Neoliberalism and The FormalTechnical Turn

7
Embedded Liberalism and Purposive Law

This chapter examines the GATT during its first two decades, with particular
attention to the operation of its legal regime, and its approach to domestic regulation. In the first section, I draw on a range of existing histories of the trade regime to
recall the way that the regimes purpose was collectively imagined during this period,
and the role which law was understood to play in the achievement of these purposes.
In the second section, I show how this ideational framework, and the institutional
processes which reflected and embodied it, produced a peculiarly limited and modest
approach to the oversight of domestic regulation during this period.
I. The nature and purpose of the post-war trade regime

A. Two philosophies and the embedded liberal compromise
The planners of the post-war international economic order were guided in large part
by a common desire to avoid a repetition of what had occurred during the interwar
years.1 International trade relations had been severely disrupted during the First
World War, as participants sought to encourage domestic production of products
essential to war, and to limit imports from and exports to their enemies.2 In the
immediate aftermath of the First World War, a number of factors combined to
entrench this trend towards protection. Given widespread perceptions of the fragility
of the international peace established in Versailles, former belligerents continued to
prioritize self-sufficiency as an economic policy, particularly in respect of products
considered essential. The same was true also of the numerous smaller European states
formed in the aftermath of the First World War after the break-up of the major
European empires, with the result that Eastern and South Eastern Europe were
transformed into an area bristling with frontiers and barriers to trade.3 In addition,
on the periphery of the international economic system, those countries which had
1
G Curzon and V Curzon, The Management of Trade Relations in the GATT in A Shonfield (ed),
International Economic Relations of the Western World, 19591971, Volume 1: Politics and Trade (OUP,
London 1976) 143, 144; RE Hudec, The GATT Legal System and World Trade Diplomacy (2nd edn,
Butterworths, Salem, NH 1990) 5.
2
See generally G Curzon, Multilateral commercial diplomacy: the General Agreement on Tariffs and
Trade and its impact on national commercial policies and techniques (Michael Joseph, London 1965) 20ff.
3
ibid 21.
World Trade Law after Neoliberalism. Andrew Lang.

Oxford University Press 2011. Published 2011 by Oxford University Press.
191
developed industrial capacity for the first time during the war were reluctant to
permit the rapid disappearance of these industries as patterns of trade returned to
normal, and resorted to trade measures as a means of protection.
Between 1920 and 1929, at a series of conferences in Brussels (1920), Genoa
(1922), and Geneva (twice in 1927), members of the League of Nations attempted to
address these developments in commercial policy. Though they neither endorsed nor
pursued full liberalization, they reiterated their formal commitment to a series of
principles, including the undesirability of exchange controls as limits on international trade, the removal of obstacles created by instability in administrative
and legal measures, the progressive suppression of import and export prohibitions,
and the transparency and stability of tariff restrictions on trade.4 In part as a result,
non-tariff barriers to tradewhich at that time essentially meant prohibitions,
exchange controls, and quantitative restrictionswere largely removed by the mid1920s, with the process continuing even as late as 1930. Tariffs, on the other hand,
were not reduced, and indeed in most countries continued to rise, sometimes
spectacularly.5 France, Germany, the United Kingdom and the United States all
markedly increased their applied duties during this period, as a way of easing the
domestic impacts of the profound economic instability of the period.
The onset of the Depression from 1929 brought these efforts to re-establish a
more liberal international economic order, such as they were, to a rapid halt. For
primary producing countries, the rapid fall in global agricultural prices, combined
with the withdrawal of US capital after the stock market crash of 1929, led to falling
incomes at the producer level and severe balance of payments problems at the
country level. Economic instability in one country rapidly spread internationally,
both because of falling global demand for world exports, and as a result of surges of
cut-price imports from countries seeking sources of foreign exchange. In the face of
very high levels of domestic unemployment, countries reluctantly sought to insulate
themselves from global economic instability through renewed resort to barriers to
trade. Most famously, the United States introduced the Smoot-Hawley Tariff Act of
1930, substantially increasing its already high tariff levels, which prompted retaliatory measures from other countries around the world which saw no other option
than to follow suit. These higher tariffs were accompanied by myriad new or
resurrected forms of more exotic trade barriers, including not only import prohibitions, but also quotas, exchange controls, local content and local mixing requirements, andespecially after the collapse of the Gold Standard in 1931
competitive currency devaluations.6 Over the next few years, three more specific
trends emerged: the development of far-reaching government intervention in the
agricultural sector, including state trading and price stabilization regimes, requiring
4
See generally WA Brown, The United States and the restoration of world trade: an analysis and appraisal
of the ITO charter and the General Agreement on Tariffs and Trade (Brookings Institution, Washington, DC
1950) 2937.
5
Curzon (1965) 22.
6
G Kenwood and A Lougheed, Growth of the International Economy 18202000 (Routledge, London
1999) 198205.
192
The Trade Regime and the Neoliberal Turn
virtually complete insulation of the sector from international trade; the establishment of closer economic relations among small groups of countries, providing
mutual preferences for one anothers products on a discriminatory basis, including
the British Imperial Preferences; and the growth of international cartels in Europe as
a means of fostering greater economic stability in certain industrial sectors.7 The
effect of all these developments was dramatic, with the value of world trade falling by
about a third in the two years after 1929, and the persistence of trade barriers
dampening global demand and hampering economic recovery throughout the
1930s.
There were efforts to reverse this catastrophic failure of commercial policy from
the mid 1930s, most prominently through the United States Reciprocal Trade
Agreements Act of 1934, and the network of twenty bilateral trade agreements
which the United States signed between 1934 and 1939 under the leadership of
US Secretary of State Cordell Hull. But these were undermined by the political
instability of the period, and it was not until the final years of the Second World War
that efforts to reconstruct a multilateral trading order began seriously to take shape.
While the post-war plannersinitially, officials in the United Kingdom and United
Stateswere united by a common desire not to return to the trading system of the
1930s, there was no clear agreement on the appropriate lessons to be learnt from the
experience of the interwar period. There were, broadly speaking, two philosophies8
which emerged at this time, with somewhat different interpretations of the causes
and consequences of commercial policy during the 1930s, and different prescriptions
for the post-war trading order. As outlined in Chapter 2, both of these philosophies
had an important influence on the post-war GATT regime.
The first philosophy, which tends to be better remembered today within legal
scholarship on the trade regime, is represented in the figure of Cordell Hull, who was
a major influence on post-war economic planning on the American side, and
brought a clear set of views about the importance of a broadly liberal international
trading order to world peace and prosperity.9 The proliferation of barriers to
international trade during the interwar years, within this philosophy, had caused
or at the very least intensified and prolonged the Depression of the 1930s by
suppressing world demand. More than that, the trade wars of the period led in a
straight line to the outbreak of the Second World War, by inflaming political
hostilities and resentment in foreign countries:
Nations have more often than not . . . raised up trade barriers with complete disregard for the
damaging effects on the trade and livelihood of other peoples . . . The resultant misery,
bewilderment and resentment, together with other equally pernicious contributory causes,
Brown (1950) 3842.

Curzon and Curzon (1976) 149.
RN Gardner, Sterling-dollar diplomacy: Anglo-American collaboration in the reconstruction of multilateral trade (Clarendon Press, Oxford 1956) 102. For a more detailed account see WR Allen, The
International Trade Philosophy of Cordell Hull, 19071933 (1953) 43 American Economic Rev 101,
116.
8
9
193
paved the way for the rise of those very dictatorships which have plunged almost the entire
world into war.10
Specifically, the practice of discriminatory tradethe formation of trading blocs, and

the use of commercial policy to reward friends and punish enemieswas seen to be
particularly problematic, and undermining of international peace. Discrimination, it
was argued, is invariably resented by the countries which are discriminated against,
and three centuries of experience demonstrates that under all circumstances [it]
operate[s] to poison international relations and to make more difficult the task of
maintaining international harmony.11 Though also shared by many British economists,12 this preference for a more open, non-discriminatory, and liberal trading
order was particularly prominent in the United States, which emerged from the war
as the worlds industrial engine, confident of its capacity to compete in virtually all
industrial sectors around the world.13 It entered negotiations over the post-war
economic order with a goal of eliminating all non-tariff barriers to tradeat least
as that term was understood at the timewith an exception only for balance of
payments measures.14
Although this represents the more recognizably liberal strand of post-war thinking
about the international economic order, writers of the period have, as noted in
Chapter 2, consistently stressed that it was far from a doctrinaire classical position on
free trade, certainly outside a few individuals in the State Department.15 As Viner
famously asserted, in a 1947 article in support of the compromised Havana Charter:
. . . there are few free traders in the present-day world, no one pays any attention to their views,
and no person in authority anywhere advocates free trade. The practical issue turns on whether
existing trade barriers should be reduced, and if so, how much . . .16
While Viner may be somewhat overstating the case here for emphasis, the point is an
important one. As Kock has noted, while domestic debate in the United States
focused on the reinstitution of free trade as part of the maintenance of the American
free enterprise system, both of those terms remained somewhat nebulous in their
meaning.17 Certainly, it is clear that free trade usually meant the removal of
quantitative restrictions, exchange controls, and the other administrative controls
of trade which had proliferated during the 1930s, rather than the total elimination of
10
US Secretary of State Sumner Wells, 1941, cited in JH Jackson, World trade and the law of GATT: a
legal analysis of the General Agreement on Tariffs and Trade (Bobbs-Merill, Indianapolis, IN 1969) 38. For a
clear expression of this general philosophy see C Wilcox, A Charter for World Trade (Macmillan Co, New
York, NY 1949) 5.
11
J Viner, Conflicts of Principle in Drafting a Trade Charter (1947) 25 Foreign Affairs 612, 616.
12
13
JK Galbraith, Economics in Perspective: A Critical History (Houghton Mifflin, Boston, MA 1987)
160.
14
Hudec (1990) 15.
15
GJ Ikenberry, A world economy restored: expert consensus and the Anglo-American postwar
settlement (1992) 46 Intl Org 289, eg 305 and generally.
16
Viner (1947) 613.
17
K Kock, International trade policy and the GATT: 19471967 (Almqvist & Wiksell, Stockholm
1969) 7.
194
tariffs.18 Moreover, free trade meant non-discriminatory trade above allthat is,
the re-establishment of the norm of unconditional MFN treatment in the post-war
trading order, and the elimination of regional blocs and preference regimes as far as
possible. The post-war planners in the liberal mould, then, were far from ideologues.
They were careful to avoid taking a doctrinaire position on trade matters, and sought
also to avoid the impression during the Havana Charter negotiations of seeking to
impose a particular vision of liberal capitalism on the rest of the world.19
The second philosophy, more prominent in the United Kingdom than in the
United States in part as a result of the influence of Keynes on economic thinking in
that country, proceeded from a somewhat different starting point. This philosophy
remembered the Depression primarily as an experience of unemployment and
economic instability. From this perspective, it was the instability of free markets,
combined with the prolonged and regular periods of unemployment that accompanied such instability, which had led to the political revolutions of the first half of
the twentieth century, and ultimately the descent into a second world war.20 The
fundamental problem facing post-war planners, on this view, was the problem of
unemployment, and the first and over-riding goal of economic policy must be the
attainment of full and stable employment. This prioritization of full employment
policies governed the attitude towards international trade of those who shared this
way of thinking. On one hand, it was fully acknowledged that international trade
would support employment growth, by opening foreign markets to domestic exports
and boosting global demand. At the same time, there were at least two serious
difficulties with a policy of free trade. First, pursuing a free trade policy opened up
domestic economies to sources of instability caused by economic fluctuations in
major trading partners. The experience of the Depression taught policy-makers to
expect and plan for bleak times,21 and they knew that they needed to maintain some
ability to insulate themselves from international sources of instability. Many
countries were particularly unwilling to expose themselves again to the notorious
instability of the American economy,22 which had proved so disastrous in the
interwar years. Secondly, the pursuit of a full employment policy through the
kinds of inflationary demand management policies developed by Keynes could
lead to serious balance of payments difficulties, particularly if other countries did
not follow suit.23 The United Kingdom in particular sought to ensure that it was able
to impose major trade restrictions in such cases, without triggering another trade war
of the kind that was seen during the early 1930s.
This second philosophy, then, by no means jettisoned freer trade as an important
goal of the post-war trading system. It was, however, based on a causal account of the
dynamics of the international economy which stressed economic integration as a
18
19
20
21
22
23
ibid 8; Viner (1947) 617.

Viner (1947) 614, 627.
See generally K Polany, Origins of our time: the great transformation (V Gollancz, London 1945).
Viner (1947) 622.
ibid 621; Kock (1969) 10.
195
cause of instability, and gave lesser weight to trade as a driver of prosperity. In

Hansens words:
We have learned that the removal of trade obstacles undertaken by itself alone is not a feasible
program. A liberal trade policy must be undertaken side by side with a program of development, expansion and full employment throughout the world . . . 24
More than that, there were important circumstances in which the goal of liberalizing
trade must, in this view, give way to the higher priority of maintaining full and stable
employment.
As noted in Chapter 2, both of these two philosophies influenced negotiations
over the shape of the post-war trading order, and both were reflected in the text of
Chapter IV of the Havana Charter, and therefore ultimately the GATT 1947.25 In
some respects, the evolution of the GATT over the following two decades can be
understood as a tug of war between the two.26 But there was enough overlap
between the two to form a compromisethe compromise of embedded liberalism,
discussed in Chapter 2.27 Key players in the negotiations shared a common view of
the legitimacy of state intervention to secure domestic stability, even if they disagreed
on the precise form and depth that that intervention should take in particular
circumstances.28 The post-war international trading order was therefore designed
in a way which safeguarded the ability of states to pursue policies of domestic
stability, while at the same time moving away from the bilateralism and discrimination of interwar commercial policy towards a more open and multilateral international trading system.
It is therefore common among historians of the trade regime to talk of the initial
GATT as being built upon, and sustained by, a shared normative commitment to
interventionist domestic policies of a broadly Keynesian kindthat is, a broadly
shared understanding of what constituted the normal and legitimate purposes for
which governments might intervene in economic life. Of course, it is important to be
clear about precisely what was shared, and to what extent. Ikenberry, for example,
argues that the negotiation of post-war monetary and trade order was enabled by the
existence of a community of policy specialists who articulated a more or less
coherent governing philosophy of postwar economic order, and that this shared
basic vision among specialists defined a middle ground in negotiations, enabling the
possibility of bridging political divisions.29 Hudec argues that the international
conferences of the interwar period had given trade policy officials a chance to wrestle
with the meaning of the typical provisions in trade treaties of the time, and in the
24
AH Hansen, Americas Role in the World Economy (WW Norton & Co, New York, NY 1945) 92;
see also Kock (1969) 10.
25
Kock (1969) 270, 2734.
26
27
JG Ruggie, International Regimes, Transactions, and Change: Embedded Liberalism in the Postwar
Economic Order (1982) 36 Intl Org 379. See pp 2930 above.
28
ibid 394.
29
Ikenberry (1992) 292ff.
196
process, to clarify among themselves what those generalities meant.30 For the most
part, he said, they found themselves in complete agreement about particular issues
of substantive policy, even where there might be underlying differences of broader
political vision.31 Furthermore, it should be stressed that this shared substantive
policy consensus as regards specific trade instruments was primarily shared among
the three or four key countries in the negotiations, and in particular the United States
and the United Kingdom, rather than all Contracting Parties to the GATT. The
maintenance of a consensus for the following two decades had much to do with the
continuing authority of these countries in the GATT during this time, and was
enabled by the fact that outsiders [to this group of key countries] had accepted the
GATT from the beginning with the knowledge that it was meant to be the proprietary instrument of the inner group.32
We should not forget, too, that the consensus underlying the post-war GATT
regime was the product of a particular political and material context. As we saw in
Chapter 2, for example, the sustainability of the embedded liberal compromise over
the first decades of the GATTs existence depended in part on the marginalization of
the interests of developing countries within the regimein particular, the effective
exclusion of agriculture and textiles from liberalizing commitments under the
GATT. The delicate compromise between international liberalization and domestic
stability could not have survived politically if the industrialized West had opened its
borders fully to imports of agricultural products and light manufactured goods from
the developing world. Indeed, as we shall see, it was in significant part the rise of the
newly industrializing countries over the course of the 1960s and 1970s which led to
the erosion of embedded liberalism around that time. Moreover, the GATT consensus was made possible precisely because GATT membership was limited almost
exclusively to a small number of countries, all of which had domestic market
economies33and, importantly, because the GATT was in part understood as a
way of maintaining Western unity during the Cold War, by placing it on a firm and
stable economic footing.34 The Cold War too, then, provided part of the material
and political basis for the continuation and acceptance of a particular collective
political commitment in the post-war GATT.
That said, the fact remains that a fairly strong set of shared understandings existed
about what the purpose and objectives of the regime were, and what its legal
provisions did and did not mean. The fundamental and primary purpose of the
post-war regime was international stability, in the specific sense of preventing a repeat
30
Hudec (1990) 8.
ibid 9.
ibid 131.
33
EH Preeg, Traders and diplomats: an analysis of the Kennedy round of negotiations under the General
Agreement on Tariffs and Trade (Brookings Institution, Washington, DC 1970) 24.
34
Curzon and Curzon (1976) 177; Preeg (1970) 5 (the GATT was a pillar of solidarity among noncommunist countries); TB Curtis and JR Vastine, The Kennedy Round and the future of American trade
(Praeger, New York, NY 1971) 3 (the solidarity and prosperity that would be promoted by the Kennedy
Round promised to strengthen the Wests ability to oppose communism and soviet imperialism); IM
Destler, American Trade Politics (Institute for International Economics, Washington, DC 2005) 7.
31
32
197
of the disastrous trade wars of the early 1930s.35 Such stability, it was understood, was a
necessary precondition of the flourishing of the social democratic welfare state. This,
broadly understood, was the historically situated collective project that participants
understood themselves to be engaging in. Thus, although the removal of trade barriers
was of course an important intermediate purpose of the regime, liberalization was in fact
a less important norm during the first two decades of the trade regimes history than is
often assumed.36 Liberalization was pursued not through the application of a rigid
principle, but only as far as states were practically able, and only as far as was consistent
with the broader norm of economic stability. This emphasis on stability, moreover,
meant that measures taken to maintain domestic economic stability and full employment took precedence over international commitments wherever they came into serious
conflict. Thus, Article XII:3(d) recognized that as a result of domestic policies directed
towards the achievement and maintenance of full and productive employment . . . a
contracting party may experience a high level of demand for imports involving a threat
to its monetary reserves, and permitted the use of quantitative trade restrictions in such
circumstances. This was highly significant, as such domestic policies could have major,
if indirect, impacts on trade.37 Similar exceptions were inserted for domestic price
stabilization programmes for agricultural products and other commodities, as well as for
the use of safeguard measures in situations of destabilizing and unforeseen import
surges.38 More generally, there were provisions for the waiver or renegotiation of tariff
and other commitments contained in the GATT, available to be activated when
necessary.39 More important than the norm of liberalization, then, at least in these
early years, were the norms of non-discrimination and reciprocity. Non-discrimination,
especially MFN treatment, helped to prevent further balkanization of international
trade and the subjugation of commercial policy to the destabilizing influences of
broader geopolitical priorities. Reciprocity helped to ensure that the priority of the
trade regime during this period was to maintain the balance of concessions between
Contracting Parties, such that when unforeseen disturbances occur[red] to upset this
balance . . . the organizations energies [were] devoted to restoring the original balance as
quickly as possible.40
This normative emphasis on stability was also reflected in the institutional form of
the GATT in the post-war period, in three primary ways. The first was through the
creation of a system for the management of trade disputes which focused on the goals
of containment and non-escalation, rather than the rigid enforcement of legal rules.
I will say more about this in a moment. The second was the self-conscious policing of
the boundary between the low politics of international trade and the high politics
35
DA Irwin, The GATT in Historical Perspective (1995) 85 American Economic Rev 323, 326.
JA Finlayson and MW Zacher, The GATT and the regulation of trade barrier: regime dynamics
and functions (1981) 35 International Organization 561, 570.
37
Viner (1947) 619.
38
See eg GATT, Arts XI, XIX.
39
GATT, Arts XXV, XXVIII.
40
G Curzon and V Curzon, GATT: Traders Club in RW Cox and HK Jacobson (eds), Anatomy of
Influence: Decision Making in International Organization (Yale University Press, New Haven, CT 1973)
298, 321.
36
198
of foreign policy more generally.41 In the light of what were seen as the destabilizing
and destructive influences that geopolitical concerns could have on the conduct of
commercial policy, the GATT and its participants cultivated an apolitical ethos.
They sought to insulate themselves from the broader world of international politics,
and resisted any attempts to connect trade issues with what were understood as
matters of high politics.42 Trade negotiations, wrote Gerard and Victoria Curzon in
the 1970s, tend to be self-contained . . . [and] attempts to use totally unconnected
political or economic levers with which to extract trade concessions are rare and
deeply resented when they do in fact occur.43 In disputes involving political
elements, the Contracting Parties made every effort to avoid involvement in the
political substance of the matter, leaving the responsibility for this to the competent
organizations.44 On those relatively rare occasions where, for example, East-West
tensions unavoidably spilled over into the GATTsuch as the breakdown of trade
relations between the United States and Czechoslovakia in the late 1940sthe
GATT contracting parties adopted, as far as possible, an avoidance strategy, pragmatically accepting the facts as they were, on the basis that it was not their role to
mediate in a political issue arising out of the cold war.45
Thirdly, this self-consciously insular sensibility was reflected also in the more
general social autonomy of the community making up the field of international trade
governance at the time. The early GATT has often been compared more . . . to a club
than an international organization or an international commercial parliament:
The discussions within four walls, the attempts at reconciliation rather than head-on conflicts
by affirmation of rights, the private meetings to talk things over are all features of this clublike atmosphere.46
Hudec, too, described the early GATT as a cluba place where like-minded
officials could communicate without having to spell things out in confrontationproducing clarity.47 More recently, Weiler has described this atmosphere in these
terms:
A very dominant feature of the GATT was its self-referential and even communitarian ethos
explicable in constructivist terms. The GATT successfully managed a relative insulation from
the outside world of international relations and established among its practitioners a closely
knit environment revolving round a certain set of shared normative values (of free trade) and
41
RN Cooper, Trade Policy is Foreign Policy (1972) 9 Foreign Policy 18.

Curzon and Curzon (1975) 328; Kock (1969) 273.
Curzon and Curzon (1976) 163, 199.
44
O Long, Law and its limitations in the GATT multilateral trade system (M Nijhoff, Boston, MA
1985) 813, giving some examples from a somewhat later period.
45
Curzon (1965). For one list of where geopolitical issues did influence GATT practice, see Long
(1985) 813; see also Curzon and Curzon (1973) 328.
46
Curzon (1965) 523.
47
RE Hudec, The GATT Legal System: A Diplomats Jurisprudence (1970) 4 J of World Trade L
615, 635.
42
43
199
shared institutional (and personal) ambitions situated in a matrix of long-term first-name

contacts and friendly personal relationships.48
The Curzons note that, while delegates were of course subject to pressures from
protectionist interest groups in their capitals, there was an absence of protectionist
lobbies in Geneva, and the comings and goings of their countrys delegates to and
from Geneva seemed to have escaped their influence.49 The point is not that this
social separationthe relatively small size and club-like feel of the GATT regime
meant that decision-makers within the regime were insulated from the influence of
protectionist lobbies, or directions from their capitals back home, and were free
agents in any simple way. There is ample evidence, if any is needed, that trade
negotiators consulted with domestic industries, even if not to the same extent as they
do today.50 The point, rather, is that this structurally insular aspect of the field is the
direct institutional and social analogue of the characteristic mindset of trade professionals of the timea mindset which was oriented towards the maintenance of stable
international commercial relations as a primary objective of the regime, and which
actively maintained a self-consciously apolitical posture as a way of managing and
containing the potentially destabilizing influence of domestic and international
political forces. It is this shared ethos and collective sense of purpose, I will argue,
that deeply structured the nature, operation, and evolution of the GATTs legal
disciplines on domestic regulation during that time.
B. The role of law: purposive law and embedded legal forms

Before I turn to explain that impact, however, there is another important part of the
ideational context of the post-war GATT that needs to be addressed. In addition to
ideas about the precise role and function of the GATT in the post-war international
order, there was also another set of ideas about the role that law could and could not
play in the achievement of the GATTs objectives. Hudec has explained that during
the negotiations over the Havana Charterwhich, as explained in Chapter 2, were
ongoing during the drafting of the GATTthere were two currents of thinking
about the place of law and legal obligations in establishing the post-war international
economic order.51 On one hand, there were those who were determined to create a
harder legal infrastructure than the ineffective multilateral conventions promulgated by the international economic conferences of the interwar period.52 They
48
JHH Weiler, The Rule of Lawyers and the Ethos of Diplomats: Reflections on the Internal and
External Legitimacy of WTO (2001) 35 J of World Trade 191, 192; see also RO Keohane and JS Nye, Jr,
The Club Model of Multilateral Cooperation and the World Trade Organization: Problems of Democratic Legitimacy in R Porter et al (eds), Efficiency, Equity and Legitimacy: The Multilateral Trading System
at the Millennium (Brookings Institution Press, Washington, DC 2001); R Howse, From Politics to
Technocracyand Back Again: The Fate of the Multilateral Trading Regime (2002) 96 AJIL 94;
Finlayson and Zacher (1981) 600.
49
Curzon (1965) 53; Curzon and Curzon (1976) 175; Curzon and Curzon (1973) 317.
50
Curzon and Curzon (1976) 159, 175, 184; Preeg (1970) 78; Curzon (1965) 129.
51
Hudec (1970) 619ff.
52
ibid 619; Hudec (1990) 7, 22, 26, 289.
200
therefore sought, at least in some respects, to spell out legal obligations with some
precision in the Havana Charter, and in early drafts made these obligations enforceable through a formal complaints procedure, and justiciable in the International
Court of Justice.53 As Jackson has also noted, one of the oft-cited advantages of the
Charter (and subsequently the GATT), at least from this perspective, was that the
rights and obligations embodied in [the] agreement are contractual in nature and
legally binding.54 This view of law was similar to that which has recently been
resurrected in some recent accounts of the trade regime55a view in which law is
seen primarily as a way of producing predictability and certainty by ensuring that the
trade policy behaviour of a state conforms to the prior expectations of its trading
partners. There was even a proposal in 1947 to establish a procedure for direct citizen
complaint in cases of violation of international trade law, though it did not receive
much support.56 The United States was always the strongest champion of this view
of law, and, as we shall see in the next chapter, was in large part responsible for its rise
to dominance from the mid-1970s onwards.
On the other hand, there was an alternative view of the role of law, which placed
much less emphasis on formally binding obligations, and which saw legal ambiguity
as a potentially constructive force. This view was based in part on a particular
understanding of the kind of judgment which was required in the context of
economic disputes, reflected in this passage in a memorandum submitted by the
United Kingdom during Charter negotiations:
. . . there are numerous provisions in the Charter which require the discretion and economic
judgment rather than precise interpretation of the terms of the Charter . . . In almost every
conceivable case arising under the Charter the issues will of their nature involve the element of
economic appraisal and assessment and will not be purely legal in character, and it will be
impossible to say where economic judgment ends and legal judgment begins.57
As Hudec notes, the issue was not just to ensure that dispute settlement was guided
by economic (not just legal) forms of expertise, and controlled by pragmatic
economic experts whose judgment was not cluttered by legalistic ritual.58 The
point was also to ensure that dispute settlement was animated by a spirit of political
compromise and discretion, which a more formalized legal machinery might discourage.59 More generally, law was primarily seen in purposive terms as an instrument of the broader purposes underlying the regime as a whole, rather than just as a
53
N Weinrichter, Perspectives on the changing spirit of GATT (1999) 3 European Integration Online
Papers <http://eiop.or.at/eiop/pdf/1999-010.pdf> accessed 15 February 2011, rightly describes the ITO
as a highly legalized international organization; see also Hudec (1990) 26ff.
54
Jackson (1969) 755.
55
See below pp 241ff.
56
Jackson (1969) 187.
57
UN Economic and Social Council, Article 86: Summary of Points Made by Mr Shackle Regarding
Interpretation and Settlement of Disputes (14 February 1947) UN Doc E/PC/T/C.6/W.77, cited in
Hudec (1970) 6223.
58
Hudec (1970) 618; Hudec (1990) 25.
59
Hudec (1970) 619, 623.
201
mechanism for producing clear rules and therefore predictable behaviour.60 This
meant that the drafters had to ensure a certain degree of ambiguity and flexibility in
the texts, forms, and procedures of the law, to allow for the mobilization and
deployment of appropriately sensitive discretion and judgment in particular cases.
Thus, the softness of the GATTs ultimate dispute settlement procedures, and the
indeterminate language of much of its texts, was notor not onlya pragmatic
concession to the reality that governments would not accept more binding commitments. It also reflected a view that these legal forms would produce a more effective
and operational dispute settlement machinery than more formal alternatives.
Both of these views of the appropriate role of law in international trade politics
and variants of themhave been in play within the trade regime from its very
beginning. Accounts differ as to the precise ebbs and flows between them over the
course of its first few decades. Hudec describes the flexible legal style as more or less
dominant in the GATT from the very beginning, through to the Tokyo Round.61
While it is true that there was a distinct change in the operation of the dispute
settlement in the 1960s, as Contracting Parties came to see formal legal proceedings
as overly aggressive,62 in Hudecs account this did not represent so much a change in
ideas about the nature of laws role in international trade politics, as an erosion of a
belief in laws efficacy in general. In any case, whatever the precise turns and twists, to
most contemporary eyes it seems clear that the GATTs early approach to law leaned
much more towards the flexible and diplomatic than the formal and juridified.63
It is useful to remind ourselves that this tendency was not just a product of the
internal politics of the trade regime itself, but of broader political and legal transformations characteristic of the period running from roughly the 1930s to the 1970s.
The emergence of the welfare state during that period was associated with a broader
transformation in the characteristic form of law, from formal rational legal systems
towards purposive . . . [or] substantively rational law, law used as an instrument of
purposive, goal-oriented intervention, particularly in the context of the collective
regulation of economic activity.64 This transformation in the nature and function of law, from formal rational law towards purposive law, was also felt at the
60
An interesting illustration of this view is Jackson (1969) 5: law . . . should be subservient to goals that
essentially come from outside the limits of law itself. Cf this statement with the view of the nature and
function of law in JH Jackson, The Birth of the GATT-MTN System: A Constitutional Appraisal (1980)
12 Law and Policy in International Business 21.
61
See generally Hudec (1970); see also Viner (1947) 615.
62
Jackson (1969) 759.
63
For a number of accounts using this or a similar concept, see A Reich, From Diplomacy to Law: The
Juridicization of International Trade Relations (1997) 17 Northwestern J of Intl L and Business 775; D
Palmeter and PC Mavrodis, Dispute Settlement in the World Trade Organization: Practice and Procedure
(2nd edn, CUP, Cambridge 2004); J Pauwelyn, The Transformation of World Trade (2005) 104
Michigan Law Rev 1.
64
G Teubner, Substantive and Reflexive Elements in Modern Law (1983) 17 Law & Society Rev
239, 240, 253ff, 268; see also D Kennedy, The Rule of Law, Political Choices, and Development
Common Sense in DM Trubek and A Santos (eds), The New Law and Economic Development: A Critical
Appraisal (CUP, Cambridge 2006) 98110; D Kennedy, Three Globalizations of Law and Legal
Thought: 18502000 in Trubek and Santos (2006) 19.
202
international level, including in the GATT. Slaughter, following Mitrany and Maier,
has persuasively argued that the post-war international order, including the GATT,
represented in important respects the international projection of the organizing
principles of political power characteristic of the welfare state of the time. This,
in her account, did not just mean a projection of its substantive commitment to
government intervention to protect the welfare of citizens. It also meant the projection and proliferation of the particular institutional forms characteristic of the
welfare statenamely, specialized administrative agencies with functional mandates
exercising independent expertise over a specific field of social life.65 It also meant,
most relevantly, the emergence of purposive legal forms, in precisely the mode
described above. The point, then, is that the embedded liberalism of the post-war
period should be understood not just as a set of substantive political commitments to
interventionism, and a set of shared ideas about the regimes collective purpose, but
also a commitment to a particular style and mode of deploying law in international
governance.
These ideas about laws proper role and function produced a GATT legal system
within the trade regime with at least five noteworthy structural features.66 First,
it was structurally subordinated to (or embedded in) the field of trade diplomacy. As
is well known, the earliest disputes within the GATT system were resolved by trade
delegations themselves, in a plenary meeting of the Contracting Parties, or in
working parties staffed by governmental representatives. Members of dispute settlement panels and working parties were drawn almost exclusively from the community
of trade diplomats. These panel members, although ostensibly acting in an individual capacity, nevertheless [were] representing a country in the GATT.67 Virtually all
significant stages of the dispute settlement process required the consent of all
delegates of GATT Contracting Parties to proceed. The result was that dispute
settlement was an extension of diplomatic processes, and the internal logic and
dynamics of the law more or less followed those of trade diplomacy. In addition,
panels and working parties did not feel that they could substantively address questions brought before them on which there was strong and ongoing disagreement
among the Contracting Parties. There was a consistent practice of pushing such
disputes back to more regular processes of diplomacy and negotiation.68
Secondly, the text of the GATT itself represented only a very partial and even
misleading window onto the normative universe of the GATTs legal system during
this period. While this is no doubt true of all legal systems to some extent, it was
65
A-M Burley [Slaughter], Regulating the World: Multilateralism, International Law, and the
Projection of the New Deal Regulatory State in J Ruggie (ed), From Multilateralism Matters (Columbia
University Press, New York, NY 1993) 125; D Mitrany, A Working Peace System (Quadrangle Books,
Chicago, IL 1966).
66
See generally Hudec (1990) ch 17; see also T Broude, International Governance in the World Trade
Organization: Judicial Boundaries and Political Capitulation (Cameron and May, London 2004) for a
useful account.
67
Jackson (1980) 42; Reich (1997) 796.
68
See eg UruguayRecourse to Article XXIII, GATT Panel Report, L/1923, adopted 16 November
1962, BISD 11S/95.
203
particularly striking in the GATT context, and frequently commented upon by

writers of the period. There was a large body of informal norms which sprang up
in and around the formal texts, which exercised a powerful pull on Contracting
Parties, and which structured the interpretation and application of the law:
Slowly rules of the game have evolved which are as important as the Articles of Agreement.
While the latter may sometimes be ignored, the former seldom are. They have never been
codified but have come to be recognized by participating countries when statesmen speak of
the GATT philosophy.69
These rules of the game can be thought of as customary norms,70 which reflect the
GATT consensus existing at the time71or at least that which passed for consensus
given the dynamics of the trade diplomacy. Sometimes, these customary norms
included the tacit acceptance of the violation of legal norms.72 To a large extent,
the GATTs legal system relied on the existence of this underlying and evolving
substratum of informal norms, without which it simply could not function
effectively.73
Thirdly, GATT dispute settlement also functioned as a venue and mechanism for
the production and mobilization of community opinion in the world of trade
diplomacy. Dispute resolution involved, in part, an intricate process of producing
and reflecting the prevailing commonsense, and applying it to the dispute at issue.
Hudecs careful account, written in 1970, of the 1952 Belgian Family Allowances
Panel Report, remains the best and subtlest description of this process in action.74
I will say more about this decision below, but for now it is enough to note Hudecs
revealing general descriptions of the function of the Panel Report in that case:
The Panel could not simply state the law . . . directly. What it could do, however, was to
suggest . . . a conclusion in terms vague enough to avoid being pinned down and yet clear
enough in the context to add some authority to the idea. These suggestions . . . could invite a
community reaction somewhat stronger than the document itself . . . Rather than being a final
resolution of the legal issue, the Panel decision is simply the first stage in a larger decisionmaking process that involves the whole community. The decision seeks to energize and
influence community opinion, but it does so in a way which requires the community, as it
were, to vote on the answer proposed.75
One lesson from Hudecs analysis is that GATT law, and GATT dispute settlement,
was profoundly shaped by the community consensus underlying the texts, and that
this informal community consensus informed the texts interpretation and application in fundamental ways. Another lesson, however, is that legal processes in the
69
Curzon (1965) 53.

Jackson (1969) 757.
Hudec (1970) 665.
72
RE Hudec, Adjudication of International Trade Disputes (Trade Policy Research Centre, London
1978) ch 2; JH Jackson, The Crumbling Institutions of the Liberal Trade System (1978) 12 J of World
Trade L 93; Jackson (1969) 757; Long (1985) 19.
73
See generally Hudec (1978); Hudec (1990) 2023; Finlayson and Zacher (1981) 588.
74
Hudec (1970).
75
ibid 657.
70
71
204
GATT had a creative function. They were an important part of the processes by
which this community consensus developed, and helped to provide the formal basis
for absorbing such developments into a regular legal framework.76 This process
recalls Dams description of GATT law as a set of procedures and [a] process
[which] serves to identify the common interest in complex situations.77 Law, in
other words, was closely tied to the processes by which particular perceptions and
ways of interpreting the world were produced within the community of trade
diplomats, and commonsense interpretive frameworks were generated and mobilized
among them.
Crucially, a corollary of this is that the operation of GATT law during this time
its politics and normative orientation as expressed in actual legal practices, rather
than textswas determined in part by the specific political and social processes by
which consensus was formed within this diplomatic world. It was here, in these
background processes through which a particular commonsense was constructed,
that the structural biases and normative orientation of the legal system were contested and reproduced. The power to direct the operation of the law derived in part
from the power to legitimately define a particular version of commonsense for the
community.
Fourthly, the characteristic legal forms associated with this legal order were
ambiguous and flexible, not only in panel reports, but also in treaty texts. Hudec
describes the legal style of the period as displaying a sophisticated and often artistic
use of ambiguity, which often seemed to represent a deliberate policy to avoid
stating even the simplest of ideas clearly and without qualification.78 Ambiguity was
important as a way of ensuring sufficient flexibility in the law to permit it to express
whatever informal community consensus existed at any particular point in time,
modulating according to circumstances. Furthermore:
. . . the ability to express ideas in this vague and cloudy manner allows the community to bring
to bear whatever degree of consensus does exist. A suggestive statement, however ambiguous,
provides something formal and official on which the community can focus its attitudes. The
statement itself will point to the preferred result, and support expressed for that statement can
create some pressure of consensus toward that end. Part of that pressure lies in the possibility
that, faced with a specific case, the community may find itself willing to take a more positive
stand than it has taken before. The ambiguous gesture invites such a development, and
furnishes a vehicle through which it may come about.79
In many early GATT panel reports, then, one searches in vain for a clear statement of
the law, or indeed a clear statement of the outcome of the dispute.80 Precisely
because of its structural subjection to the logics and dynamics of trade diplomacy,
the GATT developed [legal] forms and techniques which work[ed] to suppress the
76
ibid 630.
KW Dam, The GATT: Law and International Economic Organization (University of Chicago Press,
Chicago, IL 1970) 4, 5.
78
Hudec (1970) 630.
79
ibid.
80
ibid 650.
77
205
laws natural instinct for final decisions.81 The intentional and functional nature of
this is often under-appreciated. What has come to appear to contemporary eyes as an
inadequate legal framework of imprecise and to some extent contradictory rules
without clear and predictable consequences, was in fact well suited to the legal
consciousness of the period. Relatively vague legal forms couched in the language
of general principles gave interpreters the needed flexibility to interpret and apply the
law in accordance with the spirit of the collective project which animated the regime.
Indeed, it is no exaggeration to say that the legal forms contained in the GATT can
only be made to make sense as a coherent whole, and can only be made operable and
workable, to the extent that they are imagined and interpreted by reference to a clear
understanding of the nature and purpose of the GATT, such as that provided by the
ideology of embedded liberalism in the immediate post-war decades.
Finally, the overall purpose of dispute settlement in the GATT during this period
was to contain trade disputesto rebalance expectations, and to avoid escalation.
This is a consistent theme of all observers of the period. Long, for example, suggests
that the development of an understanding between the partiesor a mutually
acceptable solutionis the main objective of the dispute settlement procedure.82
The Curzons saw the GATT as founded upon an unwritten credo of maintaining a
balance across imports and exports between trading partners, and dispute settlement as all about the rapid restoration of that balance when it was upset.83 This
way of imagining the purpose of dispute settlement was again clearly the product of a
characteristically embedded liberal legal sensibility. For one thing, an instrumental
or purposive view of the law meant that legal processes of dispute settlement were
understood as serving the broader aims of trade diplomacy, and therefore were in
large part governed by the same powerful norm of reciprocity as was the diplomatic
process of the time. For another thing, if the substantive purpose of the trade regime
at the time was the maintenance of international stability, and the avoidance of
another protectionist trade war, then it is only natural that dispute settlement should
be understood as a pragmatic process of de-escalation and mutual resolution, rather
than the rigid application of juridical rules.
II. Approaches to domestic regulation in the GATTs early decades

We have seen, then, that when the post-war GATT regime was born, there existed a
small GATT community whose key players shared a set of ideas about the kinds of
trade policies which were and were not considered broadly acceptable, about the role
that the GATT could and should play in the maintenance of the international
trading system, and about the role of GATT law in the resolution of trade disputes.
These shared ideas were reflected in a set of characteristic institutional forms, social
81
82
83
ibid 665.
Long (1985) 76.
Curzon and Curzon (1973) 321; see also Weinrichter (1999); Weiler (2001).
206
practices, and legal structures which more or less endured over the first two decades
of the GATTs existence. This section examines the GATTs approach during this
period to the broad issue of non-tariff barriers, and demonstrates that this approach
was informed and structured by the shared ideas and institutional forms just
described.
For the architects of the post-war international trading order, the term non-tariff
barrier connoted a much narrower range of governmental measures than it does
today. Their immediate priority was to remove whatever remained of the more
restrictive, exotic commercial policy instruments which had been put in place during
the interwar period, including quotas, foreign exchange restrictions, licensing requirements, and so on.84 As a result, Article XI of the GATT outlawed, in broadly
worded terms, all prohibitions or restrictions other than duties, taxes or other
charges, whether made effective through quotas, import or export licences or other
measures. While this provision was subject to some important exceptionsprimarily those required for the purpose of maintaining domestic economic stability, as
described aboveit covered the bulk of these very distorting instruments, most of
which were eliminated quite quickly.85 Another set of non-tariff barriers which were
of concern to post-war negotiators related to the administrative treatment of foreign
products at the border, for example in respect of customs formalities and customs
valuation, as well as the imposition of foreign marks of origin. These were addressed
in Articles VII, VIII, and IX of the GATT.
While the focus was therefore squarely on border barriers, it was not possible
entirely to ignore other forms of government action, which might be characterized as
domestic, internal, or behind the border in nature. The reason for this is familiar
to trade lawyers, and simple enough to state: the benefits of any reduction in border
barriers to trade might be entirely nullified were a country to impose internal
measures which disadvantaged the ability of foreign products to compete fairly on
domestic markets once they had entered the country. One category of internal
measures which was of particular concern for some countries during the negotiations
was internal taxes and charges.86 As a result, Article III:2 of the GATT contained an
obligation not to use internal taxes and charges to discriminate against imported
products:
The products of the territory of any contracting party imported into the territory of any
other contracting party shall not be subject, directly or indirectly, to internal taxes or other
internal charges of any kind in excess of those applied, directly or indirectly, to like domestic
products. Moreover, no contracting party shall otherwise apply internal taxes or other internal
charges to imported or domestic products in a manner contrary to the principles set forth
in paragraph 1.87
84
See above p193 and surrounding; also Preeg (1970) 24.

Currency exchange restrictions were addressed in the Articles of Agreement of the IMF rather than
in the GATT.
86
Hudec (1990) 126.
87
For a drafting history of the text, see DA Irwin, PC Mavrodis and AO Sykes, The Genesis of the
GATT (CUP, Cambridge 2008) 138ff.
85
207
This was supported by Article I, which extended the MFN non-discrimination norm
to measures covered by Article III:2. A second specific category of internal measures
of interest to some negotiators was local mixing and local content requirements.
Although a prohibition of these measures was strongly resisted by developing
countries in negotiations over the Havana Charter,88 a provision to that effect was
included in both the Charter, and in Article III:5 of the GATT.89 Negotiations over
two other kinds of internal measures did not result in clear disciplines, even though
both could affect trade very significantly: anti-dumping duties and domestic subsidies. As Jackson has noted, disciplines on subsidies posed specific problems, in part
because there was no clear consensus among the Contracting Parties on where the
line should be drawn between justifiable government policies on the one hand, and
policies that constitute a dangerous and improper attempt to export ones own
problems on the other, and in part because it was recognized that subsidies were a
preferable policy as compared to tariffs and quotas.90 In the end, the 1947 text placed
very few, if any, substantive restrictions on the ability of states to put in place
domestic subsidies.91
But the negotiators also knew that these specific categories of internal measures
did not exhaust the range of possible measures that governments could put in place
to undermine their obligations under the GATT. Indeed, as noted in Chapter 6,
since virtually all actions of a government in and around its domestic market have
some direct or indirect impact on trade, this range of possible measures is limitless,
and it is impossible to specify in advance what they might be.92 Drawing on similar
efforts during the interwar years,93 the drafters of the GATT therefore developed a
three-pronged strategy to address trade-restrictive internal measures as they arose.
The first prong was to apply the two non-discrimination norms to all internal law
and regulations which had an impact on trade. Thus, the national treatment norm
was given general application to internal measures in Article III:4:
The products of the territory of any contracting party imported into the territory of any other
contracting party shall be accorded treatment no less favourable than that accorded to like
products of national origin in respect of all laws, regulations and requirements affecting their
internal sale, offering for sale, purchase, transportation, distribution or use.
The MFN norm was extended to the same category of measures by Article I:1.
Secondly, these norms were then subject to the provisions of Articles III:8(a), XX,
and XXI, which provided for exceptions for governmental procurement, as well as
88
Brown (1950) 1089.

Art III:5 states: No contracting party shall establish or maintain any internal quantitative regulation
relating to the mixture, processing or use of products in specified amounts or proportions which requires,
directly or indirectly, that any specified amount or proportion of any product which is the subject of the
regulation must be supplied from domestic sources. Moreover, no contracting party shall otherwise apply
internal quantitative regulations in a manner contrary to the principles set forth in paragraph 1.
90
See generally Jackson (1969) 367.
91
GATT, Arts XVI and III:8(b); Curzon and Curzon (1976) 195.
92
Hudec (1990) 24.
93
ibid; Hudec (1970) 617.
89
208
for governmental measures for a range of legitimate purposes, including the protection of national security, the protection of animal and human health, the conservation of exhaustible natural resources, and so on. Then, thirdly, in apparent
recognition of the fact that even this flexible legal framework might not have been
adaptable enough to respond adequately to all future challenges, a mechanism was
created through which any Contracting Party could complain if it should consider
that any benefit accruing to it directly or indirectly under [the GATT] is being
nullified or impaired as the result of the application by another contracting party of
any measure, whether or not it conflicts with the provisions of this Agreement.94 This
clause had come to be included as a standard part of trade agreements during the
years leading up to the Second World War,95 and it was included in the GATT on
the basis that it would permit the Contracting Parties to respond flexibly to any new
situation which might upset the careful balance set out in the agreement.96
It takes only a moments reflection to see how broad and intrusive these disciplines
had the potential to be. Putting to one side for a moment the limited carve-outs and
exceptions described above, there is little in these provisions which a priori excludes
any kind of domestic governmental actions from GATT oversight, even those which
apparently have nothing directly to do with trade policy.97 The non-discrimination
norms apply to all laws, regulations or requirements which affect the internal sale,
offering for sale, purchase, transportation, distribution or use of foreign goods.
There is nothing in the text which limits this to direct or significant effects, and
certainly nothing which limits it to measures explicitly targeted at imports.98 There
are few aspects of fiscal and monetary policy, employment policy, industrial policy,
controls on capital flow, even labour regulation and immigration controls which
could not be brought within this definition with a minimum of creative thinking.
Moreover, as we shall see in the next chapter, these non-discrimination norms can in
principle have significant substantive bite, particularly where they are interpreted as
prohibiting de facto, indirect or even merely accidental and fortuitous discrimination
between products in the marketplace. Furthermore, for its part, the nullification and
impairment clause had perhaps an even greater scope of potential application. As the
South African delegate noted in the negotiations over the equivalent provision in the
Havana Charter, it had the effect, at least in practice, of giving an international
institution the power to declare new obligations where none existed before:99
It appears to me that what it says is this: In this wide world of sin there are certain sins which
we have not yet discovered and which after long examination we cannot define; but there being
such sins, we will provide some sort of punishment for them if we find out what they are and
94
Art XXIII:1(b) (emphasis added).

Hudec (1990) 24; Hudec (1970) 617 and surrounding.
Hudec (1990) 40.
97
For another view of the tremendous formal breadth of these provisions see JW Evans, The Kennedy
Round in American trade policy: the twilight of the GATT? (Harvard University Press, Cambridge, MA
1971) 87; see also Hudec (1990) 121 (who describes them as virtually unlimited in scope).
98
See Panel Report, Italian Discrimination Against Imported Agricultural Machinery (adopted
23 October 1958) BISD 7S/60 (L/833) paras 1215.
99
See Hudec (1990) 39.
95
96
209
if we find anybody committing them . . . seeing that there are such sins, and in spite of the fact
that we do not know what they are, and in spite of the fact that we do not know under what
circumstances we are going to apply any punishment to them, we shall still provide a sort of
vague and general sword of Damocles, if such a thing is possible, to hang over the heads of all
the people who may possibly commit this sin.100
As a matter of purely textual interpretation, it is hard to argue with this

characterization.
Why, then, did the Contracting Parties find themselves able to agree to these
provisions, despite their potentially serious limitations on sovereignty? The answer,
in short, was that the Contracting Parties themselves could be trusted not to abuse
this powerit would be a very great pity if we could not trust an international
organization formed out of our own membership to interpret this clause intelligently
and with sufficient discretion101and that in any case the dispute settlement
mechanism ultimately rested on soft enforcement mechanisms. The workability
and acceptability of these potentially intrusive mechanisms, in other words, was
founded on the existence of a broadly shared informal consensus as to the meaning
and purpose of the GATT,102 as well as a set of institutions which ensured that this
consensus could practically shape the interpretation and application of GATT law.
This gave negotiators comfort that the virtually limitless formal application of
GATT disciplines to domestic regulation would in practice be kept within strictly
limited and reasonable boundaries.103
This expectation proved, more or less, to be well founded. This shared consensus
as to the meaning and purpose of the GATTitself deeply informed by the
normative framework of embedded liberalism, as described abovehelped to ensure
that, of all the different internal governmental actions which influenced trade flows
during the first two decades of the GATTs existence, only a few very specific
categories were made visible as barriers to trade, and brought before dispute settlement. The remainder were implicitly understood to be normal or background
regulation, never intended to be subject to GATT oversight. Thus, of the sixtytwo formal GATT disputes which Hudec lists as occurring during the first two
decades of the GATTs existence,104 twenty-four involved traditional border barriers
(quantitative restrictions and tariff charges and one case involving export fee). While
internal measures therefore constituted over half of the disputes, virtually all of
these cases were brought in relation to internal measures of the kind which were
specifically raised in the context of negotiations over the GATT 1947, and which
were therefore already clearly acknowledged by these negotiators as trade barriers,
100
UN Economic and Social Council, Verbatim Report of the Preparatory Committee of the UN
Conference on Trade and Employment (26 June 1947) UN Doc E/PC/T/B/PV/33, cited in Hudec
(1990) 42.
101
Cited in Hudec (1990) 41. The reference to an international organization is explained by the fact
that this statement was made in the context of Havana Charter negotiations.
102
ibid 297; Howse (2002) 97.
103
See Hudec (1990) 41ff.
104
ibid 302ff; see also RE Hudec, Enforcing International Trade Law: The Evolution of the Modern
GATT Legal System (Butterworth Legal Publishers, Salem, NH 1993) 375ff.
210
and made visible as such. Thus, of these thirty-eight disputes about internal measures, fifteen were brought in respect of internal tax measures, eleven involved export
subsidies, three related to trade remedies actions, and five to domestic subsidies.
Only three related to domestic measures of a miscellaneous kind, none of which
produced a panel report: the first being a dispute about a Hawaiian law requiring all
vendors of foreign eggs to display a sign to that effect; the second involving Greek
credit facilities; and the third relating to pricing practices of a British governmentowned steel industry.105 Although it is true that some of the disputes about internal
taxes and domestic subsidies raised real questions about whether the specific measure
in dispute was of the kind which was intended to be covered by the GATT, it is
nevertheless significant that the vast bulk of all disputes were brought in respect of
categories of measures which were already understood as serious trade barriers at the
time that the GATT was negotiated, and legitimately within its purview. In this
respect, Hudec also finds it noteworthy that the application of GATT disciplines in
the regimes early years focused on practices which were major target[s] of postwar
commercial policy reforms, so that transgressions in . . . these areas might have
appeared to be of greater than normal significance.106
There are, to be sure, other reasons why the practical scope of application of
GATT disciplines was kept in relatively limited bounds during these years, and was
not extended to the kinds of measures which are now routinely brought before WTO
dispute settlement. It has been argued, for example, that other kinds of behind the
border barriers were not in existence during that time, or at least were commercially
insignificant, so there was simply no pressing need to bring such cases. It has also
been suggested that claims were not brought in respect of certain measures because
they were not subject to effective GATT discipline, so that a claim against them was
less likely to succeed and therefore to be brought. Nevertheless, while these reasons
are part of the explanation, they do not provide an adequate account on their own. It
is clear that many behind the border barriers which have only recently become
subject to more frequent dispute resolutionsuch as labelling requirements, or food
safety restrictionsalready existed at this time in many countries.107 On the question of commercial significance, Hudec notes that a significant proportion of GATT
disputes were brought in respect of measures which had only minor commercial
significance.108 Moreover, it is not clear on the evidence that non-tariff barriers
actually were commercially insignificant during this period, even as compared to two
decades later.109 As to their amenability to effective GATT discipline, this explanation must deal with the problem of the broad formal scope of GATT disciplines
105
See Hudec (1990) 312 (No 37), 314 (No 48), 318 (No 62). The one dispute not counted in this
breakdown (of the 38) is Panel Report, UruguayRecourse to Article XXIII (adopted 16 November 1962)
BISD 11S/95 (L/1923).
106
Hudec (1990) 106.
107
See Brown (1950) 35; Curtis and Vastine (1971) 4, 202; Evans (1971) 87ff.
108
Hudec (1990) 1035.
109
Evans (1971) 3079.
211
described above, as well as the fact that eleven disputes were brought in respect of
export subsidies, notwithstanding the lack of effective discipline on such measures.
One of my claims, then, is that the limited practical scope of application of the
GATT during these years was in significant part the result of informal understandings among those in the GATT community about its limited legitimate scope of
scrutiny.110 These understandings were in turn the product of a number of features
of the field described above: generally shared views that many different kinds of
highly interventionist domestic policies were normal; an orientation towards stability and the containment of disputes which left GATT diplomats disinclined to test
the boundaries of the non-discrimination provisions if to do so risked destabilizing
trade relations and trespassing on jealously guarded domestic policy priorities; the
structural embedding of the GATTs legal system within the world of trade diplomacy and the separation of the world of trade diplomacy from that of high politics;
and a prevailing view that the law ought properly to be interpreted and applied in
accordance with the norms and expectations of that world. These orientations
produced a field of trade law and policy which was narrowly defined (relative to
the present day), focused on tariffs and other kinds of trade measures understood at
the time as quantitative restrictions, and unwilling to scrutinize internal measures
except in the clearest circumstances of circumvention of liberalization commitments.
Even if it was recognized that internal regulation could pose problems for traders,
that fact alone did not make all such regulation a proper subject of strict oversight
within the international trade regime. I shall return to this claim when I address the
unsuccessful turn to internal non-tariff barriers in the Kennedy Round, below.
My second claim is that these informal norms and expectations circulating among
trade diplomats and others in the trade regime also deeply shaped the interpretation
of GATT disciplines in those disputes involving internal regulations.111 When
GATT panels were asked to judge and declare the permissibility of domestic
regulations which were the subject of dispute, they did so by reference to a variety
of overlapping standards, including backgrounds norms concerning reasonable or
normal domestic regulation, particularly among its most powerful actors,112 as well
as informal understandings of what the drafters meant to say.113 The interpretation
of these norms therefore came to reflect a broader inclination of the Contracting
Parties to limit the potentially constraining impact of GATT disciplines on their
domestic regulation, especially where such constraint would work against their
shared substantive political commitment to domestic interventionism to safeguard
to economic stability and the welfare of their populations.114 Thus, they developed
an approach to the non-discrimination norm which focused primarily on de jure
110
For similar arguments, see Howse (2002); R Wai, Countering, Branding, Dealing: Using Economic and Social Rights in and around the International Trade Regime (2003) 14 EJIL 35, 46.
111
With this claim, I am following Hudec (1990); Hudec (1970); Howse (2002).
112
Hudec (1990) 1312.
113
See also Curzon (1965) 322: the philosophy of an agreement provides a measure by which an
individual countrys action is judged to be good or bad.
114
JL Dunoff, Trade and: Recent Developments in Trade Policy and ScholarshipAnd Their
Surprising Political Implications (1997) 17 Northwestern J of Intl L and Business 759; JL Dunoff,
212
discrimination and discriminatory intent as the primary criterion of illegality. And

they developed an approach to nullification and impairment which focused on the
reasonable expectations of the parties to negotiations.
A number of the early cases on internal tax and subsidy measures can serve as
illustrations of these basic dynamics.115 In August 1948, during the meeting of the
Second Session of the GATT Contracting Parties, the Chairman was asked by
Pakistans delegate to make a ruling concerning the application of Article I to an
Indian measure, according to which an excise tax on domestic products was rebated
to the producer if the goods were exported to any country other than Pakistan. The
Chairman was quick to rule that Article I was indeed applicable to the measure, and
had a strong textual basis for doing so. Article I by its terms applied to all measures
covered by paragraphs 2 and 4 of Article III, and it was certainly arguable that this
was an internal taxation measure covered by Article III:2, even though it related to
exports. To the extent that some textual ambiguity existed, this was easily disregarded. In Hudecs words:
The Chairman was, after all, one of the authors of the text in question. And so, for that matter,
were most of the Secretariat officials and the other delegates in the room. No matter how hard
they might try to look at the bare text alone in this situation, it was inevitable that everyones
interpretation would be dominated by their own highly authoritative recollection of what they
had meant to say.116
This is hardly surprising, nor perhaps is it particularly noteworthy that an international legal text is being interpreted in the light of the clear intentions of its
authors. What is more interesting, though, is that subsequent discussion of this
matter, including in the context of the drafting of a clarifying amendment to Article
I, proceeded on the basis that while paragraph 2 language applied to exports as well as
imports, paragraph 4 language did not.117 This is a perfectly understandable premise, as to have decided otherwise would have been to impose (or recognize) an
obligation of enormous reach, with little practical utility.118 But it had no textual
basis. In the end, as Hudec has argued, what was guiding the Contracting Parties in
this context was not so much their recollection of what they meant to sayit is
unlikely that they had ever really considered the question of exports at all119but
rather a fairly clear, coherent and authoritative consensus about the policy and
purposes of the General Agreement.120 Here, that consensus proved strong enough
to override the clear terms of the text, with the effect of limiting its constraining
effect.
Rethinking International Trade (1998) 19 University of Pennsylvania J of Intl Economic L 347; JL

Dunoff, The Death of the Trade Regime (1999) 10 EJIL 733.
115
116
117
118
119
120
The following pages draw heavily on Hudec (1990) and Hudec (1970).
Hudec (1990) 116.
Hudec (1990) 120.
ibid 121.
ibid.
ibid.
213
The following year, the French delegate complained about a Brazilian internal
taxation measure which imposed taxes on certain imported luxury products at a rate
double that imposed on like domestic products. Since this measure was already in
place when the GATT was signed, it was permitted. But was it permitted for Brazil to
raise the underlying tax rate, such that the absolute tax differential between imported
and domestic products was massively increased, even though the former were still
taxed at double the rate of the latter? The text was not clear. A Working Party was
formed, which included Brazil, France, the United Kingdom, and the United States
among its seven members. The latter three countriesat that stage clearly the first
among equals within the GATT regimewere evidently of the view that Brazils
actions were prohibited, but Brazil would not concede the legal point. Although this
stalemate was not formally resolved by the report of the Working Party, the reality
was clear:
Brazils legal position was unlikely to be accepted by the Contracting Parties, and Brazil must
have known this. The line-up on the other side was simply too overpowering. The Big Three
were quite sure of their position, and, having virtually written the General Agreement and the
Protocol themselves, their views would carry the greatest weight. The Secretariat was equally
convinced of the same view.121
When the matter was raised again the following year, the Working Party report was
treated as if it had definitively found that Brazils measure violated the GATTit is
not impossible that the delegates . . . simply confused their own well-settled views on
the legal issues with a not-too-well-remembered document in which those views had
been expressed122and Brazil did not object. No reference was made in subsequent
documentation to Brazils legal arguments. The inner-group consensus seemed to be
taken for granted as the authoritative word.123
The Belgian Family Allowances case, which began with a complaint by Norway and
Denmark against Belgium in 1951, concerned a tax imposed by Belgium on
imported goods which were purchased by local government bodies. The purpose
of the tax was to help to pay for Belgiums generous system of family allowances, by
supplementing the revenue raised by a domestic payroll tax. It was levied only on
those goods which came from countries which did not have a programme of family
allowances equivalent to that of Belgium itselfin other words, products would be
exempt from the tax if they came from a country which did have a roughly equivalent
programme. The tax was therefore arguably designed (in principle at least) to help to
equalize the competitive disadvantage that certain producers faced as a result of the
creation of this social welfare programme. Norway and Denmark brought the
complaint when, in September 1950, products from Sweden were granted an
exemption while their products were not. What is interesting was that Norway and
Denmark did not argue that the entire tax was in violation of Article I of the
GATTthough that argument was certainly textually open, as the tax clearly treated
identical imports from different countries differently. Instead, they accepted the
121
ibid 126.
122
ibid.
123
ibid 131.
214
basic tax regime, but argued that their family allowance programmes were at least as
generous as Swedens, and that it was therefore discriminatory to grant an exemption
for Swedish products, and not theirs. According to Hudec, both the complainants
and the members of the Panel which was subsequently convened shared . . . a hunch
that the real reason for the Swedish exemption had been that Sweden paid for the
exemption as part of the trade agreement concessions124 negotiated in September
1950. When, later, the Belgian government granted a further exemption, this time to
Switzerland, the suspicion that exemptions were being sold for commercial advantages was, rightly or wrongly, confirmed within the GATT.125
It is not clear why Denmark and Norway argued their case on the narrower
ground, rather than simply arguing that the measure as a whole was discriminatory.126 It may have been that they simply wished to have access to the benefit of the
exemption, as a commercial matter. It may have been that the justice of the situation
was clearer on the narrower ground. Outlawing a tax genuinely designed to financially and politically support an important domestic programme of social welfare
may just have been a step too far, particularly if other countries might have wished to
keep open that policy themselves. Where a discriminatory intention existed, however,
a finding of a violation of Article I was clear. In the end, the Panel produced a report
which could be read as endorsing either the broader or the narrower proposition.127
But what is clear is that the broad perception that a discriminatory intent was
behind the Belgian measure was crucial to mobilizing community opinion within the
GATT against the Belgian measure. In this respect, the case is reflective of a broader
tendency in the GATTs early years to treat intention as, in practice, the critical
element of a claim of discrimination.128 The few disputes arising under Articles I and
III at this time overwhelmingly concerned direct and explicit discrimination. In
practice, then, and despite its somewhat ambiguous and indeterminate formal face,
the non-discrimination norm seems to have been understood to be there to prohibit
directly discriminatory regulationrather than other forms of regulation which
incidentally or indirectly undermined the commercial opportunities available to
foreign products, which as we shall see became much more commonly the subject
of dispute settlement later on.
Finally, it is worth referring briefly to a brace of decisions which illustrate the
importance which panels tended to attach to norms and expectations internal to the
community of GATT diplomats in reaching their conclusions, and the easy
124
ibid 141.
Hudec (1970) 661.
126
This may also have been because the Belgian law was from 1939, and mandatory pre-GATT
legislation which would otherwise be inconsistent with the GATT was grandfathered by the Protocol of
Provisional Application. For a careful and considered review of Hudecs analysis of this case, see S
Charnovitz, Belgian Family Allowances and the challenge of origin-based discrimination (2005) 4
World Trade Rev 7.
127
Hudec (1970) 650ff.
128
Hudec (1978) 61, n 113 notes this tendency in respect of subsidy claims. It is worth noting that no
examples exist of a finding of de facto discrimination in early GATT cases, in the absence of a fairly clear
indication of discriminatory intent.
125
215
familiarity with which they evoked such norms. AustraliaAmmonium Sulfate

concerned an Australian decision to remove a subsidy on particular kinds of fertilizer,
which were predominantly imported, while maintaining it for others, which were
predominantly locally produced. In discussing whether the two products were like
one another for the purposes of Article I of the GATT, the Working Party noted that
the two products were given different tariff classifications not only by Australia but
also by many other countries around the world, and on that basis considered the
products unlike one another.129 Tariff classifications here represented the best
evidence available to the Working Party of the beliefs and expectations of those
within the field of trade policy and diplomacy about whether these two products
could legitimately be treated differently.130 The same case also involved a nullification and impairment claim, and the Working Party took the opportunity to limit
such claims to measures which could not reasonably have been anticipated by the
relevant parties at the time of the negotiations. The fact that the Working Party felt
confident in evoking the notion of reasonable anticipation is just one more indication of the closely knit consensus on which the early GATT system rested.131 Then,
in GermanySardines, the Panel again looked to expectations within the diplomatic
community for guidance on the question of the likeness of different species of
sardine, noting that:
. . . it would be sufficient to consider whether in the conduct of the negotiations at Torquay
the two parties agreed expressly or tacitly to treat these preparations as if they were like
products.132
Although this tendency to look for guidance to norms circulating within the field of
trade policy and diplomacy became less overt over time, it was a pervasive and to
some extent enduring characteristic of GATT legal interpretation on the issue of
likeness during this period.
Overall, then, what we see during the first decades of the GATTs existence is a
modest and limited approach to the disciplining of internal barriers to trade, or
what we now call domestic regulation. The kinds of measures which were the
subject of complaint were by and large limited to those which had a direct and
immediate impact on trade. They were also confined primarily to those kinds of
measures which the negotiators of the GATT had clearly and explicitly understood
to be potential targets of discipline. Moreover, the non-discrimination norm tended
to be applied in practice only in cases of de jure discrimination against imports, or in
those cases where a clear protectionist or discriminatory intent was either manifest or
129
Working Party Report, The Australian Subsidy on Ammonium Sulphate (3 April 1950) BISD II/188
(GATT/CP.4/39) para 8.
130
For a similar interpretation of the reason for drawing on tariff classifications in the context of Art I,
see generally RE Hudec, Like Products: The Differences in Meaning in GATT Articles I and III in T
Cottier and P Mavrodis (eds), Regulatory Barriers and the Principle of Non-Discrimination in World Trade
Law (University of Michigan Press, Ann Arbor, MI 2000) 101.
131
Hudec (1990) 167.
132
Panel Report, Treatment by Germany of Imports of Sardines (31 October 1952) BISD IS/53 (G/26)
para 12.
216
strongly suspected. Although this modest approach excluded from oversight a great
deal of government intervention and policy which affected trade flows in very
significant (if indirect) ways, it was very much in keeping with the spirit of
embedded liberalism described in the first part of this chapter. Recall that the
purpose of the trade regime at that time was commonly understood in relatively
limited terms: to avoid a repeat of the trade wars of the early 1930s and to reverse the
proliferation of the kinds of restrictive non-tariff measures which were instituted
during that period, and more generally to provide stability in international commercial relations, as a precondition for the achievement of domestic economic stability
and full employment. The general disciplines which the GATT imposed on internal
measures were understood to play a supportive role only. They were there to ensure
that the careful bargain which was struck in relation to traditional trade restrictions
could be resurrected and restabilized if it was undermined by an internal measure
which the GATT Contracting Parties saw as illegitimate. To use the non-discrimination norms to reshape domestic statemarket relations more actively, by challenging
all forms of internal governmental intervention which indirectly affected conditions
of competition on the domestic market, went well beyond what was understood to
be the purpose of the GATT. More than that, too intrusive an application of GATT
disciplines on internal regulation would undermine that purpose, as it would run the
risk of itself upsetting the delicate balance of concessions embodied in the original
agreement, and undermining support for the trading system as a whole. Moreover,
extending the GATTs scope of application to matters of domestic political sensitivity would risk undermining the clear mutual separation which the trade community
was so careful to maintain during this period, between technical issues of trade
policy on the one hand, and issues relating to high politics on the other. As noted
above, the institutional and social insulation of the trade field was itself a functional
expression of the priority given to international stability in the context of embedded
liberalism.
Thus, while the substantive spirit of embedded liberalism produced a limited
understanding of the meaning and purpose of the GATTs disciplines on internal
measures, the flexible institutional and legal forms associated with embedded liberalism ensured that this limited vision governed the operation of those disciplines in
practice. The flexibility and ambiguity of the text, the structural embeddedness of the
dispute settlement system in the world of trade diplomacy, and the emergence of an
interpretive style which placed greatest weight on the shared norms and expectations
of the diplomatic community, all ensured that the limiting informal norms of the
trade community itself counted more than the formal breadth of the text. This was
tolerable to the GATTs commercial constituenciesprimarily domestic exporting
firms seeking greater access to foreign marketsbecause the work that the GATT
was doing in reducing tariffs and other border barriers was still commercially highly
significant for these firms, even if it was less so than some might have liked. As trade
and economic activity rapidly expanded in the post-war decades, the shared prosperity it produced temporarily muted any concerns that some of these actors might have
had about the GATTs self-consciously limited vision.
217
Already by the middle of the 1960s, however, the material and political basis for
this shared vision was beginning to erode.133 The formation of the European
Economic Community in the late 1950s provided a substantial counterweight to
the political and economic dominance of the United States, and represented a new
and more powerful force in GATT affairs, with goals and objectives that had not
been anticipated in the original GATT blueprint.134 During the 1960s, intense
import competition from low-cost products, especially from an emerging Japan
which was gaining dominance in sectors such as textiles and clothing, placed pressure
on the ability of industrialized states to maintain their commitment to the substantive bargain embodied in the GATT. Recessions in the United States and elsewhere
at the beginning and end of the 1960s, intensified this pressure to some degree. The
emergence of more and more developing countries as members of the GATT also
put significant pressure on the regime, as described in Chapter 2. At the same time,
the immediate fear of a repeat of the trade wars of the 1930s was waning.135
As a result of all these developments, the substantive policy consensus underlying
the GATT 1947 itself began to erode. More and more legal obligations contained in
the GATT appeared unsuitable for contemporary economic problems and priorities,
and increasingly countries turned to measures which violated at least the letter of the
original agreement. The emergence of the EEC and other regional trade agreements,
for example, seriously undermined Article XXIV as a meaningful source of constraint
and oversight of preferential arrangements. Developing countries as a whole were
exempted from many of the most significant GATT disciplines, and both agricultural trade and trade in textiles and clothing were essentially removed from normal
GATT discipline.
Ruggie has rightly argued that many of these measures, and those which followed
during the economic turmoil of the 1970s, represented attempts to resurrect the
spirit of embedded liberalismthat is, to find new tools and techniques to maintain
the same basic compromise between domestic and international stability in transformed economic conditions.136 But at the same time, it is also true that this period
saw the beginning of the erosion of embedded liberalism, and the start of a shift in
the relative dominance of the different philosophies which had always been present
and in tension within the GATT system. Certainly, by the 1970s, it was possible to
talk of a schism in [the] sense of common purpose underlying the regime.137
It will come as no surprise that, just as the shared political commitment to
embedded liberalism began to erode, there were significant efforts to shift the
focus of the GATTs attention to domestic regulation, and to describe more areas
of governmental activity as potential obstacles to international trade. Non-tariff
133
D Harvey, A Brief History of Neoliberalism (OUP, Oxford 2007) 12ff.

Hudec (1990) 211.
135
ibid.
136
Ruggie (1982); JG Ruggie, Embedded Liberalism Revisited: Institutions and Progress in International Economic Relations in E Adler and B Crawford (eds), Progress in Postwar International Relations
(Columbia University Press, New York, NY 1991) 202.
137
Hudec (1990) 297.
134
218
barriers had become increasingly visible to exporters as barriers to access to foreign

markets, and some began to agitate for their inclusion on the GATT agenda. These
efforts were most powerful within the United States. As that country found its
economic dominance increasingly challenged, and as its trade balance deteriorated
towards the second half of the 1960s, the claim became increasingly common that
US exports were unfairly and disproportionately hindered by foreign non-tariff
barriers that had not been subject to removal through the GATT.138 As a result,
numerous publications began to emerge which identified and categorized a range of
non-tariff barriers which had either never been recognized before, or had received
very little attention in practice within post-war trade policy circles.139 These
included: labelling requirements; safety standards for farm machinery; inspection
requirements for imported seeds and livestock; sanitation and health standards;
testing requirements for drugs; consumer and environmental protection standards;
government subsidy of research and development; border tax adjustments; government procurement practices; differential application of domestic regulations; and
even the mere fact of disparate standards in different foreign markets.140 By the late
1960s, these efforts had gained some traction within the United States, where the
above measures were all regularly included in standard lists of non-tariff barriers
produced at the timea term which had by now taken on an entirely different
meaning from that understood just two decades earlier.141 In addition, it had
become received wisdom, especially in Washington political circles and more
broadly within the US trade policy community, that these non-tariff barriers had
become the most commercially significant barriers to US exports, and that sooner or
later it was imperative that the GATT should address them if it was to continue to
have American support.142 The proliferation of such barriers, it was said, had upset
the balance of concessions which formed the heart of the post-war agreement.
But incorporating internal non-tariff barriers of this sort on the GATT agenda
proved to be difficult. In significant part as a result of US action on the issue, the
GATT Ministerial Declarations in 1963the year before the commencement of the
Kennedy Roundasked the GATTs Trade Negotiations Committee to set rules to
govern and methods to be employed in the treatment of non-tariff barriers.143
However, the concept of non-tariff barrier was not defined, and in the end the
138
See Evans (1971) 304ff.

Administrative and Technical Regulations: The Lesser Known Obstacles to Trade, The OECD
Observer, No 4, June 1963; MS Massel, Non-tariff barriers as an obstacle to world trade (Brookings
Institution, Washington, DC 1965); WB Kelly, Jr, Non-Tariff Barriers in BA Balassa and ME Kreinin
(eds), Studies in International Trade Liberalization: Problems and Prospects for the Industrial Countries
(Johns Hopkins Press, Baltimore, MA 1967) 265314; R Baldwin, Non-Tariff Distortions of International
Trade (George Allen & Unwin, Sydney 1970); Curtis and Vastine (1971) 202ff; Evans (1971) 86ff. Much
significant work was done in the OECD, see Curzon and Curzon (1975) 312.
140
These examples are taken from the material cited in n 139 above.
141
See below pp 223ff.
142
Though it is clear that their commercial significance was increasing, Evans (1971) 3079 disputes
the factual basis for this strong claim.
143
Cited in Curtis and Vastine (1971) 204. There had been an earlier attempt to include non-tariff
barriers in the Dillon Round which had been resoundingly rejected by the EEC, see Kock (1969) 101.
139
219
Kennedy Round only dealt with a limited number of specific internal measures.
Perhaps most significantly, this round of negotiations produced the International
Anti-Dumping Code, which was effectively an embryonic international standard for
the drafting of domestic anti-dumping laws, and provided something of a model for
the subsequent codes negotiated during the Tokyo Round.144 Apart from that,
however, there were only piecemeal negotiations on isolated individual measures.
The United States agreed to remove its American Selling Price regulation, whereby
customs duties on a small range of goods were levied on a base determined by the
price of domestic competitor goods.145 In return, a number of European countries
agreed to modify their systems of road tax, which imposed a higher tax burden on
large vehicles of the kind used and sold by American companies.146 But even these
agreements did not survive scrutiny by the US Congress, which vetoed the two nontariff barrier agreements which imposed restrictions on US trade legislation and
policy. A number of major non-tariff issues raised during the course of negotiations
were hardly addressed at all, perhaps the most prominent being the problem of how
border tax adjustments should be dealt with appropriately under the GATT, in the
light of a recent rethinking of the theory underpinning the original GATT compromise on the issue.147 All in all, the results of the Kennedy Round in respect of nontariff barriers are generally agreed to have been meagre.
A number of different reasons have been offered to explain the reluctance of
GATT negotiators to take up internal non-tariff barriers at this time. Gerard and
Victoria Curzon argue that incorporating negotiations on non-tariff barriers did not
fit comfortably with the self-consciously self-contained nature of trade negotiations,
and was initially resented by some as an illegitimate linkage of trade issues with issues
outside the traditional field of trade.148 They also suggest that the inclusion of nontariff barriers was made more difficult by the strength of the norm of reciprocity:
rules on non-tariff barriers made it more difficult for negotiators to assess the degree
of reciprocity in a negotiated package, as it was not clear who would benefit from the
inclusion of a new general discipline on a particular kind of domestic regulation, and
by how much.149 This was particularly the case since the incidence of non-tariff
barriers was not equally distributed among participants in the Kennedy Round,
making reciprocity an even trickier issue to assess and manage. But perhaps most
importantly, it mattered a great deal that there was simply no substantive agreement
at the level of policy about what counted as an illegitimate obstacle to trade, and what
was a legitimate domestic regulation.150 Since these measures had not been part of
the original bargain struck by the post-war architects, and since it had never been the
purpose of the GATT to subject them to strict oversight, GATT participants had no
144
145
146
147
148
149
150
Preeg (1970) 166ff; Curtis and Vastine (1971) 205ff.

Evans (1971) 90ff.
Curtis and Vastine (1971) 204; Curzon and Curzon (1976) 1912.
Preeg (1970) 1356; Curtis and Vastine (1971) 235; Hudec (1990) 2312; Evans (1971) 95ff.
ibid 164; Evans (1971) 2567; Finlayson and Zacher (1981) 575; Weinrichter (1999) 3.
Evans (1971) 256, 260.
220
shared view as to the nature of normal or appropriate regulation in these new areas,
nor any collectively authorized forms of knowledge or principles of evaluation by
which to create and legitimate such a shared view.
Precisely the same problems made GATT dispute settlement ill-suited to address
disputes over non-tariff barriers. Hudec argues that the erosion of the substantive
policy consensus underlying the post-war GATT regime was one of the most
significant reasons that GATT dispute settlement was hardly used at all from 1963
to the end of that decade.151 Since, as described above, dispute settlement worked
precisely by mobilizing and reflecting that community consensus among trade
diplomats, it was not equipped for situations in which such underlying consensus
did not exist or could not be subtly manufactured. This was paradigmatically the case
with respect to non-tariff barriers. Precisely because the dispute settlement system
had been imagined in purposive terms as an instrument of collective purpose, and
had as a result been deeply embedded within the diplomatic world, it was singularly
poorly equipped to play the role of objective and neutral arbiter in difficult disputes
about internal non-tariff barriers.
What the experience of non-tariff barrier negotiations in the Kennedy Round
makes clear, then, is that the trade regimes turn to behind the border barriers
during the late 1960s and 1970s was not simply the continuation of its old agenda in
new areasas if dealing with the problems caused by internal regulations was simply
the next stage of a single process of trade liberalization which had begun in the postwar era with tariff and non-tariff border barriers. The collective ideas, norms, social
structures, and institutions which had been created during the embedded liberal
period acted as an obstacle to the scrutiny of internal measures, not as a prelude to
such scrutiny. Thus, if the trade regime were to address internal non-tariff barriers in
a serious and rigorous way, it would require a root-and-branch transformation of the
regimea re-imagining of the economic and political purposes for which it existed,
a new definition of the problems it was designed to address, a new understanding of
the proper role and function of law in the achievement of those purposes, and a new
set of institutions and social practices to reflect this new collective project. In the end,
this transformation was at least two decades in the making, and came to fruition only
with the Uruguay Round of negotiations. The story of this transformation is taken
up in the next chapter.
151
Hudec (1990) 287ff; Hudec (1978) 43; Long (1985) 88; see also Jackson (1969) 759.

Neoliberalism and The FormalTechnical Turn

Uploaded by

Copyright:

Available Formats

Neoliberalism and The FormalTechnical Turn

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Neoliberalism and The FormalTechnical Turn

Uploaded by

Copyright:

Available Formats

7

Embedded Liberalism and Purposive Law

I. The nature and purpose of the post-war trade regime

World Trade Law after Neoliberalism. Andrew Lang.

Embedded Liberalism and Purposive Law

The Trade Regime and the Neoliberal Turn

Brown (1950) 3842.

Embedded Liberalism and Purposive Law

Specifically, the practice of discriminatory tradethe formation of trading blocs, and

The Trade Regime and the Neoliberal Turn

ibid 8; Viner (1947) 617.

Embedded Liberalism and Purposive Law

cause of instability, and gave lesser weight to trade as a driver of prosperity. In

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

The Trade Regime and the Neoliberal Turn

RN Cooper, Trade Policy is Foreign Policy (1972) 9 Foreign Policy 18.

Embedded Liberalism and Purposive Law

shared institutional (and personal) ambitions situated in a matrix of long-term first-name

B. The role of law: purposive law and embedded legal forms

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

particularly striking in the GATT context, and frequently commented upon by

Curzon (1965) 53.

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

II. Approaches to domestic regulation in the GATTs early decades

The Trade Regime and the Neoliberal Turn

See above p193 and surrounding; also Preeg (1970) 24.

Embedded Liberalism and Purposive Law

Brown (1950) 1089.

The Trade Regime and the Neoliberal Turn

Art XXIII:1(b) (emphasis added).

Embedded Liberalism and Purposive Law

As a matter of purely textual interpretation, it is hard to argue with this

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

The Trade Regime and the Neoliberal Turn

discrimination and discriminatory intent as the primary criterion of illegality. And

Rethinking International Trade (1998) 19 University of Pennsylvania J of Intl Economic L 347; JL

Embedded Liberalism and Purposive Law

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

familiarity with which they evoked such norms. AustraliaAmmonium Sulfate

The Trade Regime and the Neoliberal Turn

Embedded Liberalism and Purposive Law

D Harvey, A Brief History of Neoliberalism (OUP, Oxford 2007) 12ff.

The Trade Regime and the Neoliberal Turn

barriers had become increasingly visible to exporters as barriers to access to foreign

See Evans (1971) 304ff.

Embedded Liberalism and Purposive Law

Preeg (1970) 166ff; Curtis and Vastine (1971) 205ff.

The Trade Regime and the Neoliberal Turn

You might also like