What's New?
 - Sitemap - Calendar
Trade Agreements - FTAA Process - Trade Issues 

español - français - português
Search

WORLD TRADE
ORGANIZATION

WT/DS62/AB/R
WT/DS67/AB/R
WT/DS68/AB/R
5 June 1998
(98-2271)
  Original: English

EUROPEAN COMMUNITIES - CUSTOMS CLASSIFICATION
OF CERTAIN COMPUTER EQUIPMENT

AB-1998-2

Report of the Appellate Body


(continued)

V. "Legitimate Expectations" in the Interpretation of a Schedule

74. The European Communities also submits that the Panel erred in interpreting Schedule LXXX, in particular, by:

(a) reading Schedule LXXX in the light of the "legitimate expectations" of an exporting Member; and

(b) considering that Article II:5 of the GATT 1994 confirms the interpretative value of "legitimate expectations".

Subordinately, the European Communities submits that the Panel erred in considering that the "legitimate expectations" of an exporting Member with regard to the interpretation of tariff concessions should be based on the classification practices for individual importers and individual consignments, or on the subjective perception of a number of exporting companies of that exporting Member.

75. Schedule LXXX provides tariff concessions for ADP machines under headings 84.71 and 84.73 and for telecommunications equipment under heading 85.17. The customs duties set forth in Schedule LXXX on telecommunications equipment are generally higher than those on ADP machines.50 We note that Schedule LXXX does not contain any explicit reference to "LAN equipment" and that the European Communities currently treats LAN equipment as telecommunications equipment. The United States, however, considers that the EC tariff concessions on ADP machines, and not its tariff concessions on telecommunications equipment, apply to LAN equipment. The United States claimed before the Panel, therefore, that the European Communities accords to imports of LAN equipment treatment less favourable than that provided for in its Schedule, and thus has acted inconsistently with Article II:1 of the GATT 1994. The United States argued that the treatment provided for by a concession is the treatment reasonably expected by the trading partners of the Member which made the concession.51 On the basis of the negotiating history of the Uruguay Round tariff negotiations and the actual tariff treatment accorded to LAN equipment by customs authorities in the European Communities during these negotiations, the United States argued that it reasonably expected the European Communities to treat LAN equipment as ADP machines, not as telecommunications equipment.


76. The Panel found that:

... for the purposes of Article II:1, it is impossible to determine whether LAN equipment should be regarded as an ADP machine purely on the basis of the ordinary meaning of the terms used in Schedule LXXX taken in isolation. However, as noted above, the meaning of the term "ADP machines" in this context may be determined in light of the legitimate expectations of an exporting Member.52

77. In support of this finding, the Panel explained that:

The meaning of a particular expression in a tariff schedule cannot be determined in isolation from its context. It has to be interpreted in the context of Article II of GATT 1994 ... It should be noted in this regard that the protection of legitimate expectations in respect of tariff treatment of a bound item is one of the most important functions of Article II.53

The Panel justified this latter statement by relying on the panel report in European Economic Community Payments and Subsidies Paid to Processors and Producers of Oilseeds and Related Animal-Feed Proteins54 ("EEC - Oilseeds"), and stated that:

The fact that the Oilseeds panel report concerns a non-violation complaint does not affect the validity of this reasoning in cases where an actual violation of tariff commitments is alleged. If anything, such a direct violation would involve a situation where expectations concerning tariff concessions were even more firmly grounded.55

78. The Panel also relied on Article II:5 of the GATT 1994, and stated that:

Although Article II:5 is a provision for the special bilateral procedure regarding tariff classification, not directly at issue in this case, the existence of this provision confirms that legitimate expectations are a vital element in the interpretation of Article II and tariff schedules.56

79. Finally, the Panel observed that its proposition that the terms of a Member's Schedule may be determined in the light of the "legitimate expectations" of an exporting Member:

... is also supported by the object and purpose of the WTO Agreement and those of GATT 1994. The security and predictability of "the reciprocal and mutually advantageous arrangements directed to the substantial reduction of tariffs and other barriers to trade" (expression common in the preambles to the two agreements) cannot be maintained without protection of such legitimate expectations. This is consistent with the principle of good faith interpretation under Article 31 of the Vienna Convention.57

80. We disagree with the Panel's conclusion that the meaning of a tariff concession in a Member's Schedule may be determined in the light of the "legitimate expectations" of an exporting Member. First, we fail to see the relevance of the EEC - Oilseeds panel report with respect to the interpretation of a Member's Schedule in the context of a violation complaint made under Article XXIII:1(a) of the GATT 1994. The EEC - Oilseeds panel report dealt with a non-violation complaint under Article XXIII:1(b) of the GATT 1994, and is not legally relevant to the case before us. Article XXIII:1 of the GATT 1994 provides for three legally distinct causes of action on which a Member may base a complaint; it distinguishes between so-called violation complaints, non-violation complaints and situation complaints under paragraphs (a), (b) and (c). The concept of "reasonable expectations", which the Panel refers to as "legitimate expectations", is a concept that was developed in the context of non-violation complaints.58 As we stated in India - Patents, for the Panel to use this concept in the context of a violation complaint "melds the legally-distinct bases for 'violation' and 'non-violation' complaints under Article XXIII of the GATT 1994 into one uniform cause of action",59 and is not in accordance with established GATT practice.

81. Second, we reject the Panel's view that Article II:5 of the GATT 1994 confirms that "legitimate expectations are a vital element in the interpretation" of Article II:1 of the GATT 1994 and of Members' Schedules.60 It is clear from the wording of Article II:5 that it does not support the Panel's view. This paragraph recognizes the possibility that the treatment contemplated in a concession, provided for in a Member's Schedule, on a particular product, may differ from the treatment accorded to that product and provides for a compensatory mechanism to rebalance the concessions between the two Members concerned in such a situation. However, nothing in Article II:5 suggests that the expectations of only the exporting Member can be the basis for interpreting a concession in a Member's Schedule for the purposes of determining whether that Member has acted consistently with its obligations under Article II:1. In discussing Article II:5, the Panel overlooked the second sentence of that provision, which clarifies that the "contemplated treatment" referred to in that provision is the treatment contemplated by both Members.

82. Third, we agree with the Panel that the security and predictability of "the reciprocal and mutually advantageous arrangements directed to the substantial reduction of tariffs and other barriers to trade" is an object and purpose of the WTO Agreement, generally, as well as of the GATT 1994.61 However, we disagree with the Panel that the maintenance of the security and predictability of tariff concessions allows the interpretation of a concession in the light of the "legitimate expectations" of exporting Members, i.e., their subjective views as to what the agreement reached during tariff negotiations was. The security and predictability of tariff concessions would be seriously undermined if the concessions in Members' Schedules were to be interpreted on the basis of the subjective views of certain exporting Members alone. Article II:1 of the GATT 1994 ensures the maintenance of the security and predictability of tariff concessions by requiring that Members not accord treatment less favourable to the commerce of other Members than that provided for in their Schedules.

83. Furthermore, we do not agree with the Panel that interpreting the meaning of a concession in a Member's Schedule in the light of the "legitimate expectations" of exporting Members is consistent with the principle of good faith interpretation under Article 31 of the Vienna Convention. Recently, in India - Patents, the panel stated that good faith interpretation under Article 31 required "the protection of legitimate expectations".62 We found that the panel had misapplied Article 31 of the Vienna Convention and stated that:

The duty of a treaty interpreter is to examine the words of the treaty to determine the intentions of the parties. This should be done in accordance with the principles of treaty interpretation set out in Article 31 of the Vienna Convention. But these principles of interpretation neither require nor condone the imputation into a treaty of words that are not there or the importation into a treaty of concepts that were not intended.63

84. The purpose of treaty interpretation under Article 31 of the Vienna Convention is to ascertain the common intentions of the parties. These common intentions cannot be ascertained on the basis of the subjective and unilaterally determined "expectations" of one of the parties to a treaty. Tariff concessions provided for in a Member's Schedule -- the interpretation of which is at issue here -- are reciprocal and result from a mutually-advantageous negotiation between importing and exporting Members. A Schedule is made an integral part of the GATT 1994 by Article II:7 of the GATT 1994. Therefore, the concessions provided for in that Schedule are part of the terms of the treaty. As such, the only rules which may be applied in interpreting the meaning of a concession are the general rules of treaty interpretation set out in the Vienna Convention.

85. Pursuant to Article 31(1) of the Vienna Convention, the meaning of a term of a treaty is to be determined in accordance with the ordinary meaning to be given to this term in its context and in the light of the object and purpose of the treaty. Article 31(2) of the Vienna Convention stipulates that:

The context, for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:

(a) any agreement relating to the treaty which was made between all the parties in connexion with the conclusion of the treaty;

(b) any instrument which was made by one or more parties in connexion with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.

Furthermore, Article 31(3) provides that:

There shall be taken into account together with the context:

(a) any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;

(b) any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;

(c) any relevant rules of international law applicable in the relations between the parties.

Finally, Article 31(4) of the Vienna Convention stipulates that:

A special meaning shall be given to a term if it is established that the parties so intended.

86. The application of these rules in Article 31 of the Vienna Convention will usually allow a treaty interpreter to establish the meaning of a term.64 However, if after applying Article 31 the meaning of the term remains ambiguous or obscure, or leads to a result which is manifestly absurd or unreasonable, Article 32 allows a treaty interpreter to have recourse to:

... supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion.

With regard to "the circumstances of [the] conclusion" of a treaty, this permits, in appropriate cases, the examination of the historical background against which the treaty was negotiated.65

87. In paragraphs 8.20 and 8.21 of the Panel Report, the Panel quoted Articles 31 and 32 of the Vienna Convention and explicitly recognized that these fundamental rules of treaty interpretation applied "in determining whether the tariff treatment of LAN equipment ... is in conformity with the tariff commitments contained in Schedule LXXX".66 As we have already noted above, the Panel, after a textual analysis,67 came to the conclusion that:

... for the purposes of Article II:1, it is impossible to determine whether LAN equipment should be regarded as an ADP machine purely on the basis of the ordinary meaning of the terms used in Schedule LXXX taken in isolation.68

Subsequently, the Panel abandoned its effort to interpret the terms of Schedule LXXX in accordance with Articles 31 and 32 of the Vienna Convention.69 In doing this, the Panel erred.

88. As already discussed above, the Panel referred to the context of Schedule LXXX70 as well as to the object and purpose of the WTO Agreement and the GATT 1994, of which Schedule LXXX is an integral part.71 However, it did so to support its proposition that the terms of a Schedule may be interpreted in the light of the "legitimate expectations" of an exporting Member. The Panel failed to examine the context of Schedule LXXX and the object and purpose of the WTO Agreement and the GATT 1994 in accordance with the rules of treaty interpretation set out in the Vienna Convention.

89. We are puzzled by the fact that the Panel, in its effort to interpret the terms of Schedule LXXX, did not consider the Harmonized System and its Explanatory Notes. We note that during the Uruguay Round negotiations, both the European Communities and the United States were parties to the Harmonized System. Furthermore, it appears to be undisputed that the Uruguay Round tariff negotiations were held on the basis of the Harmonized System's nomenclature and that requests for, and offers of, concessions were normally made in terms of this nomenclature. Neither the European Communities nor the United States argued before the Panel72 that the Harmonized System and its Explanatory Notes were relevant in the interpretation of the terms of Schedule LXXX. We believe, however, that a proper interpretation of Schedule LXXX should have included an examination of the Harmonized System and its Explanatory Notes.

90. A proper interpretation also would have included an examination of the existence and relevance of subsequent practice. We note that the United States referred, before the Panel, to the decisions taken by the Harmonized System Committee of the WCO in April 1997 on the classification of certain LAN equipment as ADP machines.73 Singapore, a third party in the panel proceedings, also referred to these decisions.74 The European Communities observed that it had introduced reservations with regard to these decisions and that, even if they were to become final as they stood, they would not affect the outcome of the present dispute for two reasons: first, because these decisions could not confirm that LAN equipment was classified as ADP machines in 1993 and 1994; and, second, because this dispute "was about duty treatment and not about product classification".75 We note that the United States agrees with the European Communities that this dispute is not a dispute on the correct classification of LAN equipment, but a dispute on whether the tariff treatment accorded to LAN equipment was less favourable than that provided for in Schedule LXXX.76 However, we consider that in interpreting the tariff concessions in Schedule LXXX, decisions of the WCO may be relevant; and, therefore, they should have been examined by the Panel.

91. We note that the European Communities stated that the question whether LAN equipment was bound as ADP machines, under headings 84.71 and 84.73, or as telecommunications equipment, under heading 85.17, was not addressed during the Uruguay Round tariff negotiations with the United States.77 We also note that the United States asserted that:

In many, perhaps most, cases, the detailed product composition of tariff commitments was never discussed in detail during the tariff negotiations of the Uruguay Round ...78 (emphasis added)

and that:

The US-EC negotiation on Chapter 84 provided an example of how two groups of busy negotiators dealing with billions of dollars of trade and hundreds of tariff lines relied on a continuation of the status quo.79 (emphasis added)

This may well be correct and, in any case, seems central to the position of the United States. Therefore, we are surprised that the Panel did not examine whether, during the Tokyo Round tariff negotiations, the European Communities bound LAN equipment as ADP machines or as telecommunications equipment.80

92. Albeit, with the mistaken aim of establishing whether the United States "was entitled to legitimate expectations"81 regarding the tariff treatment of LAN equipment by the European Communities, the Panel examined, in paragraphs 8.35 to 8.44 of the Panel Report, the classification practice regarding LAN equipment in the European Communities during the Uruguay Round tariff negotiations. The Panel did this on the basis of certain BTIs and other decisions relating to the customs classification of LAN equipment, issued by customs authorities in the European Communities during the Uruguay Round.82 In the light of our observations on "the circumstances of [the] conclusion" of a treaty as a supplementary means of interpretation under Article 32 of the Vienna Convention83, we consider that the classification practice in the European Communities during the Uruguay Round is part of "the circumstances of [the] conclusion" of the WTO Agreement and may be used as a supplementary means of interpretation within the meaning of Article 32 of the Vienna Convention. However, two important observations must be made: first, the Panel did not examine the classification practice in the European Communities during the Uruguay Round negotiations as a supplementary means of interpretation within the meaning of Article 32 of the Vienna Convention84 ; and, second, the value of the classification practice as a supplementary means of interpretation is subject to certain qualifications discussed below.

93. We note that the Panel examined the classification practice of only the European Communities,85 and found that the classification of LAN equipment by the United States during the Uruguay Round tariff negotiations was not relevant.86 The purpose of treaty interpretation is to establish the common intention of the parties to the treaty. To establish this intention, the prior practice of only one of the parties may be relevant, but it is clearly of more limited value than the practice of all parties. In the specific case of the interpretation of a tariff concession in a Schedule, the classification practice of the importing Member, in fact, may be of great importance. However, the Panel was mistaken in finding that the classification practice of the United States was not relevant.

94. In this context, we also note that while the Panel examined the classification practice during the Uruguay Round negotiations, it did not consider the EC legislation on customs classification of goods that was applicable at that time. In particular, it did not consider the "General Rules for the Interpretation of the Combined Nomenclature" as set out in Council Regulation 2658/87 on the Common Customs Tariff.87 If the classification practice of the importing Member at the time of the tariff negotiations is relevant in interpreting tariff concessions in a Member's Schedule, surely that Member's legislation on customs classification at that time is also relevant.

95. Then there is the question of the consistency of prior practice. Consistent prior classification practice may often be significant. Inconsistent classification practice, however, cannot be relevant in interpreting the meaning of a tariff concession. We note that the Panel, on the basis of evidence relating to only five out of the then 12 Member States,88 made the following factual findings with regard to the classification practice in the European Communities:

To rebut the presumption raised by the United States, the European Communities has produced documents which indicate that LAN equipment had been treated as telecommunication apparatus by other customs authorities in the European Communities.89 (emphasis added)

.. it would be reasonable to conclude at least that the practice [regarding classification of LAN equipment] was not uniform in France during the Uruguay Round.90

Germany appears to have consistently treated LAN equipment as telecommunication apparatus.91

... LAN equipment was generally treated as ADP machines in Ireland and the United Kingdom during the Uruguay Round.92 (emphasis added)

As a matter of logic, these factual findings of the Panel lead to the conclusion that, during the Uruguay Round tariff negotiations, the practice regarding the classification of LAN equipment by customs authorities throughout the European Communities was not consistent.

96. We also note that in paragraphs 8.44 and 8.60 of the Panel Report, the Panel identified Ireland and the United Kingdom as the "largest" and "major" market for LAN equipment exported from the United States. On the basis of this assumption, the Panel gave special importance to the classification practice by customs authorities in these two Member States. However, the European Communities constitutes a customs union, and as such, once goods are imported into any Member State, they circulate freely within the territory of the entire customs union. The export market, therefore, is the European Communities, not an individual Member State.

97. For the reasons set out above, we conclude that the Panel erred in finding that the "legitimate expectations" of an exporting Member are relevant for the purposes of interpreting the terms of Schedule LXXX and of determining whether the European Communities violated Article II:1 of the GATT 1994. We also conclude that the Panel misinterpreted Article II:5 of the GATT 1994.

98. On the basis of the erroneous legal reasoning developed and the selective evidence considered, the Panel was not justified in coming to the conclusion that the United States was entitled to "legitimate expectations" that LAN equipment would be accorded tariff treatment as ADP machines in the European Communities93 and, therefore, that the European Communities acted inconsistently with the requirements of Article II:1 of the GATT 1994 by failing to accord imports of LAN equipment from the United States treatment no less favourable than that provided for in Schedule LXXX.94

99. In the light of our conclusion that the "legitimate expectations" of an exporting Member are not relevant in determining whether the European Communities violated Article II:1 of the GATT 1994, we see no reason to examine the subordinate claim of error of the European Communities relating to the evidence on which the "legitimate expectations" of exporting Members were based.

To continue with VI. Clarification of the Scope of Tariff Concession

Return to Table of Contents


50. See Panel Report, paras. 2.10 and 8.1.

51. See Panel Report, para. 5.15.

52. Panel Report, para. 8.31.

53. Panel Report, para. 8.23.

54. Adopted 25 January 1990, BISD 37S/86, para. 148.

55. Panel Report, para. 8.23.

56. Panel Report, para. 8.24.

57. Panel Report, para. 8.25.

58. See Appellate Body Report, India - Patents, adopted 16 January 1998, WT/DS50/AB/R, paras. 36 and 41.

59. Adopted 16 January 1998, WT/DS50/AB/R, para. 42.

60. See Panel Report, para. 8.24.

61. See Panel Report, para. 8.25.

62. Panel Report, India - Patents, adopted 16 January 1998, WT/DS50/R, para. 7.18.

63. Appellate Body Report, India - Patents, adopted 16 January 1998, WT/DS50/AB/R, para. 45.

64. R. Jennings and A. Watts (eds.), Oppenheim's International Law, 9th ed., Vol. I (Longman, 1992), p. 1275.

65. I. Sinclair, The Vienna Convention on the Law of Treaties, 2nd ed., (Manchester University Press, 1984), p. 141:

... the reference in Article 32 of the Convention to the circumstances of the conclusion of a treaty may have some value in emphasising the need for the interpreter to bear constantly in mind the historical background against which the treaty has been negotiated.

66. Panel Report, para. 8.22.

67. See Panel Report, para. 8.30.

68. Panel Report, para. 8.31.

69. As discussed above in paragraphs 76-84, the Panel relied instead on the concept of "legitimate expectations" as a means of treaty interpretation.

70. See Panel Report, paras. 8.23-8.24.

71. See Panel Report, para. 8.25.

72. We recall, however, that in reply to our questions at the oral hearing, both the European Communities and the United States accepted the relevance of the Harmonized System and its Explanatory Notes in interpreting the tariff concessions of Schedule LXXX. See paras. 13 and 38 of this Report.

73. See Panel Report, para. 5.12.

74. As noted in para. 6.34 of the Panel Report, Singapore pointed out, before the Panel, that:

... the WCO's HS Committee had recently decided that LAN equipment was properly classifiable in heading 84.71 of the HS. The HS Committee had specifically declined to adopt the position advanced that heading 85.17 was the appropriate category ... The EC had suggested that the HS Committee decision was intended solely to establish the appropriate HS classification for future imports. It ignored that the language interpreted by the HS Committee was the same language appearing in the EC's HS nomenclature and in the EC's concession schedule at the time of the negotiations and afterwards.

75. Panel Report, para. 5.13.

76. See Panel Report, para. 5.3.
 

77. See Panel Report, para. 5.28.

78. Appellee's submission of the United States, para. 26.

79. Panel Report, para. 5.31.

80. We note that in paragraph 8 of its third participant's submission, Japan stated that: "[i]n particular, the classification of the LAN equipment among the Members of the EC was not identical before the Uruguay Round".

81. Panel Report, para. 8.60.

82. The lists of the BTIs and classification decisions in the form of a letter, submitted by the parties and considered by the Panel, were attached to the Panel Report as Annex 4 and Annex 6 thereof.

83. See para. 86 of this Report.

84. It examined the actual classification practice to determine whether the United States could have "legitimate expectations" with regard to the tariff treatment of LAN equipment.

85. See Panel Report, paras. 8.36-8.44.

86. See Panel Report, para. 8.60. We note that in paragraph 8.58 of the Panel Report, the Panel stated that the classification of LAN equipment by other WTO Members was not relevant either.

87. Title I, Part I of Annex I of Council Regulation (EEC) No. 2658/87 of 23 July 1987, Official Journal No. L 256, 7 September 1987, p. 1.

88. With regard to the manner in which the Panel evaluated the evidence regarding classification practice during the Uruguay Round tariff negotiations, we note that in paragraph 8.37 of the Panel Report, the Panel accepted certain BTIs submitted by the United States as relevant evidence, while in footnote 152 of the Panel Report, it considered similar BTIs submitted by the European Communities to be irrelevant.

89. Panel Report, para. 8.40.

90. Panel Report, para. 8.42.

91. Panel Report, para. 8.43.

92. Panel Report, para. 8.41. In this paragraph, the Panel stated that the only direct counter-evidence against the claim of the United States that customs authorities in Ireland and the United Kingdom consistently classified LAN equipment as ADP machines during the Uruguay Round negotiations is a BTI issued by the UK customs authority to CISCO, classifying one type of LAN equipment (routers) as telecommunications apparatus. The Panel dismisses the value of this BTI as evidence on the basis that it "became effective only a week or so before the conclusion of the Uruguay Round negotiations [15 December 1993]". Similarly, in footnote 152 of the Panel Report, the Panel did not consider other BTIs issued by the UK customs authorities to be relevant because they became valid after the conclusion of the Uruguay Round negotiations. We note, however, that all of these BTIs became valid in December 1993 or February 1994, i.e., before the end of the verification process, to which all Schedules were submitted and which took place between 15 February 1994 and 25 March 1994 (MTN.TNC/W/131, 21 January 1994). Therefore, in our view, the Panel should have considered these BTIs.


93. See Panel Report, para. 8.60.

94. See Panel Report, para. 9.1.