is that, taken a face value, it allows no room at all for a category of ‘market-enabling’ or ‘market-defining’ governmental actions, and therefore in principle is at odds with the basic idea that governments (or their populations) have a right to establish the institutional parameters for markets.
Importantly, this problem is not just conceptual, but practical. If any kind of state action can in principle ground a determination that a market is distorted, and there is no room for taking ‘background institutional choices’ as given, then it is clear that every WTO Member is vulnerable. Where one country finds itself subject to trade defences on the basis that its markets are significantly distorted, one rational response to would be tit-for-tat retaliation—imposing trade defences in return on the basis of significant state ‘intervention’ in its trading partner. When one is determined enough, it is not difficult to create a catalogue of state actions in any country that affect market prices significantly—in the trivial sense that prices would be different were they not there. The problem with ‘kitchen-sink-ism’ is that it provides no conceptual tools to respond to such claims, and to explain why some state actions affecting prices can be considered ‘distortions’ while others cannot. It therefore provides no meaningful check on endless and escalating tit-for-tat retaliation.
In some cases before WTO dispute settlement, there has, in fact, been a recognition of the problems of ‘kitchen-sink-ism’, and an acknowledgement of the need to draw distinctions between the different kinds of governmental action set out in the previous section. Most recently, the Appellate Body noted:
the concept of ‘price distortion’ is not equivalent to any impact on prices as a result of any government intervention … the determination of whether in-country prices are distorted must be made on a case-by-case basis, taking into account the characteristics of the market being examined, and the nature, quantity, and quality of the information on the record. 80
This is a welcome and potentially important statement of principle. However, it is fair to say that, the relevant cases have so far failed to articulate a consistent—or indeed explicit—set of principles and premises for making this distinction. They have, apparently at least, by and large been based on assertions of common-sense.
One example comes from the landmark subsidies case of Canada—Renewable Energy. 81 As is well known, this case concerned a challenge to a feed-in-tariff scheme established by the Ontario government in an attempt to encourage a greater proportion of its population’s electricity needs to be supplied from renewable sources, and to eliminate the need for coal-fired electricity plants by 2014. To determine whether or not payments under the feed-in-tariff scheme constituted subsidies, the Panel was required to compare these payments with a benchmark ‘market price’ for electricity. For the purposes of this paper, the important point is that the Panel determined that the broad energy policy goals defined by the Ontario government were not themselves interventions in (or distortions of) the relevant market, but merely one of the background conditions or parameters in which the market operated. It is necessary, the Panel noted, to test the FIT payments ‘against the types of arm’s length purchase transactions that would exist in a wholesale electricity market whose broad parameters are defined by the Government of Ontario’. 82 These ‘parameters’ included the elimination of coal-fired electricity plants by 2014, and the promotion of increased use of renewable energy as a minimum defined percentage of supply. 83
On appeal, the Appellate Body disagreed with the Panel for other reasons, but nevertheless adopted the same approach to the domestic policy environment. Like the Panel, it noted that the Ontario government in that case had adopted a set of general strategic goals regarding the electricity sector, 84 and agreed that the appropriate benchmark market price should take into account these goals:
an appropriate benefit benchmark for wind power and solar PV electricity generation in Ontario should be one that, within the parameters of the Government of Ontario’s definition of the energy supply mix, reflects what a market benchmark would yield for wind- and solar PV-generated electricity. 85
Indeed, it is clear from other statement that both the Panel and the Appellate Body clearly understood the need to make the sorts of conceptual distinctions described in Section IV above. ‘A distinction should be drawn’, the Appellate Body famously noted:
between, on the one hand, government interventions that create markets that would otherwise not exist and, on the other hand, other types of government interventions in support of certain players in markets that already exist, or to correct market distortions therein. 86
The notion of government action that ‘creates markets’ has been controversial, and is somewhat different from the distinctions I set out above, but can certainly be profitably understood as a way of articulating the idea that some governmental actions set the enabling conditions in which market competition takes place. The Panel, for its part, made a similar distinction in its decision:
Governments regularly intervene in markets for a variety of reasons including in order to avoid outcomes that are believed to be socially unacceptable, or to address various market failures. For instance, governments may decide to limit the availability of certain products because of human health and environmental concerns, or as the Government of Ontario has done, choose to end the use of a particular production technology (coal-fired electricity generation) for the same reasons. These kinds of actions are designed to internalize the social costs (in the case of negative externalities) and benefits (in the case of positive externalities) of certain actions in the production and consumption decisions of economic agents. However, where such government intervention is limited to defining the broad parameters of a market, significant scope will remain for private actors to operate within those parameters on the basis of commercial considerations. 87
Where government action is of this latter kind—defining the broad parameters of a market—it is appropriate, according to the Panel, to take it as given for the purposes of market-benchmarking under WTO law. 88 The Appellate Body explicitly agreed. 89
A broadly similar set of distinctions was in play in the Panel decision in US—AD/CVDs. 90 Recall that in that case, one of the claims concerned the provision of finance at allegedly below-market rates by Chinese state-owned banks to certain producers of a range of products including steel pipes, other metal pipes and tubes, woven sacks and tyres. Again, in order to determine whether these financing arrangements constituted subsidies, it was necessary to compare them to benchmark market rates for equivalent financing. In this context, the US investigating authority had determined that domestic Chinese interest rates were too distorted to provide an adequate basis for a market benchmark, and instead used an alternative benchmark based on a constructed proxy. This determination rested on a range of actions of the Chinese government in relation to the financial sector. 91 Before the Panel, China argued that all financial markets are, necessarily and inherently, shaped by state action and state policies, if only through the setting of benchmark interest rates by central banks. 92 For China, all interest rates are therefore ‘distorted’. In response, the Panel asserted that there is a ‘clear distinction’
between, on the one hand, the government as the setter and implementer of the general monetary policy of a country; and, on the other hand, the government participating as a lender and/or otherwise intervening in the lending market as such, in a way and to an extent that effectively it is the government, and not the market, that establishes the lending rates. 93
The ‘simple implementation of monetary policy’, the Panel suggested, ‘by definition cannot give rise to distortion’, and must simply be taken as given for the purposes of defining a benchmark interest rate. On the other hand, China’s interventions could not be ‘accepted “as is” as the underlying “market” condition’. 94 This, it seems, was because they were of a different ‘kind and degree’, 95 but the Panel offered no more detailed basis for the distinction, treating it instead as largely self-evident in this case.
The point of drawing attention to these passages is not to agree or disagree substantively with the choices made by the adjudicators in them. It is easy to think of many justifications, and indeed many criticisms, of both of them. As it happens, the distinctions drawn in Canada—Renewable Energy proved to be controversial among close observers of WTO jurisprudence, while the Panel’s decision on this point in US—AD/CVDs did not. 96 The point is simply to note that, as the jurisprudence on market distortions continues to develop, it is inevitable that distinctions such as these will have to be made, and that adjudicators will be forced to articulate and defend the unstated premises on which they are based. It is inadequate to treat them as self-evident or a matter of common-sense. In the absence of a more explicit considered approach, it will be hard to avoid the suspicion that investigating authorities will often find markets to be distorted, ‘merely because they do not accord with an investigating authority’s conception of how markets should be structured and regulated’. 97 And it will be equally difficult for panels to articulate an adequate basis for identifying the circumstances in which they impermissibly do so.
The third confusion concerns the use that has at times been made of world market prices. In some circumstances, the determination that domestic prices are distorted is made on the basis of a significant difference between international and domestic prices. One example comes from the Biodiesel cases, in which the EU’s antidumping authority purported to reject the actual input costs (of soybeans) for producers of biodiesel from Indonesia and Argentina, on the basis that such costs were distorted by export bans placed on domestically produced soybeans. 98 This practice was famously rejected in those cases—but here I am interested in the underlying reasoning, which was not rejected, that the difference in domestic and international prices of soybeans demonstrated both the existence and the magnitude of a price ‘distortion’ in domestic soybean markets in Indonesia and Argentina.
It is one thing to observe that the export bans in place in Indonesia and Argentina affect domestic prices—this seems relatively uncontroversial on the facts of the case. But it is quite another to characterize this effect as indicating a ‘distortion’ of domestic prices. This is a much trickier and more complicated question that it initially appears. It is crucial to remember that world prices are themselves the product in part of a set of institutional conditions, which govern the conditions of competition between participants in world markets. These institutional conditions include not only domestic laws (of all major producer and consumer countries) with cross-border effects, but also international rules governing the relations between them. Once one accepts the specificity and contingency of these institutional arrangements—that is to say, once one accepts that they represent only one way of organizing world markets, rather than an ideal free market—then it becomes clear that there is no reason to think that a deviation between domestic and international prices necessarily and entirely derives from distortions on the domestic side. Moreover, it becomes clear that international prices might not be an appropriate comparator at all. Recall that for the purposes of WTO trade remedies law, in making a determination that a market is relevantly distorted, we are not comparing it to an imagined ‘perfectly competitive’ market in the abstract, but to an undistorted market with the basic structure and characteristics of the domestic market in question. The real question, then, is whether international prices represent a good approximation of what prices would be in that undistorted domestic market.
The answer to this question in turn depends in part on whether or not we take the degree of international openness of that market as given for the purposes of the analysis. If we do, and if the market is relatively closed, then there is very little reason to place much significance on the existence of a differential between international and domestic prices. In the absence of international competitive pressure, there is simply no reason to assume that world prices need necessarily bear any resemblance to the prices that would be produced by an undistorted domestic market with its own quite distinct institutional conditions. If we do not—that is to say, if our reference point is a domestic market with the same institutional characteristics, but fully open to international competition—then the existence of a difference may well be relevant. But it is not at all clear which approach is the right one to take to this issue. Which of a country’s myriad trade barriers ought to be taken as part of the given institutional conditions of its domestic market for the purposes of a distortion analysis, and which should not? Given that no market anywhere is truly characterized by a complete absence of barriers, and given indeed that world market prices reflect in part the prevalence of such barriers, what degree of international openness should be assumed as part of a distortion analysis? A simple comparison of domestic and world prices does not permit us to approach this question in a coherent manner.
The fourth and final problem has to do with a later stage of the analysis, after a domestic market has been found to be relevantly distorted. Recall that where domestic markets are found to be relevantly distorted, then investigating authorities may take into account world market prices, prices in foreign markets, prices constructed from data on the production costs of domestic producers, or indeed other sources of data, in constructing an appropriate benchmark. 99 The current confusion about the notion of distortion has left this area of practice in a poor state.
One issue is that the choice of third country markets rarely takes into account the institutional similarities and dissimilarities between that third country and the domestic market in question. Factors such as market size, level of development, and geographic region are sometimes taken into account, but rarely institutional conditions. One attempt to do so simply sought to control for institutional ‘quality’ based on international indicators of political stability, government effectiveness, and the rule of law. 100 Probably the only exception—and in some respects the exemplar here—is the benchmark analysis in Canada—Renewable Energy, in which proposed benchmarks from electricity markets in other Canadian provinces, as well as some US states, were explicitly rejected as appropriate comparators on the basis of the very different institutional choices that had been made in establishing those markets. 101
In principle, a poor choice of third country market might be corrected by an appropriate adjustment. Recall that, whenever out-of-country data are used, the investigating authority is ‘under an obligation to ensure that the resulting benchmark relates or refers to, or is connected with, prevailing market conditions in the country of provision’. 102 Put another way, the investigating authority must ‘do its best to identify a benchmark that approximates the [domestic] market conditions that would prevail in the absence of the distortion’. 103 Out of country prices are only to be used as far as they can help to generate an ‘estimate of’ (or ‘proxy for’ 104 ) the ultimate benchmark, which is the domestic market of the subsidizing Member—as it would be, absent the distorting interventions of that Member. As a result, adjustments should be made to prices deriving from world markets or foreign markets so as to replicate as far as possible the prevailing market conditions in the domestic market of the exporting member. 105 This should include identifying those background institutional conditions of the domestic market, which should be taken as given for the purposes of benchmark analysis, and adjusting the constructed benchmark so that it reflects the prices that would be established in a market with such institutional conditions.
This simply does not happen at present in any rigorous way. 106 In some cases, domestic investigating authorities simply use out-of-country benchmarks without any significant adjustment. In at least one instance, this was considered acceptable by a WTO panel with little explanation. 107 In other cases the process of adjustment is give only cursory attention. 108 This can at times have bizarre results. Recall, for example, that in US–AD/CVDs, US investigating authorities in the investigations in question had used data on the price for land-use rights in certain regions of Thailand as proxies for the market value of Chinese land-use rights. In response, China argued that USDOC had not even attempted to account for a range of factors that might legitimately lead to differences in the value of land-use rights between China and Thailand: tax treatment of land purchases; the quality of local schooling, hospitals and transport infrastructure, as well as rules concerning their availability and cost; zoning and planning laws; prevailing interest rates and regulatory rules governing the provision of credit for land acquisitions; and so on. 109 Whatever one thinks of the means of distributing land-use rights in China, the underlying logic of China’s complaint here seems reasonable. It is heroic to imagine that land prices in Thailand even approximately reflect prevailing domestic Chinese conditions, when China’s choices on all of these matters are properly taken into account. But the Panel was not convinced:
while there can be no question that the Thailand prices and constructed annual rents relied upon by the USDOC do not provide a perfect picture of what prices for land-use rights would be in China in the absence of the distortions that the USDOC found to exist, it nevertheless is not clear that the fact of adjusting those benchmarks, as such, would ensure a closer approximation of the counterfactual situation (an undistorted land-use rights market in China). 110
In any case, the Panel continued, ‘China has not demonstrated that the USDOC failed to make any specific adjustment which it was required to make or which the Government of China or interested parties had identified, before the USDOC or before us’. 111
Other examples could easily be given. 112 The problem here is not just that the alternative benchmarks used are unrealistic—though that is certainly a real issue. In practice, the use of unadjusted foreign prices can fundamentally transform the market benchmark test from an internal to an external yardstick, judging a state’s actions against norms of market conduct derived from markets, which reflect political and institutional choices that are not that country’s own. Qin has argued persuasively that permitting the use of unadjusted foreign prices as benchmarks in this way is seriously problematic, because it can negate the comparative advantage of the exporting state, and because it fails to take account of the sovereign right of that state to design its market institutions (for her, the focus of attention is on the domestic regime of property ownership) as it sees fit. 113 The latter point seems exactly right. A failure adequately to adjust foreign market prices to reflect ‘prevailing market conditions’ in the exporting state runs directly counter to the idea that all states and their populations have the right to set the institutional conditions in which their domestic markets operate.
Faced with these and other difficulties, a number of the most thoughtful observers of international trade law have suggested discarding the market distortion test altogether, and starting over. 114 One can easily understand why: conceptually, the problems with the concept of ‘market distortion’ are apparently insoluble, at least at the level of principle. On this view, one way of proceeding would simply be to prohibit the use of trade remedies altogether. This is periodically proposed in the literature, and is clearly the favoured approach of some scholars, but it is of course at present not politically feasibly—nor in my view would it be desirable in the absence of a replacement ‘buffer’ mechanism. A more easily imagined way ahead would simply be to pursue transactional solutions to the problem. On this approach, the general rules would simply prohibit the differential treatment of distorted and non-distorted markets in the application of trade defences—but this would be coupled with a parallel process by which states may contract out of these rules on an agreed basis. This may, most obviously, take the form of bilateral or plurilateral agreements by which one state agrees to be treated differently in the context of the application of trade defences, in return (presumably) for other concessions of one kind or another. It is essentially the model of China’s Accession Protocol, but applied on a bilateral, and perhaps permanent, basis. Despite a range of legal difficulties associated with this approach, it may be that we are headed in this direction, at least in the context of US-China relations.
Close observers of the GATT/WTO system will know that ad hoc or exceptional arrangements of this kind have at least informally played an important role in the system in the past, particularly during periods of stress. They can act as safety valves of a sort—practical mechanisms for resolving tensions that might otherwise threaten the system as a whole—and there is no question that they have successfully performed that function in the past. To the extent that they make the pursuit of fairness costly, they may in fact be preferable in some ways to a system of apparently general rules, which is not general in practice: that is to say, a system that permits the costless denial of some of the practical benefits of membership to some Members on the grounds of their legitimate institutional choices. But of course such mechanisms also come with considerable costs and risks. The most obvious risks relate to the general threat such arrangements pose to the idea of a multilateral rules-based order—the idea of a single system of rules, applicable to all WTO Members equally, which governs trading relations on a non-discriminatory basis. But there is another risk that is just as important and perhaps less visible. I noted above that preserving a generalized capacity for institutional experimentation of an open-ended kind is an important systemic good for the global trading order. To the extent that the frictions caused by institutional diversity are resolved through bilateral means, there is a distinct possibility that this global systemic interest is lost—or at least given inadequate weight—in the specific bargains struck between pairs of trading nations. The risk, then, would be the incremental creation of an order, which did not preserve that systemic capacity as far as it should.
The remainder of this section, then, asks what a legal solution to the problem might be, and in particular whether the notion of ‘market distortion’ can be made legally operable despite its conceptual flaws. But it is important to be clear at the start about what we can reasonably expect a legal ‘solution’ of this sort to look like. The problem of institutional diversity is fundamentally a meta-institutional problem. That is to say, it is not a question of the right rules to define fair and unfair competition in a single global market order, but of the right rules to govern competition between differently-instituted market orders, each organized around their own conceptions of fair and unfair competition. It would, at least in present circumstances, 115 be naïve to imagine that a single, shared conception of what constitutes ‘fair’ competition is achievable among the diverse WTO membership. The task is, therefore, neither to produce such a shared conception, nor to reduce an existing shared conception into legal rules. Furthermore, it would be equally naïve to imagine that a solution can be found simply by finding the crafting the right rules and interpreting them in the right way. There is simply no single set of logical or normative principles by which we can give definite content to the notion of ‘market distortion’ without trespassing upon deeply sensitive and contested questions, and we should not imagine that our job is to find such principles. That is just not what a ‘solution’ to this sort of problem looks like.
Instead, what the history of the GATT/WTO system reveals is that the frictions caused by intra-capitalist diversity have in the past been addressed—provisionally, incompletely, but adequately—through a series of messy compromises, expressed in legal rules and concepts that are just ‘good enough’ to help hold together a multilateral system, which its members perceive to be on balance in their own long-term interests. The question for this section is whether the concept of ‘market distortion’—understood as a practical legal tool operationalized in the specific context of the adjudication of international trade disputes—can be made to perform that modest but important task. It seems to me that the answer to this question might be yes, but with the crucially important qualification that this will only be so if it is interpreted and applied on the basis of a fully dereified understanding of the institutionally embedded character of markets. Without this, it is likely to do more harm than good, and to exacerbate the problems to which it is offered as a solution.
We can learn a lot about how to approach the challenge of institutional diversity by looking at the way the WTO dispute settlement system approached the problem of regulatory diversity in its first decades. The bedrock principle of this approach has been that states have the right to set their own regulatory objectives and determine the level of protection they seek in respect of regulatory risks. At every stage, the dispute settlement system has refused to draw bright lines defining the boundary between legitimate and illegitimate regulation, or to given substantive content to any particular understanding of ‘right’ or ‘optimal’ regulation. 116 At the same time, it has, incrementally, and somewhat experimentally, developed a body of open-ended tests and forms of review—such as anti-arbitrariness norms, proportionality review, procedural review, basic tests of good faith and so on—which provide practical assistance in the resolution of regulatory frictions, without trespassing on those bedrock principles. Although no one to my knowledge would suggest that WTO adjudicators have always got it right, this messy mix of legal tools and techniques has to my mind so far done a ‘good enough’ job of addressing regulatory disputes in a sensitive and practical manner.
It seems to me that adopting a similar approach would be profitable in response to the challenge of institutional diversity. The core features of this approach are easy to describe. In this context, the bedrock principle would be that WTO Members can legitimately establish the institutional conditions for their own markets. This principle is already part of the system: as noted above, it is visible throughout the history of the GATT/WTO, and is expressed in the deep structure of WTO law, especially the law relating to trade defences. 117 As a consequence, a clear legal distinction would be drawn between a market ‘distortion’ and a governmental measure, which formed part of the background institutional conditions of that Member’s market. Where a measure is properly characterized as a background institutional condition, it would not be permissible for an investigating authority to use it as evidence of price distortion. In addition, where surrogate prices are used as benchmark for other reasons, this alternative benchmark must be adjusted, as far as reasonable and possible, to ensure that it reflects a market with such background institutional conditions.
Similar to the regulatory context, it would be crucial that the WTO dispute settlement system avoid drawing bright lines around the notion of ‘market distortion’, lest it be seen to be defining the ‘right’ or ‘proper’ institutional arrangements for competitive markets, or filling in substantive notion of ‘fairness’ on which the membership has not agreed. It is important in this respect to recall that, at least in the context of the application of trade defences, it will of course be the importing state’s investigating authority, which makes the determination that a foreign market, or foreign price, is ‘distorted’. It will therefore be for the investigating authority to characterize a foreign government’s measure as either ‘distortive’ or a ‘background institution’ in the first instance. The role for the WTO adjudicative bodies would merely be to review that assessment. In a recent statement of the standard of review applicable to WTO review of investigating authorities’ decision, the AB noted:
it is the role of panels to assess whether the investigating authority’s explanation for its determination is reasoned and adequate by critically reviewing that explanation, in depth, and in light of the facts and explanations presented by the interested parties. 118
In the specific context of a distortion analysis, it noted, ‘panels have to review whether the competent authority’s explanation of how government intervention actually results in price distortion in the markets in question fully addresses the nature and complexities of the data in the record, and whether it appears adequate in light of alternative methods, data, and explanations of that data presented by the parties’. 119 Under the Antidumping Agreement, a separate standard of review is defined in Article 17.6. For factual matters, the panel’s task is to ‘determine whether the authorities’ establishment of the facts was proper and whether their evaluation of those facts was unbiased and objective’. 120 Accordingly, where an investigating authority determines that a foreign market is so distorted that in-country prices are unsuitable as benchmarks, the role of the panel is not to assess whether that determination is ‘correct’ according to some universal standard of ‘distortion’, but rather to assess whether the determination is adequate according to the above standard of review.
This does not mean that the substantive content of the notion of ‘market distortion’ is entirely for each investigating authority to determine for itself according to its own values and standards of fairness. It would be perfectly permissible—and entirely in keeping with the approach WTO jurisprudence has taken in the regulatory context—for international tribunals to develop a non-exhaustive list of indicative factors to be taken into account in such determinations. These would no doubt include, for example: the generality of the measure in question; the objective intention (if any) behind its enactment; the historical and political conditions of its emergence; its systemic functional importance in the economy in question and its interaction with other structural elements of that economy; and so on. But it does mean that the process of WTO review could, and probably often would, focus on matters other than the substantive correctness of the investigating authority’s characterization. For example, indications of partiality—perhaps where such determinations are made inconsistently as between different trading partners, or in an obviously discriminatory manner—may be of particular significance. So too may procedural matters: an assessment may be made, for example, of the adequacy of the opportunities give to a foreign producer to challenge the finding of distortion, and of the seriousness and justifiability of the responses given by the investigating authority to such challenges. Where a Member without justification applies a different standard to its own market as compared to that of a foreign market, this may also in some circumstances be an indication of a failure to meet a standard of objectivity, impartiality and coherence. Similarly, WTO dispute settlement may be particularly concerned to ensure that investigating authorities’ decisions in this regard are scrupulously informed and outward-looking: based as far as possible on an informed and careful understanding of the institutional foundations of foreign markets, rather than an unthinking and parochial projection of ‘normal’ market relations based on their own home context. Those familiar with WTO jurisprudence will recognize almost all of these moves immediately: they have been an important part of the approach adopted by the Appellate Body, for example, in the context of the adjudication of scientific disagreements in the context of the application of the SPS Agreement, 121 as well as its regulatory jurisprudence more generally. They are part of the standard toolset that WTO adjudicators have crafted over time to help them review sensitive and contested questions: simultaneously permitting the possibility of meaningful oversight where appropriate, while avoiding impossibly contentious or practically undecidable issues where necessary.
This approach may also go some way towards ameliorating, in practice, the problems associated with the process of adjustment. I suggested above that a core aspect of the proposed approach would be for investigating authorities to be required rigorously to adjust surrogate prices, wherever they were used, so that the constructed benchmark reflects prices that would exist in a market with any measure it determines to constitute a ‘background institutional condition’ in the exporting foreign market. This is, as noted, essentially what is already required under both the Antidumping Agreement and the Subsidies Agreement, even if it does not seem to be particularly rigorously applied in practice at present. The biggest difficulty with its application, of course, is likely to be its practicality. This is in part because the full range of domestic institutional conditions, which in principle, ought to be taken into account is potentially extremely wide. 122 Importantly, for the purposes of this argument, they can also be political and legal—that is to say, they include at least some aspects of the domestic legal, regulatory, and policy environment that establish the general conditions in which market activity takes place, as long as they can be shown to lead to identifiable and reasonably predictable price differentials between foreign and domestic markets. With such a huge multiplicity of institutional arrangements that could in principle be relevant, it is clear that the process of adjustment could very quickly become unwieldy and unreliable, and certainly not amenable to precise economic modelling. 123 This is especially the case where the distortion is alleged to derive from the complex interaction of different institutional elements.
Again, however, the problem appears more tractable when the standard of review is properly considered. It is for the investigating authority in the first instance to identify the range of institutional aspects that demand adjustment, and to carry out the adjustment in accordance with a methodology of its own choosing. It is for the WTO dispute settlement system to review those actions in accordance with the standard of review set out earlier. Applying this standard of review, it is unlikely that a WTO Panel would require the investigating authority to take into account all conceivable institutional conditions—it would be perfectly permissible to determine whether the subset of institutional conditions actually considered was reasonable and defensible in all the circumstances, taking into account the available information, the arguments of the parties, and other practical constraints. Similarly, a Panel would not be seeking to determine whether the adjustment methodology used produced the right result, but rather whether it represented a ‘reasoned and adequate’ approach to the problem. There is, to be sure, a level of imprecision, uncertainty, and disagreement, which inevitably would have to be tolerated. At the same time, it is not too burdensome to ask that this process of adjustment be evidence-based, coherent, and properly reasoned. It should not be acceptable for an investigating authority simply to declare that adjustment is too complex, and therefore to refuse even to attempt it. 124 And it is perfectly practical to require investigating authorities to make every effort to take the institutional choices of foreign states into account in their analysis of market distortions, and to respond adequately and fully to arguments made in domestic proceedings that they have not properly done so. This is the approach taken in the context of review of scientific risk assessments under the SPS agreement—assessments that can be at least as complex, uncertain, and subject to evidentiary problems, as the process of benchmark adjustment. It has worked reasonably well there.
Of course, an approach of this kind will have both advantages and disadvantages. Its great strength is its flexibility—that is to say, the ability it gives to judicial bodies to craft a basis for decision, which is responsive to the peculiar context in which it is applied, including its political and economic sensitivities. It is an approach appropriate for those situations in which a rigid and unbending standard is unworkable. At the same time, these strengths are also weaknesses. An approach such as that outlined here relies heavily for its acceptability and practical functionality on the adjudicators who apply it—their sensitivity to institutional context, the limits of their own mandates, the ability to understand and respond practically to the specificities of the challenge before them. As such, it is an approach that is most suitable in a dispute resolution system, which is characterized by a high degree of trust in, and deference to, such judicial decision-makers. At least the latter condition (and possibly both) appears less certain in relation to the WTO than it once used to, and its adoption may place great weight on the institution’s credibility at time when it is under the greatest strain.
That said, it seems to me that this approach—with all its flaws, messiness and ambiguity—is hugely preferable to an alternative that appears to be emerging in the jurisprudence. Given the indeterminacy and malleability of the concept of ‘market distortion’, a clear danger of its incorporation into WTO trade remedies jurisprudence is that it opens the door to the proliferation of (protectionist) trade defences, without a clear means of controlling them. One clear response to this concern has been to attempt to limit the applicability of the ‘distortions’ jurisprudence to a very narrow set of circumstances. Recall that, in the earliest stages of this jurisprudence, the Appellate Body confined itself to formulating a very narrow principle, which apparently limited resort to out-of-country benchmarks only where: the government is the only supplier of the particular goods in question; the government administratively controls all the prices for those goods in the country; or domestic private prices are distorted because of the predominant role of the government in the market as a provider of the same or similar goods. 125 It also made clear that this principle should be cautiously and rarely applied. 126 More recently, as the door has opened to arguments based on the existence of distortions of an unspecified but clearly broader kind than those originally contemplated, a different way of narrowing the application of the distorted markets methodology has been emphasized. This to limit its use to only those cases in which the distortions are of a sufficient magnitude or significance, or of a particularly serious or widespread character. As noted above, the EU has adopted the same approach in its new antidumping regulation with its concept of a ‘significant’ market distortion. 127 Furthermore, the Appellate Body has already indicated that this is likely to be part of its preferred approach, noting that ‘the market from which a benchmark is selected for the purpose of a benefit analysis need not be completely undistorted or free [from] any government intervention.’ 128
The aspirations behind this approach are clear. By limiting its application to apparently ‘extreme’ cases of particular obvious or widespread intervention, one hope is no doubt that it will garner more support. Where the anticipated targets are few in number and predictable, it may be thought that fewer countries will fear the consequences for their own exports, and that the systemic impact of trade defences triggered by ‘distortions’ will be kept within reasonable bounds. It may also be thought that limiting its application to only very clear or substantial examples of serious distortion will limit the need to engage with the hard, borderline cases, and thereby reduce the risk of being drawn into the thankless task of defining the notion of market distortion in a rigorous or precise way.
However, it seems to me the reality is likely to be quite different. On one hand, if the out-of-country methodology were limited to cases involving the circularity problem, it would have very limited utility in resolving the broader tensions between institutional variants of capitalism. On the other hand, limiting its application to only very significant distortions risks being counterproductive, and is unlikely to lead to the benefits imagined. There are two possible trajectories, each with its own associated problem. It may be that adopting this approach would lead in fact to the use of a substantially different trade defence regime for probably only a small handful of countries—similar to the existing NME regime, but based on different and less secure foundations. This would risk reinforcing the sense that, in practice, the concept of ‘distorted markets’ is a tool for disciplining only radically new and different or unfamiliar market forms, or for neutralizing the competitive threat from geopolitical rivals. This would legitimately give rise to the complaint that it serves primarily to justify differential treatment of a relatively small set of countries, whose primary shared characteristic is their institutional distance from a perceived understanding of normal market relations. Alternatively, limiting the application of the principle to ‘significant’ distortions may do little to limit the tit-for-tat retaliation, which I have argued above is likely to occur in response to the application of the market distortion test. If this were the case, it would become clear that this approach does not, in truth, avoid the problem of having to define the nature and boundaries of the concept of ‘distortion’. This is a conceptual problem that is almost certain to arise one way or another, whether or not it is qualified by the language of ‘significance’.
One final point is worth mentioning. Taking some elements of a Member’s legal environment as ‘given’ for the purposes of market distortion does not mean freeing them entirely from scrutiny. It is better to think of it as something approaching a proportionality test, which has been tried and tested in other areas of WTO jurisprudence. The clearest illustration of this comes from the Appellate Body’s reasoning in Canada—Renewable Energy. As noted above, the Appellate Body suggested that the appropriate benchmark to establish whether the feed-in-tariffs constituted subsidies should take into account the broader strategic goals of the Ontario government. Accordingly, to repeat the passage cited earlier:
an appropriate benefit benchmark for wind power and solar PV electricity generation in Ontario should be one that, within the parameters of the Government of Ontario’s definition of the energy supply mix, reflects what a market benchmark would yield for wind- and solar PV-generated electricity. 129
The appropriate benchmark, that is to say, is the price produced by a market in which renewable energy constitutes the proportion of supply as defined in the Ontario government’s strategic goals. Put another way, the question of whether or not a subsidy existed was equivalent to the question whether the Ontario government set prices for renewable energy higher than was strictly necessary in order to achieve their preferred energy supply mix. Note that while this approach did take Ontario’s strategic goals as a given part of the prevailing market conditions, this was not equivalent to simply accepting the subsidy measures themselves as part of those conditions—in fact, the Appellate Body itself hinted at a number of points that the Ontario government may in fact have set prices excessively high, even taking into account its legitimate policy goals. 130
In the language of this paper, the determination in this case was that fundamental policy context of Ontario’s electricity market—that is to say, the Ontario government’s supply mix goals—was treated as a ‘background institutional condition’ for that market, and was therefore required to be reflected in the market benchmark used for comparison. Conceptually, this has much to commend it: any other possibility is both nonsensical (because there is no reason to think that the government would give up on those objectives even in the absence of the specific measure in question) and incorrect (because nothing in the WTO agreements ought to constrain outright the pursuit of such objectives, provided that they are legitimate, and genuinely public). As a consequence, the relevant benchmark analysis effectively tested whether the support program had raised prices higher than was strictly required to meet public policy objectives for the sector. In this way, institutionally sensitive market-benchmarking was applied in a way which approached the means-end tests associated with proportionality review. 131 This is far more comfortable ground for an adjudicative body than being asked to delineate precisely the boundaries of such a fraught and contested concept as ‘market distortion’. Pragmatically, it brings the jurisprudence on market distortions closer in its substance to the substantial body of jurisprudence on trade-restrictive regulatory measures for legitimate public policies—a smart move, given that this latter jurisprudence is more evolved, and more generally accepted, than the former.
Admittedly, it will not always be possible to perform this sort of analysis. For one thing, it will very often be the case that the relevant institutional conditions of the domestic market do not take the form of overt, legitimate public policy goals as they did in the Canada—Renewable Energy case. Most of the core institutional conditions—labour market arrangements, regimes of property rights, mechanisms for allocating finance, corporate governance system, and so on—cannot realistically be treated as embodying discrete and identifiable public policy goals, and as a result are simply not amenable to the application of this sort of means-end proportionality test. For another thing, this approach will only be practical where it does not lead to the circularity problem. That is to say, it is likely to provide an acceptable solution only where a meaningful distinction can be drawn between the precise public policy measure, which is being challenged (the effect of which must be tested), and the broader public policy goals which that measure expresses (which ought to be taken as given, and controlled for). This will not always be possible.
I have argued in this paper that contemporary trade frictions should be understood, in part, as being driven by radical changes in the institutional foundations of global capitalism since the 1990s. In particular, they reflect the emergence since that time of a wide range of new ‘varieties of capitalism’, of which China’s hybrid market economy is only one. The emergence of these new market forms, and their deeper integration, is a sign of the strength of the system, but it is also a source of serious friction. Such frictions have emerged before, and the GATT/WTO system has evolved a series of practical techniques—expressed and embedded in its legal framework—for addressing them, in a more or less effective manner. The expiration of section 15(a)(ii) of China’s Accession Protocol, coupled with a range of other long-simmering concerns, has placed this existing legal and institutional framework under pressure. This has in turn given rise to new legal techniques to manage this pressure, one of the most important of which is the concept of ‘market distortion’ that has emerged, quite explicitly, as a response to a perceived Chinese threat. We have already proceeded some distance down the road of embedding this concept as a central feature of international trade governance, but in my view we have done so with an inadequate appreciation of the difficulties it will necessarily lead us into. The concept can be a useful one, but it brings with it certain risks. The potential result is a system of trade defences targeted in a discriminatory and even punitive manner against heterodox institutional forms, in ways which can only dis-incentivize institutional experimentation. Instead of being a blunt but potentially legitimate instrument to pursue goals of stability, cost-sharing or distributional equity, as Jackson originally described, trade defences would in this future be (or, perhaps, continue to be) a more questionable instrument of competitive commercial policy or ideological projection. In response, I have argued for an approach to the interpretation and application of this concept, which proceeds from an understanding of the institutionally embedded character of markets. This does not take the form of a readily available ‘solution’, but rather a messy and evolving set of legal techniques that, in the best case, will form the legal basis of a practical and justifiable approach to the frictions caused by institutional diversity. A toolkit of legal techniques of this kind cannot, of course, take the place of a more foundational political settlement of some sort, but it is a necessary accompaniment to it, if we are to preserve the aspiration towards a genuinely non-discriminatory and rules-based global economic order.
As I explain further below, my focus on legal techniques does not imply that contemporary trade instability is resolvable through the application of legal techniques, but rather that finding an adequate set of such techniques is a necessary element if a durable solution is to be found, which preserves the aspiration towards a rules-based trading order.
The literature is voluminous. For one comprehensive recent overview of China’s response to WTO litigation see Zhou, China’s Implementation of the Rulings of the World Trade Organization (Oxford: Hart, 2019). One very useful set of perspectives on China’s transformation, now a decade old but still important, can be found in Brandt and Rawski (eds), China’s Great Transformation (Cambridge: CUP, 2008).
Within legal scholarship see, eg, Wu, ‘The “China, Inc.” Challenge to Global Trade Governance’, 57(2) Harvard International Law Journal 261–324 (2016); Du, ‘China’s State Capitalism and World Trade Law’, 63(2) International and Comparative Law Quarterly 409–448 (2014). See also, Ferguson, ‘We Are All State Capitalist Now’, Foreign Policy (February 2012); Wooldridge, ‘The rise of state capitalism: something quite new’, The Economist (19 January 2012). McNally’s informed account uses the term ‘Sino-capitalism’: McNally, ‘Sino-capitalism: China’s Reemergence and the International Political Economy’, 64(4) World Politics 741–776 (2012).
Statement by Ambassador Dennis Shea, ‘Views on China’s trade-disruptive economic model and implications for the WTO’, WTO General Council, Geneva, 26 July 2018, available at https://geneva.usmission.gov/2018/07/27/55299/, last accessed 30 September 2019. See also the statements to be found in China’s most recent Trade Policy Review, Trade Policy Review: China, Minutes of the Meeting, WTO Doc WT/TPR/M/375, 21 November 2018, and Joint Statement on Trilateral Meeting of the Trade Ministers of the United States, Japan, and the European Union, New York, 25 September 2018, available at https://ustr.gov/about-us/policy-offices/press-office/press-releases/2018/september/joint-statement-trilateral, last accessed 30 September 2019.
Statement by H.E. Ambassador Dr. Zhang Xiangchen at the WTO General Council Meeting, Geneva, 26 July 2018, available at http://wto2.mofcom.gov.cn/article/chinaviewpoins/201807/20180702770676.shtml, last accessed 30 September 2019.
See the literature cited below nn 17–18.
Fligstein and Zhang, ‘A new agenda for research on the trajectory of Chinese capitalism’, 7 Management and Organization Review 39–62 (2011), at 47.
See, eg, Hall and Soskice, D. (eds), Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press, 2001).
Bown rightly makes the point that, in principle, it is not sufficient to object to a trading partner’s institutional choices without a clear account of the specific cross-border costs that they impose, and has provided a helpful way of thinking about the different costs, which might arise: Bown, ‘The 2018 US-China Trade Conflict After 40 Years of Special Protection’, 12(2) China Economic Journal 109 (2019), especially 126. For the purposes of my argument, it seems to me that Bown’s first category (adjustment costs) is relevant, as is the notion that institutional choices in one country may impact on the sustainability of institutional arrangements in another, but probably most importantly that core notions of fairness of competitive conditions have a residual importance in trade policy over and above concerns about the precise nature of cross-border economic externalities.
See, eg, Hall and Soskice, above n 8.
Mavroidis and Sapir also helpfully characterize the debate in this way in their excellent recent paper, Mavroidis and Sapir, ‘China and the WTO: Towards a Better Fit’, Bruegel Working Paper, Issue 06, 11 June 2019, available at https://bruegel.org/wp-content/uploads/2019/06/WP-2019-06-110619_.pdf, last accessed 30 September 2019, at 6.
For accounts of the GATT’s relations with socialist economies, see, for example, Kostecki, East-West Tade and the GATT System (Basingstoke: Palgrave Macmillan, 1979); Reuland, ‘GATT and State-Trading Countries’, 9 Journal of World Trade Law 318 (1975); Grzybowski, ‘Socialist Countries in GATT’, 28 American Journal of Comparative Law 539–554 (1980); Grzybowki, ‘East-West Trade Regulations in the United States, The 1974 Trade Act, Title IV’, 11 Journal of World Trade Law 506 (1977); Irwin, Mavroidis and Sykes, The Genesis of the GATT (Cambridge, Massachusetts: Cambridge University Press, 2008). More recently, see Mavroidis and Sapir, above n 11.
Key contributions to this debate include Wu, above n 3, and more recently Mavroidis and Sapir, above n 11. See also Lang, ‘Heterodox market orders in the global trade system’ in Santos, Thomas and Trubek (eds), Globalization Reimagined: A Progressive Agenda for World Trade and Investment Law (Anthem, 2019), Chapter 8.
Mavroidis and Sapir, above n 11, at 5 and generally. Though note that Mavroidis, writing with Janow, also rightly recalls the GATT’s openness to different systems as one of its ‘central strength[s]’: see Mavroidis and Janow, ‘Free Markets, State Involvement, and the WTO: Chinese State-Owned Enterprises in the Ring’, 16:4 World Trade Review 571–581 (2017), at 572.
See, eg, UNCTAD, ‘“The Non-Market Economy” Issue In International Trade In The Context Of WTO Accessions’, UNCTAD/DITC/TNCD/MISC.20, 9 October 2002, available at http://unctad.org/en/docs/ditctncdmisc20_en.pdf, last accessed 30 September 2019; Michalopoulos, ‘World Trade Organization Accession for Transition Economies: Problems and Prospects’ 36 (2) Russian & East European Finance and Trade 63–86 (2000); Geraets, Accession to the World Trade Organization: A Legal Analysis (Cheltenham, UK: Edward Elgar, 2018), and more recently Mavroidis and Sapir, above n 11, at section 3.1. More generally, for an insightful account of club dynamics in the history of the GATT, see Nicolas Lamp, ‘The Club Approach to Multilateral Trade Lawmaking’ 49 Vanderbilt Journal of Transnational Law 107–190 (2016).
WTO, Accession of the People’s Republic of China—Decision of 10 November 2001, WT/L/432 (23 November 2001) (‘China Accession Protocol’), eg, sections 5, 6, 7(3), 9. See generally Qin, ‘“WTO-Plus” Obligations and Their Implications for the WTO Legal System’, 37(3) Journal of World Trade 483–522 (2003).
I refer here generally to the voluminous literature, which has grown since the late 1990s on comparative capitalisms. See generally, eg, Peck and Theodore, ‘Variegated capitalism’, 31 Progress in Human Geography 731–772 (2007); Wood and Lane (eds), Capitalist Diversity and Diversity within Capitalism (London: Routledge, 2012); Hall and Soskice, above n 8; Lane and Myant (eds), Varieties of Capitalism in Post-Communist Countries (Basingstoke: Palgrave Macmillan, 2007); Jackson and Deeg, ‘Comparing capitalisms: understanding institutional diversity and its implications for international business’ (2008) 39 Journal of International Business Studies 540–61; Amable, The Diversity of Modern Capitalism, (Oxford, UK: Oxford University Press, 2003).
In respect of Japan (and other East Asian capitalisms) see for example: Witt and Redding, The Oxford Handbook of Asian Business Systems (Oxford: Oxford University Press, 2014); Witt and Redding, ‘Asian business systems: institutional comparison, clusters and implications for varieties of capitalism and business systems theory’, 11(2) Socio-Economic Review 265–300 (2013); Boyer, Uemura and Isogai (eds) Diversity and Transformations of Asian Capitalisms (Abingdon, UK: Routledge, 2012); Tipton, ‘Southeast Asian Capitalism: History, Institutions, States, and Firms’, 26 Asia Pacific Journal of Management 401–434 (2009); Amable, above n 17; Berger and Dore (eds) National Diversity and Global Capitalism (Ithaca, NY: Cornell University Press, 1996); Storz, Amable, Casper and Lechavalier, ‘Bringing Asia into the comparative capitalism perspective’, 11(2) Socio-Economic Review 217–232 (2013); Whitley, Business Systems in East Asia: Firms, Markets and Societies (London: Sage, 1992); Hundt and Uttam, Varieties of Capitalism in Asia: Beyond the Developmental State (Basingstoke: Palgrave Macmillan, 2018); Walter and Zhang (eds), East Asian Capitalism: Diversity, Continuity and Change (Oxford: Oxford University Press, 2012).
This story told more fully in Lang, above n 13.
John Jackson, The World Trading System: Law and Policy of International Economic Relations, 1 st ed., (Cambridge, MA: MIT Press, 1989), at 218.
Jackson, above n 20, at 218; also Jackson, Davey and Sykes, Legal Problems of International Economic Relations, 3rd ed. (Eagan, Minn: West Publishing, 1995), at 668–72, 1140–42. For discussion of this view, see Kennedy, ‘The International Style in Postwar Law and Policy: John Jackson and the Field of International Economic Law’, 10(2) American University International Law Review 671–716 (1995), esp 707 and surrounding.
Jackson, above n 20, at 244.
See, eg, Bown and McCulloch, ‘U.S.-Japan and U.S.-China Trade Conflict: Export Growth, Reciprocity, and the International Trading System’, 20(6) Journal of Asian Economics 669–687 (November 2009), available at https://www.chadpbown.com/wp-content/uploads/2019/01/Bown-McCulloch-JAE-2009.pdf,, last accessed 30 September 2019; Bayard and Elliott, Reciprocity and Retaliation in U.S. Trade Policy (Washington, DC: Institute for International Economics, 1994); Zeng, Trade Threats, Trade Wars: Bargaining, Retaliation, and American Coercive Diplomacy (University of Michigan Press, 2004).
China Accession Protocol, above n 16, section 15, esp paras (a) and (b).
Tarullo, ‘Beyond Normalcy in the Regulation of International Trade’ (1987) 100(3) Harvard Law Review 546–628. See also generally the literature referred to in n 49 below.
Arguably, it is the decreasing functionality of the latter—especially trade defences—which has helped to precipitate the present crisis. On one side, there is the sense that WTO law on subsidies and antidumping no longer permits states to take adequate responsive action to what they perceive as the unfairness of emergent forms of East Asian capitalism. I am referring here not only to the operation of section 15(a)(ii) of China’s Accession Protocol (though the effect of this is unclear given the suspension of proceedings in European Union—Measures Related to Price Comparison Methodologies, DS516, which effectively permits the continued use of an NME methodology against China at least for the present), but also to the well-known controversies around the interpretation of ‘public body’ in Article 1 of the SCM Agreement, and around the use of various forms of ‘zeroing’ under the Antidumping Agreement. For an important recent discussion, which calls into question the prevailing wisdom that WTO jurisprudential prudential developments have in fact constrained the use of trade defences against China, see Zhou and Gao, ‘China’s SOE Reform: Using WTO Rules to Build a Market Economy’ 68(3) International and Comparative Law Quarterly (2019) (forthcoming). It is also in part because the application of trade defences to Chinese products now in many cases leads only to trade diversion, rather than neutralising the underlying competitive threat they pose: see eg, Bown, above n 9. On the other side, there is the opposite perception—that the rules have for too long unfairly permitted the disproportionate targeting of China on account of its institutional choices, which are formally permitted under the rules, and that China is paying too high a price for its heterodox institutional experimentation.
See, eg, European Commission, ‘On Significant Distortions in the Economy of the People’s Republic of China for the Purposes of Trade Defence Investigations’, Commission Staff Working Document, 19 December 2017, SWD(2017) 483 final/2, available at http://trade.ec.europa.eu/doclib/docs/2017/december/tradoc_156474.pdf, last accessed 30 September 2019; USTR, ‘Findings Of The Investigation Into China’s Acts, Policies, And Practices Related To Technology Transfer, Intellectual Property and Innovation under Section 301 Of The Trade Act Of 1974’, 22 March 2018, available at https://ustr.gov/sites/default/files/Section%20301%20FINAL.PDF, last accessed 30 September 2019; OECD, Competitive Neutrality: Maintaining a Level Playing Field between Public and Private Business (Paris: OECD Publishing, 2012); OECD, State-Owned Enterprises and the Pinciple of Competitive Neutrality, OECD Doc DAF/COMP(2009)37, (Paris: OECD, 2009).
The language has become important, for example, in the context of competition law, state aid, OECD rules relating to export subsidies, among others. For one contribution which proceeds from a broader view of the importance of this principle generally, see Virtanen and Valkama, ‘Competitive Neutrality and Distortion of Competition: A Conceptual View’ 32(3) World Competition 393–407 (2009).
Uruguay Round Agreement on Subsidies and Countervailing Measures (SCM Agreement), Art 1.
WTO Appellate Body Report, Canada—Measures Affecting the Export of Civilian Aircraft (Canada—Aircraft), WT/DS70/AB/R, adopted 20 August 1999, para 157.
SCM Agreement, Articles 1, 14. Appellate Body Report, United States—Final Countervailing Duty Determination with Respect to Certain Softwood Lumber from Canada (US—Lumber IV), WT/DS257/AB/R, adopted 17 February 2004, para 90ff.
SCM Agreement, Article 14.
Appellate Body Report, US—Lumber IV, above n 31, para 100 and surrounding.
Appellate Body Report, United States—Definitive Anti-Dumping and Countervailing Duties on Certain Products from China (US—AD/CVD), WT/DS379/AB/R, adopted 25 March 2011, para. 429ff; Appellate Body Report, United States—Countervailing Measures on Certain Hot-Rolled Carbon Steel Flat Products from India (US—Carbon Steel (India)), WT/DS436/AB/R, adopted 19 December 2014, para. 4.155 and surrounding; Panel Report, United States–Antidumping and Countervailing Measures on Certain Coated Paper from Indonesia (US—Coated Paper), WT/DS491/R, adopted 22 January 2018; Appellate Body Report, United States—Countervailing Duty Measures on Certain Products from China (US—CVD (China)), WT/DS437/AB/R, adopted 16 January 2015, para. 4.50 and surrounding; Appellate Body Report, United States—Countervailing Duty Measures on Certain Products from China, Recourse to Article 21.5 of the DSU by China (US—CVD (China)(21.5)), WT/DS437/AB/RW, adopted 15 August 2019, para 5.137ff.
Appellate Body Report, US—CVDs (China)(21.5), above n 34, para 5.141.
Eg, Panel Report, United States—Definitive Anti-Dumping and Countervailing Duties on Certain Products from China (US—AD/CVDs), WT/DS379/R, adopted 25 March 2011, as modified by Appellate Body Report WT/DS379/AB/R, para 10.187.
Appellate Body Report, US—Lumber IV, above n 31, paras 89, 96, 106, 108; Appellate Body Report, US—Carbon Steel (India), above n 34, para 4.158; Panel Report, US—AD/CVDs, above n 37, para 10.187.
China Accession Protocol, above n 16, section 15(b): ‘In proceedings under Parts II, III and V of the SCM Agreement, when addressing subsidies described in Articles 14(a), 14(b), 14(c) and 14(d), relevant provisions of the SCM Agreement shall apply; however, if there are special difficulties in that application, the importing WTO Member may then use methodologies for identifying and measuring the subsidy benefit, which take into account the possibility that prevailing terms and conditions in China may not always be available as appropriate benchmarks. In applying such methodologies, where practicable, the importing WTO Member should adjust such prevailing terms and conditions before considering the use of terms and conditions prevailing outside China’.
For a selection of the relevant literature, see, eg, Shadikhodjaev, ‘Non-Market Economies, Significant Market Distortions, and the 2017 EU Anti-Dumping Amendment’, 21(4) Journal of International Economic Law 885–905 (2018); Shadikhodjaev, ‘Input Cost Adjustments and WTO Anti-Dumping Law: A Closer Look at the EU Practice’, 18(1) World Trade Review 81–107 (2019); Kluttig, Tietje and Franke, ‘Cost of Production Adjustments in Anti-Dumping Proceedings: Challenging the Raw Material Inputs Dual Pricing Systems in EU Anti-dumping Law and Practice’, 45(5) Journal of World Trade 1071–1102 (2011); Vermulst, Sud and Evenett, ‘Normal Value in Anti-Dumping Proceedings against China Post-2016: Are Some Animals Less Equal Than Others?’ 11(5) Global Trade and Customs Journal 212–228 (2016); Du, above n 3; Bown, above n 9; Zhou, ‘Appellate Body Report on EU-Biodiesel: The Future of China’s State Capitalism under the WTO Anti-Dumping Agreement’, 17(4) World Trade Review 609 (2018); Noel and Zhou, ‘Replacing the Non-Market Economy Methodology: Is the European Union’s Alternative Approach Justified under the World Trade Organization Anti-Dumping Agreement?’, 11–12 Global Trade and Customs Law 559 (2016); Bungenberg et al. (eds), The Future of Trade Defence Instruments: Global Policy Trends and Legal Challenges (Switzerland: Springer, 2018).
Uruguay Round Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade (Anti-Dumping Agreement), Article 2.1.
See GATT 1994, Ad note Article VI:1(2): ‘It is recognized that, in the case of imports from a country which has a complete or substantially complete monopoly of its trade and where all domestic prices are fixed by the State, special difficulties may exist in determining price comparability for the purposes of paragraph 1, and in such cases importing contracting parties may find it necessary to take into account the possibility that a strict comparison with domestic prices in such a country may not always be appropriate.’
China Accession Protocol, above n 16, section 15(a).
As noted above, at time of writing, proceedings have been suspended in European Union—Measures Related to Price Comparison Methodologies, DS516, which effectively permits the continued use of an NME methodology against China at least for the present.
For example, such domestic transactions may not be ‘in the ordinary course of trade’, or there may be a ‘particular market situation’, or there may be in too low a volume of sales in the domestic market: Anti-Dumping Agreement, Articles 2.1, 2.2.
See, eg, Appellate Body Report, European Union—Anti-Dumping Measures on Biodiesel from Argentina (EU—Biodiesel (Argentina)), WT/DS473/AB/R and Add.1, adopted 26 October 2016.
I am grateful to Weihuan Zhou for the observation that there is a degree of incoherence in the antidumping law here, because if the institutional choice of a foreign government can properly be said to cause an unfair distortion, why should the pricing practices of private parties be relevant? Why, in other words, should only those sales which can be characterized as ‘dumping’ be offset, rather than all sales affected by the unfair distortion?
Schwartz and Harper, ‘The Regulation of Subsidies Affecting International Trade’ 70 Michigan Law Review 831 (1972). See also Goetz, Granet and Schwartz, ‘The Meaning of “Subsidy” and “Injury” in the Countervailing Duty Law’ 6(1) International Review of Law and Economics 17 (1986).
See, for a selection, Barcelo, ‘Subsidies and Countervailing Duties—Analysis and a Proposal’, 4 Law and Policy of International Business 779 (1977); Diamond, ‘A Search for Economic and Financial Principles in the Administration of United States Countervailing Duty Law’, 21 Law and Policy of International Business 507 (1990); Trebilcock, ‘Is the Game Worth the Candle—Comments on a Search for Economic and Financial Principles in the Administration of United States Countervailing Duty Law’, 21 Law and Policy of International Business 723 (1990); Alford, ‘When is China Paraguay? An Examination of the Application of the Antidumping and Countervailing Duty Law of the United States to China and Other Nonmarket Economy Nations’ 61 Southern California Law Review 79 (1987); Tarullo, n 25; Lantz, ‘The Search for Consistency: Treatment of Nonmarket Economies in Transition Under United States Antidumping and Countervailing Duty Laws’, 10(3) American University International Law Review 993 (1995).
Tarullo, above n 25.
Alford, above n 49.
Some literature which does do so, includes, (probably most prominently) Sykes, ‘The Questionable Case for Subsidies Regulation: A Comparative Perspective’, 2(2) Journal of Legal Analysis 473 (2010); Langille, ‘General Reflections on the Relationship of Trade and Labor (Or: Fair Trade is Free Trade’s Destiny)’ in Bhagwati and Hudec (eds), Fair Trade and Harmonization: Prerequisites for Free Trade? (Cambridge, Mass.: MIT Press, 1996); Gagne, ‘Policy Diversity, State Autonomy, and the US-Canada Softwood Lumber Dispute: Philosophical and Normative Aspects’, 41(4) Journal of World Trade 699 (2007); Qin, ‘Market Benchmarks and Government Monopoly: The Case of Land and Natural Resources under Global Subsidies Regulation’, 40 University of Pennsylvania Journal of International Law 575 (2019); also related is Zheng, ‘The Pitfalls of the (Perfect) Market Benchmark: The Case of Countervailing Duty Law’, 19(1) Minnesota Journal of International Law 1 (2010). See also generally, some of the concerns expressed in WTO, Committee on Anti-Dumping Practices—Minutes of the Regular Meeting Held on 27 April 2017, G/ADP/M/52, 28 July 2017, paras 7.1ff, as well as subsequent meetings of the same Committee.
See above nn 7, 17–18.
The terms takes on shades of meaning which derive, in part, from different ways of understanding what is meant by ‘market-determined’ behaviours, outcomes and prices. Sometimes, especially in technical discourse, we mean those behaviours, outcomes and prices which would be produced in a perfectly competitive market. Sometimes we mean those behaviours, outcomes and prices that would be produced a market which is characterized by some looser notion of ‘free and fair’ competition. And sometimes we simply mean those behaviours which would have been produced by the market in the absence of that governmental act—that is to say, something is a market distortion if it changes what market participants would otherwise do.
It is also possible, although uncommon, to conceptualize bona fide regulations for a public purpose as ‘justified distortions’. As it happens nothing turns on that for the purpose of the present argument.
For another way of expressing this point, see Tarullo, above n 25, at 553, 558–9. See also, importantly, Kennedy, A World of Struggle: How Power, Law and Expertise Shape Global Political Economy (Princeton University Press, 2016), pp42-44 and generally.
Langille, above n 52, at 247; Tarullo, above n 25.
For a clear expression of this point, see Vanberg, The Constitution of Markets: Essays in Political Economy (London and New York: Routledge, 2001), 21.
A classic statement is Sunstein, ‘Lochner’s Legacy’, 87 Columbia Law Review 873 (1987).
See n 82 below, and accompanying text.
It will be clear to those who are familiar with the very schools of institutionalism that I am taking a specific position here in debates within that field. I am setting out here what Jackson and Deeg call the comparative capitalisms approach to the economic function of institutions: Jackson and Deeg, above n 17, esp 549.
On different meanings of ‘constitutive’ in this context, see Lang, ‘Market Anti-naturalisms’ in Desautels-Stein and Tomlins (eds), Searching for Contemporary Legal Thought (CUP, 2017), at 312–329.
Or, equivalently, we at least we need to conceptually distinguish between the influence of a measure on the nature of competition and its role in restricting or interfering with competition (if, as may be the case, the two tend to go together). Vanberg’s way of putting this distinction is that between the ‘how’ and the ‘how much’ of competition; see Vanberg, above n 58, fn 17 to Chapter 1.
See also Langille, above n 52; Gagne, above n 52.
Tarullo, above n 25.
Among those who share a broadly similar vision of what is ‘normal’, for example, this approach can be mostly unobjectionable, which may be why it is so often adopted uncritically in domestic policy discourse. And it may also not matter tremendously in the context of a debate in which all participants share the view that an existing market functions well, is characterized by essentially free and fair competition, or for other substantive reasons should as far as possible be kept as it is.
This is, as I understand it, the significance of the notion of ‘outside factor’ in Virtanen and Valkama, above n 28.
I take this example from Tarullo, above n 25, at 562.
Take, for example, the complaints which have been made regarding China’s currency policy, corporate governance system and intellectual property rules. There is no guarantee that a measure which is implemented to enable market competition does so in a non-discriminatory or neutral manner.
Tarullo, above n 25, at 561.
For one view, see Hale, ‘Coercion and distribution in a supposedly non-coercive state’, 38 Political Science Quarterly 470–494 (1923); Hale, ‘Force and the state: a comparison of ‘political’ and ‘economic’ compulsion,” 35 Columbia Law Review 149–201 (1935); Hale, ‘Bargaining, duress, and economic liberty’, 43 Columbia Law Review 603–628 (1943).
See generally, Appellate Body Report, United States—Tax Treatment for ‘Foreign Sales Corporations’, WT/DS108/AB/R, adopted 20 March 2000.
For a classic statement of this risk in the context of the application of trade defences, see Tarullo, n 25.
European Commission, above n 27.
See paragraph 6a to Article 2 of Regulation (EU) 2017/2321 of 12 December 2017 amending Regulation (EU) 2016/1036 on protection against dumped imports from countries not members of the European Union and Regulation 2016/1037 on protection against subsidized imports from countries not members of the European Union.
See, for example, the range of investigations at issue in Panel Report, US—AD/CVDs, above n 37, which in return relied in significant part on the United States’ 2007 Coated Free Sheet Paper investigation, as well as the earlier Lined Paper investigation.
Panel Report, US—AD/CVDs, above n 37, eg paras 10.111, 10.133ff, 10,144ff. See also Appellate Body Report, US—AD/CVDs, above n 34, para 503.
Panel Report, US—AD/CVDS, above n 37, at para 10.79.