Opening Statement of Australia - Second Session of the Oral Hearing

(AB-2011-3/DS353)

Geneva, 11 October 2011

Download print version

United States - Measures Affecting Trade in Large Civil Aircraft (Second Complaint) [PDF 60 KB]

Members of the Division,

1. Thank you for the opportunity to present the views of Australia with respect
   to the adverse effects issues raised by this appeal.
2. Australia will not comment on every issue raised by the Participants. Rather,
   we will focus on two issues of systemic importance to adjudicating claims
   under Articles 5 and 6.3 of the Agreement on Subsidies and Countervailing
   Measures (the SCM Agreement). The first issue concerns the requirements
   for determining that a subsidy caused adverse effects, and the United
   States' argument that the Panel was required to consider (as part of its counterfactual
   analysis) whether Boeing would have funded the research performed under the
   aeronautics R&D subsidies, had those subsidies not been granted. The second
   issue concerns the circumstances in which it is appropriate to aggregate
   the effects of subsidies, and the European Union's argument that the Panel
   erred in declining to assess the cumulative effects of the groups of subsidies
   at issue in this dispute.

Determining that a subsidy has caused adverse effects

1. The Panel found that the aeronautics R&D subsidies caused
   adverse effects to the European Union through enabling Boeing to launch the
   787 as and when it did.¹ This finding was based
   on the Panel's determination that the aeronautics R&D subsidies, through
   their structure, design and operation, "contributed in a genuine and
   substantial way to Boeing's development of technologies for the 787".² The Panel determined that, "absent the aeronautics R&D subsidies,
   Boeing would not have been able to launch an aircraft incorporating all of
   the technologies that are incorporated on the 787 in 2004, with promised deliveries
   commencing in 2008."³ The Panel consequently
   found that the alleged market phenomena - that is, the significant price suppression,
   significant lost sales and threat of displacement and impedance of exports
   from third country markets with respect to the 200-300 seat wide-body LCA
   product market -were the effect of the aeronautics R&D subsidies, because
   these market phenomena were attributable to the introduction of the 787 in
   2004 (which would not have occurred without the aeronautics R&D subsidies).⁴
2. The United States argues that the Panel erred in reaching this finding because
   it did not conduct a proper counterfactual analysis that took account of the
   fact that "Boeing had sufficient funds to achieve the same learning and
   experience provided by the government's aeronautics R&D expenditure at
   issue".⁵ The United States argues that a
   "proper counterfactual" would have considered that, "in the
   absence of subsidies, Boeing would have funded this 'critical' research itself;
   and this "establishes that the aeronautics R&D subsidies are not
   a genuine and substantial cause of adverse effects".⁶
3. Australia has two key concerns with this argument. First, it detracts from
   the proper focus of adjudicating claims under Articles 5 and 6.3 of the SCM
   Agreement, which is on whether or not a subsidy has caused adverse effects to the interests of another Member.⁷ Second, it
   misconstrues the proper role and scope of a counterfactual analysis.
4. In relation to Australia's first concern, the implication of the United
   States' argument is that a subsidy cannot be found to have caused adverse
   effects where some other (hypothetical) factor could have happened
   had the subsidy not been granted, and could have had the same effect
   as that subsidy. Australia notes that Articles 5 and 6.3 of the SCM Agreement
   do not require a complaining Member to establish that, had the subsidy not
   been granted, there were no alternative ways in which the same market phenomenon could possibly have arisen.
5. Further, while evaluating the effects of other actual factors may
   attenuate the causal link between the subsidy and the alleged market phenomena,
   speculating about the effect a hypothetical factor may have had in
   the place of the subsidy does not assist in determining whether or not the
   challenged subsidy did in fact cause adverse effects (which is the proper
   focus of an adverse effects analysis).
6. Therefore, in Australia's view: (i) it is irrelevant for a responding
   Member to speculate that, had the subsidy not been granted, alternative sources
   of funding could have given rise to the same effects as the challenged subsidy;
   and (ii) it is incorrect to conclude that this possibility displaces
   a finding that the subsidy in fact caused the alleged market phenomena.
7. Australia's second concern with the United States' argument is twofold.
   First, as we will explain, it misconstrues the proper role of a counterfactual
   analysis, which is to determine the effect of the subsidy on the market.
   Second, it misconstrues the scope of a counterfactual analysis by
   introducing additional hypothetical factors (other than the assumption that
   the challenged subsidy was not paid).
8. The Appellate Body's previous guidance establishes that: a counterfactual
   analysis is undertaken "to isolate and properly identify the effects"
   of a subsidy;⁸ and "entails comparing the
   actual market situation that is before the adjudicator with the market
   situation that would have existed in the absence of the challenged subsidies".⁹ This enables a panel ultimately to determine whether or not a subsidy caused
   the market phenomena alleged.
9. Australia observes that, in establishing adverse effects claims under Articles
   5 and 6.3 of the SCM Agreement, there will likely be links in the causal chain between the grant of the subsidy at issue and the particular market
   phenomena alleged - that is, the grant of a subsidy may cause the particular
   market phenomena through a particular effect or mechanism,¹⁰ rather than directly giving rise to the alleged market phenomena.
10. Therefore, the causation analysis required to determine whether or not a
    subsidy causes adverse effects involves a number of discrete analytical steps.
    In Australia's view, a panel must first determine the effect that
    a subsidy has had, through considering the nature of the subsidy (its structure,
    design and operation), taking care not to attribute to the subsidy the effects
    of other factors. A panel can then undertake a counterfactual analysis
    to examine the effect of that subsidy on the market - this involves
    determining what the market would have looked like absent the subsidy (or,
    more precisely, absent the effect of the subsidy) and comparing this with
    the actual market situation before the panel. Together, these analytical steps
    allow a panel ultimately to determine whether or not the alleged
    market phenomena are caused by, or are the result of, the subsidy
    - in other words, these collective steps establish whether or not there is
    "a genuine relationship of cause and effect between the subsidy and the
    alleged market phenomena".¹¹
11. In Australia's view, the United States' argument that a "proper counterfactual"
    would have established that the aeronautics R&D subsidies did not cause
    adverse effects (because Boeing could have funded the relevant research itself)¹² improperly suggests that the Panel was required to conduct a counterfactual analysis to determine the effect that the subsidies had - that is, to determine
    whether or not the subsidies contributed to Boeing's ability to develop and
    launch the 787 as and when it did. However, this was the Panel's first analytical step. The Panel concluded that, due to their structure, design
    and operation, the aeronautics R&D subsidies "contributed in a genuine
    and substantial way to Boeing's development of technologies for the 787".¹³ In Australia's view, a counterfactual analysis has no role to play
    in this step of the causation analysis.
12. Rather, in Australia's view, the proper role of a counterfactual analysis
    is to determine what the market would have looked like absent the subsidy.
    Therefore, in this case, the proper role of a counterfactual was to determine
    what the market would have looked like absent the aeronautics R&D subsidies
    (and, consequently, absent Boeing's ability to launch the 787 as and when
    it did, which the Panel had attributed to the effect of the subsidies) - not
    whether "in the absence of subsidies, Boeing would have funded this 'critical'
    research itself".¹⁴
13. Australia further considers that the United States' argument misconstrues
    the scope of a counterfactual analysis. In Australia's view, the
    only relevant hypothetical factor in a counterfactual analysis is the assumption
    that the subsidy was not granted. The consideration of hypothetical
    factors other than this undermines the purpose of the counterfactual
    analysis, because taking account of additional variables does not "isolate
    and properly identify" the effects of the challenged subsidy.¹⁵ Rather, in order for a counterfactual analysis to properly assess the effects
    of a subsidy, a panel should examine what the market would have looked like
    had the subsidy not been granted holding all else equal.
14. Australia therefore considers that, even if the proper role of
    a counterfactual analysis were to determine whether "in the absence of
    subsidies, Boeing would have funded this 'critical' research itself',¹⁶ examining whether alternative sources of funding could, hypothetically, have taken the place of those subsidies is beyond the proper scope of
    such an analysis.

Aggregating the effects of subsidies

1. We now turn to the issue of aggregating the effects of challenged
   subsidies.
2. In Australia's view, there are two key questions directly before the Appellate
   Body. First, whether or not it is permissible to aggregate the effects of
   subsidies with a sufficient nexus with the same subsidised product and the
   same market phenomena alleged under Article 6.3 of the SCM Agreement, even
   where such subsidies operate through different "causal mechanisms".¹⁷ Second, whether or not a panel is required to aggregate the effects of subsidies where the conditions for aggregation are met.
3. In relation to the first question, Australia recalls the approach taken
   in United States – Subsidies on Upland Cotton (which we will subsequently
   refer to as Cotton). The panel found that the reference in Article
   6.3(c) to "the effect of the subsidy" (singular) did not
   mean that "a serious prejudice analysis of price suppression must clinically
   isolate each individual subsidy and its effects".¹⁸ The panel noted that, if such an approach were "[t]aken to an extreme,
   this could mean that separate dispute settlement proceedings, or at least
   separate claims, would need to be brought with respect to the serious prejudice
   caused by each and every individual subsidy, even where these subsidies exist
   contemporaneously and interact in concert in respect of a single subsidized
   product to produce a single result in the form of a price phenomenon".¹⁹
4. Instead, the panel in Cotton determined that it was legitimate
   to conduct "an integrated examination of effects of any subsidies with
   a sufficient nexus with the subsidised product and the particular effects-related
   variable under consideration". ²⁰ It found
   that "in our price suppression analysis under Article 6.3(c), we examine
   one effects-related variable - prices - and one subsidized product - upland
   cotton. To the extent that a sufficient nexus with these exists among the
   subsidies at issue so that their effects manifest themselves collectively,
   we believe that we may legitimately treat them as a 'subsidy' and group them
   and their effects together".²¹
5. The Panel in the present dispute referred to this approach in Cotton,²² and appears to have interpreted this to mean that "in order to conduct
   an aggregated analysis of the effects of subsidies in the context of this
   dispute, it should be possible to discern from their structure, design and
   operation that they affect Boeing's behaviour in a similar way".²³ The Panel therefore declined to aggregate the effects of the B&O tax subsidies
   on Boeing's pricing of the 787 with the effects of the aeronautics R&D
   subsidies, on the basis that "the two groups of subsidies operate through
   entirely distinct causal mechanisms".²⁴
6. In Australia's view, this approach is incorrect. In Cotton, the
   panel's analytical focus on the particular "effects-related variable"
   (that of price) stemmed from its implicit assumption that only measures
   with a sufficient nexus with price could cause price suppression (the particular market phenomenon at issue). The Appellate Body rightly questioned
   this in its review of the panel's decision, noting that "[w]e do not
   exclude the possibility that challenged subsidies that are not 'price-contingent'...
   could have some effect on production and exports and contribute to price suppression".²⁵ When viewed within the context of the panel's erroneous assumption, the panel's
   references to the 'effects-related variable' essentially serve as a proxy
   for the particular market phenomenon at issue - that is, if only price-contingent
   subsidies could cause price suppression, then it follows that only the effects
   of measures with a sufficient nexus with price could be aggregated
   for the purposes of a price suppression analysis.
7. In this way, the panel in Cotton did not find that it could not
   aggregate the effects of the price-contingent subsidies and the non-price
   contingent subsidies because they operated through different 'causal mechanisms'.
   Rather, the panel was not satisfied that the non-price contingent subsidies
   had a sufficient nexus with the particular market phenomenon at issue
   (price suppression). As the panel expressly stated, it declined to aggregate
   the effects of the two groups of subsidies because, in its view, Brazil had
   "not established that.. .significant price suppression.. .was 'the effect'
   of these non-price contingent subsidies".²⁶
8. However, as the panel itself acknowledged, its approach was "tailored
   to the particular facts and circumstances of this dispute and of the measures
   before [it]".²⁷ Further, the panel had indicated
   that it considered that aggregating the effects of subsidies would be appropriate
   "where these subsidies exist contemporaneously and interact in concert
   in respect of a single subsidized product to produce a single result in the form of a price phenomenon."²⁸
9. Australia recalls that the Appellate Body's recent finding in European
   Communities – Measures Affecting Trade in Large Civil Aircraft further
   supports the view that a panel may assess the cumulative effects of two groups
   of subsidies where both subsidies have a sufficient nexus with the same subsidised
   product and the same alleged market phenomena (regardless of whether
   both subsidies operate through the same 'causal mechanism').²⁹
10. In that case, the Appellate Body found that, "[o]nce the Panel determined
    that LA/MSF subsidies were a substantial cause of the observed displacement
    and lost sales, it was not necessary to establish that non-LA/MSF subsidies
    were also substantial causes of the same phenomena".³⁰ Rather, "it was permissible and sufficient for the Panel to assess whether
    a genuine causal connection between non-LA/MSF subsidies and the same market
    phenomena existed such that these non-LA/MSF subsidies complemented or
    supplemented the effects of LA/MSF".³¹ A
    cumulative assessment of the relevant subsidies was permitted "provided
    that a genuine causal link between the non-LA/MSF subsidies and the market
    phenomena alleged under Article 6.3 is established".³² In other words, if multiple subsidies cause the same market phenomena,
    a cumulative assessment of their effects is permitted.
11. Australia therefore agrees with the European Union that the Panel did not
    need to find that the B&O tax subsidies would "on their own"
    have such an effect on Boeing's prices of the 787 as to cause significant
    price suppression, significant lost sales and threat of displacement and impedance
    of exports from third country markets, with respect to the 200-300 seat wide-body
    product market.³³ Rather, the Panel could have
    considered whether the B&O tax subsidies "complemented or supplemented"
    the effects of the aeronautics R&D subsidies (which it had already found
    caused these market phenomena in respect of that product market).³⁴ The fact that the two groups of subsidies operated through "distinct
    causal mechanisms" was irrelevant to the Panel's analysis.³⁵
12. For the same reasons, Australia also agrees with the European Union that
    it would have been permissible for the Panel to have assessed whether there
    was a genuine causal link between the Remaining Subsidies and significant
    price suppression, significant lost sales, and displacement and impedance
    of exports in third country markets, such that the effects of the Remaining
    Subsidies "complemented or supplemented" the effects of the Tax
    Subsidies with respect to the 100-200 seat single aisle product market and
    the 300-400 seat wide-body product market.
13. In relation to the second key question before the Appellate Body - whether
    a panel is required to aggregate the effects of subsidies - Australia
    notes that previous panels have described their authority to aggregate the
    effects of subsidies in permissive language.³⁶ Similarly, the Appellate Body has previously described a panel's authority
    to consider whether the effect of one group of subsidies "complemented
    or supplemented" the effects of another group of subsidies in permissive
    language.³⁷This suggests that it is within a
    panel's discretion to determine on a case-specific basis whether the effects
    of different subsidies should be aggregated.
14. However, Australia notes that these previous 'permissive' descriptions were
    made in the context of a panel's decision to aggregate the effects
    of different subsidies. Given that a panel's decision to decline to
    aggregate the effects of different subsidies could have significant consequences
    for a complaining Member's adverse effects claim, and could lead to a continuation
    of measures that may 'complement or supplement' the adverse effects claimed,
    Australia would welcome the Appellate Body's guidance as to whether or not
    a panel is required to aggregate the effects of subsidies that it finds
    have a sufficient nexus with the same subsidised product and the same alleged
    market phenomena.
15. Finally, we would like to note that although Australia's written submission
    and this statement do not address every issue raised by the European Union,
    the United States, and the other Third Participants, this should not be regarded
    as an indication that Australia considers that the issues it has not addressed
    are not important. Nor does it indicate agreement or otherwise with any particular
    argument of the Participants or other Third Participants in these appellate
    proceedings.
16. Thank you.

• ¹Panel Report, paragraph 7.1775.
• ²Ibid, paragraph 7.1773.
• ³Ibid, paragraph 7.1775.
• ⁴ Panel Report, paragraph 7.1780 and 7.1854.
• ⁵Other Appellant Submission of the United States, paragraph 268.
• ⁶ Ibid, paragraph 270.
• ⁷ See Article 5 of the Agreement on Subsidies and Countervailing
  Measures.
• ⁸European Communities – Measures Affecting Trade in Large Civil Aircraft (Appellate Body), paragraph 1110.
• ⁹ Ibid.
• ¹⁰ Panel Report, paragraph 7.1596.
• ¹¹ US – Subsidies on Upland Cotton (Appellate Body),
  paragraph 438; US – Subsidies on Upland Cotton (Recourse to Article 21.5
  of the DSU by Brazil) (Appellate Body), paragraph 374; European Communities
  – Measures Affecting Trade in Large Civil Aircraft (Appellate Body); paragraph
  1231.
• ¹² Other Appellant Submission of the United States.
• ¹³ Panel Report, paragraph 7.1773. Australia does not
  express a view on whether this conclusion is sufficiently supported by the facts
  on the record.
• ¹⁴ Other Appellant Submission of the United States, paragraph 270.
• ¹⁵ European Communities – Measures Affecting Trade
  in Large Civil Aircraft (Appellate Body), paragraph 1110.
• ¹⁶ Other Appellant Submission of the United States, paragraph 270.
• ¹⁷ Panel Report, paragraph 7.1805.
• ¹⁸ United States – Subsidies on Upland Cotton (Panel), paragraph 7.1192.
• ¹⁹ Ibid, footnote 1307.
• ²⁰ Ibid, paragraph 7.1192.
• ²¹ Ibid, paragraph 7.1192.
• ²² Panel Report, paragraph 7.1804.
• ²³ Ibid, paragraph 7.1805.
• ²⁴Panel Report, paragraph 7.1824.
• ²⁵ United States – Subsidies on Upland Cotton (Appellate
  Body), paragraph 450, footnote 589.
• ²⁶ United States – Subsidies on Upland Cotton (Panel),
  paragraph 7.1350.
• ²⁷ Ibid, paragraph 7.1192, footnote 1308.
• ²⁸ United States – Subsidies on Upland Cotton (Panel),
  paragraph 7.1192, footnote 1307.
• ²⁹ European Communities – Measures Affecting Trade in Large
  Civil Aircraft (Appellate Body), paragraph 1378. However, Australia notes
  that the Appellate Body stated (in paragraph 1374) that it did not consider
  that the panel had aggregated the effects of the subsidies in assessing
  whether the effects of one group of subsidies "complemented and supplemented"
  the effects of other subsidies.
• ³⁰ Ibid.
• ³¹ Ibid.
• ³² Ibid.
• ³³ Panel Report, paragraph 7.1824.
• ³⁴ Ibid, paragraph 7.1794.
• ³⁵ Ibid, paragraph 7.1824.
• ³⁶ For example, US – Subsidies on Upland Cotton (Panel),
  paragraph 7.1192.
• ³⁷ European Communities – Measures Affecting Trade in Large Civil Aircraft (Appellate Body), paragraph 1378.