2. Claim under Article 2.2 � reasonability (claim number 4)
(a) Parties' arguments
6.88 India submits that the European Communities acted
inconsistently with Article 2.2 by applying the SG&A and profit incorrectly
determined under Article 2.2.2(ii) even though they were clearly not
"reasonable". For India, Article 2.2.2 lays down how the �amounts for
administrative, selling and general costs and for profits� are to be determined.
It does not, however, explain how the reasonable amounts for SG&A and for
profits are to be determined. For India, the word reasonable in Article 2.2 has
a separate function, and the �reasonable� test of Article 2.2 is an independent,
over-arching requirement in addition to the requirements of Article 2.2.2,
rather than a rule concretised by Article 2.2.2. �Reasonable� must thus be
interpreted as a substantive requirement: whatever method under Article 2.2.2 is
used, Article 2.2 requires that the result must be �reasonable�. Moreover, India
argues that Article 2.2.2(iii) contains an implicit definition of the notion of
�reasonable�, which can be used to test the results reached under the methods
set out in the chapeau and paragraphs (i) and (ii) of Article 2.2.2.
6.89 In respect of the specific factual situation in the bed
linen proceeding, India recalls that some producers did have sales of other
products in the same general category (textiles). The average same-category
domestic profit rate of these other producers was 7.04% and the average overall
profit rate of these other producers was 5.41%. The average profit rate for bed
linen was found to be 6.1% for the other countries in the investigation, Egypt
and Pakistan. India recalls that the reasonable profit rate imputed to the EC
industry was 5%. Finally, India notes that the profit rate determined by the
European Communities on the basis of the profitable sales of one Indian company,
and applied in calculating normal value for all other Indian producers was
18.65% . It is evident, in India' view, that in comparison with all the other
profit rates that were relevant in the context of the bed linen proceeding, the
figure of 18.65% stands out as a complete anomaly and does not reflect the
profits actually realised by bed linen producers inside and outside India. The
figure is three times higher than the average profit rates determined for the
other two countries involved in the investigation as well as that of the
European Communities� own bed linen industry. India submits that, if the word
reasonable is defined by reference to the criteria set out in the
Article 2.2.2(iii), it is obvious that the profit rate established for other
Indian producers is unreasonable.
6.90 The European Communities is of the view that the methods
of calculating SG&A and profits that are set out in Articles 2.2.2(i)-(iii)
provide for the determination of �a reasonable amount for administrative,
selling and general costs and for profits�. Those options represent, for the
European Communities, particular and detailed formulations of what constitutes
"reasonable" amounts. The European Communities also believes that the limitation
set out in the third option � �provided that the amount for profit so
established shall not exceed . . .� � applies only to the third option, and not
to the other two. Had the drafters wished to apply this proviso to all the
options, the European Communities submits they would have attached it to the
chapeau of Article 2.2.2.
6.91 The European Communities rejects India�s argument that
option (iii) defines what is reasonable. Options (i) and (ii) are, in the
European Communities' view, formulae that produce reasonable solutions. The
European Communities further considers that it was obviously the intention of
the drafters that the application of these formulae would always produce figures
for SG&A and for profits that meet the standard of reasonability specified in
the last sentence of the chapeau of Article 2.2. The words �any other reasonable
method� in option (iii) clearly refer to methods other than those
described in the preceding options (i) and (ii), which are in themselves
reasonable and do not need to be qualified as such. The wording of these options
at least implies that the results obtained through the application of options
(i) and (ii) are presumed to satisfy the standard of reasonability. The relevant
question, the European Communities proposes, would then be: What kind and weight
of evidence would be required to overturn the presumption? The European
Communities asserts that India has presented no relevant evidence to rebut the
presumption that the results obtained through the application of option (ii) in
the case were reasonable.
6.92 The European Communities suggests that the three options
in Article 2.2.2 are intended to produce approximations of the amounts that
would emerge from applying the formula in the chapeau, that is to say the SG&A
and profits of a producer selling the like product in its own market. This, in
its turn, is intended to allow investigating authorities to construct a normal
value that is as close as possible to the normal value that would have been
established on the basis of domestic prices, had there been sufficient
comparable sales in the ordinary course of trade. The European Communities
points out that Bombay Dyeing has representative sales in the Indian market.
That a single producer can have 80% of the domestic market for bed linen and
make a profit of over 18%, while numerous other producers ignore this market and
devote themselves to exports, may be an uncommon situation, but that does not
make the results arising from the use of data from this company ipso facto
unreasonable. Rather, the European Communities is of the view that it would have
been unreasonable to ignore this company and choose another source, which would
inevitably be less typical of sellers in that market.
6.93 The United States, as third party, maintains that
Articles 2.2 and 2.2.2 of the Agreement set forth the requirements for
calculating profit when normal value is based on constructed value instead of
prices. Article 2.2 provides, inter alia, for the addition to cost of
production of a reasonable amount for profit. Article 2.2.2 then sets forth
several explicit options for how a reasonable profit may be determined. The
United States disagrees with the view that there is a limitation on the amount
for constructed value profit. In the US view, no such limitation exists in the
Agreement. With one exception � subpart (iii) � the methodologies in
Article 2.2.2 limit how the authorities may determine the profit amount,
not the amount of the profit itself. The �profit cap� in subpart (iii) is
necessary to impose some limitations on �other� �reasonable� methodologies for
determining profit not specifically articulated in the Agreement. Significantly,
subpart (iii) does not expressly or implicitly impose a similar
limitation upon the preferred profit methodology in the chapeau or the
alternatives in subparts (i) or (ii).
(b) Findings
6.94 Having concluded that the European Communities could
properly apply the option set out in Article 2.2.2(ii), and that it acted
consistently with that provision in making its calculation on the basis of
information for one other producer for its sales of the like product in the
ordinary course of trade, the next issue before us is whether the results of a
proper calculation under Article 2.2.2(ii) are subject to a separate test of
�reasonability� before they may be used in constructing a normal value for other
producers.
6.95 We first consider the text of the provision. The chapeau
of Article 2.2.2 begins with the phrase �For the purpose of paragraph 2�, and
provides that the amounts for, inter alia, profits "shall be based on
actual data pertaining to production and sales in the ordinary course of trade
of the like product by the exporter or producer under investigation. The second
sentence of Article 2.2.2 specifies that if the chapeau methodology cannot be
used, these amounts "may be determined on the basis of" subparagraphs (i)-(iii).
Article 2.2, which is referred to in the first sentence, establishes the basic
principle that when a constructed value is used, it shall include, inter alia,
a reasonable amount for profit.
6.96 The text thus indicates that the methodologies set out
in Article 2.2.2 are outlined �for the purpose� of calculating a reasonable
profit amount pursuant to Article 2.2. There is no specific language
establishing a separate reasonability test, or indicating how such a test should
be conducted. In these circumstances, we consider that there is no textual basis
for such a requirement. Thus, the ordinary meaning of the text indicates that if
one of the methods of Article 2.2.2 is properly applied, the results are by
definition "reasonable" as required by Article 2.2.
6.97 Further, we note that Article 2.2.2(iii) provides for
the use of "any other reasonable method", without specifying such method,
subject to a cap, defined as "the profit normally realized by other exporters or
producers on sales of products of the same general category in the domestic
market of the country of origin". To us, the inclusion of a cap where the
methodology is not defined indicates that where the methodology is defined, in
subparagraphs (i) and (ii), the application of those methodologies yields
reasonable results. If those methodologies did not yield reasonable results,
presumably the drafters would have included some explicit constraint on the
results, as they did for subparagraph (iii).
6.98 Thus, we conclude that the text indicates that, if a
Member bases its calculations on either the chapeau or paragraphs (i) or (ii),
there is no need to separately consider the reasonability of the profit rate
against some benchmark. In particular, there is no need to consider the
limitation set out in paragraph (iii). That limitation is triggered only when a
Member does not apply one of the methods set out in the chapeau or
paragraphs (i) and (ii) of Article 2.2.2. Indeed, it is arguably precisely
because no specific method is outlined in paragraph (iii) that the limitation on
the profit rate exists in that provision.
6.99 We note further that the methodology set out in the
chapeau of Article 2.2.2, as well as those in subparagraphs (i) and (ii), rely
on actual data from the books of the producer(s) or exporter(s) being used as
sources. India, however, argues that even where the chapeau methodology is
applied, which requires the use of actual data concerning the product under
investigation sold by the producer being investigated, the results are subject
to a separate test of reasonability.42 To test for reasonability results arrived
at through the use of actual data for the production of the like product
by the producer/exporter being investigated does not, in our view, serve any
meaningful purpose. Whatever the argument about the "reasonability" of a
particular result � a 50% profit rate, for instance � if it is based on actual
data and properly calculated, then that is the reality. An important object and
purpose of Article 2.2.2, as discussed above, is to base the calculation of the
profit amount on actual data. Similarly, while the methods set out in paragraphs
(i) and (ii) are derivatives of the chapeau methodology, where actual data are
used as required and the calculation is correct, the results obtained themselves
reflect objective reality. Thus, the use of actual data itself ensures that
subjective judgments about the reasonability of the results do not affect the
calculation of constructed normal value. We consider that no purpose would be
served by testing the results obtained under the chapeau and subparagraphs (i)
and (ii) against some arbitrary or subjective standard of reasonability.
6.100 In this regard, we note that the standard of
reasonability proposed by India � the Article 2.2.2(iii) profit cap � is, in our
view, arbitrary in the context of the "reality" of the results obtained under
paragraphs (i) and (ii). The other benchmarks suggested by India in respect of
the specific factual situation in the bed linen proceeding seem equally
arbitrary and subjective. India asserts that the average category domestic
profit rate of other producers on the same category of goods was 7.04% and the
average overall profit rate of these other producers was 5.41%. India also
indicates that the average profit for bed linen producers in Egypt and Pakistan
was determined to be 6.1% during the investigation. There is no objective basis
we can discern for concluding that these amounts are more "reasonable" than the
amount determined on the basis of the actual data. The only factor common to
these figures is that they are all lower than 18.65%, the profit rate determined
by the European Communities on the basis of the profitable sales of one Indian
producer and applied in the construction of normal value for the other Indian
companies.43 India puts forth no reason as to why a "reasonable" profit rate in
this case should be defined by reference to these data, except to emphasise the
difference between the profit rate actually used and these suggested benchmark
profit rates. Merely that these other profit rates are lower does not, in our
opinion, make them more "reasonable" than the rate actually calculated and
applied by the European Communities.
6.101 We, therefore, conclude that Article 2.2.2(ii), when
applied correctly, necessarily yields reasonable amounts for profits, and that
the AD Agreement does not require consideration of a separate reasonability test
in respect of results arrived at through the use of that methodology. The
European Communities did not, therefore, act inconsistently with the
requirements of Article 2.2 by not having applied such a test to the results
that it obtained under Article 2.2.2(ii).
3. Claim under Article 2.4.2 - "zeroing" (claim number 7)
6.102 The practice of "zeroing" arises in situations where an
investigating authority makes multiple comparisons of export price and normal
value, and then aggregates the results of these individual comparisons to
calculate a dumping margin for the product as a whole. In this case, the
European Communities compared weighted averages of export prices and normal
value for each of several models or product types of bed linen. India has no
complaint about this aspect of the EC determination.44 The comparisons for the
different models in some cases showed the export price to be lower than the
normal value, and in some cases showed the export price to be higher than the
normal value. The results of the latter comparisons are referred to as
"negative" margins. The European Communities then calculated a weighted average
dumping margin for the product at issue, cotton-type bed linen, on the basis of
the results obtained in the comparisons by model. In the course of this part of
the calculation, the European Communities summed up the total value of the
dumping � the total "dumping amount" � on the investigated imports. The European
Communities calculated the dumping amounts by multiplying the value of the
imports of each model by the margin of price difference for each model. The
European Communities counted as zero the dumping amount for those models where
the margin was negative. The European Communities then divided the total dumping
amount by the value of the exports involved, including the value of those models
for which the individual margin was negative, and the dumping amount was thus
counted as zero. It is this aspect of the calculation, the assigning of a value
of zero to the comparisons yielding a "negative" margin, which constitutes the
challenged practice of zeroing which is the subject of India's claim under
Article 2.4.2.
(a) Parties' arguments
6.103 India argues that the European Communities acted
inconsistently with Article 2.4.2 of the AD Agreement by zeroing "negative
dumping" amounts for certain types of bed linen in calculating the overall
weighted average dumping margin for the like product bed linen. According to
India, the European Communities effectively averaged only within a model, and
not between models, and thus did not compare a weighted average normal value to
a weighted average of prices of all comparable export transactions, as required
by Article 2.4.2 of the AD Agreement. In India's view, Article 2.4.2 provides
for three possibilities to establish a dumping margin:
A comparison of a weighted average normal value with a
weighted average of prices of all comparable export transactions;
A comparison of normal value and export prices on a
transaction‑to‑transaction basis; or
A comparison of the normal value established on a
weighted average basis to prices of individual export transactions (in
certain specific cases).
India asserts that the European Communities opted to apply
the first option in establishing the dumping margin in this case, but did not
properly make this comparison by engaging in the practice of zeroing.45
6.104 In India's view, the practice of zeroing is not
consistent with the requirement set forth in Article 2.4.2 that the comparison
take into account the "weighted average of prices of all comparable export
transactions". India asserts that this language precludes excluding certain
amounts from the calculation simply because they showed "negative" dumping.
India argues that, given the use of the words �weighted average� in
Article 2.4.2 and the definition of the word �average�, there is clearly no
justification for excluding certain amounts in establishing an average. An
�average� relates to the total of given amounts and not to a number of given
amounts from which a selection can be made as to which ones are to be averaged.
The use of the word �all� in Article 2.4.2 underlines this idea. And, finally,
India posits that the practice of attributing a zero value to "negative dumping"
for the eventual calculation of overall dumping margins is contrary to the
concept of weighting and in fact distorts the process of actually weighting
dumping margins. Moreover, India maintains that this EC method always will lead
to a higher dumping margin compared to the method India asserts is envisaged by
the AD Agreement. India acknowledges that in the situation where all models are
dumped, the results would be the same, but argues that this situation did not
occur in the bed linen case. However, India asserts that in this case, because
all models in the bed linen proceeding were not dumped, the zeroing of "negative
dumping" margins calculated for certain product types resulted in the
overstatement of the dumping margins for four companies, and for one company a
finding of dumping where dumping did not exist.
6.105 The European Communities maintains that its practice in
calculating the dumping margin is consistent with the requirements of
Article 2.4.2. In the European Communities' view, the practice of "zeroing" as
applied in this case recognizes that the process of calculating dumping margins
is directed at dumping, and therefore the European Communities' methodology
focuses on those product types where dumping has been found. In the case of any
product types for which there is no dumping (i.e. the margin is zero or less
than zero ("negative dumping")), the European Communities treats this margin as
zero. However, the types of products that are found to have margins less than
zero (and which therefore are not being dumped), are nevertheless kept in the
calculation (albeit at notional zero margins), on a weighted average basis, of
the overall dumping margin for the like product, and thereby reduce the overall
weighted average dumping margin determined for that product.
6.106 The European Communities focuses on the need to
consider all �comparable� export transactions, which it asserts is done in its
practice, which observes the principle of comparing weighted averages for those
products that are comparable. Moreover, the European Communities argues,
Article 2.4.2 refers to �the existence of margins of dumping�, making clear that
the process of comparing weighted averages will normally conclude with more than
one dumping margin. However, the process of determining a single dumping margin,
on which the collection of the duty is based, from these margins does not, in
the European Communities' view, fall within the express terms of Article 2.4.2,
but is left to the discretion of Members. The European Communities also disputes
India's contention that its methodology will always lead to a higher dumping
margin than would have been the case if "zeroing" had not taken place.
6.107 Egypt argues that the European Communities manipulated
the calculation of the overall dumping margin for Egyptian producers by zeroing
negative dumping amounts on a per-type basis, in violation of Article 2.4.2 of
the AD Agreement. In Egypt's view, had the Commission followed strictly its own
established practice, the outcome would have been different. In failing to do
that, it is, for Egypt, clear that the European Communities was determined to
have bigger dumping margins.
6.108 Japan asserts that the EC practice of "zeroing" is not
consistent with the requirements of Article 2.4.2. In Japan's view, this
provision explicitly calls for dumping margins to be based on a comparison of a
weighted average normal value with a weighted average of prices of all
comparable export transactions, and a proper weighted average does not
arbitrarily raise some of the numbers in the average in an effort to increase
the final result of the weighted average. Japan maintains that the term
"comparable" as used in Article 2.4.2 cannot justify the European Communities'
practice. In Japan's view, the term speaks only to the need to make the
comparison on an "apples-to-apples" basis, and does not authorize zeroing. Japan
argues that the European Communities seems to believe that if it properly
weight-averages once within the product type, then it need not properly
weight-average in the next stage, aggregation across the various product
types. In Japan's view, this interpretation ignores the plain meaning of
Article 2.4.2, which requires the comparison of a weighted average based on all
comparable export transactions, not just those transactions found to be dumped.
The EC approach, including the volume of the non-dumped product types in the
overall average, considers only part of the export transactions, the volume
element, but ignores the price element. By setting the value of the non-dumped
product type to zero, Japan asserts that the European Communities essentially
changed the prices of the underlying export transactions. In Japan's view, the
text of Article 2.4.2 explicitly calls for a weighted average of the prices,
not of some actual prices and some arbitrarily adjusted prices. In addition,
Japan asserts that Article 2.4 creates an overall obligation of fair comparison
for the calculation of dumping margins. Japan maintains that it is not fair to
skew a weighted average by adjusting upward some prices used in the calculation
of that weighted average.
6.109 The United States maintains that Article 2.4.2 does not
prohibit the practice of zeroing.46 In the United States' view, all that
Article 2.4.2 requires is that, in making comparisons between the export price
and the normal value of the like product in an investigation, each comparison
shall be made either on a weighted-average-to-weighted-average basis or a
transaction-to-transaction basis. This requirement of comparing
weighted-average-to-weighted-average figures or transaction-to-transaction
figures is explicitly made subject to the requirements of Article 2.4. Thus, it
is clear that the weight-averaging normally is not to involve transactions which
are distinct in terms of physical characteristics of the products, conditions
and terms of sale, and other differences affecting price comparability. In the
United States' view, the �zeroing� practice applied by the European Communities
in this case is not covered by Articles 2.4 and 2.4.2 because it arises at a
step subsequent to the comparison of export price and normal value, when
the individual, model-specific margins were combined into an overall average
rate of dumping. The United States asserts that its view is confirmed by the
fact that Article 2.4.2 explicitly permits transaction-to-transaction
comparisons without providing a methodology for combining margins calculated
pursuant to that methodology either. The United States points out that when this
stage of combining the results of the actual comparisons is reached, the
individual, product-specific differences between normal value and export price
may be positive or negative. If positive, they represent the aggregate amount of
dumping duties that the importing country is permitted to collect for that
product or group of transactions. If negative, they represent the amount by
which the export price exceeded the normal value. However, the United States
asserts that the AD Agreement imposes no requirement on the importing country to
make payments with respect to a lack of dumping of the merchandise in question.
The negative difference between normal value and export price simply means there
is no dumping;, i.e., the dumping margin for that product or group of
transactions is zero. Thus, for such products with no dumping margins, the
amount of dumping duties which the importing country is permitted to collect is
properly considered to be zero. The United States argues than when the
investigating authority calculates an overall, average rate of dumping, no
provision of the AD Agreement requires that more credit be given for "negative
dumping" amounts than if the dumping duties were to be collected on a
product-specific basis, which it asserts would be the result if India�s
interpretation of Article 2.4.2 were accepted. The United States also argues
that India�s reading of Article 2.4.2 would fail to give meaning to the
requirements of Article 2.4, which contemplate that comparisons be made at least
on a product-specific basis in order to account for physical and other
differences which affect price comparability. Finally, the United States notes
that Article 2.4.2 was introduced to the AD Agreement during the Uruguay Round
to address the concern of certain Members with the practice of some Members,
including the European Communities and the United States, of comparing
individual export price transactions to weighted-average normal values.
Article 2.4.2 was included in the Agreement to provide that, except in the case
of targeted dumping, margin calculations in an investigation would be made on a
consistent basis, i.e., weight-average to weight-average or transaction
to transaction.47 Thus, the United States asserts, the intent was to eliminate
transaction-to-average comparisons in investigations, not to alter the manner in
which authorities calculated overall margins after all appropriate comparisons
were made.
(b) Findings
6.110 Article 2.4.2 of the AD Agreement provides:
�Subject to the provisions governing fair comparison in
[Article 2.4], the existence of margins of dumping during the investigation
phase shall normally be established on the basis of a comparison of a
weighted average normal value with a weighted average of prices of all
comparable export transactions or by a comparison of normal value and export
prices on a transaction‑to‑transaction basis. A normal value established on
a weighted average basis may be compared to prices of individual export
transactions if the authorities find a pattern of export prices which differ
significantly among different purchasers, regions or time periods, and if an
explanation is provided as to why such differences cannot be taken into
account appropriately by the use of a weighted average‑to‑weighted average
or transaction‑to‑transaction comparison.�
6.111 As background, we note that Article 2.4 of the
AD Agreement (which is not the subject of a claim by India in this dispute)
provides, in pertinent part:
"A fair comparison shall be made between the export price
and the normal value. This comparison shall be made at the same level of
trade, normally at the ex‑factory level, and in respect of sales made at as
nearly as possible the same time. Due allowance shall be made in each case,
on its merits, for differences which affect price comparability, including
differences in conditions and terms of sale, taxation, levels of trade,
quantities, physical characteristics, and any other differences which are
also demonstrated to affect price comparability.7. . . ".
___________
7 It is understood that some of the above factors may
overlap, and authorities shall ensure that they do not duplicate adjustments
that have been already made under this provision.
The two subparagraphs of Article 2.4 deal with specific
aspects of the comparison of normal value and export price. Article 2.4.1 (which
is not at issue in this dispute) provides rules for the conversion of
currencies, when such conversion is necessary for the purposes of a comparison
under Article 2.4. Article 2.4.2 establishes that, subject to the provisions of
Article 2.4 governing fair comparison, dumping margins should normally be
established on the basis of an average-to-average comparison or a
transaction-to-transaction comparison. These provisions are new to the
AD Agreement - the Tokyo Round AD Code contained no similar provisions.
6.112 In accordance with the provisions of the Vienna
Convention governing treaty interpretation, we look first to the ordinary
meaning of the phrase "a comparison of a weighted average normal value with a
weighted average of prices of all comparable export transactions", in its
context and in light of its object and purpose, in determining whether the
practice of zeroing is permitted under Article 2.4.2. Looking first at the text,
we note that Article 2.4.2 requires that normally, except in circumstances not
applicable here, the existence of "margins of dumping" is to be established on
the basis of "a comparison of a weighted average normal value with a weighted
average of prices of all comparable export transactions" or on the basis of
comparison of individual transactions.
6.113 The European Communities argues that this provision
simply does not address the question of what to do with "multiple" margins
determined on the basis of comparisons for different models within the like
product. This "subsequent stage" of the calculation simply does not fall within
the scope of Article 2.4.2 in the European Communities' view, and therefore the
methodology to be applied in arriving at the dumping margin for the like product
as a whole, in a case where multiple comparisons are made, is within the
discretion of the Member conducting the investigation.
6.114 We cannot agree. The language of Article 2.4.2
specifically establishes the permissible bases for establishing the "existence
of margins of dumping". "Dumping" is defined in Article 2.1 of the
AD Agreement, which states that
"For the purpose of this Agreement, a product is to be
considered as being dumped, i.e., introduced into the commerce of
another country at less than its normal value, if the export price of the
product exported from one country to another is less than the comparable
price in the ordinary course of trade, for the like product when destined
for consumption in the exporting country".
The succeeding provisions of Article 2 of the AD Agreement,
which is entitled "Determination of Dumping" set forth, in some detail, various
information and methodologies to be used in the determination of whether dumping
exists. Article 2.4.2 sets out the permissible bases for comparison of normal
value and export price in order to establish the existence of margins of
dumping. In light of Article 2.1 of the AD Agreement, we consider that the
"margins of dumping" established under Article 2.4.2, based on the comparison
methodologies set forth, must relate to the ultimate question being addressed:
whether the product at issue is being dumped. Thus, in our view, a margin of
dumping, that is, a determination that there is dumping, can only be established
for the product at issue, and not for individual transactions concerning that
product, or discrete models of that product.
6.115 We note also that Article 2.4.2 specifies that the
weighted average normal value shall be compared with " a weighted average of
prices of all comparable export transactions". In this case, the European
Communities' calculation of the final weighted average dumping margin for the
product did not, in fact, rest on a comparison with the prices of all comparable
export transactions. By counting as zero the results of comparisons showing a
"negative" margin, the European Communities, in effect, changed the prices of
the export transactions in those comparisons. It is, in our view, impermissible
to "zero" such "negative" margins in establishing the existence of dumping for
the product under investigation, since this has the effect of changing the
results of an otherwise proper comparison. This effect arises because the
zeroing effectively counts the weighted average export price to be equal to the
weighted average normal value for those models for which "negative" margins were
found in the comparison, despite the fact that it was, in reality, higher than
the weighted average normal value. This is the equivalent of manipulating the
individual export prices counted in calculating the weighted average, in order
to arrive at a weighted average equal to the weighted average normal value. As a
result, we consider that an overall dumping margin calculated on the basis of
zeroing "negative" margins determined for some models is not based on
comparisons which fully reflect all comparable export prices, and is therefore
calculated inconsistently with the requirements of Article 2.4.2.
6.116 We recognize that Article 2.4.2 does not, in so many
words, prohibit "zeroing". However, this does not mean that the practice is
permitted, if it produces results inconsistent with the obligations set forth in
that Article, as we believe it does. We consider that the requirements of
Article 2.4.2 must be understood to apply to the entire process of determining
the existence of margins of dumping for the product, and that there is no
"subsequent stage" which escapes entirely from the strictures of Article 2.4.2.
6.117 We do not mean to suggest that anything in
Article 2.4.2 prohibits an investigating authority from undertaking multiple
comparisons of weighted average normal value and a weighted average of prices of
all comparable export transactions. To the contrary, we agree with the European
Communities and India that Article 2.4.2 allows investigating authorities to
make multiple comparisons on a model basis within the like product being
investigated, as the European Communities did in this case. In this regard, we
note the word comparable in Article 2.4.2. Read in light of the obligation in
the Article 2.4 to make a fair comparison, the specific requirements to make
comparisons at the same level of trade and at as nearly as possible at the same
time, and the obligation to make due allowance for differences affecting price
comparability, the use of the word comparable in Article 2.4.2 indicates to us
that investigating authorities may insure comparability either by making
necessary adjustments under Article 2.4, or by making comparisons for models
which are, themselves, comparable. However, in arriving at a conclusion whether
the product as a whole is being dumped, we consider that Article 2.4.2 obligates
an investigating authority to make its determination in a way which fully
accounts for the export prices on all comparable transactions. The
European Communities' methodology, which focuses on those models which are, in
its view, dumped, and takes less than full account of those models where the
comparison results in a negative margin, does not accomplish this goal.
6.118 We note that the European Communities argues that
Article 2.4.2 refers to the establishment of �the existence of margins of
dumping� in the plural, asserting that it is thus clear that the process of
comparing weighted averages will normally conclude with more than one dumping
margin. As discussed above, however, we consider that a dumping margin is
established for the product under investigation, and not for individual
models being compared as the basis of the establishment of the dumping
margin. Thus, in our view, the fact that Article 2.4.2 refers to the existence
of margins of dumping in the plural is a general statement, taking account of
the fact that, as is made clear in Articles 6.10 and 9 of the AD Agreement,
individual dumping margins are determined for each producer or exporter under
investigation, and for each product under investigation. While the comparisons
required under Article 2.4.2 yield margins of price difference, these are not,
properly speaking, margins of dumping to the extent that they relate to discrete
models of or transactions concerning the product under investigation, rather
than the product under investigation as a whole.
6.119 Based on the foregoing, we conclude that the European
Communities acted inconsistently with Article 2.4.2 of the AD Agreement in
establishing the existence of margins of dumping on the basis of a methodology
which included zeroing negative price differences calculated for some models of
bed linen.
D. CLAIMS UNDER ARTICLE 3
6.120 India's claims as to injury and causation raise
multiple issues relating to the interpretation of several provisions of
Article 3 � in particular, Articles 3.1, 3.4 and 3.5.48 Although India has broken
the issues down into multiple discrete claims and arguments, there are
essentially three questions under Article 3 before us: 1) whether the European
Communities violated its obligations under Articles 3.1, 3.4, and 3.5 by
considering all imports from India (and Egypt and Pakistan) to be dumped in
carrying out its analysis of injury caused by dumped imports (India's claims
numbers 8, 19, and 20), 2) whether the European Communities violated its
obligations under Article 3.4 by failing to evaluate "all relevant economic
factors and indices having a bearing on the state of the industry", (India's
claim number 11) and 3) whether the European Communities violated its
obligations under Article 3.4 by considering information for various groups of
EC producers in its analysis of the impact of dumped imports on the domestic
industry (India's claim number 15).
1. Claims under Articles 3.1, 3.4, and 3.5 -
consideration of all imports from India (and Egypt and Pakistan) as dumped
in the analysis of injury caused by dumped imports (claims numbers 8, 19,
and 20)
(a) Parties' arguments
6.121 India asserts that the European Communities assumed,
for the purposes of the injury determination, that all imports of the product
concerned during the investigation period (1 July 1995-30 June 1996) were
dumped. In addition, India asserts that the European Communities assumed that
imports during the entire period of the injury investigation (1 January 1992-30
June 1996), as well as imports prior to that period, were dumped. India asserts
that, with respect to the first assumption, much of the bed linen exported from
India during the investigation period was not, or should have been found not to
be, dumped, and that with respect to the second assumption, there was no
investigation covering those periods on the basis of which a finding of dumping
could have been made, and thus imports before the period of the dumping
investigation clearly can not be assumed to be dumped. In India's view,
inclusion of non-dumped imports in the analysis of injury and causation is
inconsistent with Articles 3.1, 3.2, 3.4, 3.5 and 3.6.49
6.122 India notes that the European Communities cumulated the
volume of all imports from the three countries under investigation � Egypt,
India and Pakistan � and not just the volume of imports that was the subject of
dumped transactions. (Exhibits India-23 and India-52). India further asserts
that if the European Communities had not zeroed, the imports from one company
would have been found not to be dumped at all. This company accounted for 28.5%
of the volume of Indian bed linen exports by sampled companies. Thus, India
maintains that it is clear that the total amount of non-dumped imports accounted
for more than one-third of India�s exports. India asserts that, assuming the
percentages of non-dumped imports from Egypt and Pakistan are of a similar order
of magnitude, this indicates that the total market share of dumped imports was
overstated by more than a fifth. With respect to the imports during years prior
to the dumping investigation period, India maintains that, as no finding of
dumping was ever made for any imports during this period, it was incorrect for
the European Communities to consider imports of bed linen from India in the
years preceding the dumping investigation period as dumped.
6.123 India also maintains, with reference to Article 3.5 of
the AD Agreement, that the European Communities failed to determine to what
extent injuries caused by other factors (such as, for instance, contraction in
demand or changes in consumption patterns) were responsible for the injury
allegedly suffered by the domestic industry. Consequently, India argues that the
establishment of the facts considered under Article 3.5 was not proper and/or
the evaluation of those facts was not unbiased and objective. In particular,
India argues that the term �dumped imports� in Article 3.5 has the same meaning
as in Article 3.4. Consequently, in India's view, the European Communities acted
inconsistently with Article 3.5 by automatically considering all imports of bed
linen from India between 1992 and 30 June 1995 as dumped.
6.124 The European Communities considers that the term
�dumped imports� as used in Article 3 of the AD Agreement includes all the
imports of the product in question from the country that is found to be dumping,
as opposed to only those transactions that are dumped, as suggested by India.
For the European Communities, the interpretation of the term "dumped imports"
proposed by India raises doubts because of its uncertainty. If each transaction
is to be allocated to a dumped or non-dumped classification, there is no
provision to cope with the situation where an exporter conceals the volume of
dumping by varying the prices from one consignment to another, perhaps in
collusion with the importer. There would need to be sub-categorisation by
exporter in that case.
6.125 The European Communities cites Articles 2.1, 3.1 and
5.7of the AD Agreement as contextual support for its view that dumping and
injury-causation issues are to be analysed on a product and country, rather than
transaction, basis. Article 2.1, in the European Communities' view, makes clear
that the existence of dumping is to be determined for a country at the
level of the product under investigation, referred to as the �like product�.
While Article 2 allows, or may even require, that the product under
investigation from a country be divided up by exporter and type in calculating
the margin of dumping, the determination of dumping is still made for the
product under investigation and the country. Further, the European Communities
argues, Article 3.1 requires that a determination of injury caused by dumped
imports has to be made for the domestic market for, and the domestic producers
of, the like product. In the European Communities' view, it is not possible to
isolate the effects of individual transactions in a single product market, and
the market situation is determined by the overall impact of imports. The
European Communities also maintains that Article 5.7 requires that �[t]he
evidence of both dumping and injury shall be considered simultaneously (a) in
the decision whether or not to initiate an investigation, and (b) thereafter,
during the course of the investigation, starting on a date not later than the
earliest date on which in accordance with the provisions of this Agreement
provisional measures may be applied�. Since injury has to be investigated before
it is established which transactions are dumped, it is clear, to the European
Communities, that the term �dumped products� used in connection with the injury
provisions of Article 3 must refer to all imports of the product under
investigation (although a finding of injury is of course conditional upon
dumping being found).
6.126 Finally, the European Communities indicates that a
consideration of the object and purpose of Article 3 supports its interpretation
of the term �dumped imports�. The European Communities states that the
unlikelihood of Article 3 pursuing its object and purpose with the needlessly
complex notion of �dumped imports� forwarded by India is reinforced by an
examination of the first sentence of Article 3.2, which requires consideration
whether there have been significant increases in dumped imports. In the European
Communities' view, the AD Agreement evidently intends national authorities to
gather information covering a lengthy period, since the investigation period
used to assess dumping (typically a year) would hardly be enough to assess
trends in the volume of imports. Article 3.2 is manifestly for the benefit of
exporters, according to the European Communities, because it sets conditions
that must be satisfied before causation is established. Nevertheless, on India�s
interpretation, in order to apply this provision, the exporters would have to
provide not just one, but several, years� price data in order to establish
whether dumping was occurring throughout the longer period for which import
volumes are considered. Far from benefiting exporters, such an interpretation
would in many cases make this provision unworkable.
6.127 The European Communities rejects India�s assertion that
it assumed imports prior to the dumping investigation period to be dumped and
found injury caused by those imports. The European Communities maintains that
there is nothing in either the EC Regulation, or any other statement by EC
authorities that expressly or implicitly supports the view that it reached such
a conclusion. The European Communities maintains that imports in years preceding
the dumping investigation period were examined in order to put the situation
during that period into context. The phrase �injury investigation period� is
used by the European Communities to refer to the longer period over which the
condition of the industry is evaluated, but this does not imply dumping during
that period.
6.128 In response to the Indian contention that the European
Communities �at several instances puts great emphasis on companies allegedly
disappeared from the EC market in the period 1992-POI�, the European Communities
draws attention to the statement in the Regulations that the principal basis for
the finding of material injury was the reduced profitability and price
suppression of the Community industry as observed among the sampled companies.
The information on the contraction in the number of producers showed that what
might otherwise have seemed a contradiction was in fact a realistic scenario.
Otherwise put, the EC authorities found injury principally because of the
domestic industry�s reduced profitability and price suppression, and the data of
the disappeared companies was relevant to explaining the improved position of
the industry with regard to sales and market share.
6.129 Egypt submits that by failing to separate out dumped
exports and those that were not dumped, the European Communities acted contrary
to Articles 3.1, 3.2, 3.4, 3.5 and 3.6 of the AD Agreement. Egypt focuses on the
use of the words �dumped imports� in Article 3.1. Moreover, Egypt argues, the
European Communities failed to properly consider whether other factors could
have caused the injury, as required by Article 3.5 of the AD Agreement.
6.130 Japan believes that the language �dumped imports� in
various places in Article 3 means that the injury determination set forth in
Article 3 must reflect the authorities� assessment of only �dumped imports�, and
not imports that were not found to have been �dumped�. In Japan's view, if the
authorities find that some imports were "dumped" and others were not, then they
must distinguish between the two in making their injury determination. Japan
asserts that this obligation extends to imports that are cumulated, as the
investigating authority is obliged to make the injury determination only for the
�dumped� portion of the cumulated imports. Allowing �dumped� imports to taint
all imports from a company seriously skews the injury analysis required under
Article 3.
6.131 The United States disagrees with the European
Communities' reasoning that dumping is determined for countries, and therefore
that it is entitled to consider all imports from a country found to be dumping
as dumped imports for purposes of the injury investigation. However, the United
States submits that, even assuming the European Communities did treat all
subject imports during the injury assessment period as dumped, that treatment
would have been consistent with the AD Agreement. The reasons supporting the
European Communities� view that it acted consistently with the Agreement in
treating all subject imports as dumped during the period of investigation
similarly apply with respect to the treatment of subject imports during the
portion of the injury assessment period that was prior to the dumping
investigation period.50 In the United States' view, the requirements of the injury
determination necessarily oblige Members to gather and consider information for
a period longer than the period of the dumping investigation, in order to
evaluate volume and price changes. This disparity of time periods has been an
element of dumping investigations since long before the current AD Agreement was
negotiated and came into effect. However, there is, in the United States' view,
no reasonable way to eliminate this disparity, as it would not be meaningful to
assess injury only over the period of the dumping investigation, and it would be
unduly burdensome to investigating authorities and exporters to extend the
period of the dumping investigation. The United States points out that in the
Salmon cases51 the United States had considered all imports during the injury
investigation period to be dumped (and subsidised). The Panels reviewing that
injury determination concluded that the United States had properly considered
whether there had been a significant increase in the volume of dumped (and
subsidised) imports under the Tokyo Round Anti-Dumping and Subsidies Codes. The
relevant text of the AD Agreement is the same as that of the Tokyo Round
Anti-Dumping Code.
(b) Findings
6.132 India�s claim regarding the European Communities�
treatment of all imports of the product concerned during the investigation
period as dumped is primarily characterised as a claim of inconsistency with
Article 3.1.
6.133 Article 3.1 states:
�A determination of injury for purposes of Article VI
of GATT 1994 shall be based on positive evidence and involve an
objective examination of both (a) the volume of the dumped
imports and the effect of the dumped imports on prices in the domestic
market for like products, and (b) the consequent impact of these
imports on domestic producers of such products.�
Article 3.1, which requires consideration of the volume,
price, and consequent impact of dumped imports on the domestic industry sets out
the general requirements for a determination of injury, and the succeeding
sections of Article 3 provide more specific guidance for such determinations.
Articles 3.4 and 3.5 similarly require consideration of dumped imports.
6.134 There is no dispute between the parties as to the facts
with respect to the consideration of imports as dumped during the period of the
dumping investigation. The European Communities explicitly acknowledges that it
considered all imports from the three countries investigated, India, Egypt and
Pakistan, as dumped, and considered the volume and price effects of all imports
from those countries during that period in evaluating whether injury was caused
by dumped imports. India asserts that the European Communities was only entitled
to consider as dumped imports in its injury analysis those imports attributable
to specific transactions as to which dumping was actually found during the
period of the dumping investigation.
6.135 Thus, we are faced with the question of the
interpretation of the term �dumped imports� in Articles 3.1, 3.4, and 3.5 of the
AD Agreement, rather than an assessment of the facts as such. If we were to
conclude that the term �dumped imports� may be understood to comprise the volume
of imports of the product in question from the country for which an affirmative
determination of dumping has been made then we must, under the standard of
review set forth in Article 17.6(ii), find in favor of the European Communities
on this issue, at least with respect to the consideration of imports during the
period of the dumping investigation. On the other hand, to sustain India's
position, we would have to conclude that the phrase "dumped imports" must be
understood to refer only to imports which are the subject of transactions in
which export price was below normal value, which India considers to be "dumping"
transactions.
6.136 However, consideration of the ordinary meaning of the
phrase "dumped imports" in its context, and in light of the object and purpose
of Article 3 of the AD Agreement, leads us to the conclusion that the
interpretation proposed by India is not required. As discussed above52, we
consider that dumping is a determination made with reference to a product
from a particular producer/exporter, and not with reference to individual
transactions. That is, the determination of dumping is made on the basis of
consideration of transactions involving a particular product from particular
producers/exporters. If the result of that consideration is a conclusion that
the product in question from particular producers/exporters is dumped, we are of
the view that the conclusion applies to all imports of that product from such
source(s), at least over the period for which dumping was considered. Thus, we
consider that the investigating authority is entitled to consider all such
imports in its analysis of "dumped imports" under Articles 3.1, 3.4, and 3.5 of
the AD Agreement.
6.137 We note that Article 9.2 of the AD Agreement, which may
be considered relevant context for our analysis, provides:
"When an anti‑dumping duty is imposed in respect of any
product, such anti‑dumping duty shall be collected in the appropriate
amounts in each case, on a non‑discriminatory basis on imports of such
product from all sources found to be dumped and causing injury, except as to
imports from those sources from which price undertakings under the terms of
this Agreement have been accepted. The authorities shall name the supplier
or suppliers of the product concerned".
We consider that this provision lends support to our
conclusion that all imports from any producer/exporter found to be dumping may
be considered as dumped imports for purposes of injury analysis.
6.138 In this regard, we note that, although the European
Communities found de minimis margins for four Pakistani exporters of bed
linen, it did not make a negative determination of dumping with respect to any
producer or exporter subject to the investigation. India, of course, has made no
claim with respect to the treatment of Pakistani imports as dumped. India does
argue that, had the European Communities properly calculated the dumping margins
for Indian producers, it would have come to the conclusion that imports from one
company were not dumped. We have found above that the European Communities did
act inconsistently with its obligations under Article 2.4.2 of the AD Agreement
in its calculation of dumping margins for Indian producers. It is possible that
a calculation conducted consistently with the AD Agreement would lead to the
conclusion that one or another Indian producer should be attributed a zero or
de minimis margin of dumping. In such a case, it is our view that the
imports attributable to such a producer/exporter may not be considered as
"dumped" for purposes of injury analysis. However, we lack legal competence to
make a proper calculation and consequent determination of dumping for any of the
Indian producers � our task is to review the determination of the EC
authorities, not to replace that determination, where found to be inconsistent
with the AD Agreement, with our own determination. In any event, we lack the
necessary data to undertake such a calculation. Thus, while the treatment of
imports attributable to producers or exporters found to not be dumping is
an interesting question, it is not an issue before us and we reach no
conclusions in this regard.
6.139 Our conclusion that investigating authorities may treat
all imports from producers/exporters for which an affirmative determination of
dumping is made as "dumped imports" for purposes of injury analysis under
Article 3 is bolstered by our view that the interpretation proposed by India,
which entails the conclusion that the phrase "dumped imports" refers only
to those imports attributable to transactions in which export price is
below normal value, would lead to an unworkable result in certain cases. One of
the objects and purposes of the AD Agreement is to establish the conditions
under which Members may impose anti-dumping duties in cases of injurious
dumping. An interpretation which would, in many cases, make it impossible to
assess one of the necessary elements, injury, is not consistent with that object
and purpose.
6.140 An assessment of the volume, price effects, and
consequent impact, only of imports attributable to transactions for which a
positive margin was calculated would be, in many cases, impossible, or at least
impracticable. Attempting to segregate individual transactions as to whether
they were "dumped" or not, even assuming it could be done, would leave
investigating authorities in a quandary in cases in which the dumping
investigation is undertaken for a sample of companies or products. Such sampling
is specifically provided for in the AD Agreement, yet it would not be possible,
in such cases, accurately to determine the volume of imports attributable to
"dumped" transactions.53 Similarly, if dumping is determined on the basis of a
comparison of weighted average normal value to weighted average export price,
there would be no comparisons concerning individual transactions which could
serve as the basis for segregating imports in "dumped" and "not-dumped"
categories.
6.141 We note, in this context, the findings of the GATT
Panels in the Salmon cases. While the specific issue raised here was
neither raised nor addressed in that case, the Panel in Salmon - Anti-Dumping
Duties was considering the question of whether the United States had
properly determined that the imports caused material injury to the domestic
industry "through the effects of dumping". The Panel found that this language,
which is found in Article 3.5 of the AD Agreement, did not require that the
volume, price, and impact "effects" to be considered be those of the dumping,
but rather those of the dumped imports, that is, the "effects of the dumping"
were equated by the Panel with "the effects of the dumped imports".54 In that
case, the "dumped imports" included all imports from all producers in the
country without distinction by transactions. In our view, this conclusion is
consistent with an interpretation of the phrase "dumped imports" as referring to
all imports of the product from producers/exporters as to which an
affirmative determination of dumping has been made.
6.142 We therefore conclude that the European Communities,
having made an affirmative determination of dumping with respect to imports from
all producers/exporters in this case, did not act inconsistently with
Articles 3.1, 3.4, and 3.5 of the AD Agreement by considering all imports from
India (and Egypt and Pakistan) in its evaluation of the volume, price effects,
and consequent impact of dumped imports.
6.143 With respect to the question whether the European
Communities improperly considered imports before the dumping investigation
period as dumped, we note the European Communities' explanation that it did not
determine injury caused by dumped imports for any period before the dumping
investigation period. Since we have concluded, as discussed below, that the
European Communities' determination of injury was not made consistently with its
obligations under Article 3.4, we do not consider it necessary or appropriate to
decide this question.
6.144 Finally, with respect to India's claim that the
European Communities failed to properly consider "other factors" which might
have been causing injury to the domestic industry, as required by Article 3.5 of
the AD Agreement, we note that, with the exception of the argument concerning
improper consideration of "dumped" imports, India has made no other arguments in
support of this claim. Having rejected India's position in that regard, we
consider that India has failed to present a prima facie case in this
regard.
2. Claim under Article 3.4 - failure to evaluate "all
relevant economic factors and indices having a bearing on the state of the
industry" (claim number 11)
(a) Parties' arguments
6.145 India considers that the European Communities failed to
consider all injury factors mentioned in Article 3.4 of the AD Agreement for the
purpose of its determination of the impact of the dumped imports on the domestic
industry concerned. In particular, India asserts that the European Communities
did not consider the following elements: productivity; return on investments;
utilisation of capacity; magnitude of margin of dumping; cash flow; inventories;
wages; growth; and ability to raise capital or investments. In India�s opinion,
the European Communities, therefore, acted inconsistently with Article 3.4.
6.146 India emphasises the use of the word "shall" in
Article 3.4, arguing that it follows that the evaluation mentioned in
Article 3.4 shall by necessity include "all relevant . . . factors". The word
"all" indicates that all relevant factors must be included in this "evaluation".
The word "all", according to India, is given further meaning by the word
"including". In India's view, it follows from the word "including" that, at a
minimum, the factors and indices listed after the word "including" must be
evaluated.
6.147 The European Communities presents three defences to
India�s claim, which it characterises as relying on the supposedly compulsory
nature of the evaluation of the factors listed in Article 3.4 and not on the
argument that, because of the circumstances of this particular case, the listed
factors should be evaluated. First, the European Communities asserts that the
factors listed in Article 3.4 were evaluated during the investigation.55 For
several of the factors, the data could be derived from the exporters� accounts
and, for others, from the questionnaire sent to the domestic producers. The
European Communities also indicates the evaluation accorded to each of the
factors.
6.148 Second, the European Communities asserts that the
factors listed in Article 3.4 are negative in character and, as such, were
properly evaluated during the investigation. The European Communities stresses
that the one feature that stands out in a close examination of the terms of
Article 3.4 is that the listed factors are explicitly concerned with indications
of injury, not the absence of injury. Of fifteen factors, only two are not
qualified by the words �decline� or �negative effects�. The opening clause of
Article 3.4 � which speaks of the �impact of the dumped imports� �
reinforces this interpretation of the listed factors. The purpose of the
examination under Article 3.4 is to determine what is wrong with the domestic
industry, not what is right with it. The European Communities does not wish to
suggest that non-negative factors have no relevance. However, in the view of the
European Communities, India seeks to establish that Article 3.4 requires
investigating authorities to evaluate in an explicit fashion all the fifteen
listed factors. However, the wording of Article 3.4 refers almost exclusively to
negative factors and, consequently, according to the European Communities, what
might be called the �comprehensive evaluation� requirement, if it exists,
applies only to such factors. Profits and prices were the two principal negative
factors identified by the EC authorities in the bed linen proceeding, and these
were thoroughly examined and evaluated.
6.149 Third, the European Communities puts forward various
reasons for concluding that Article 3.4 does not require that every one of the
listed factors need be evaluated in every investigation. The European
Communities points to the use of the words �relevant� and �have a bearing on the
state of the industry� in Article 3.4 as well as the last sentence of the
provision, which states: �This list is not exhaustive, nor can one or several of
these factors necessarily give decisive guidance.� The European Communities also
underlines the use of the word �including� and emphasises what it calls the
�nature� of the list, explaining that it is �broken into parts by semicolons,
and the word �or� is used to indicate that not all of the factors need be
considered�. Not only do the factors differ in importance from case to case,
but, for the European Communities, it is possible to deduce that certain of them
are inherently likely to be more significant than others and that findings on
some may make findings on others superfluous.
6.150 The European Communities argues that the obligation in
Article 3.4 to consider injury factors does not exist in isolation and, in
particular, account must be taken of Articles 6.13 and 6.14, which deal with the
difficulties experienced by interested parties � particularly small companies �
in supplying information requested and the need for a Member to proceed
expeditiously in its investigation, respectively. In this context, aspects of
the evaluation required by Article 3.4 may have to be limited in order to
observe the spirit of Article 6.13. The European Communities posits that the
decision on the limits to be set on the obligation in Article 3.4 is a matter of
judgement that must be exercised by the investigating authorities.
6.151 Japan, as third party, asserts that the language of
Article 3.4 requires all listed factors to be considered, and the list of
factors is the minimum that must be evaluated by the investigating authorities.
The degree of importance of each factor may vary from case to case, but all of
the listed factors must be fully considered and evaluated in each case.
Authorities may not exclude certain factors because they deem these to be
irrelevant. Japan is of the opinion that this interpretation finds support in
the change in the language of this provision over time. The change from the
phrase �such as� in the comparable provision of the Tokyo Round Anti-Dumping
Code to the word �including� in the AD Agreement underscores the interpretative
significance of the word �including�. Because �including� means �part of a
whole�, the factors after the word �including� must be viewed as a subset of a
potentially larger group of factors that must be evaluated by the authorities.
Had the drafters intended this list of factors to be a discretionary checklist
from which authorities may pick and choose, they would not have changed the
words �such as� to �including�. The drafters would also have used language to
more clearly provide that authorities could consider as many or as few of these
factors as they wished.
6.152 The United States, as third party, takes the position
that while, in light of Article 12.2, investigating authorities are not required
in each case to make a specific finding on each enumerated factor in
Articles 3.2 and 3.4, it should be discernible from the authorities�
determination that they evaluated each of the enumerated factors. This objective
may be achieved when a determination, through its demonstration of why the
authorities relied on the specific factors they found to be material in the
case, thereby discloses why other factors on which they do not make specific
findings were accorded little weight. In the current case, however, the United
States shares some of India�s concerns about the inadequacy of the European
Communities� findings, because the European Communities� specific findings on
the factors it addressed do not elucidate why it did not give weight to factors
it did not discuss. The United States does not agree with the European
Communities� argument that some of the Article 3.4 factors are negative in
character. The United States points out that the European Communities ignores
that Article 3.5 refers to �the effects of dumping, as set forth in
paragraphs 2 and 4 of Article 3�. The �relevance� of the Article 3.4 factors
extends beyond supporting an injury determination. Article 3.4 states that �all
relevant economic factors and indices having a bearing on the state of the
industry� must be evaluated. Thus, even if a factor does not lend support to an
affirmative injury determination, the authority must evaluate it so long as it
sheds light on the condition of the domestic industry.
(b) Findings
6.153 India's claim raises a number of issues. The most basic
of these is the interpretation of Article 3.4, i.e., whether the list of
factors set out in that provision is illustrative or mandatory and, if
mandatory, whether there are only four groups of �factors� represented by the
subgroups separated by semicolons that must be evaluated, or whether each
individual factor listed must be considered. We must also consider the nature of
the evaluation of the factors that is required, how the �relevance� of a given
factor is to be determined, and the extent to which the final determination must
reflect the required consideration, whatever its nature. Finally, we must
consider the facts.
6.154 Article 3.4 provides:
�The examination of the impact of the dumped imports on
the domestic industry concerned shall include an evaluation of all relevant
economic factors and indices having a bearing on the state of the industry,
including actual and potential decline in sales, profits, output, market
share, productivity, return on investments, or utilisation of capacity;
factors affecting domestic prices; the magnitude of the margin of dumping;
actual and potential negative effects on cash flow, inventories, employment,
wages, growth, ability to raise capital or investments. This list is not
exhaustive, nor can one or several of these factors necessarily give
decisive guidance.�
The use of the phrase "shall include" in Article 3.4 strongly
suggests to us that the evaluation of the listed factors in that provision is
properly interpreted as mandatory in all cases. That is, in our view, the
ordinary meaning of the provision is that the examination of the impact of
dumped imports must include an evaluation of all the listed factors in
Article 3.4.
6.155 The European Communities emphasises the use of the
terms �relevant�, �having a bearing on the state of the industry� and
�including� in Article 3.4 in arguing that not all factors need be evaluated in
all cases. We do not consider that these textual elements affect the conclusion
that the ordinary meaning of Article 3.4 is properly understood as requiring the
evaluation of all the listed factors in all cases. We note that the terms
"relevant" and the phrase "having a bearing on the state of the industry"
precede the introduction of the list of factors. In our view, the text of
Article 3.4 indicates that the listed factors are a priori "relevant"
factors "having a bearing on the state of the industry", and therefore must be
evaluated in all cases.56
6.156 With regard to the use of the word "including", we
consider that this simply emphasises that there may be other "relevant economic
factors and indices having a bearing on the state of the industry" among "all"
such factors that must be evaluated. We recall that, in the Tokyo Round AD Code,
the same list of factors was preceded by the phrase �such as�, which was changed
to the word �including� that now appears in Article 3.4 of the AD Agreement. The
term �such as� is defined, inter alia, as �Of the kind, degree, category
being or about to be specified; for example�.57 By contrast, the verb �include� is
defined, inter alia, to mean �enclose�; �contain as part of a whole or as
a subordinate element; contain by implication, involve�; or �place in a class or
category; treat or regard as part of a whole�.58 We thus read the phrase �shall
include an evaluation of all relevant economic factors and indices having a
bearing on the state of the industry, including . . .� as introducing a
mandatory list of relevant economic factors which must be evaluated in every
case. The change in the wording that was introduced in the Uruguay Round in our
view supports an interpretation of the current text of Article 3.4 as setting
forth a list that is mandatory, and not merely indicative or illustrative.59
6.157 The European Communities also focuses on the semicolons
in the list of factors in Article 3.4. However, in our view, neither the
presence of semicolons separating certain groups of factors in the text of
Article 3.4, nor the presence of the word "or" within the first and fourth of
these groups, serves to render the mandatory list in Article 3.4 a list of only
four "factors". We further note that the two "ors" appear within � rather than
between � the groups of factors separated by semicolons. Thus, we consider that
the use of the term �or� here does not detract from the mandatory nature of the
textual requirement that �all relevant economic factors� shall be evaluated.
With respect to the second �or,� it appears in the phrase "ability to raise
capital or investments", which clearly indicates that the factor that an
investigating authority must examine is the "ability to raise capital" or the
"ability to raise investments", or both.
6.158 Finally, we consider the European Communities'
assertion that not all factors listed in Article 3.4, �being solely negative in
character�, need to be evaluated. This characterisation of the European
Communities of the factors listed in Article 3.4 is somewhat perplexing to us.
Each of the factors to be evaluated may be found to indicate material injury, or
not, to the industry. We fail to see the purpose of describing them as �negative
factors�, or factors having "negative character". Nor are we able to reconcile
the European Communities� statement that �the listed factors are explicitly
concerned with indications of injury, not the absence of injury� with its
statement that �[t]he European Communities does not wish to suggest that
non-negative factors have no relevance�. On the contrary, the European
Communities� comment that �[the] wording [of Article 3.4] refers almost
exclusively to negative factors and, consequently, what might be called the
�comprehensive evaluation� requirement, if it exists, applies only to such
factors� suggests that the European Communities believes that only factors
indicating material injury to the industry must be evaluated. Such an
interpretation of Article 3.4 clearly runs counter to the requirement to
properly establish a factual basis in support of a well-reasoned and meaningful
analysis of the state of the industry and a finding of injury as well as to the
requirement of an unbiased and objective evaluation as provided for in
Articles 3.1 and 17.6(i) of the AD Agreement.
6.159 Based on the foregoing, we conclude that each of
the fifteen factors listed in Article 3.4 of the AD Agreement must be
evaluated by the investigating authorities in each case in examining the impact
of the dumped imports on the domestic industry concerned.
6.160 We note that our conclusion is the same as that reached
by the Panel on Mexico - HFCS on this specific issue.60 The Panel stated61:
�The text of Article 3.4 is mandatory:
�The examination of the impact of the dumped imports on
the domestic industry concerned shall include an evaluation of all
relevant economic factors and indices having a bearing on the state of
the industry, including. . . � (emphasis added by HFCS panel)
In our view, this language makes it clear that the listed
factors in Article 3.4 must be considered in all cases. There may be other
relevant economic factors in the circumstances of a particular case,
consideration of which would also be required."
6.161 Turning to the question of the nature of the evaluation
of each factor that is required, we note the views of the Mexico-HFCS
panel on this question:62
"But consideration of the Article 3.4 factors is required
in every case, even though such consideration may lead the investigating
authority to conclude that a particular factor is not probative in the
circumstances of a particular industry or a particular case, and therefore
is not relevant to the actual determination. Moreover, the consideration of
each of the Article 3.4 factors must be apparent in the final determination
of the investigating authority.602
_________________________
602In this regard, we note the text of Article 12.2.2,
which provides:
�A public notice of conclusion or suspension of an
investigation in the case of an affirmative determination providing for the
imposition of a definitive duty or the acceptance of a price undertaking
shall contain, or otherwise make available through a separate report, all
relevant information on the matters of fact and law and reasons which have
led to the imposition of final measures . . .��
6.162 In other words, while the authorities may determine
that some factors are not relevant or do not weigh significantly in the
decision, the authorities may not simply disregard such factors, but must
explain their conclusion as to the lack of relevance or significance of such
factors. We agree. Thus, we are of the view that every factor in Article 3.4
must be considered, and that the nature of this consideration, including whether
the investigating authority considered the factor relevant in its analysis of
the impact of dumped imports on the domestic industry, must be apparent in the
final determination.
6.163 We now consider the implications of a ruling that
consideration of each factor must be apparent in the final determination, and
that the relevance or lack thereof of each factor must be explained. The
European Communities objects to the concept of a "checklist" and argues that the
relevance of some factors may be apparent early in the investigation. Even
granting that the investigating authority may be aware early in the proceeding,
or indeed from the outset, that a certain factor is not relevant to the
examination of the impact of dumped imports on the domestic industry, it must
nonetheless be possible for a panel, reviewing that determination, to be able to
assess whether all of the Article 3.4 factors have been evaluated. We consider
that an authority, in discussing why it found certain factors relevant, may at
the same time make apparent why it did not deem other factors to be material
and, thus, without following a checklist approach, make it possible for a
reviewing panel to determine whether it complied with the requirements of
Article 3.4. Thus, we conclude that, as long as the lack of relevance or
materiality of the factors not central to the decision is at least implicitly
apparent from the final determination, the Agreement's requirements are
satisfied. While a checklist would perhaps increase an authority�s and a panel's
confidence that all factors were considered, we believe that it is not a
required approach to decision-making under Article 3.4.
6.164 Having concluded that Article 3.4 requires an
evaluation of all listed factors, which might result in a conclusion that some
of the listed factors are not relevant to the examination of the impact of
dumped imports on the domestic industry in the particular circumstances of the
investigation at hand, and that the consideration of all the Article 3.4 factors
must be apparent from the final determination, the question before us is whether
the European Communities' determination is consistent with this obligation.
6.165 The European Communities submits that it evaluated the
factors listed in Article 3.4 during the investigation. In paragraphs 81-91 of
the Provisional Regulation,63 the European Communities, under the heading
"Situation of the Community industry", addresses the following factors: (a)
production; (b) sales by volume; (c) sales by value; (d) market share; (e) price
development; (f) profitability; and (g) employment. The other factors listed in
Article 3.4 � productivity; return on investments; utilisation of capacity; the
magnitude of the margin of dumping; cash flow; inventories; wages; growth;
ability to raise capital or investments � are not even referred to in this
section.
6.166 We further note that, in paragraph 62 of the
Provisional Regulation, the European Communities states:
"Data for the examination of injury caused to the
Community industry was collected and analysed at three different levels, as
follows:
- at the level of the entire Community (EU-15) for trends
concerning production, consumption in the Community, imports, exports and
market share . . .
- at the level of the Community industry . . . for trends
concerning production, sales by value and employment.
- at the level of the sampled Community producers, for
the factors mentioned above and also for trends concerning prices and
profitability."
6.167 It appears from this listing that data was not even
collected for all the factors listed in Article 3.4, let alone evaluated by the
EC investigating authorities. Surely a factor cannot be evaluated without the
collection of relevant data. While some of the data collected for the factors
that are mentioned in the Provisional Regulation by the EC authorities may have
included data for the factors not mentioned, we cannot be expected to assume
that this was the case without some indication to that effect in the
determination. Nor is the relevance or lack thereof, as assessed by the EC
authorities, of the factors not mentioned under the heading �Situation of the
Community industry� at all apparent from the determination.
6.168 The European Communities explains that certain factors
were evaluated, but were "found not to be a significant independent factor". In
response to a question from the Panel regarding the meaning of the phrase �not a
significant independent factor� as used by the European Communities, the
European Communities states: �The interpretation to be given to the phrase �an
evaluation of all relevant economic factors and indices� must be flexible enough
to cope with the enormous variety of circumstances that arise in investigations
into injury and injury causation. Relevance is a matter of degree rather than of
�yes or no�.�64 While we certainly agree with the European Communities as to the
"enormous variety of circumstances" that may arise in investigations, and the
need to be flexible in the evaluation of relevant factors, we fail to see how
relevance of a factor to the determination of injury in a particular
investigation can be a matter of degree. That is to say, it is clear that
not all factors will be, or will be equally, "relevant", in the sense of bearing
on the state of the industry, in all cases. Nonetheless, it would seem to us
that in a particular case, a particular factor either is or is not
relevant to the determination of whether there is injury, depending on the
particular facts and circumstances of the industry in question. Indeed, it is
precisely because, as the European Communities states, �the process of
determining the relevance of a factor may be little different from that of
evaluating it� that the authorities� assessment of the lack of relevance of a
factor, that is, the conclusion that it has no (or little) bearing on the
determination of injury, should that be the case, must be as apparent from the
determination as the authorities� evaluation of a factor that does bear on the
determination of injury. Otherwise, it becomes impossible to determine which of
the many factors that have a bearing on the state of the industry actually were
considered to weigh in the determination of injury and were evaluated by the
investigating authority. We find that, where factors set forth in Article 3.4
are not even referred to in the determination being reviewed, if there is
nothing in the determination to indicate that the authorities considered them
not to be relevant, the requirements of Article 3.4 were not satisfied. A
conclusion that a factor is not relevant which must be assumed from the absence
of any discussion of it is, in our view, simply not tenable.
6.169 Based on the foregoing, we conclude that the European
Communities did not conduct �an evaluation of all relevant economic factors and
indices having a bearing on the state of the industry� and, therefore, failed to
act consistently with its obligations under Article 3.4 of the AD Agreement.
3. Claim under Article 3.4 - consideration of information
for various groupings of EC producers in analysis of the state of the
domestic industry (claim number 15)
(a) Parties' arguments
6.170 India submits that the European Communities acted
inconsistently with Article 3.4 by considering information relating to different
groupings of EC producers of bed linen in evaluating certain of the factors
under Article 3.4. India asserts that the European Communities, after defining
the "Community industry" as a group of 35 producers, selected a sample of 17 of
those 35 for purposes of the injury investigation. However, India argues, the
European Communities did not consistently base the injury analysis on this
sample group. India argues that the European Communities' reliance on
information for companies outside this group, specifically by considering
information for the "Community industry" as a whole, and for all EC producers of
bed linen, in determining injury, was a violation of Article 3.4. Moreover,
India maintains, the European Communities' choice of which group of producers to
consider with respect to different aspects of its analysis was without any
apparent reason other than a goal-oriented �picking and choosing� in order to
find injury.
6.171 The European Communities argues that Article 4.1
provides Members with two options for defining the domestic industry, either the
"domestic producers as a whole" of the like product, or "those of them [the
domestic producers] whose collective output of the products constitutes a major
proportion of the total domestic production of those like products". In EC
practice, a "major proportion" is defined by reference to the standing
requirements of Article 5.4 of the Agreement, that is, producers accounting for
at least 25 per cent of domestic production. The European Communities states
that in the bed linen investigation, it applied the second option, defining as
the "Community industry" a group of 35 producers of bed linen supporting the
application whose collective output constituted more than 25 per cent of EC
production of the like product. Because of the number of companies in the
Community industry, the European Communities decided to resort to sampling. An
initial list of 19 companies was decided upon for inclusion in the sample, which
was subsequently reduced to 17 companies. The European Communities collected and
analysed data for the examination of injury to the Community industry at three
levels, i.e., for the sampled companies, for the Community industry, and
for all EC producers of bed linen.65
6.172 The European Communities notes that the conclusions
drawn from evidence must ultimately concern the domestic industry as defined in
the investigation, but argues that there is no intrinsic limit to the types of
evidence that may be used to arrive at such conclusions. In particular, the
European Communities submits that it cannot be excluded ab initio that
the condition of EC producers of bed linen as a whole may provide evidence of
the condition of those producers who comprise the domestic industry. The
European Communities emphasises that the principal basis for the finding of
material injury was the reduced profitability and price suppression of the
Community industry as observed among the sampled companies.
6.173 In Egypt's view, there is no textual support in the
AD Agreement for the approach adopted by the European Communities, which is
incompatible with Article 3.1, as well as Articles 3.3, 3.4, 3.5 and 3.6, of the
AD Agreement. These provisions require the European Communities to determine
whether the domestic industry has suffered injury and do not permit an injury
determination based on data relating to companies not belonging to the domestic
industry. As an ancillary matter, Egypt states that the Commission, by ignoring
the results of its own sample of the domestic industry, failed to make an
unbiased and objective assessment of the facts and thus acted inconsistently
with Article 6.10 in conjunction with Articles 3.4, 3.5 and 3.6. In Egypt's
view, in assessing whether the domestic industry suffered material injury, the
European Communities improperly considered both the Community industry and the
total EC production, drawing conclusions regarding the situation of the
Community industry on the basis of information concerning total EC production.
Egypt submits that the evidence shows that the domestic industry was in a
healthy state, and that in any event, the European Communities failed to take
into account in its analysis the fact that consumption of the like product in
the European Communities decreased over the relevant period.
6.174 In the United States' view, the parties' arguments on
this issue miss an important underlying point. The United States is of the
opinion that the European Communities, in applying its Regulation on definition
of the domestic industry, has defined the domestic industry in this case in a
manner which violates Article 4 of the AD Agreement, and that therefore the
entire injury analysis is based on a flawed premise. In the United States' view,
the European Communities' position that Article 4.1 allows two equally valid
options for defining the domestic industry - either producers as a whole, or
producers of a major proportion of domestic production, is wrong. The United
States believes that the second option is not a separate basis for defining
industry, but is a provision which allows a determination to be made in
situations where information for the industry as a whole is not available, so
long as that information relates to producers of a major proportion of domestic
production. The United States asserts that the European Communities� industry
definition limited the domestic industry to those producers that came forward to
affirmatively pursue the investigation, and thus was fundamentally skewed. A
proper definition of the domestic industry under Article 4.1 would have required
the European Communities to define the �domestic industry� as all EC producers
of the like product, and obtain information from that universe of producers, or
at least from a sample drawn from that universe of producers. Thus, with respect
to India�s claim that the European Communities acted impermissibly by
considering some information concerning all EC producers, the United States, in
contrast, believes that the European Communities acted inconsistently with the
AD Agreement by not including all EC producers of bed linen in the
domestic industry for the purposes of evaluating factors such as price and
impact under Articles 3.1, 3.2, 3.4 and 3.5. Moreover, the United States argues
that the European Communities' definition of domestic industry conflates the
�domestic industry� definition of Article 4.1 with the standing determination
under Article 5.4, and thus misconstrues the relationship between the two
provisions. If Article 4.1 were intended to define the domestic industry as
those producers who expressly supported the petition, an injury investigation
would be mostly a pro forma exercise in which the authorities would
simply check whether petitioning firms really were materially injured.
Article 5.4 does not provide a basis for the creation of such a self-selecting
industry and does not purport to define the term �a major proportion� as used in
Article 4.1. The United States adds that Article 3.1 reflects that Article 4.1
establishes a preference for basing an injury determination on examination of
the domestic producers as a whole. Further, Articles 3.4 and 3.5 specifically
direct that an injury analysis shall concern �the domestic industry�.
These provisions accordingly do not contemplate that an authority will at its
discretion use one industry definition in a determination examining injury and
another definition in that determination for other purposes.
(b) Findings
6.175 We first note that the issue raised by the United
States regarding the European Communities� interpretation of �domestic industry�
is an interesting one, and raises questions regarding the proper application of
that term in this case. However, India has made no claim under Article 4 of the
AD Agreement in this dispute regarding the European Communities' definition of
the domestic industry. Our analysis and finding relate only to the claim before
us, whether having defined the Community industry as a group of 35 producers and
resorted to a sample of those producers, the European Communities was precluded
from considering information relating to producers not within that sample, or
not within the Community industry. We express no opinion as to the correctness
vel non of the European Communities' interpretation of Article 4 of the
AD Agreement or its application in this case.
6.176 India's claim raises two related questions. First, we
must consider whether the fact that the European Communities considered
information for different groupings of producers of bed linen, the 17 producers
in the sample, the 35 producers comprising the "Community industry", and all EC
producers of bed linen, constitutes a violation of Article 3.4. The second
question is whether, assuming that the consideration of different data sets with
regard to analysis of the state of the domestic industry under Article 3.4 is
permissible, the European Communities explained how its consideration of the
information at the different levels supported its determination.
6.177 India's claim rests on premises concerning the correct
definition of domestic industry and sampling which are outside the scope of our
terms of reference. Focusing solely on the claim that is before us, we note that
the European Communities defined the domestic industry by starting with the list
of companies which supported the application. After eliminating seven found not
to be complainants66, and excluding several others for various reasons, the
European Communities arrived at a group of 35 companies whose production of bed
linen the European Communities considered to constitute a "major proportion" of
total EC production of the like product. The European Communities defined this
group as the "Community industry".67 The European Communities decided to establish
a sample of this Community industry, and in consultation with the complainant
Eurocoton, an initial list of 19 producers was arrived at, which was
subsequently reduced to 17 producers. These 17 companies represented 20.7 per
cent of total EC production of bed linen, and 61.6 per cent of the production of
the Community industry (i.e., the 35 producers referred to above). The EC
investigating authorities considered this sample to be representative of the
domestic industry.
6.178 As noted above, the European Communities collected
information concerning injury with respect to three groups of companies � all EC
producers of bed linen (referred to in the Provisional and Definitive
Regulations as the "EU-15") for trends concerning production, consumption,
imports, exports, and market share; the Community industry for trends concerning
production, sales by value, and employment; and the sample for the factors
mentioned above and for trends concerning prices and profitability.68 In its
analysis of factors regarding the state of the domestic industry, the EC
authorities considered data for the three levels where available for the various
factors.
6.179 To succeed, India's claim requires us to determine
that, having selected a sample, the European Communities was precluded as
a matter of law from considering, in its analysis under Article 3.4, any
information for any factor for any producers of bed linen not included in the
sample. One aspect of this claim relates to those producers of bed linen who,
while not included in the sample set selected by the investigating authorities,
were members of the "Community industry" as defined by the European Communities.
A second aspect of India's claim relates to those EC producers of bed linen who
were not members of the "Community industry" as defined by the European
Communities.
6.180 There is simply no basis in the AD Agreement for the
first aspect of India's claim. Keeping in mind that India has made no claim
regarding the constitution of the sample, and no claim regarding the definition
of the domestic industry, the only basis for India's position is that the
findings of the European Communities were not reached in an objective manner
based on properly established facts. However, India has not challenged the facts
themselves, but rather the European Communities' choices as to which facts,
among those it had gathered, it would consider in evaluating the Article 3.4
factors. There may be inadequate explanation or analysis of those facts, which
might constitute a violation of Article 12.2.2 or a failure adequately to
evaluate the Article 3.4 factors. However, this does not answer the question
whether consideration of evidence for domestic producers outside the selected
sample but within the domestic industry constitutes, ipso facto, a
violation of Article 3.4.
6.181 It is clear from the language of the AD Agreement, in
particular Articles 3.1, 3.4, and 3.5, that the determination of injury has to
be reached for the domestic industry that is the subject of the
investigation. Article 3.4 specifically requires that "The examination of the
impact of the dumped imports on the domestic industry concerned shall
include an evaluation of all relevant economic factors and indices having a
bearing on the state of the industry. . . " (emphasis added). In this case, the
European Communities defined the domestic industry as 35 producers of the like
product. In our view, it would be anomalous to conclude that, because the
European Communities chose to consider a sample of the domestic industry, it was
required to close its eyes to and ignore other information available to it
concerning the domestic industry it had defined. Such a conclusion would be
inconsistent with the fundamental underlying principle that anti-dumping
investigations should be fair and that investigating authorities should base
their conclusions on an objective evaluation of the evidence. It is not possible
to have an objective evaluation of the evidence if some of the evidence is
required to be ignored, even though it relates precisely to the issues to be
resolved. Thus, we consider that the European Communities did not act
inconsistently with Articles 3.1, 3.4, and 3.5 of the AD Agreement by taking
into account in its analysis information regarding the Community industry as a
whole, including information pertaining to companies that were not included in
the sample.
6.182 However, our conclusion with respect to the second
aspect of India's claim is different. As we have noted, the determination of
injury has to be reached for the domestic industry as defined by the
investigating authorities, in this case the 35 producers comprising the
"Community industry" as defined by the European Communities. In our view,
information concerning companies that are not within the domestic industry is
irrelevant to the evaluation of the "relevant economic factors and indices
having a bearing on the state of the industry" required under Article 3.4. This
is true even though those companies may presently produce, or may have in the
past produced, the like product, bed linen. Information concerning the
Article 3.4 factors for companies outside the domestic industry provides no
basis for conclusions about the impact of dumped imports on the domestic
industry itself. If other present or former bed linen producers had been
considered part of the domestic industry, the fact that some of them went out of
business would be relevant to the evaluation of the impact of dumped imports on
the domestic industry. But given that the European Communities defined the
domestic industry as 35 producers of bed linen, information concerning other
companies does not inform the evaluation of "factors and indices having a
bearing on the state of the industry" under Article 3.4 of the AD Agreement, and
thus cannot serve as the basis of findings regarding the impact of dumped
imports on the domestic industry.
6.183 We therefore conclude that, by relying on information
concerning producers not part of the domestic industry in its evaluation of the
impact of dumped imports on the domestic industry under Article 3.4 of the
AD Agreement, the European Communities failed to act consistently with that
provision.69
42 In response to a question from the Panel, India confirms that
"[t]he criterion of reasonability, as laid down in Article 2.2, instructs the
chapeau of Article 2.2.2 and its subparagraphs". Response of India to Question 1
from the Panel following the first meeting of the Panel, Annex 1-6.
43 We note in this context that India has not challenged
the reasonability of this rate, calculated on the basis of the chapeau
methodology, as applied to Bombay Dyeing itself, but only its use as the rate
applied, pursuant to Article 2.2.2(ii) for other Indian producers.
44 Indeed India appears to acknowledge that comparisons for
individual product types within a single like product are appropriate in the
context of an anti-dumping investigation, that the weighted average normal value
and weighted average export price for each model at issue in this case were
properly calculated by the European Communities, and that a fair comparison was
made with respect to each product type considered.
45 India asserts that the European Communities does not always
follow the practice of zeroing as applied in the bed linen proceeding, referring
to a document disclosing the calculation in another EC anti-dumping proceeding.
In that other case, India maintains that the "negative" dumping found for
certain models was offset against the dumping found for other models which were
dumped. We do not consider that the European Communities' practice in other
investigations has any relevance to our decision here, as India has made no
claim of discriminatory treatment in this dispute.
46 The United States notes that nothing in its argument on this
issue should be construed as expressing agreement or disagreement with the
European Communities� actual calculation of the dumping margin in this case,
because the United States does not have access to the specific factual
information considered by the European Communities.
47 The United States cites, in this regard, Stewart, Terence P.,
ed., The GATT Uruguay Round: A Negotiation History (1986-1992), Kluwer
Law International, The Hague, pp. 155-61 (discussing the negotiations
concerning, inter alia, weighted-average comparisons), and EC -
Anti-Dumping Duties on Audio Tapes and Cassettes Originating in Japan, Panel
Report, ADP/136, 28 April 1995 (unadopted), para. 348 (discussing the European
Communities� prior practice of comparing individual export prices to a weighted
average normal value).
48 In its response to the European Communities' request for
preliminary rulings, India withdrew any putative claims under Article 3.6, a
fact of which we have taken note, stating that we would issue no ruling on any
such claim. See paragraph 6.18 above.
49 As noted above, India withdrew its Article 3.6 claim in this
regard. The Article 3.4 claim in this regard is the subject of the European
Communities' preliminary objection based on failure to sufficiently identify the
claim in the request for establishment, which we have denied. See paragraph 6.28
above. In addition, we note that while the parties have made reference to
Article 3.2 in their arguments, India has made no claim in respect of that
Article, and we therefore make no findings under Article 3.2 of the
AD Agreement.
50 The United States stated, in response to questions from the
Panel, that its own practice is to exclude from the volume and price effects
analysis imports from producers or exporters for which a finding of no dumping
is made. These are considered in its injury investigation in the category of
"non-dumped" imports, as are imports from third countries not subject to the
investigation. However, while the United States clearly disputes the principle
espoused by the European Communities that dumping is determined for countries,
it does not assert that its own practice is required by the AD Agreement.
51 Imposition of Anti-Dumping Duties on Imports of Fresh and
Chilled Atlantic Salmon from Norway ("Salmon - Anti-Dumping Duties"),
Panel Report, ADP/87, adopted 26 April 1994, BISD 41S/228; Imposition of
Countervailing Duties on Imports of Fresh and Chilled Atlantic Salmon from
Norway ("Salmon - Countervailing Duties"), Panel Report, SCM/153,
adopted 28 April 1994, BISD 41S/576.
52 See section VI.C.3 (zeroing), supra.
53 India's argument suggests that the proportion of imports
attributable to dumped transactions for one producer or country could be applied
to determine the volume of dumped imports for a different producer or country.
We do not consider that such a practice would satisfy the general requirements
of the AD Agreement for consideration of positive evidence and objective
decision-making.
54 Salmon - Anti-Dumping Duties, Panel Report,
paras. 565-571; Salmon - Countervailing Duties, Panel Report, paras.
328-340 ("effects of the subsidy" and "effects of the subsidised imports").
55 See First Submission of the European Communities,
Annex 2-1, Table 4.
56 We note, in this regard, that the Panel in Korea - Dairy
Safeguard, considered the language of Article 4.2 of the Agreement on
Safeguards, which provides that, in making a determination of serious injury or
threat thereof in a safeguard investigation, the investigating authority:
"shall evaluate all relevant factors of an objective and
quantifiable nature having a bearing on the situation of that industry, in
particular, �"
The Panel concluded that the text of this provision made it
clear that:
"among "all relevant factors" that the investigating
authorities "shall evaluate", the consideration of the factors listed is
always relevant and therefore required, even though the authority may
later dismiss some of them as not having a bearing on the situation of
that industry". Korea - Dairy Safeguard, Panel Report,
para. 7.55.
See also, Argentina - Safeguard Measures on Imports of
Footwear, Appellate Body Report, WT/DS121/AB/R, adopted 12 January 2000,
para. 136. The similarities between the drafting of the provisions is obvious,
and we consider that the same conclusion is appropriate in interpreting
Article 3.4 of the AD Agreement. While the standard for injury in safeguards
cases ("serious injury") is different from that applied to injury determinations
in the anti-dumping context ("material injury"), the same type of analysis is
provided for in the respective covered agreements, i.e., evaluation or
examination of a listed series of factors in order to determine whether the
requisite injury exists.
57 The New Shorter Oxford English Dictionary, Clarendon
Press, Oxford, 1993.
58 Id.
59 Article 3.2 of the DSU directs panels to clarify the
provisions of the covered agreements "in accordance with customary rules of
interpretation of public international law", which are set out in Articles 31
and 32 of the Vienna Convention. See, e.g., Japan
� Taxes on Alcoholic Beverages, Appellate Body Report,
WT/DS8/AB/R�WT/DS10/AB/R�WT/DS11/AB/R, adopted 1 November 1996, pp.10-12. Here,
we look to negotiating history pursuant to Article 32 of the Vienna
Convention in order to confirm the meaning resulting from the application of
the general rule of interpretation in Article 31 of the Vienna Convention.
60 While that panel was examining the application of the
requirements of Article 3.4 in a case involving "threat of material injury", we
consider that its views on Article 3.4 are also relevant in this case, dealing
with material injury.
61 Mexico - HFCS, Panel Report, para. 7.128.
62 Id.
63 Exhibit India-8.
64 See Response of the European Communities to Question
20 from the Panel following the first meeting, Annex 2-5.
65 Provisional Regulation, Exhibit India-8, para. 62; see
para. 6.166, supra.
66 We are somewhat at a loss to understand how the European
Communities could find that companies listed in the complaint were nonetheless
not complainants, but this is not a question to be resolved in this case.
67 In EC proceedings, the "Community industry" is the domestic
industry for purposes of the AD Agreement.
68 Provisional Regulation, Exhibit India-8, para. 62.
69 Having found a violation of the AD Agreement in this regard,
and having found a violation in the failure of the European Communities to
evaluate all the Article 3.4 factors, we do not consider it necessary to
consider questions regarding the European Communities' evaluation of information
on Article 3.4 factors for the different groupings of producers.