CANADA - MEASURES AFFECTING THE EXPORT
OF CIVILIAN AIRCRAFT



Recourse by Brazil to Article 21.5 of the DSU




Report of the Panel

(Continuation)

4. Evaluation by the panel

(a) Scope of the disagreement between the parties

5.9 Brazil's primary³ claim concerns the measures taken by Canada to restructure the TPC programme insofar as it will apply in the future to the Canadian regional aircraft industry. In particular, Brazil's primary claim raises issues concerning the substance of the prospective implementation action undertaken by Canada. With respect to Brazil's primary claim, therefore, there is no disagreement between the parties resulting from the fact that, in order to implement the recommendation of the DSB concerning TPC assistance to the Canadian regional aircraft industry, Canada has taken prospective action. The parties agree that to "withdraw" the subsidy in this case requires some sort of prospective action on the part of Canada.

5.10 We recall that Article 21.5 disputes arise "[w]here there is disagreement as to the existence or consistency with a covered agreement of measures taken to comply with the recommendations and rulings"⁴ of the DSB. Since there is no disagreement between the parties⁵ that, in order to implement the recommendation of the DSB concerning TPC assistance to the Canadian regional aircraft industry, Canada is required to take some form of prospective action, we do not consider it necessary to provide a comprehensive interpretation of what is required for an implementing Member to "withdraw" a prohibited export subsidy. Rather, it is sufficient to conclude (and we note that the parties seem to agree with this) that a Member cannot be understood to have withdrawn a prohibited subsidy if it has not ceased to provide such a subsidy, as that Member therefore would not have ceased to violate its WTO obligations in respect of such a subsidy. In our view, therefore, Canada's obligation arising from the DSB's recommendation in this dispute includes the obligation to cease providing prohibited export subsidies to the regional aircraft sector under the TPC. We note that in the circumstances of this Article 21.5 proceeding concerning TPC, such an assessment is by nature forward-looking. Accordingly, we shall focus on Canada's restructuring of the TPC programme insofar as it will apply to the Canadian regional aircraft industry in the future. If necessary, we shall then examine Brazil's alternative claim regarding past TPC assistance to the regional aircraft industry.

5.11 With regard to the future, Brazil claims that Canada should abolish / withdraw the TPC programme in respect of the Canadian regional aircraft industry. At a minimum, though, Brazil asserts that Canada must "ensure" that de facto export subsidies cannot be granted to the regional aircraft industry, and not merely that they might not be granted.⁶ According to Brazil, if Canada maintains funding to the Canadian regional aircraft industry under the "new" TPC, Canada must ensure that the program will operate in full compliance with the SCM Agreement.⁷ Canada denies that it is required to abolish / withdraw the TPC programme in respect of the Canadian regional aircraft industry, but asserts that it "has taken the steps within Canada's control to ensure that any assistance that TPC may provide in the future to the Canadian regional aircraft industry will not be contingent on export performance in law or in fact".⁸

5.12 Thus, Brazil and Canada effectively agree on the need for Canada to satisfy Brazil's minimum implementation standard, i.e., to "ensure" that future TPC assistance to the Canadian regional aircraft industry will not be de facto contingent on export performance. The parties disagree, however, on whether Canada has taken sufficient steps to satisfy that standard. To resolve this disagreement, we must examine whether or not Canada has taken sufficient steps to ensure that future TPC assistance to the regional aircraft industry will not be de facto contingent on export performance.

(b) Burden of proof

5.13 In examining this issue, we note that "Brazil recognises that it bears the burden of showing that Canada has failed to implement. � It then becomes Canada's burden to explain how Brazil was wrong and how Canada's purported changes actually constitute effective implementation."⁹ Canada agrees that the initial burden of proof falls on Brazil.

5.14 We agree that Brazil, as the complaining party, bears the burden of proof in this proceeding. We agree with the Appellate Body's statement in EC - Hormones that "[t]he initial burden lies on the complaining party, which must establish a prima facie case of inconsistency �",¹⁰ and consider that this should apply in the context of Article 21.5 proceedings. Since the burden is on Brazil (i.e., the complaining party) to show that Canada has failed to implement the recommendation of the DSB (by reference to the minimum implementation standard agreed on by the parties), Brazil must establish that Canada has failed to "ensure" that future TPC assistance to the Canadian regional aircraft industry will not be de facto contingent on export performance.

(c) Substantive analysis

5.15 Brazil considers that it has discharged its burden of proof by demonstrating "that all the essential elements of the [TPC] program remain unchanged, and that many of these elements will never change".¹¹ In this regard, Brazil claims that the facts surrounding the "new" TPC still support an inference of de facto export contingency. In particular, Brazil refers to the following four factors which, in its opinion, lead to an inference that future TPC assistance to the Canadian regional aircraft industry continues to be de facto export contingent:

• eligible industries remain "specifically targeted" because of their export orientation;
   
• eligible activities continue to betray an interest in near-market projects;
   
• export performance is an implicit selection and assessment criterion; and
   
• many TPC documents have not yet been replaced or amended.

We shall examine each of these factors in turn.

(i) Eligible industries remain specifically targeted because of their export orientation

5.16 Brazil argues that the continued de facto export contingency of TPC may be inferred from the fact that the Canadian regional aircraft industry continues to be "specifically targeted" for TPC assistance because of its undisputed export orientation.¹² Brazil asserts that "[n]othing, in short, has changed - neither the industries eligible for TPC contributions, nor the recognized export-orientation of the industry that enjoys the lion's share of those contributions, nor the significance of that industry's export orientation to Canadian government officials, nor that industry's prospects for continued dominance of TPC's treasury. None of these factors is destined for change."¹³ Brazil asserts that, to maintain the export orientation of the Canadian regional aircraft industry, "the Canadian aerospace industry receives the vast majority of the rapidly increasing pool of TPC finds available".¹⁴ Brazil further argues that "in choosing which industry would receive the lion's share of 'old' and 'new' TPC funds, Canada was not casually indifferent to the trading patterns of that industry. Instead, Canada chose, as TPC's showcase, an industry that exports significantly more than others, because it exports significantly more than others. The 'new' TPC retains a focus on contributions to the aerospace industry."¹⁵ According to Brazil, "the targeted industries of the 'old' TPC are the same recipients under the 'new' TPC".¹⁶

5.17 Thus, we understand Brazil to argue that "nothing has changed" because TPC assistance continues to be "specifically targeted" at the Canadian aerospace or regional aircraft industries, in the sense that these industries will continue to receive the "vast majority", or "lion's share", of TPC assistance. In addressing this argument, we recall that the "specific targeting" concept (in those or other words) did not form part of our reasoning regarding contingency in fact on export performance in that dispute. While we do not exclude the possibility that, in a given case, a factual circumstance of "specific targeting" might be considered by a panel to be part of the totality of facts leading to an inference of export contingency, this was not the case in the original Canada - Aircraft dispute. That is, of the factual considerations enumerated by us at para. 9.340 of our Report, none concerned the alleged targeting of the Canadian aerospace industry generally, or the Canadian regional aircraft industry in particular, by TPC, none concerned the amount of total TPC funding directed at the Canadian aerospace or regional aircraft industries,¹⁷ and none concerned the fact that the aerospace or regional aircraft industries were eligible for TPC assistance. Arguing a failure to implement on the grounds that there has been no change in alleged factual circumstances, which themselves were not part of our original ruling, is of questionable merit and logic. Indeed, we consider that the question of whether TPC assistance is "specifically targeted" to the aerospace and regional aircraft industries is not relevant to the present dispute, which concerns the issue of whether or not Canada has implemented the DSB recommendation on TPC assistance to the Canadian regional aircraft industry. That recommendation cannot have required Canada to take implementation action to ensure that TPC assistance is not "specifically targeted" at the aerospace and regional aircraft industries, because such alleged "specific targeting" did not form part of the basis for the finding of de facto export contingency that gave rise to that recommendation.¹⁸ The fact that "nothing has changed" concerning the alleged "specific targeting" of the aerospace and regional aircraft industries therefore has no bearing on the present dispute.

5.18 For these reasons, we do not consider it necessary to examine Brazil's argument that "nothing has changed" because TPC assistance continues to "specifically target" the Canadian aerospace and regional aircraft industries.

(ii) Interest in near-market projects

5.19 Brazil argues that the de facto export contingency of future TPC funding to the Canadian regional aircraft industry should be inferred from the fact that the available descriptions of eligible activities under the "new" TPC betray an interest in "near market" projects with high commercialization potential. Brazil also argues that essentially the same projects continue to be eligible for "new" TPC contributions as were eligible under the "old" TPC, such that "if funding for the development of commercial products was available in the 'old' TPC, it is similarly available in the 'new' TPC, and as it did before contributes to an inference of de facto export contingency".¹⁹

5.20 We recall that our earlier findings in Canada - Aircraft were based in part on the express recognition in the 1996/1997 TPC Business Plan that "TPC's 'approach' in the aerospace and defence sector is to '[d]irectly support the near market R & D projects with high export potential'".²⁰

5.21 In its review of our findings, the Appellate Body asserted that, if a panel takes the "nearness-to-the-export-market factor" into account, "it should treat it with considerable caution". � [T]he mere presence or absence of this factor in any given case does not give rise to a presumption that a subsidy is or is not de facto contingent upon export performance". Accordingly, we shall proceed with caution when addressing Brazil's arguments regarding the alleged nearness-to-the-export-market of "new" TPC projects in the regional aircraft sector.

5.22 We note that the 1996-1997 TPC Business Plan, which contained the aforementioned reference to "near market R & D projects with high export potential" is no longer valid, and no longer exists for the purposes of TPC as it is now constituted.²¹ The 1996-1997 TPC Business Plan is therefore irrelevant when considering whether future TPC assistance to the Canadian regional aircraft industry will be de facto contingent on export performance.²²

5.23 In order to substantiate its claim that eligible activities for "new" TPC funding betray an interest in near-market projects, Brazil states that, according to "new" TPC documentation, TPC will fund "projects 'aimed at the discovery of knowledge, with the objective that such knowledge may be useful in developing new products,' and those projects leading to 'translation of industrial research findings into a plan, blueprint or design for new, modified or improved products �'".²³ In response, Canada asserts that the inclusion of Industrial Research as an Eligible Activity "permits TPC to support earlier stage research and development that is further removed from the production and sale of specific products. The pre-competitive development category of eligible activity enables TPC to support the development of horizontal technologies that cut across the operations of recipient firms � rather than the development of specific products".²⁴

5.24 In our view, the mere fact that the results of a project may in the future be useful in the development of new products, or the modification / improvement of existing products, does not by itself render the project near-market. This view is confirmed by former footnotes 28 and 29 to former Article 8.2(a) of the SCM Agreement²⁵, concerning non-actionable subsidies, which appears to have strongly influenced Canada's choice of wording in the "new" TPC documents cited by Brazil.²⁶ In our view, the non-actionable subsidy projects referred to in former footnotes 28 and 29 concerned "industrial research" and "pre-competitive development activity" projects that were sufficiently removed from the market to suggest that their impact on the market was likely to be minimal. As a result, it would be incongruous for us to find similarly defined TPC projects to be "near-market".

5.25 Brazil has also argued that "new" TPC eligible activities betray an interest in "near market" projects because they are similar to "old" TPC eligible activities which were found to be "near market". In this regard, Brazil relies exclusively on a description of "old" TPC activities contained in a January 1998 TPC website excerpt.²⁷ According to Brazil, the description of "new" TPC eligible activities is similar to the description of "old" TPC eligible activities found in the January 1998 TPC website excerpt. We do not consider it necessary to pursue this argument, since our original finding that TPC funding in the aerospace & defence sector (and therefore in the regional aircraft industry component thereof) was focused on "near market" projects was based on the aforementioned statement in the 1996-1997 TPC Business Plan, and not the January 1998 TPC website excerpt. We therefore do not see the relevance of comparing the "new" description of eligible activities with the "old" description of TPC eligible activities contained in the January 1998 TPC website excerpt. Of far greater relevance, however, is the fact that the 1996/1997 TPC Business Plan, which contained the explicit reference to "near market" projects ("with high export potential") is no longer valid for the "new" TPC. Aerospace & defence activities eligible for "new" TPC funding will necessarily differ from aerospace & defence activities eligible for funding under the "old" TPC, since - as provided for in the 1996/1997 TPC Business Plan - "old" TPC funding in the aerospace & defence sector was explicitly and exclusively focused on "near-market projects", which - on the basis of the evidence before us - is not the case for "new" TPC funding in the aerospace & defence sector²⁸.

5.26 Accordingly, Brazil has failed to demonstrate that the available descriptions of eligible activities under the "new" TPC betray an interest in "near market" projects with high commercialization potential, or that activities eligible for funding under the "new" TPC are essentially the same as those eligible for funding under the "old" TPC.²⁹

(iii) Export performance as an implicit selection and assessment criterion

5.27 Brazil notes that the goals and objectives of the "new" TPC, like those of the "old" TPC,³⁰ concern the creation of Canadian jobs, the increase of Canadian economic growth, or the increase of Canadian wealth. Brazil asserts that an inference of de facto export contingency may be drawn from an intimate "link" between (1) the fulfilment of these goals and objectives and (2) exports. Brazil asserts that, because of this "link", TPC assistance to the regional aircraft industry will be implicitly conditioned on, or tied to, export performance.

5.28 Canada notes that the mandate and overall programme objective of the restructured TPC provide that "TPC is a technology investment fund established to contribute to the achievement of Canada's objectives of increasing economic growth, creating jobs and wealth, and supporting sustainable development".³¹ According to Canada, the restructured TPC's mandate and objectives do not encompass the enhancement of exports or Canada's export base.

5.29 We recall that our original findings were based in part on the Terms and Conditions of the "old" TPC, which stated that the Aerospace & Defence component of the TPC would be "directed to projects that will maintain and build upon the � export base extant in the aerospace and defence sector".³² We note that the Terms & Conditions of the "new" TPC no longer explicitly direct the Aerospace & Defence component thereof at projects that maintain and build upon the "export base" of the aerospace & defence sector. It is presumably for this reason that Brazil refers to the alleged implicit conditionality between the grant of "new" TPC assistance to the Canadian regional aircraft industry and the export performance of that industry.

5.30 While it is certainly true that the provision of funds on the basis of the "new" TPC's mandate and objectives could result in additional exports by funded sectors, we recall the Appellate Body's ruling that the mere knowledge, or anticipation, that exports will result from a subsidy does not by itself render that subsidy de facto contingent on export performance, because it does not by itself demonstrate conditionality.

5.31 Brazil has argued that the requisite conditionality may be inferred from the fact that recipients of "new" TPC assistance implicitly "commit to export performance". In this regard, Brazil has sought to draw an analogy with the facts of the Australia - Leather case³³ where the panel, in Brazil's own words, "determined that requesting undifferentiated sales performance targets led to an inference of de facto export contingency because the Australian government knew that in order to reach those targets, the recipient would have to export".³⁴ According to Brazil, "[t]he same logic applies in the case of the Canadian regional aircraft industry; the Canadian government knows that the industry exports virtually all its products, and thus to reach sales forecasts, it must export".³⁵

5.32 Without taking any view on the findings of the Australia - Leather panel, we note that Brazil has adduced no evidence that "new" TPC assistance to the Canadian regional aircraft industry will be conditioned on the fulfilment of sales targets (as was found to be the case in Australia - Leather³⁶ ). Brazil claims instead that the grant of "new" TPC assistance to the regional aircraft industry is contingent on the fulfilment of sales forecasts. However, in response to a question from the Panel, Brazil was unable to adduce any evidence to substantiate its claim of contingency on sales forecasts. Brazil only adduced evidence to the effect that sales forecasts will be used in the context of the "new" TPC programme.³⁷ In response, Canada explicitly denied that the granting of TPC assistance to the regional aircraft sector is contingent on the fulfilment of sales forecasts.³⁸ In the absence of any evidence to the contrary, we see no reason to doubt Canada's explicit denial. Furthermore, although sales forecasts may be used in the context of "new" TPC assistance to the regional aircraft industry, as they were under the "old" TPC,³⁹ this does not by itself mean that "new" TPC assistance to the Canadian regional aircraft industry will be contingent on fulfilment of those sales forecasts. The fact that a subsidy repayment schedule may be based on royalties from forecast sales does not mean that compliance with the sales forecast becomes a condition for the bestowal of the subsidy; it simply means that a sales forecast was used to fix the repayment schedule.⁴⁰ This situation is different from that before the Australia - Leather panel, where the relevant payments were "conditioned on Howe's agreement to satisfy, on the basis of best endeavours, the aggregate performance targets".⁴¹ For these reasons, we disagree with Brazil that the logic of the Australia - Leather panel applies in the present case.

5.33 Furthermore, we note that Part 4 of the "new" TPC Investment Decision Document requires TPC administrators to record the "[b]enefits [of the project] to Canada". Section 5 of the "new" TPC Investment Application Guide defines "technological and net economic benefits to Canada" as "increasing economic growth, creating jobs and wealth, and supporting sustainable development". Nowhere in these documents is increased export performance identified as a "technological" or "net economic benefit" to Canada. Indeed, Part 4 of the aforementioned Investment Decision Document explicitly provides that "TPC will not accept or consider information concerning the extent to which a company does or may export". The only conclusion that we can reach from the face of these documents is that projects will be compared against one another, and eventually selected for funding, on the basis inter alia of the amount of technological and/or net economic benefits to which they are expected to give rise. While it is clear that for some projects, these benefits will derive largely or exclusively from exports, there is no factual basis in the documents (which are at this point the only available evidence) on which to conclude that projects generating the most exports will be those selected for funding. Indeed, the documents indicate that the administrators simply will not have specific information about the volume of exports that might result from any project for which TPC funding is sought. Thus, whereas TPC assistance is conditional on a project having certain technological or net economic benefits to Canada, in our view this simply cannot be assumed to be synonymous with export performance, and therefore it does not mean ipso facto that such assistance is contingent on export performance. This remains true even though TPC administrators know that fulfilment of net economic benefits in certain cases may be likely to result in increased exports. The fact that they will have no concrete quantifiable information on exports in our view will act in practical terms to limit their discretion to select projects on the basis of export performance.

3 Only in the alternative does Brazil raise any claims concerning past TPC assistance to the regional aircraft industry. However, Brazil has explicitly stated that a remedy concerning (exclusively) future TPC assistance to the regional aircraft industry is preferred (see para. 5.45 below).

4 Emphasis supplied.

5 We recall that we are not, at this juncture, addressing Brazil's alternative claim regarding repayment of past TPC assistance to the Canadian regional aircraft industry.

6 Brazil could be understood to have proposed an impossible implementation standard, since no sovereign state will ever be able to provide an absolute guarantee that it will not in the future provide de facto export subsidies. Any such guarantee would effectively eliminate the totality of a state's discretionary authority. Brazil acknowledges this point, by stating that "[o]bviously, a sovereign state cannot [eliminate all of its discretionary authority] and remain a sovereign state" (Brazil's reply to TPC question 1(a) from the Panel). In light of Brazil's acknowledgement, we understand Brazil to argue that Canada need only ensure that de facto export subsidies cannot be granted to the regional aircraft industry within the context of the "new" TPC programme. This understanding is confirmed by Brazil's assertion that "Canada must ensure that the program will operate in full compliance with the [SCM] Agreement" (Second written submission of Brazil (Annex 1-2), para. 19, underline emphasis supplied).

7 Id.

8 Canada's reply to the Panel's TPC question 2, para. 57 (Annex 2-4), emphasis supplied.

9 Brazil's reply to the Panel's TPC question 1(a) (Annex 1-5).

10 EC - Hormones, WT/DS26/AB/R, WT/DS48/AB/R, para. 98, adopted 13 December 1998.

11 Brazil's reply to TPC question 1(a) from the Panel (Annex 1-5) (emphasis supplied).

12 Brazil's reply to TPC question 2 from the Panel (Annex 1-5).

13 Brazil's first written submission, para. 22, Annex 1-1).

14 Brazil's first written submission, para. 23 (Annex 1-1).

15 Brazil's second written submission, paras 32 and 33 (Annex 1-2).

16 Brazil's concluding remarks, para. 9 (Annex 1-4).

17 We recall that, in our original findings, we referred to the fact that three specific transactions examined by us accounted for approximately 68 per cent of TPC contributions to the aerospace and defence sector during the period 1996-1997 (see para. 9.307 of Canada - Aircraft, WT/DS70/R). However, we made this factual reference in the context of our original findings on subsidization. This factual reference played no part whatsoever in our original findings on de facto export contingency.

18 We note the statement of the Australia - Leather Article 21.5 panel (which Brazil has quoted in its reply to the Panel's TPC question 5 (see Annex 1-5)) that "[t]he specific details of the factual evidence underlying the conclusion that the subsidies were in fact contingent upon export performance � do not, in our view, determine what is required in order to 'withdraw the subsidy' within the meaning of Article 4.7 of the SCM Agreement" (WT/DS126/RW, adopted 11 February 2000, note 24). We do not understand this statement to mean that factual considerations underlying a panel's finding that a subsidy is de facto export contingent are irrelevant for determining what action must be taken to remove that de facto export contingency. Indeed, the context in which that statement was made by the Australia - Leather Article 21.5 panel was altogether different. In that case, the question of implementation of the DSB's recommendation was addressed, in the first instance by the parties, on the basis of the "subsidy" element, rather than the "export contingency" element, of the prohibited subsidy. Specifically, the parties both made arguments concerning the amount of the subsidy that should be repaid, and Australia based its arguments concerning this point on its interpretation of the panel's original finding of export contingency. The quoted statement of the panel was made in addressing this argument, and we believe was intended to express the view that the basis for the original finding of de facto export contingency was not useful or relevant for calculating the amount of the subsidy to be repaid.

19 First written submission of Brazil (Annex 1-1) at para. 30.

20 Canada - Aircraft, WT/DS70/R, para. 9.340, emphasis in original findings.

21 Oral statement of Canada (Annex 2-3) at para. 45.

22 For the reasons set forth at para. 5.40, we see no reason to draw any inferences concerning Canada's failure to provide the 2000/2001 - 2001/2002 TPC Business Plans, which are still under development.

23 Second written submission of Brazil (Annex 1-2) at para. 35 (emphasis in Brazil's submission). Brazil is referring to the "new" TPC Terms and Conditions, and the "new" TPC Investment Application Guide, at this juncture.

24 First written submission of Canada (Annex 2-1) at para. 34.

25 Pursuant to Article 31 of the SCM Agreement, Articles 8 and 9 of that Agreement applied for an initial period of five years, ending 31 December 1999, and could have been extended beyond that date on the basis of a consensus by the SCM Committee. As of 31 December 1999, no such consensus had been reached.

26 Former footnote 28 provided:
The term "industrial research" means planned search or critical investigation aimed at discovery of new knowledge, with the objective that such knowledge may be useful in developing new products, processes or services, or in bringing about a significant improvement to existing products, processes or services.