WT/DS70/R
14 April 1999
(99-1398)
Original: English
The report of the Panel on Canada – Measures Affecting the Export of Civilian Aircraft is being circulated to all Members, pursuant to the DSU. The report is being circulated as an unrestricted document from 14 April 1999 pursuant to the Procedures for the Circulation and Derestriction of WTO Documents (WT/L/160/Rev.1). Members are reminded that in accordance with the DSU only parties to the dispute may appeal a panel report. An appeal shall be limited to issues of law covered in the Panel report and legal interpretations developed by the Panel. There shall be no ex parte communications with the panel or Appellate Body concerning matters under consideration by the Panel or Appellate Body.
Note by the Secretariat: This Panel Report shall be adopted by the Dispute Settlement Body (DSB) within 30 days after the date of its circulation unless a party to the dispute decides to appeal or the DSB decides by consensus not to adopt the report. If the Panel Report is appealed to the Appellate Body, it shall not be considered for adoption by the DSB until after the completion of the appeal. Information on the current status of the Panel Report is available from the WTO Secretariat.
WT/DS70/R
Page i
TABLE OF CONTENTS
Page
INTRODUCTION 1
FACTUAL ASPECTS 2
FINDINGS AND RECOMMENDATIONS REQUESTED BY THE PARTIES 2
REQUESTS FOR PRELIMINARY RULINGS 4
CANADA’S REQUEST REGARDING THE PANEL’S JURISDICTION 4
General issues 4
Arguments of Canada 4
The Panel’s responsibility to decide the scope of its jurisdiction 4
Request that the Panel should rule before the first submissions of the Parties are due 5
Arguments of Brazil 6
Consistency with Article 4.4 of the SCM Agreement of Brazil’s Request for the Establishment of a Panel with respect to financing by the Export Development Corporation 6
Arguments of Canada 6
Applicable law 7
Whether Brazil requested consultations on financing provided by the Export Development Corporation ...9 (b) Arguments of Brazil 10
(c) Response of Canada. 12
Consistency with Article 6.2 of the DSU and Brazil's request for a Panel in respect of certain claims 12
Arguments of Canada 12
Applicable law 13
The matters at issue 13
Brazil’s request for the establishment of a panel does not meet the requirements of Article 6.2 of the DSU.
....................................................................................................................................................................15
Arguments of Brazil 16
EDC “financing” 17
Definition of “civil aircraft industry” 18
TPC “predecessor programmes” 18
enefits provided under the Subsidiary Agreements and SDI 19
REQUEST OF CANADA FOR PRELIMINARY FINDING ON JURISDICTION: THE SCM AGREEMENT DOES NOT APPLY TO CONTRIBUTIONS AND TRANSACTIONS THAT TOOK PLACE BEFORE THE ENTRY INTO FORCE OF THE WTO AGREEMENT 19
Request of Canada 19
Response of Brazil 20
BRAZIL’S REQUEST REGARDING ADDITIONAL FACT FINDING 20
Arguments of Brazil. 20
Arguments of Canada 21
Background to Brazil’s motion: the consultations, “transaction-specific” information and confidentiality 21
Brazil’s request is inconsistent with WTO law and practice 21
Brazil’s request further supports Canada’s preliminary submissions 25
Response of Brazil 26
Rebuttal of Canada 27
CANADA’S REQUEST FOR A DEADLINE FOR THE SUBMISSION OF NEW EVIDENCE 27
Arguments of Canada 27
Arguments of Brazil. 28
BRAZIL’S REQUEST REGARDING DEADLINE FOR SUBMISSION OF NEW EVIDENCE 28
Arguments of Brazil. 28
Arguments of Canada 29
Actions of the Panel 29
Response of Brazil 29
Proposal by the Panel 31
Joint letter from Brazil and Canada, and Panel action 31
WT/DS70/R
Page ii
ARGUMENTS CONCERNING THE SUBMISSION OF EVIDENCE IN RESPONSE TO PANEL QUESTIONS POSED IN CONNECTION WITH THE SECOND MEETING 31
Comment of Canada 31
Comment of Brazil. 32
PROCEDURES FOR PROTECTION OF BUSINESS CONFIDENTIAL INFORMATION 34
Procedure Proposed by Canada 34
Arguments of Brazil. 36
Advisors of a Party should be considered an approved person 36
Review of Confidential Business Information should be permitted outside the WTO premises 36
Third Parties should be granted access to Confidential Business Information 37
A complete administrative record should be maintained for the Appellate Body 37
Action by the Panel 37
Statements of Canada 38
ARGUMENTS OF THE PARTIES REGARDING LEGAL ISSUES CONCERNING SCM ARTICLES 1 AND 3 39
GENERAL 39
Arguments of Canada 39
Principles of Treaty Interpretation 39
Burden of Proof 40
A complainant must present a prima facie case in respect of each element of its claim 41
The principles applicable to the evaluation of evidence 41
Judicial Economy 42
Response of Brazil 43
“SUBSIDY” PER SCM ARTICLE 1 44
Arguments of Canada 44
The meaning of “financial contribution” 45
The meaning of “benefit” 45
Ordinary meaning 45
Context 45
The object and purpose of the SCM Agreement 46
Arguments of Brazil. 46
Response of Canada 48
“CONTINGENT, IN LAW OR IN FACT, … ON EXPORT PERFORMANCE” 49
Arguments of Canada 49
Brazil’s interpretation of Article 3.1(a) 49
Article 3.1(a) prohibits subsidies that are conditional on or tied to export performance 49
The ordinary meaning of “contingent ... upon export performance” 50
The context of Article 3.1(a) 51
The object and purpose of the SCM Agreement 52
The negotiating history of the SCM Agreement 53
The implications of Brazil’s interpretation 54
The OECD Arrangement on Guidelines for Officially Supported Export Credits (OECD Consensus) 55
The United States misunderstands Canada’s interpretation of Article 3 56
Arguments of Brazil. 58
Response of Canada 61
ARGUMENTS OF PARTIES REGARDING CANADIAN MEASURES ALLEGED BY BRAZIL TO BE PROHIBITED EXPORT SUBSIDIES IN THE SENSE OF SCM ARTICLES 1 AND 3.. 63
EXPORT DEVELOPMENT CORPORATION (“EDC”) 63
General arguments of the parties 63
Alleged concessionary terms of EDC financing; EDC risk level and performance 63
Arguments of Brazil 63
Arguments of Canada 66
Rebuttal of Brazil on EDC’s risk level and performance 70
Response of Canada on EDC’s risk level and performance 72
Contingency on export performance 73
Arguments of Brazil 73
Arguments of Canada 73
Response of Brazil 74
EDC Debt Financing 74
Arguments of Brazil 74
WT/DS70/R
Page iii
Arguments of Canada 76
Substantive issues 76
Arguments regarding whether Brazil has presented a prima facie case against EDC financing 80
Response of Brazil 81
Loan guarantees 82
Arguments of Brazil 82
Arguments of Canada 83
Equity infusions into corporations established to facilitate the export of civil aircraft 84
Arguments of Brazil 84
Arguments of Canada 88
Response of Brazil 90
Residual value guarantees 91
Arguments of Brazil 91
Arguments of Canada 92
Response of Brazil 92
Canada Account 93
Arguments of Brazil 93
Arguments of Canada 95
Comments of Brazil 98
TECHNOLOGY PARTNERSHIPS CANADA (“TPC”) AND PREDECESSOR DEFENCE INDUSTRY PRODUCTIVITY PROGRAMME (“DIPP”) 99
General arguments of the parties 99
Arguments of Brazil 99
Arguments of Canada 100
Subsidy 100
Arguments of Brazil 100
Arguments of Canada 104
Comments of Brazil 106
TPC’s anticipated or expected return is below market 107
TPC’s expected return - “expected” at the time the loan commitment was made - was insufficient, even under Canada’s “cost of funds” test 107
Export contingency 109
Arguments of Brazil 109
Arguments of Canada 111
Comments of Brazil 117
SALE BY ONTARIO OF OWNERSHIP SHARE IN DE HAVILLAND 118
Subsidy 118
Arguments of Brazil 118
Arguments of Canada 120
Response of Brazil 121
Comments of Canada 122
Export contingency 122
Arguments of Brazil 122
Arguments of Canada 123
CANADA-QUÉBEC SUBSIDIARY AGREEMENT ON INDUSTRIAL DEVELOPMENT 124
Subsidy 124
Arguments of Brazil 124
Arguments of Canada 126
Comments of Brazil 126
Export contingency 127
Arguments of Brazil 127
Arguments of Canada 128
Response of Brazil 129
SOCIÉTÉ DE DÉVÉLOPPEMENT INDUSTRIEL DU QUÉBEC (“SDI”) 129
Subsidy 129
Arguments of Brazil 129
Arguments of Canada 130
Comments of Brazil 131
Export contingency 131
Arguments of Brazil 131
Arguments of Canada 132
ARGUMENTS REGARDING THE FINAN REPORT SUBMITTED BY BRAZIL 133
Arguments of Canada 133
WT/DS70/R
Page iv
Response of Brazil 135
The Time Frames Over Which Subsidies are Quantified and Aggregated 135
EDC Equity Infusions 135
Can$87 Million “Investment” in the 70-seat CRJ-700 135
Can$57 Million “’Investment‘” in the Dash 8-400 136
an$100 Million “Investment” in Turboprop Engines 136
The Acquisition and Restructuring of de Havilland 136
The Comparative Treatment of Repayable and Non-Repayable Contributions 136
The Comparative Treatment of Actual and Anticipated Benefits 136
Rebuttal by Canada 137
ARGUMENTS REGARDING THE CLARK REPORT 137
Arguments of Canada 137
Saturation of the Canadian regional turboprop market 137
The 26 CRJs delivered to Air Canada 138
Response by Brazil 139
Rebuttal of Canada 140
Rebuttal of Brazil 143
ARGUMENTS OF THIRD PARTIES 143
EUROPEAN COMMUNITIES 143
Export contingency 143
Annex I (k), first paragraph SCM Agreement 145
SUBMISSION OF THE UNITED STATES 147
Applicability of the SCM Agreement to subsidies provided prior to the entry into force of the WTO Agreement 147
Canada’s interpretation of “export subsidy” 149
The relevance of item (k) of the Illustrative List to export financing activities of the EDC 151
INTERIM REVIEW 152
FINDINGS 154
INTRODUCTION 154
PRELIMINARY ISSUES 155
Jurisdiction of the Panel 155
Article 4.4 of the SCM Agreement 155
Arguments of the parties 155
Evaluation by the Panel 156
Specificity of Brazil's request for establishment of a panel 158
Arguments of the parties 158
Evaluation by the Panel 160
Temporal application of the SCM Agreement 163
Fact-finding by the Panel 164
Arguments of the parties 164
Evaluation by the Panel 165
Procedures governing Business Confidential Information 166
Arguments of the parties 166
Evaluation by the Panel 166
Deadline for the submission of new evidence or allegations 169
Submission of new evidence 170
Submission of new allegations 170
Deadline for the submission of affirmative defences 170
Panel's right to seek information in respect of defences Canada has not made 171
Panel's right to seek information in the absence of a preliminary ruling on the establishment of a prima facie case 172
Panel's right to seek information in the absence of an allegation by the complaining party 173
DEFINITION OF "SUBSIDY" WITHIN THE MEANING OF ARTICLE 1 OF THE SCM AGREEMENT 174
Arguments of the parties 174
Interpretation by the Panel 176
EXPORT DEVELOPMENT CORPORATION 178
Is the EDC programme per se a prohibited export subsidy? 179
WT/DS70/R
Page v
Does the EDC programme as applied provide prohibited export subsidies? 180
EDC debt financing 180
Arguments of the parties 180
Evaluation by the Panel 186
EDC loan guarantees 190
EDC residual value guarantees 191
EDC equity financing 192
CANADA ACCOUNT 193
Is the Canada Account programme per se a prohibited export subsidy? 194
Does the Canada Account programme as applied provide prohibited export subsidies? 195
Is there a prima facie case against Canada Account financing in the regional aircraft sector? 196
Has Canada rebutted the prima facie case? 197
Is the Canada Account debt financing contingent on export? 197
SALE TO BOMBARDIER BY THE ONTARIO AEROSPACE CORPORATION OF A 49 PER
CENT INTEREST IN DE HAVILLAND INC 198
BENEFITS PROVIDED UNDER THE CANADA-QUEBEC SUBSIDIARY AGREEMENTS ON INDUSTRIAL DEVELOPMENT 202
Arguments of the parties 202
Evaluation by the Panel 204
SOCIETE DE DEVELOPPEMENT INDUSTRIEL DU QUEBEC 206
Arguments by the parties 206
Evaluation by the Panel 208
Assistance provided under the IQ programme 208
Assistance provided under the SDI programme 208
TECHNOLOGY PARTNERSHIPS CANADA AND DIPP 209
Does the application of the TPC programme result in subsidies for the Canadian regional aircraft industry? 209
Arguments of the parties 209
$87 million to Bombardier 210
$57 million TPC contribution to de Havilland; and $100 million TPC contribution to Pratt & Whitney .210
$12.7 million TPC contribution to Allied Signal; and $9.9 million TPC contribution to Sextant
Avionique 211
Additional information on the specific TPC contributions identified by Brazil 211
$87 million to Bombardier 211
$57 million TPC contribution to de Havilland; and $100 million TPC contribution to Pratt & Whitney .213
$12.7 million TPC contribution to Allied Signal; and $9.9 million TPC contribution to Sextant
Avionique 214
Evaluation by the Panel 215
Is there a prima facie case that TPC assistance to the Canadian regional aircraft industry takes the
form of subsidies? 215
Has Canada rebutted the prima facie case of subsidization? 215
Is TPC assistance to the Canadian regional aircraft industry contingent on export performance? 217
nterpretation of "contingent … in fact … upon export performance" and "in fact tied to …
anticipated exportation or export earnings" 221
Application of the "but for" test 223
) Conclusion 227
Assistance provided under the DIPP programme 227
CONCLUSIONS AND RECOMMENDATION 227
ANNEX I 229
ANNEX II 235
WT/DS70/R
Page 1
INTRODUCTION
On 10 March 1997, Brazil requested consultations with the Government of Canada pursuant to Article 4 of the Agreement on Subsidies and Countervailing Measures (“SCM Agreement”) regarding certain alleged subsidies granted by the Government of Canada or its provinces that support the export of civilian aircraft from Canada.
Brazil and Canada held consultations on 30 April 1997, but failed to reach a mutually satisfactory solution.
On 10 July 1998, pursuant to Article 4.4 of the SCM Agreement and Article 6 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (“DSU”), Brazil requested the immediate establishment of a panel with standard terms of reference.
At its meeting on 23 July 1998, the DSB established a panel pursuant to the request by Brazil (WT/DS70/2).
At that DSB meeting, parties agreed that the Panel should have standard terms of reference. The terms of reference of the Panel are the following:
"To examine, in the light of the relevant provisions of the covered agreements cited by Brazil in document WT/DS70/2, the matter referred to the DSB by Brazil in that document and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements".
On 16 October 1998, Brazil requested the Director-General to determine the composition of the Panel, pursuant to paragraph 7 of Article 8 of the DSU. This paragraph provides:
"If there is no agreement on the panelists within 20 days after the date of the establishment of a panel, at the request of either party, the Director-General, in consultation with the Chairman of the DSB and the Chairman of the relevant Council or Committee, shall determine the composition of the panel by appointing the panellists whom the Director-General considers most appropriate in accordance with any relevant special or additional rules or procedures of the covered agreement or covered agreements which are at issue in the dispute, after consulting with the parties to the dispute. The Chairman of the DSB shall inform the Members of the composition of the panel thus formed no later than 10 days after the date the Chairman receives such a request."1
On 22 October 1998, the Director-General accordingly composed the panel as follows:
1 Paragraph 12 of Article 4 of the SCM Agreement provides:
"For purposes of disputes conducted pursuant to this Article, except for time-periods specifically prescribed in this Article, time-periods applicable under the DSU for the conduct of such disputes shall be half the time prescribed there."
WT/DS70/R
Page 2
Chairman: Mr. David de Pury Members: Mr. Maamoun Abdel Fattah
Mr. Dencho Georgiev
The European Communities and the United States reserved their rights to participate in the Panel proceedings as third parties.
The Panel met with the parties on 26 and 27 November 1998 and 12 and 13 December 1998. It met with the third parties on 27 November 1998.
The Panel submitted its interim report to the parties on 17 February 1999. Although on 25 February 1999 both parties requested the Panel to review precise aspects of the interim report, neither party requested an additional meeting with the Panel. The Panel submitted its final report to the parties on 12 March 1999.
FACTUAL ASPECTS
This dispute concerns various Canadian measures which Brazil alleges are subsidies inconsistent with Canada’s obligations under Articles 3.1(a) and 3.2 of the SCM Agreement in that they are contingent in law or in fact, whether solely or as one of several other conditions, upon export performance.
The measures as identified in Brazil’s request for a panel are financing and loan guarantees provided by the Export Development Corporation including equity infusions into corporations established to facilitate the export of civil aircraft; support provided to the civil aircraft industry by the Canada Account; funds provided to the civil aircraft industry by Technology Partnerships Canada and predecessor programmes; the sale by the Ontario Aerospace Corporation, an agency or instrumentality of the Government of the Province of Ontario, of a 49 per cent interest in a civil aircraft manufacturer to another civil aircraft manufacturer on other than commercial terms; benefits provided under the Canada-Québec Subsidiary Agreement on Industrial Development; and benefits provided by the Government of Québec under the Société de Développement Industriel du Québec.
FINDINGS AND RECOMMENDATIONS REQUESTED BY THE PARTIES
Brazil requests the Panel to find that the following are subsidies contingent in law or in fact upon export performance and, therefore, are inconsistent with the requirements of Article 3 of the SCM Agreement:
“a) Financing and loan guarantees provided by the EDC, including equity infusions made by the EDC into corporations established to facilitate the export of civil aircraft;
Support provided to the civil aircraft industry by the Canada Account;
Funds provided to the civil aircraft industry by TPC and its predecessor programmes;
The sale to a civil aircraft manufacturer (Bombardier Inc.) by OAC, an agency or instrumentality of the Government of the Province of Ontario, of a 49-per cent interest in another civil aircraft manufacturer (de Havilland, Inc.) on other than commercial terms;
WT/DS70/R
Page 3
Benefits provided under the Canada-Québec Subsidiary Agreement on Industrial Development;
Benefits provided by the Government of Québec under SDI”.
Brazil requests that the panel recommend that “the Dispute Settlement Body request Canada to withdraw all of these subsidies without delay and bring its federal and provincial programmes with regard to support for the civilian aircraft industry into conformity with its obligations under the Agreement on Subsidies and Countervailing Measures”.
Canada requests that the Panel make preliminary findings prior to the deadline for Brazil’s first written submission that:
“Brazil’s request for a Panel with respect to financing provided by the Export Development Corporation is inconsistent with SCM Agreement Article 4.4 because no consultations were requested with respect to this claim, and therefore, Brazil’s claims regarding financing provided by the Export Development Corporation fall outside the jurisdiction of the Panel”;
“Brazil’s request for Panel does not ‘identify the specific measures at issue’, as required by Article 6.2 of the DSU with respect to the following items:
financing provided by the Export Development Corporation;
funds provided to the civil aircraft industry by Technology Partnerships Canada and predecessor programmes;
benefits provided under the Canada-Québec Subsidiary Agreement on Industrial Development; and
benefits provided by the Government of Québec under the Société de Développement Industriel du Québec.
Therefore, Brazil’s claims pertaining to these measures fall outside the jurisdiction of the Panel.”
Canada additionally requests that the Panel rule that:
“the complaining Party may not adduce new allegations or evidence after the end of the first substantive meeting of the Panel with the Parties”;
“pursuant to Article 28 of the Vienna Convention on the Law of Treaties (‘Vienna Convention’), contributions made prior to the entry into force of the WTO Agreement fall outside the jurisdiction of the Panel. Specifically, Canada requests that the Panel find and conclude that contributions made in 1989 by the Defence Industry Productivity Programme (DIPP) and the Canada-Québec Subsidiary Agreement fall outside the jurisdiction of the Panel”.
Canada also requests that the Panel dismiss Brazil’s claims.
WT/DS70/R
Page 4
REQUESTS FOR PRELIMINARY RULINGS2
CANADA’S REQUEST REGARDING THE PANEL’S JURISDICTION
General issues
Arguments of Canada
Canada requested preliminary rulings with respect to two issues regarding the Panel’s jurisdiction. The first issue is whether certain measures identified in Brazil’s request for a panel had been the subject of a request for consultations, and thus could be the subject of a panel request under SCM Article 4.4. The second issue is whether certain of the measures identified in Brazil’s request for a panel were sufficiently specific as required by Article 6.2 of the DSU. Canada asked the Panel to issue the requested rulings prior to the deadline for the parties’ first submissions.
The Panel’s responsibility to decide the scope of its jurisdiction
Canada states that a Panel’s terms of reference establish the jurisdiction of the panel by defining the precise claims at issue in the dispute,3 and notes that in the present case, the terms of reference are set out in the complainant’s request for the establishment of the panel. Canada argues that the Panel must examine Brazil’s request to ensure that it complies with the DSU Article 6.2, since it determines the scope of the Panel’s jurisdiction, and quotes the Appellate Body:
“… it is incumbent upon a panel to examine the request for the establishment of the panel very carefully to ensure its compliance with both the letter and the spirit of Article 6.2 of the DSU. It is important that a panel request be sufficiently precise for two reasons: first, it often forms the basis for the terms of reference of the panel pursuant to Article 7 of the DSU; and, second, it informs the defending party and the third parties of the legal basis of the complaint.”4 [emphasis added by Canada]
Canada submits that this reasoning applies equally to compliance with Article 4 of the SCM Agreement, as Article 4 also limits the jurisdiction of the Panel in this dispute. For Canada, the Panel may not look to the submissions of the complainant to cure uncertainties in the panel request,5 nor may the Panel adopt a claim pursuant to its own ruling:6 a Panel may not assume jurisdiction that it
2 Except as otherwise noted, the footnotes and citations, and the emphasis in the text are as contained in the parties’ submissions.
3 Brazil – Measures Affecting Desiccated Coconut (Complaint by the Phillippines) (1997), WTO Doc.
WT/DS22/AB/R at 21 (Appellate Body Report) hereinafter Brazil – Desiccated Coconut.
4 European Communities - Regime for the Importation, Sale and Distribution of Bananas (Complaints by Ecuador, Guatemala, Honduras, Mexico and the United States) (1997) WTO Doc. WT/DS27/AB/R at
para. 142 (Appellate Body Report) hereinafter European Communities - Bananas.
5 Id.. at para. 143:
“We do not agree with the Panel that ‘even if there was some uncertainty whether the panel request had met the requirements of Article 6.2, the first written submission of the Complainants “cured” that uncertainty because their submissions were sufficiently detailed to present all the factual and legal issues clearly.’ Article 6.2 of the DSU requires that the claims, but not the arguments, must all be specified sufficiently in the request for the establishment of a panel in order to allow the defending party and any third parties to know the legal basis of the complaint. If a claim is not specified in the request for the establishment of a panel, then a faulty request cannot be subsequently ‘cured’ by a complaining party’s argumentation in its first written submission to the panel or in any other submission or statement made later in the panel proceeding.”
6India - Patent Protection for Pharmaceutical and Agricultural Chemical Products (Complaint by the United States) (1997) WTO Doc. WT/DS50/AB/R at para. 92 (Appellate Body Report) [hereinafter India – Pharmaceuticals]:
WT/DS70/R
Page 5
does not have. Canada states that in accordance with the views of the Appellate Body, recent panels have examined their jurisdiction and ruled that certain claims did not fall within it.7
Request that the Panel should rule before the first submissions of the Parties are due
Canada argues that although there are no explicit panel working procedures that govern preliminary requests for findings,8 there is an evolving practice of WTO panels and the Appellate Body concerning preliminary rulings. According to Canada, preliminary issues of an organisational nature have been dealt with before the first substantive hearing of the parties with the panel or the Appellate Body: in European Communities - Bananas, the panel issued a preliminary ruling on the participation of third parties prior to the first substantive meeting;9 and more recently, in the case of United States – Import Prohibition of Certain Shrimp and Shrimp Products,10 the Appellate Body issued a preliminary decision four days after an objection was raised by the appellees in their written submissions concerning the admissibility of amicus curiae briefs. 11
Canada submits that Indonesia –Automobiles provides the most recent example, where in its first written submission, Indonesia argued as a preliminary matter that a claim made in the first submission of the United States was not within the terms of reference of the panel. Canada recalls that the panel considered and ruled on Indonesia’s objection at the first substantive meeting with the parties.12
For Canada, the common theme in these decisions is that many kinds of preliminary issues should be decided at an early stage in the panel proceedings. Canada submits that the issues of
“We note also the Panel's statement that it "ruled, at the outset of the first substantive meeting held on 15 April 1997, that all legal claims would be considered if they were made prior to the end of that meeting; and this ruling was accepted by both parties". We do not find this statement at all persuasive in advancing the argument made by the United States on this issue. Nor do we find this statement consistent with the letter and the spirit of the DSU. Although panels enjoy some discretion in establishing their own working procedures, this discretion does not extend to modifying the substantive provisions of the DSU. To be sure, Article 12.1 of the DSU says: "Panels shall follow the Working Procedures in Appendix 3 unless the panel decides otherwise after consulting the parties to the dispute". Yet that is all that it says. Nothing in the DSU gives a panel the authority either to disregard or to modify other explicit provisions of the DSU. The jurisdiction of a panel is established by that panel's terms of reference, which are governed by Article 7 of the DSU. A panel may consider only those claims that it has the authority to consider under its terms of reference. A panel cannot assume jurisdiction that it does not have. In this case, Article 63 was not within the Panel's jurisdiction, as defined by its terms of reference. Therefore, the Panel had no authority to consider the alternative claim by the United States under Article 63.”
7 See for example, Japan – Measures Affecting Consumer and Photographic Film and Paper
(Complaint by the United States) (1998), WTO Doc. WT/DS44/R at para. 10.21 (Panel Report) hereinafter Japan - Film; Indonesia – Certain Measures Affecting the Automobile Industry (Complaints by Japan, European Communities, United States) (1998), WTO Doc. WT/DS54,55,59,64/R at para. 14.3 (Panel Report)
hereinafter Indonesia - Automobiles.
8 In European Communities – Bananas, at para. 144, the Appellate Body noted:
“We note, in passing, that this kind of issue could be decided early in panel proceedings, without causing prejudice or unfairness to any party or third party, if panels had detailed, standard working procedures that allowed, inter alia, for preliminary rulings.”
9 European Communities - Regime for the Importation, Sale and Distribution of Bananas (Complaints by Ecuador, Guatemala, Honduras, Mexico, United States) (1997) WTO Doc. WT/DS27/ECU at para. 7.6
(Panel Report).
10 United States – Import Prohibition of Certain Shrimp and Shrimp Products (Complaints by India, Malaysia, Pakistan, Thailand) (1998), WTO Doc. WT/DS58/AB/R hereinafter United States - Shrimp.
11 Id.. at para. 83.
12 Indonesia –Automobiles, at para. 14.3.
WT/DS70/R
Page 6
jurisdiction identified here are precisely these kinds of issues because an early decision would best fulfil the objectives of due process, quoting the Appellate Body:
“… terms of reference fulfil an important due process objective -- they give the parties and third parties sufficient information concerning the claims at issue in the dispute in order to allow them an opportunity to respond to the complainant's case.”13
Canada submits that certain claims made by Brazil in its request for the establishment of a Panel are not within the Panel’s jurisdiction because they are not identified with sufficient precision or they were not the subject of consultations. Canada argues that it must be given sufficient notice and specificity concerning Brazil’s claims to allow Canada an opportunity to respond to Brazil’s case, and that it follows that if certain claims are not properly within the Panel’s jurisdiction, the Panel should so rule before the Parties’ first submissions are due, to give both Canada and Brazil adequate notice of the actual claims in dispute, and to allow them adequate time to prepare their submissions accordingly.
Arguments of Brazil
Brazil agrees with Canada’s statements, consistent with the observations of the Appellate Body in "Brazil - Measures Affecting Desiccated Coconut"14 and "European Communities - Regime for the Importation, Sale and Distribution of Bananas",15 that it is for the Panel to ensure compliance with Article 6.2 of the DSU and, by analogy, Article 4 of the SCM Agreement. Brazil also agrees that it is within the Panel’s authority to make preliminary rulings on objections raised by the parties. In Brazil’s view, however, there is no specific requirement nor any established practice, requiring a panel to rule on preliminary objections concerning the Panel’s jurisdiction before the first submissions of the parties. Brazil considers, however, that it may be in the best interests of the parties for the Panel to expedite its ruling on the questions raised by Canada.
Consistency with Article 4.4 of the SCM Agreement of Brazil’s Request for the Establishment of a Panel with respect to financing by the Export Development Corporation
Arguments of Canada
Canada states that although Brazil’s request for the establishment of a panel includes financing provided by the Export Development Corporation, no previous consultations were requested with respect to this item. Canada requests, therefore, that the Panel make a preliminary finding that any claim relating to financing provided by the EDC is not within its terms of reference, and therefore cannot be addressed by the Panel.
13 Brazil – Desiccated Coconut, at 21.
14 (1997) WTO Doc. WT/DS22/AB/R.
15 (1997) WTO Doc. WT/DS27/AB/R.
WT/DS70/R
Page 7
Applicable law
Canada states that as required by the SCM Agreement, a request for a panel must be in respect of a matter that has already been the subject of consultations. Canada notes that Article 4, paragraphs 1 to 4 provide that:
Whenever a Member has reason to believe that a prohibited subsidy is being granted or maintained by another Member, such Member may request consultations with such other Member.
A request for consultations under paragraph 1 shall include a statement of available evidence with regard to the existence and nature of the subsidy in question.
Upon request for consultations under paragraph 1, the Member believed to be granting or maintaining the subsidy in question shall enter into such consultations as quickly as possible. The purpose of the consultations shall be to clarify the facts of the situation and to arrive at a mutually agreed solution.
If no mutually agreed solution has been reached within 30 days of the request for consultations, any member party to such consultations may refer the matter to the Dispute Settlement Body (‘DSB’) for the immediate establishment of a panel, unless the DSB decides by consensus not to establish a panel. (emphasis added)
Canada states that in Article 4.4, the ordinary meaning of “the matter” in the light of the context and object and purpose of the SCM Agreement can only refer to an alleged “prohibited subsidy” on which a Member has requested consultations in accordance with Article 4.1 and the “subsidy in question” referred to in Articles 4.2 and 4.3. The “matter”, Canada submits, cannot be measures that have not been the subject of consultations. Rather, for Canada, the “matter” referred to in Article 4.4 is one on which “such consultations” have been requested, and not simply any other matter on which a WTO Member may consider worthy of adjudication by the panel.
Canada notes that under the DSU, the consultations step in the dispute settlement process is not a mere formality,16 and that SCM Agreement Article 4.3 provides that the responding Member shall enter into consultations, for the purpose of arriving at a mutually agreed solution. Canada views consultations under the SCM Agreement Article 4 as more onerous than under the DSU: Article 4.2 requires that in its request for consultations the Member must provide a statement of available evidence with regard to the existence and nature of the subsidy in question.
For Canada, the question is the nature, and strength, of the connection that has to exist between an alleged “prohibited subsidy” subject to a request for consultations and a “matter” subject to a request for a panel.17 Canada submits that there must be a rational connection between the
16 DSU Article 4.3 provides, inter alia, that if a request for consultations is made under a covered agreement, the Member to which the request is made shall enter into consultations in good faith, with a view to reaching a mutually satisfactory solution.
17 The issue here is not what took place during the consultations between Canada and Brazil: WTO panels have no mandate to inquire into the adequacy of consultations. The issue of the adequacy of consultations arose before the panels in European Communities - Bananas, and more recently in Korea – Taxes on Alcoholic Beverages (Complaints by European Communities and United States) (1998), WTO Doc.
WT/DS75,84/R (Panel Report)hereinafter Korea – Liquor Taxes. In the latter case, the Panel noted the report in European Communities - Bananas, and agreed that it was impossible for a Panel to adjudicate on what occurred in the consultations (at para. 10.19):
“In our view, the WTO jurisprudence so far has not recognized any concept of ‘adequacy’ of consultations. The only requirement under the DSU is that consultations were in fact held, or were at least requested, and that a period of sixty days has elapsed from the time consultations were requested
WT/DS70/R
Page 8
request for consultations (the “subsidy in question”) and the “matter” to be the subject of dispute settlement; the “matter” must necessarily follow from the “subsidy in question.” For Canada, the SCM Agreement explicitly requires these connections, as do the implicit requirements of due process. Canada cites the Appellate Body:
“All parties engaged in dispute settlement under the DSU must be fully forthcoming from the very beginning both as to the claims involved in a dispute and as to the facts relating to those claims. Claims must be stated clearly. Facts must be disclosed freely. This must be so in consultations as well as in the more formal setting of panel proceedings. In fact, the demands of due process that are implicit in the DSU make this especially necessary during consultations. For the claims that are made and the facts that are established during consultations do much to shape the substance and the scope of subsequent panel proceedings.”18 [emphasis added by Canada]
For Canada, this analysis is equally applicable to requests for consultations; Article 4.4 is not a licence for a fishing expedition. In Canada’s view, the SCM Agreement is not observed, and due process is not served, if Canada is forced to respond to a “matter” on which no consultations were requested and none took place.
In response to a question from the Panel, Canada further argues that whether a matter has been raised in the consultations is not relevant to the determination of whether it is properly within the request for a panel in accordance with Article 4.4 of the SCM Agreement, for two reasons, the first being its argument (para. 4.13) that the measures identified in the request for a panel must be rationally connected to the “prohibited subsidy” identified in the request for consultations and must flow directly from that alleged prohibited subsidy, and that the scope of the request for consultations is determined, in turn, by the requirements of Article 4.2 of the SCM Agreement.
In view of the above, according to Canada, whether a “matter” is raised during consultations is immaterial to the question of whether a request for a panel is consistent with Article 4.4. It is not the content or the context of the consultations that is governed by Article 4.2 and 4.4, but rather the content of the request for consultations, which launches a dispute, and the content of the request for a panel, which establishes the Panel’s terms of reference.
Canada’s second reason is that WTO panels have consistently held that the substantive merits of consultations are impossible to police and therefore to discipline by the WTO19, as the WTO Secretariat is not present during these consultations and, indeed, consultations are generally confidential. In Canada’s view, introduction of evidence relating to consultations would have three consequences that would be deleterious to the dispute settlement system of the WTO:
to the time a request for a panel was made. What takes place in those consultations is not the concern of a panel. The point was put clearly by the Panel in Bananas III, where it was stated:
‘[7.19] Consultations are . . . a matter reserved for the parties. The DSB is not involved; no panel is involved; and the consultations are held in the absence of the Secretariat. While a mutually agreed solution is to be preferred, in some cases it is not possible for parties to agree upon one. In those cases, it is our view that the function of a panel is only to ascertain that the consultations, if required, were in fact held. ...’
We do not wish to imply that we consider consultations unimportant. Quite the contrary, consultations are a critical and integral part of the DSU. But, we have no mandate to investigate the adequacy of the consultation process that took place between the parties and we decline to do so in the present case.”
18 India - Pharmaceuticals, at para. 94.
19 European Communities - Regime for the Importation, Sale and Distribution of Bananas (Complaints by Ecuador, Guatemala, Honduras, Mexico, United States) (1997) WTO Doc. WT/DS27/ECU at paras. 7.17-
(Panel Report). See also Korea - Taxes on Alcoholic Beverages (1998) WTO Doc. WT/DS75,84/R (Panel Report) at paras 10.17-10.19.
WT/DS70/R
Page 9
in the event of a disagreement between the two sides as to what took place, a panel would be put in the impossible position of determining which set of notes taken by the Parties to a dispute represent a closer version of the truth;
in the event of multiple complainants that have held separate consultations, a panel would be put in the difficult position of determining which statements during which set of consultations have more probative weight; and
consultations are the first step in litigation in the WTO dispute settlement process. They are also, in many instances, the last step in the process of diplomatic negotiations to avoid litigation. The introduction of evidence of the consultations would, in Canada’s submission, inhibit a full discussion of the facts and issues and thereby render consultations a formality.
Whether Brazil requested consultations on financing provided by the Export Development Corporation
Canada notes that Brazil’s request for consultations dated 10 March 1997, mentioned in part:
“1. Export Development Corporation (EDC) equity infusions into corporations specially established to facilitate the export of aircraft;
EDC loan guarantees for exported aircraft; …”20
Canada states that in a letter of 19 March 1997 acknowledging Brazil’s request for consultations, Canada stated:
“…as the Government of Brazil is aware, to ensure meaningful consultations and to encourage Members to reach mutually satisfactory solutions to disagreements, Article 4.2 of the [SCM] Agreement requires that a request for consultations ‘include a statement of available evidence with regard to the existence and nature of the subsidy in question.’ The Government of Canada notes that the request from the Government of Brazil relies only on ‘information available’ to the Government of Brazil that merely ‘suggests’ inconsistencies; the request from the Government of Brazil provides no statement of evidence in respect of any of the measures or programmes listed.
Accordingly, with a view to clarifying the facts of the situation and arriving at a mutually agreed solution, the Government of Canada would respectfully request that the Government of Brazil provide to the Government of Canada the available evidence required to be provided under Article 4.2 of the Agreement in advance of the consultations.”21
According to Canada, Brazil’s response of 2 April 199722 mentioned only “[EDC] equity infusions into special purpose corporations” and EDC loan guarantees. Thus, regarding financing activity by the EDC, Canada argues, Brazil provided no notice at all, let alone a statement of evidence as required by SCM Article 4.2.
20Request for consultations by Brazil, WTO document WT/DS70/1, G/SCM/D12/1, 14 March 1997.
21Letter from the Government of Canada to the Government of Brazil dated 19 March 1997. [TAB D, 23 October 1998 preliminary submission of Canada]
22 Letter from the Government of Brazil to the Government of Canada dated 2 April 1997. [TAB E 23 October 1998 preliminary submission of Canada]
WT/DS70/R
Page 10
Canada maintains that EDC’s financing activities (direct lending) are distinct and different from EDC’s guarantee activities, stating that loans and loan guarantees are clearly distinguished in EDC’s annual reports, which are public documents and were readily available to Brazil at the time it made its request for consultations.23 Canada asserts that Brazil recognised the distinction between financing and loan guarantees at the time it made its request for the establishment of a panel, but had elected to request consultations on one and not the other activity of EDC.
Canada states that similarly, equity infusions are distinct from loans,24 and that therefore financing provided by the EDC, though related to EDC’s other activities, is legally and factually distinct from loan guarantees or equity infusions.
Canada therefore submits that financing provided by the EDC was not “a prohibited subsidy” on which Brazil requested consultations in accordance with SCM Agreement Article 4.1, and was not a “subsidy in question” within the meaning of SCM Agreement Articles 4.2 and 4.3. Consequently, for Canada financing provided by the EDC cannot be the “matter” referred to the DSB under SCM Agreement Article 4.4, and as such is not within the Panel’s terms of reference and falls outside the Panel’s jurisdiction.
Arguments of Brazil
Brazil requests that the Panel reject Canada’s request to find Brazil’s request for review of EDC financing inconsistent with Article 4.4 of the SCM Agreement. Brazil contends that its identification of the EDC in its consultation request, combined with the language of that request (“ . . . certain subsidies granted by the Government of Canada or its provinces that support the export of civilian aircraft from Canada”), covered all subsidy aspects of the EDC. In Brazil’s view, a request for consultations need not set forth an exhaustive list of questionable measures, particularly where, as here, the very non-transparency of the system makes compilation of such a list virtually impossible. To Brazil, such a requirement would reward non-transparent practices and place an unreasonable burden on other Members.
Moreover, Brazil asserts, the full range of EDC support for exports of civilian aircraft, including what Canada now terms as “EDC’s financing activities (direct lending),” in fact was subject to consultations. According to Brazil, in the consultations it explicitly asked about all aspects of EDC support, including financing, and therefore any claim that Canada’s due process rights will now somehow be violated if it is “forced to respond” to claims concerning EDC “financing activities (direct lending)” is without any basis in fact.
In response to a Panel request for evidence to substantiate the range of subjects on which consultations were held, Brazil notes that direct evidence of the range of actual discussion during consultations usually is not available, and is not available in this case. Transcripts of consultation meetings are not kept, and were not in fact kept in these proceedings. Brazil directs the Panel’s attention to an article submitted by Brazil25 which Brazil asserts was the specific impetus behind Brazil’s request and, along with information from EMBRAER (the Brazilian aircraft manufacturer), informed Brazil’s agenda for consultations with Canada in April 1997. Brazil notes that the article refers, with respect to EDC, to: “’. . . direct loans and other export assistance available to non- Canadian purchasers of Canadian aircraft . . .’”; “’. . . export credit insurance, guarantee services and buyer credits; the latter usually by means of direct loans. . . .’”; and “’. . . bridging the gap between equity and senior debt available from other sources. . . .’”
23 Export Development Corporation, 1996 Annual Report at 38, 43, 68. [TAB F 23 October 1998 preliminary submission of Canada]
24 SCM Agreement Article 1.1(a)(1)(i) lists loans and equity infusions as distinct forms of financial contribution.
25 Exh. BRA-18
WT/DS70/R
Page 11
Brazil indicates that it also had at its disposal EDC’s response to an inquiry of Ministry, submitted by Brazil,26 which detailed “loans and loan guarantees to foreign buyers in support of the sale of Canadian goods and services by” Bombardier, de Havilland, Canadair, Boeing of Canada and others.27 Brazil maintains that this evidence formed the basis for Brazil’s consultation request, and framed Brazil’s agenda for the consultations themselves, and that it is evidence of the range of issues discussed at consultations, which included the full range of EDC financing, including debt, equity and guarantees.
According to Brazil, the fact that its request focused upon the use of equity financing and loan guarantees as particularly egregious infringements of Canada’s obligations under the SCM Agreement speaks only to the broad press coverage granted EDC’s incorporation of a new subsidiary, Exinvest, Inc., as a new vehicle for using these devices to support exports of civilian aircraft. As an example, Brazil provides an article from Transport Finance which illustrates and describes the structure and operation of EDC’s equity financing vehicle.28 According to Brazil, while the request for consultations may have emphasized this aspect of EDC’s activities, the request and the consultations covered all aspects of EDC financing, including direct lending.
Additionally, Brazil contends, although Canada claims in its preliminary submission that there is a marked legal and factual distinction between loan guarantees, equity infusions and “financing,”29 recourse to the common definition of the term proves instructive. Black’s Law Dictionary defines the verb “finance” in the following terms:
to supply with funds through the payment of cash or issuance of stocks, bonds, notes, or mortgages; to provide with capital or loan money as needed to carry on business.
In Brazil’s view, this definition suggests that “financing” does not imply a distinct form of “direct lending,” but is a broader, more general term encompassing direct lending, debt and equity support. For Brazil, all of these types of support together constitute “financing.” Brazil cites the Panel in Japan - Measures Affecting Consumer Photographic Film and Paper that “the requirements of Article 6.2 would be met in the case of a ‘measure’ that is subsidiary or so closely related to a ‘measure’ specifically identified, that the responding party can reasonably be found to have received adequate notice of the scope of the claims asserted by the complaining party.”30 Brazil contends that all of these types of EDC support are “so closely related to one another” that, coupled with the fact that they were all discussed in consultations, Canada must be held to have received “adequate notice” of Brazil’s claims.
Brazil also notes Canada’s own statement that it “does not suggest that the initial request for consultations and the request for a Panel must be identical” (para. 4.55), and recalls Canada’s observation of the Appellate Body’s statement in India - Patent Protection for Pharmaceutical and Agricultural Chemical Products that “the claims that are made and the facts that are established during consultations do much to shape the substance and the scope of subsequent panel proceedings.”31 For Brazil, the whole point of consultations, according to the Panel in Korea - Taxes on Alcoholic Beverages, “is to enable the parties to gather correct and relevant information,”32 and the fact that Brazil would refine its knowledge of EDC’s operations as a result of consultations with
26 Exh. BRA-28
27 Exh. BRA-69, which is an extract from the Canadian Parliamentary reporter, Hansard, demonstrates that responses to the inquiry of Ministry included as Exh. BRA-28 were tabled in Parliament on 16 September 1996.
28 Transport Finance No. 74, 3 October 1996, Exh. BRA-18.
29 Government of Canada, Preliminary Submission, dated 23 October 1998, at para. 34.
30 (1998) WTO Doc. WT/DS44/R, at para. 10.8.
31 (1997) WTO Doc. WT/DS50/AB/R, at para. 94.
32 (1998) WTO Doc. WT/DS75, 84/R, at para. 10.23.
WT/DS70/R
Page 12
Canada should come as no surprise, and in fact is consistent with the very purpose for those consultations.
Response of Canada
Canada rejects Brazil’s argument that:
“its identification of the EDC in its consultation request, combined with the language of that request (“ … certain subsidies granted by the Government of Canada […] that support the export of civilian aircraft for Canada”) covered all subsidy aspects of the EDC.”
Canada argues that under the SCM Agreement, requests for consultation are subject to specific requirements. Article 4.2 provides, for example, that a “request for consultations under paragraph 1 shall include a statement of available evidence with regard to the existence and nature of the subsidy in question.” The “prohibited subsidy” in Article 4.1 is the subsidy with respect to which such a statement must be made.
For Canada, if Brazil is correct that its general statement about “certain subsidies” and EDC practices established the scope of its request for consultation – so that EDC “financing” would be captured – then, in view of its letter of 2 April 1997 its entire request for consultation is inconsistent with Article 4.2 and the case must be dismissed in its entirety.
Canada also rejects Brazil’s argument that “financing” does not imply a distinct form of “direct lending”, but is “a broader, more general term encompassing direct lending, debt and equity support,” and that its request for a panel on “financing by the EDC” is consistent with Article 4.4. For Canada, if Brazil’s reference to equity infusion and loan guarantees in its request for consultations were also references to EDC financing, then its request for consultations would be completely inconsistent with the requirements of Article 4.2, as it did not include a statement of evidence, as required by that Article, of “financing” by the EDC, and therefore Brazil’s case, based on its own arguments, must fail ab initio.
Canada does not request such an action, but rather that Brazil’s case be narrowed to include only those prohibited subsidies on which consultations had been requested – that is, only those with respect to which Brazil had provided Canada with a statement of evidence as required by Article 4.2.
Responding at the request of the Panel to Brazil’s statement that “the full range of EDC support … was subject to consultations”, Canada reiterates that Brazil’s initial request for consultations consisted of a series of vague allegations and therefore was not consistent with the requirements of Article 4.2 of the SCM Agreement. Canada agreed to enter into consultations, but requested that Brazil provide it a statement of evidence as required by Article 4.2 of the SCM Agreement. Canada argues that Brazil’s letter of 2 April 1997 did not in any way include a statement of evidence relating to the “full range of EDC support for the exports of civilian aircraft.” Accordingly, such full range of support was not, in Canada’s view, subject to consultations.
Consistency with Article 6.2 of the DSU and Brazil's request for a Panel in respect of certain claims
Arguments of Canada
Canada submits that Brazil’s request for Panel is inconsistent with Article 6.2 of the DSU in that it does not “identify the specific measures at issue”, as required by Article 6.2 of the DSU with respect to the following items:
WT/DS70/R
Page 13
financing and loan guarantees provided by the Export Development Corporation … [emphasis added by Canada]
funds provided to the civil aircraft industry by Technology Partnerships Canada and predecessor programmes; …[emphasis added by Canada]
benefits provided under the Canada-Québec Subsidiary Agreement on Industrial Development; [emphasis added by Canada]
benefits provided by the Government of Québec under the Société de Développement Industriel du Québec. [emphasis added by Canada]
For Canada, these items in the request are too vague to provide Canada with sufficient information concerning the claims at issue, prejudicing Canada’s due process right to know the case against it. Canada requests, in accordance with Article 6.2 of the DSU and principles of due process, that the Panel make a preliminary finding that Brazil’s request is inconsistent with Article 6.2 of the DSU with respect to these items.
Applicable law
Canada cites language from DSU Article 6.2:
“The request for the establishment of a panel shall … identify the specific measures at issue and provide a brief summary of the legal basis of the complaint sufficient to present the problem clearly.” [emphasis added by Canada]
Canada recalls the due process objectives the Appellate Body identified in Brazil – Desiccated Coconut, and European Communities – Bananas (paras. 4.2 and 4.6), and states that in the latter case, the Appellate Body also found that a panel could not cure defects in a panel request (see footnote 5, supra).
Canada submits that the reasoning of the Appellate Body is applicable to the case where a defending party is not given specific information as to the factual basis of the claims against it, and that this is so a fortiori in a process subject to an accelerated time-table. The time allotted to a defending Party is simply not adequate from a due process perspective, in Canada’s view, for the preparation of a credible defence against a vague accusation, and any clarification that might be provided in Brazil’s first submission cannot, therefore, “cure” the fundamental inconsistency of the request for the establishment of a panel with Article 6.2.
The matters at issue
Canada states that with respect to the items identified in paragraph 4.38, Brazil’s request does not provide any precision, and Canada simply does not know the case against it.
“Financing” provided by the Export Development Corporation
Canada claims to be at a loss as to the scope of this heading of the request for the establishment of a panel, having noted that this was not a matter on which consultations had been requested. For Canada, nothing in the course of consultations could therefore guide it as to the nature of the case against it.
Canada notes the less-than-two-week period for it to respond to Brazil’s first submission, arguing that given the accelerated procedure of Article 4, it is manifestly contrary to Canada’s due
WT/DS70/R
Page 14
process rights to be forced to respond to a legal claim that could potentially cover hundreds of clients, many thousands of financing transactions over several years, and a portfolio of Can$10 billion.
“Civil aircraft industry”
Canada submits that in the particular circumstances of this case, especially in view of the accelerated time-table under which this panel operates, “civil aircraft industry” does not meet the requirement for specificity under Article 6.2 of the DSU.
For Canada, the civil aircraft industry is characterised by a complex network of forward and backward relationships between firms in many different industrial sectors that, nevertheless, because of specialisation, may be considered part of the civil aircraft sector.33 It thus includes firms ranging from machine shops and metal treatment facilities to those involved in advanced instrumentation and communications equipment.34 In Canada, this comprises over 200 enterprises employing 38,000 workers.35
Canada states that Brazil is not unaware of this complexity, as during the last set of bilateral negotiations to resolve this dispute that immediately preceded the request for the establishment of this panel, Brazil provided Canada with an all-encompassing definition:
“Aircraft Industry means the industry producing any and all types of aircraft and its related suppliers of engines, systems, parts, components, materials and services through a company or any other related entity engaged in such industry of a Party, including but not limited to corporations, joint ventures, special purpose companies (SPCs) and any related ‘off-balance sheet’ companies, as well as subsidiaries and affiliates and any companies that are directly or indirectly controlled by said companies or entities or that are under its direct or indirect control. This definition does not apply to helicopters, {military aircraft} nor to any item related here above which individually costs represents [sic] less than 2 per cent of the standard sales price of the aircraft in which such item is used.”36
For Canada, this over-broad description puts Canada in an impossible position in preparing a defence to this claim.
Canada cites the Korea – Liquor Taxes37 dispute, in which Korea argued that the request for the establishment of a panel--which referred to products under HS heading 2208--was not specific enough to satisfy the requirements of Article 6.2, and that each distilled alcoholic beverage had to be specifically identified. The complainants responded that the appropriate imported product was all distilled beverages and simply identifying the HS heading was appropriate.
33 Industry Canada, Aerospace and Defence: Part C, online: Industry Canada homepage http://strategis.ic.gc.ca/SSG/ad01475e.html (last modified: 8 December 1995). [TAB G 23 October 1998 preliminary submission of Canada]
34 Id..
35 Industry Canada, Aerospace and Defence: Part B, online: Industry Canada homepage http://strategis.ic.gc.ca/SSG/ad01541e.html (last modified: 16 May 1997). [TAB H 23 October 1998 preliminary submission of Canada]
36 Government of Brazil, Proposed Memorandum of Understanding between the Government of Brazil
and the Government of Canada on Exchange of Information on the Aircraft Industry, 26 June 1998. [TAB J 23 October 1998 preliminary submission of Canada]
37 Korea – Liquor Taxes..
WT/DS70/R
Page 15
According to Canada, the panel, noting the importance of ensuring that the panel request met the criteria set out in Article 6.2, observed that the issue required the weighing of evidence, particularly in view of the previous findings made in Japan – Taxes on Alcoholic Beverages38:
“While it is possible that in some cases, the complaint could be considered so vague and broad that a respondent would not have adequate notice of the actual nature of the alleged discrimination, it is difficult to argue that such notice was not provided here in light of the identified tariff heading and the Appellate Body decision in the Japan - Taxes on Alcoholic Beverages II.”39
For Canada, in the present dispute there is no identified tariff heading and no previous decision to guide Canada as to what Brazil might mean by the “civil aircraft industry”, and accordingly, in view of the short time-lines of the Article 4 process and in view of the nature of Article 3 as an outright prohibition, Canada submits that it is contrary to its due process rights as protected by the DSU to be forced to respond to cases of which it has been given no specific notice, contrary to the explicit requirements of the DSU.
“Predecessor programmes”
Canada submits that “predecessor programmes” does not identify with adequate specificity the programmes that, Brazil alleges, are inconsistent with Article 3. Canada states that it does not know which “predecessor” programmes to Technology Partnerships Canada are being challenged, nor which activities or transactions are being challenged.
Benefits provided under the Canada-Québec Subsidiary Agreement and the Société de Développement Industriel du Québec
Canada argues that the last two headings of the request are also inconsistent with Article 6.2, in that “benefits provided” does not provide any measure of specificity as required by Article 6.2. For Canada, it is not clear which aspect of these programmes, or which activities or transactions under these programmes, Brazil considers to have conferred a “benefit”, and thus the request does not give Canada any clue as to the scope of the challenge against it.
Brazil’s request for the establishment of a panel does not meet the requirements of Article 6.2 of the DSU
Canada argues that Brazil’s request for a Panel is far vaguer than its initial request for consultation.. Canada does not suggest that the initial request for consultations and the request for a Panel must be identical, recalling the observation by the Appellate Body that “the claims that are made and the facts that are established during consultations do much to shape the substance and the scope of subsequent panel proceedings,”40 and by the Panel in Korea – Taxes on Alcoholic Beverages that “in our view, the very essence of consultations is to enable the parties to gather correct and relevant information.”41
According to Canada, consultations thus afford the Parties the opportunity to give more precision to the dispute, not less; additional facts gathered in the consultations phase should give substance to a complaining Party’s allegations, not make them less substantial, as Brazil’s claims have become.
38 Japan - Taxes on Alcoholic Beverages (Complaints by European Communities, Canada and the United States) (1996), WTO Doc. WT/DS8,10,11/AB/R (Appellate Body Report).
39 Korea – Liquor Taxes, para. 10.16.
40 India - Pharmaceuticals, at para. 94.
41 Korea – Liquor Taxes, at para. 10.23.
WT/DS70/R
Page 16
For Canada, the integrity of the procedures set out in the DSU and amplified in successive Panel and Appellate Body Reports goes to the root of the Dispute Settlement Body’s principal objective of “providing security and predictability to the multilateral trading system.”42 In Canada’s view, due process, fairness, procedural justice: these inexact concepts permeate the DSU -- indeed any legal order -- and though implicit, they give meaning and embellish the explicit procedural guarantees provided for in the DSU, one of the most important of which is Article 6.2, the effective grant of jurisdiction to the panel.
Canada submits that the importance put on Article 6.2 of the DSU by the Appellate Body arises from the recognition that security and predictability evaporate when a party attempts to conduct its case as “‘trial by ambush.’”43
Arguments of Brazil
Brazil maintains that its request for establishment of a panel is consistent with Article 6.2 of the DSU. Brazil notes, concerning the factual basis of Brazil’s claims, its request in a preliminary submission that the Panel ask Canada to reveal the complete details of all operations of EDC, the Canada Account, the Technology Partnerships Canada (“TPC”) and its predecessor programmes, the Canada-Québec Subsidiary Agreements on Industrial Development (the “Subsidiary Agreements”), and the Société de Développement Industriel du Québec (“SDI”) with regard to the civil aircraft industry. (paras. 4.79 - 4.83 )
Before addressing Canada’s arguments concerning particular problems with the specificity of Brazil’s request for the establishment of a panel, Brazil considers it crucial to address Canada’s failure to participate meaningfully in the consultation process by offering factual information regarding the programmes at issue. Brazil cites the Appellate Body in India - Patent Protection for Pharmaceutical and Agricultural Chemical Products:
All parties engaged in dispute settlement under the DSU must be fully forthcoming from the very beginning both as to the claims involved in a dispute and as to the facts relating to those claims. Claims must be clearly stated. Facts must be disclosed freely. This must be so in consultations as well as in the more formal setting of panel proceedings. In fact, the demands of due process that are implicit in the DSU make this especially necessary during consultations. For the claims that are made and the facts that are established during consultations do much to shape the substance and the scope of subsequent panel proceedings.44
Brazil argues that due process flows two ways, and that it would violate all notions of basic procedural fairness to blame Brazil for any alleged lack of specificity in stating the factual bases for its claim stemming from Canada’s failure to disclose those facts crucial to the process envisioned in the DSU. For Brazil, Canada, in making its own request to the panel regarding procedures for handling confidential information (See paras. 4.152 - 4.161) implicitly admits that it failed to disclose those facts. Brazil accepts that Canada’s concern for protecting confidential business information is genuine, and that this genuine concern motivated the refusal to disclose. But for Brazil this refusal, however genuinely motivated, cannot then be turned on Brazil in the form of an accusation of failure to be specific.
42 Article 3.2 of the DSU.
43Redfern, A. and M. Hunter, Law and Practice of International Commercial Arbitration, 2nd ed.
(London: Sweet & Maxwell, 1991) at 320-21. The authors were referring to the withholding of principal arguments in the written submission, only to make them in the oral hearing; in the context of this case, however, and in view of Article 6.2, their observation is apposite. [TAB I 23 October 1998 preliminary submission of Canada]
44 (1997) WTO Doc. WT/DS50/AB/R, at para. 94.
WT/DS70/R
Page 17
Brazil further notes the Appellate Body’s statement in European Communities - Regime for the Importation, Sale and Distribution of Bananas that “Article 6.2 of the DSU requires that the claims, but not the arguments, must all be specified sufficiently in the request for the establishment of a panel . . .” 45 According to Brazil, the Appellate Body’s decision emphasized that Article 6.2 requires that a panel request be “sufficiently precise” because (1) it serves as the basis for the panel’s terms of reference, and (2) “it informs the defending party and the third parties of the legal basis of the complaint.”46
For Brazil, Canada’s contention that Article 6.2 is infringed by an alleged failure to provide “specific information as to the factual basis of the claims against it” (para. 4.42) is irrelevant; the Appellate Body in European Communities - Regime for the Importation, Sale and Distribution of Bananas required only that “legal basis” of the complaining party’s claim be “sufficiently precise.”
Brazil contends that the claims stated in its panel request meet the test of Article 6.2 of the DSU, as applied by the Appellate Body in European Communities - Regime for the Importation, Sale and Distribution of Bananas. Brazil notes that its panel request states that “Canada grants or maintains an extensive array of subsidies for the Canadian industry producing civil aircraft which are inconsistent with Canada’s obligations under Article 3.1(a) and Article 3.2 of the SCM Agreement in that they are contingent in law or in fact, whether solely or as one of several other conditions, upon export performance,”47 and specifies those programmes exemplifying Canada’s breach of its obligations under Articles 3.1(a) and 3.2 of the SCM Agreement. Brazil maintains that as such it identified the two components of a claim essential to maintain Canada’s due process rights pursuant to Article 6.2: (1) the particular provisions of the SCM Agreement with which Canada is charged with failure to comply; and (2) the particular Canadian measures demonstrating that failure. Brazil contends that it has therefore satisfied the test of identifying its claim with “sufficient precision,” as required by Article 6.2 and the Appellate Body’s decision in European Communities - Regime for the Importation, Sale and Distribution of Bananas.48
EDC “financing”
Brazil recalls its view (see paras. 4.61, 4.73) that it would be contrary to the purposes of Article 6.2 of the DSU and Article 4.4 of the SCM Agreement for the Panel to reward Canada for its failure to participate meaningfully in the type of disclosure required by the consultation process, as discussed by the Appellate Body in India - Patent Protection for Pharmaceutical and Agricultural Chemical Products.49 Brazil argues that Canada maintains a veil of secrecy around EDC and the other export subsidy programmes provided to the civilian aircraft industry which makes it impossible for Brazil to perceive what Canada contends are clear “legal[] and factual[]” distinctions between the various forms of EDC financing (para. 4.22), and that this should not stand in the way of Brazil’s right to pursue its claim. Brazil maintains that “EDC’s financing activities (direct lending)” ( para. 4.21) were discussed in consultations, and Canada was therefore on notice of Brazil’s claim.
Brazil also notes that Canada purports to know precisely what Brazil means in its use in its panel request of the term “financing” by the EDC; Canada defined this term to mean “EDC’s financing activities (direct lending).” For Brazil, this indicates that Canada appears to understand fully the meaning of the term “financing,” calling into question how the use of that term in its panel request prejudices Canada’s “due process right to know the case against it.”50
45 (1997) WTO Doc. WT/DS27/AB/R, at para. 143.
46 Id. at para. 142 (emphasis added by Brazil).
47 Government of Brazil, Request for the Establishment of a Panel.
48 (1997) WTO Doc. WT/DS27/AB/R, at paras. 139-144.
49 (1997) WTO Doc. WT/DS50/AB/R, at para. 94.
50 Id. at para. 37.
WT/DS70/R
Page 18
Definition of “civil aircraft industry”
Regarding the definition of “civil aircraft industry”, Brazil contends that this term is more specific than “aircraft,” the term used by Canada in its panel request concerning Brazil’s export credit programme.51 Brazil views as implausible Canada’s contention that it does not have notice of the meaning of that term, one of the primary purposes behind the requirement of specificity in Article 6.2, as Canada provides the definition of “aircraft industry” proposed by Brazil during bilateral negotiations preceding the establishment of this Panel (para. 4.48). In Brazil’s view, Canada therefore had notice of the meaning of that term in its panel request, and cannot now claim that its “due process right to know the case against it” is somehow prejudiced (para. 4.39).
Regarding Canada’s apparent contention that this working definition is “over-broad” and therefore fails to meet the specificity requirement of Article 6.2 of the DSU, (para. 4.49) Brazil recalls an Appellate Body statement in European Communities - Customs Classification of Certain Computer Equipment, that an infringement of Article 6.2 is not created because of a “lack of precision of the terms . . . in the request for the establishment of a panel . . .”52 For Brazil, there is no requirement in Article 6.2 that a complaining party somehow circumscribe a narrow category of products to which the challenged measures apply, particularly in the context of export subsidies, which are considered de facto specific within the meaning of Article 2.3 of the SCM Agreement. Rather, if the products subject to the defending party’s export subsidies fall within a broad category, then it is a complaining party’s right to challenge the full effect of those export subsidies, as long as the due process or notification purpose underlying Article 6.2 is met. In Brazil’s view, if it were otherwise, Article 6.2 would serve as an incentive for Members to structure export subsidies to benefit as broad a range of products as possible, in order to escape a panel’s jurisdiction.
Brazil objects to Canada’s citation of the Panel in Korea - Taxes on Alcoholic Beverages as support for Canada’s argument that Brazil’s failure to identify the tariff headings relevant to the “civil aircraft industry” suggests that Brazil has failed to meet the specificity requirement of Article 6.2 (para. 4.50). In Brazil’s view, Canada neglects to note that Panel’s citation to European Communities
- Customs Classification of Certain Computer Equipment, in which the Appellate Body found that “a panel request based on a broader grouping of products [than those included in an identified tariff heading] was sufficiently specific for the purposes of Article 6.2.”53
Brazil also argues that here, as in Korea - Taxes on Alcoholic Beverages, “[t]he issue of the appropriate categories of products to compare is important to this case,”54 and that determining the scope of the definition of “civil aircraft industry” is an issue “that requires a weighing of evidence” and that is therefore “not an issue appropriate for a preliminary ruling . . .”55
TPC “predecessor programmes”
Regarding Canada’s argument that the reference in Brazil’s panel request to TPC’s “predecessor programme” does not meet the specificity requirement of Article 6.2, Brazil asserts that Canada is well aware that there is one and only one predecessor programme to TPC, i.e., the Defence Industry Productivity Programme (“DIPP”), as evidenced in numerous statements concerning the
51 Brazil – Export Financing Programme for Aircraft, WT/DS/46/5.
52 (1998) WTO Doc. WT/DS62, 67/AB/R, at para. 70.
53 Korea - Taxes on Alcoholic Beverages, (1998) WTO Doc. WT/DS75, 84/R, at para. 10.16, citing European Communities - Customs Classification of Certain Computer Equipment, (1998) WTO Doc.
WT/DS62,67/AB/R, at paras. 58-73.
54 (1998) WTO Doc. WT/DS75, 84/R, at para. 10.16.
55 Id.
WT/DS70/R
Page 19
launching of TPC in 1996. Brazil refers to a response to a March 1996 Parliamentary inquiry56 in which Industry Canada (the Canadian government agency overseeing TPC) announced that “no new commitments are being made under DIPP,” and that “[a]ny future funding will be considered under the auspices of the new programme, Technology Partnerships Canada.” Brazil argues that Canada’s professions of ignorance concerning TPC’s predecessor programme must therefore be rejected. In response to a question from the Panel, Brazil indicates that its claims with respect to “funds provided to the civil aircraft industry by TPC and its predecessor programmes” is restricted to funds provided by TPC and DIPP.
Benefits provided under the Subsidiary Agreements and SDI
Regarding Canada’s argument that Brazil’s panel request with regard to “benefits provided” to the civil aircraft industry under the Subsidiary Agreements and SDI does not meet the specificity requirement of Article 6.2, Brazil notes that these programmes were explicitly included in its request for consultations and were on the table for discussion during consultations. Brazil argues that Canada was therefore on notice of these matters, and their inclusion in Brazil’s panel request therefore satisfies the requirements of Article 6.2 of the DSU.
In Brazil’s view, the Panel should not reward Canada for its failure to uphold its duty, as identified by the Appellate Body in India - Patent Protection for Pharmaceutical and Agricultural Chemical Products, to be “fully forthcoming” in consultations. Brazil contends that the facts of Canada’s provision of export subsidies under the Subsidiary Agreements and SDI are within Canada’s exclusive control, and that it would be contrary to the letter and spirit of both the consultation provisions and Article 6.2 of the DSU to reward Canada’s failure to “disclose[] freely” the facts of those programmes by limiting the scope of the Panel’s jurisdiction.
REQUEST OF CANADA FOR PRELIMINARY FINDING ON JURISDICTION: THE SCM AGREEMENT DOES NOT APPLY TO CONTRIBUTIONS AND TRANSACTIONS THAT TOOK PLACE BEFORE THE ENTRY INTO FORCE OF THE WTO AGREEMENT
Request of Canada
Canada argues that claims by Brazil concerning certain 1989 funding by the Defense Industry Productivity Programme “(DIPP”) and the Société de Développement Industriel (“SDI”) for development of the 50-seat regional jet are outside the Panel’s jurisdiction because they were provided prior to 1 January 1995, the date of entry into force of the WTO Agreement. Canada asks the Panel to so rule.
According to Canada, pursuant to Article 28 of the Vienna Convention, the SCM Agreement is not applicable to these contributions, as it provides that a treaty shall not be applied retroactively ‘unless a different intention appears from the treaty or is otherwise established’.57
Canada maintains that there is no contrary intention expressed in either the SCM Agreement or the WTO Agreement that demonstrates the SCM Agreement would apply to an event that took place in 1989. Therefore, according to Canada, this claim lies outside the jurisdiction of the Panel and should be dismissed.
56 Exh. BRA-28
57 Brazil – Measures Affecting Desiccated Coconut, WTO Doc. WT/DS22/AB/R Report of the Appellate Body adopted on 20 March 1997, at 15.
WT/DS70/R
Page 20
Response of Brazil
Brazil agrees that, with regard to prohibited subsidies, the SCM Agreement is not intended to apply retroactively. Thus, for Brazil, this 1989 DIPP and SDI subsidy, even though it was inconsistent with Canada’s obligations under the Tokyo Round Code, which applied at the time the subsidy was granted, is not subject to the current SCM Agreement. Accordingly, Brazil concedes that the Panel may disregard DIPP contributions made prior to 1 January 1995.
The Panel notes that because Brazil has dropped these claims, the substantive arguments of the parties concerning them are not included in this report.
BRAZIL’S REQUEST REGARDING ADDITIONAL FACT FINDING
Arguments of Brazil
Brazil, in a preliminary submission, requests the Panel to engage in additional fact-finding by requesting Canada to present to the Panel and the parties, at the first meeting of the Panel, the complete details of all operations of the Export Development Corporation, the Canada Account, the Technology Partnerships Canada and its predecessor programmes, the Canada-Québec Subsidiary Agreement on Industrial Development, and the Société de Développement Industriel du Québec with regard to the civil aircraft industry, including all grants, loans, equity infusions and loan guarantees, or any other direct or indirect financial contribution of any kind.
According to Brazil, public information is available to show that the measures listed in its request for establishment of a panel (1) involve a financial contribution by government; (2) to the Canadian manufacturer of regional civil aircraft; (3) conferring a benefit on that manufacturer; (4) contingent in law or in fact on export. Brazil argues that its arguments in the case would be handicapped because Canada declined to provide transaction-specific details concerning these measures during consultations, on the stated grounds of confidentiality.
For Brazil, Canada's position is at odds with the teachings of the Appellate Body in India – Patent Protection for Pharmaceutical and Agricultural Chemical Products:
All parties engaged in dispute settlement under the DSU must be fully forthcoming from the very beginning both as to the claims involved in a dispute and as to the facts relating to those claims. Claims must be stated clearly. Facts must be disclosed freely. This must be so in consultations as well as in the more formal setting of panel proceedings. In fact, the demands of due process that are implicit in the DSU make this especially necessary during consultations.58
Brazil asserts that this statement makes clear that Canada was not justified in declining to disclose relevant details during consultations. Brazil further argues that there is no justification for declining to produce confidential information at the panel stage of dispute settlement. Article 18.2 of the Understanding on Rules and Procedures Governing the Settlement of Disputes permits Members to designate information submitted to the Panel as confidential and requires other Members to respect that designation. Brazil notes that Members that believe the procedures of Article 18.2 are inadequate are free to propose an alternative procedure to the Panel.59
Brazil recalls that in India Pharmaceuticals the Appellate Body specified that, "If, in the aftermath of consultations, any party believes that all the pertinent facts relating to a claim are, for any
58 AB-1997-5, WT/DS50/AB/R (19 December 1997) para. 94.
59 Indonesia – Certain Measures Affecting the Automobile Industry, WT/DS54/R, WT/DS55/R, WT/DS59/R, WT/DS64/R (2 July 1998) para. 14.7.
WT/DS70/R
Page 21
reason, not before the panel, then that party should ask the panel in that case to engage in additional fact-finding".60 Brazil argues that the result of Canada's refusal to disclose the details of the operations of these measures with regard to regional civil aircraft would be that not all of the pertinent facts relating to its claim would be before the Panel.
Arguments of Canada
Background to Brazil’s motion: the consultations, “transaction-specific” information and confidentiality
Canada responds that it began providing information to Brazil long before formal WTO consultations were requested in this case. According to Canada, Brazil raised concerns about “secret” accounts and “vast arrays of subsidies” in the course of consultations and negotiations in the case launched by Canada in 1996, Brazil - Export Financing Programme for Aircraft. Canada states that at the time it pointed out that much of the information requested by Brazil was public; and that it nevertheless provided considerable public documentation to Brazil. For Canada, the questions posed by Brazil throughout the negotiations in the Brazil - Export Financing Programme for Aircraft case and then during the consultations in this case, indicated that Brazil was already in possession of significant detailed information.
With respect to “transaction-specific” information requested by Brazil in its letter, Canada states that it repeatedly noted during consultations the difficulty of responding to a shotgun request for documents on “all transactions” relating to the mentioned programmes. Canada asserts that it asked Brazil to narrow its request to specific transactions and to identify its claim in respect of such transactions so that it might be made aware of the case against it and so that it might provide Brazil with more information. According to Canada, no such specificity was forthcoming.
Further, Canada argues, throughout the consultations and negotiations, Canada underlined the commercial confidentiality of the “transaction-specific” information that might be covered by Brazil’s shotgun requests. In Canada’s view, its efforts in good faith to protect the interests of private enterprises not parties to this dispute were fully consistent with the principles underlying the DSU.61 Canada maintains that its proposal on procedures governing confidential business information (paras. 4.152- 4.161) is intended to enable all necessary disclosure while protecting the interests of private persons not party to this dispute.
Brazil’s request is inconsistent with WTO law and practice
Canada argues that Brazil’s request, without having first presented a prima facie case before the Panel, is inconsistent with the WTO Agreement and international law and practice. Canada notes the statement by the Appellate Body in India- Pharmaceuticals II62 relied upon by Brazil. According to Canada, in that case, the Appellate Body was discussing the panel’s error in deciding on a US claim that was not within the panel’s terms of reference. There, the United States alleged that it had not made the relevant claim in its panel request because it had no way of knowing that India would rely on a particular defence not disclosed to the United States during consultations. Canada states that the
60 AB-1997-5, WT/DS50/AB/R (19 December 1997) Para. 94
61 The impact of a particular interpretation of Article 6.10 on private persons was an element for
consideration by the Appellate Body in United States - Restrictions on Imports of Cotton and Man-made Fibre Underwear (Cotton Underwear), WT/DS24/AB/R, Report of the Appellate Body Adopted on 25 February 1997, at 15.
62 India - Patent Protection for Pharmaceutical and Agricultural Chemical Products (India - Pharmaceuticals II), WT/DS50/AB/R, Report of the Appellate Body adopted on 16 January 1998.
WT/DS70/R
Page 22
Appellate Body found that, nevertheless, there was no basis in the DSU for a complaining party to make an additional claim, outside the scope of the panel’s terms of reference.63
Canada notes that the Appellate Body then commented on the importance of parties engaged in dispute settlement being “…fully forthcoming from the very beginning, both as to the claims involved in a dispute, and as to the facts relating to those claims,”64 then described the importance of meeting the demands of due process in consultations, and then stated that:
“If, in the aftermath of consultations, any party believes that all the pertinent facts relating to a claim are, for any reason, not before the panel, then that party should ask the panel in that case to engage in additional fact-finding.”65 [emphasis added by Canada]
For Canada, it is well-established practice for a WTO panel, having received the first submissions and evidence of the parties and having heard their first substantive oral arguments, to ask the parties for information additional to that submitted by the parties. In Canada’s view, there is also precedent for a party to the dispute to ask a panel to seek further information at this stage, and for a panel to reject that request if the panel considers the information requested is irrelevant or unnecessary.66
According to Canada, there is no support in the statement of the Appellate Body, however, or indeed in the DSU, WTO practice, or international law and practice for turning the panel process into something akin to a commission of inquiry;67 there is, more important, no provision in the DSU and no precedent in GATT or WTO jurisprudence for subjecting a responding party to a discovery process.68
63 Id.. at para. 93.
64 Id.. at para. 94.
65 Id.. at para. 94.
66 See, for example, European Communities – Measures Concerning Meat and Meat Products (Hormones), WT/DS48/R/CAN, Report of the Panel adopted on 13 February 1998, at paras. 8.5 and 8.11.
67 J.G. Merrills states that ‘inquiry’ can be defined as: “a specific institutional arrangement which
states may select in preference to arbitration or other techniques, because they desire to have some disputed issue independently investigated.” [emphasis added] See J.G. Merrills, International Dispute Settlement, 2nd ed. (Cambridge: Grotius Publications Ltd., 1991), at 43. TAB A 23 October 1998 preliminary submission of Canada According to Merrills, commissions of inquiry were introduced by the Hague Convention for the Pacific Settlement of International Disputes, 1899. Among the limitations on their mandate was that they should handle only questions of fact and not of law, and that their findings should not be seen as obligatory. See Id.. at 44. TAB A October 1998 preliminary submission of Canada This is manifestly at odds with the objectives and the nature of WTO dispute settlement.
68 The basic instruments to ensure that an arbitral tribunal has the authority to order a discovery of evidence are “through certain specific stipulations.” See V.S. Mani, International Adjudication: Procedural
Aspects, (The Hague: Martinus Nijhoff Publishers, 1980), at 212. TAB B October 1998 preliminary submission of Canada The power to order evidentiary discovery is therefore not inherent in arbitral tribunals.
For example, evidentiary discovery has been explicitly provided for in the Statute of the International Court of Justice. [Annex to the Charter of the United Nations, 26 June 1945, 15 U.N.C.I.O. 335 at 355 (entered into force 24 October 1945)] Article 49 provides that “[t]he Court may, even before the hearing begins, call upon the agents to produce any document or to supply any explanations. Formal note shall be taken of any refusal.” TAB C October 1998 preliminary submission of Canada The Optional Rules for Arbitrating Disputes between Two States of the Permanent Court of Arbitration TAB D October 1998 preliminary submission of Canada provides, in its Article 24.3, that:
At any time during the arbitral proceedings the arbitral tribunal may call upon the parties to produce documents, exhibits or other evidence within such a period of time as the tribunal shall determine. The Tribunal shall take note of any refusal to do so as well as any reasons given for such refusal.
WT/DS70/R
Page 23
In this respect Canada notes the ruling of the panel in Indonesia - Automobiles concerning the request of Indonesia “to require the United States to submit its confidential data immediately:”69
“We encourage all parties to submit relevant data to the Panel as early as possible. However, we have concluded that it would not be appropriate for us to require the United States to submit the information in question in the context of this meeting. It is a matter for each party to decide when and if to submit information and argumentation within the schedule set forth by the Panel. In this respect, we note that there is no rule in the DSU or our working procedures that requires parties to submit all factual information in their first submissions. In fact, factual information is often provided in second submissions or in response to questions from a panel as the issues in the case come into sharper focus. We see no reason to deviate from that approach in this case.”70 [emphasis added by Canada]
For Canada, “additional fact-finding”, if a panel deems it necessary, is precisely that: it is additional to facts presented in support of a particular claim. In Canada’s view, there is neither precedent in WTO practice nor provision in the DSU for a panel to engage in fact-finding when it has not had an opportunity to examine the evidence submitted by the parties in their first submissions and the relevance of that evidence to or support for the claims of the disputing parties, or to request that a party submit information in respect of defences it has not yet raised in response to claims that have yet to be made out.
Canada asserts that it cannot be expected to provide the information requested by Brazil when it does not yet know what the scope of Brazil’s claims is, when Canada has not yet put its defence before the Panel, when it is impossible for the Panel to determine what are the “pertinent facts” given the lack of specificity of Brazil’s request for a panel and Brazil’s subsequent motion for discovery, and when the Panel has not had the chance to examine whether Brazil has a prima facie case that would require Canada to present any evidence in its own defence.
Canada notes that it is the responsibility of Brazil to present its case; it must adduce “’evidence sufficient to raise a presumption‘”71 that its claims are true. In Canada’s view, if Brazil does not have such evidence, there is no obligation on the part of Canada to provide additional information: Brazil’s claims must fail; if, on the other hand, Brazil has such information, then the
The specific grants of authority noted above may be contrasted with Article 13 of the DSU, which provides that:
Each panel shall have the right to seek information and technical advice from any individual or body which it deems appropriate. However, before a panel seeks such information or advice from any individual or body within the jurisdiction of a Member it shall inform the authorities of that Member. A Member should respond promptly and fully to any request by a panel for such information as the panel considers necessary and appropriate. Confidential information which is provided shall not be revealed without formal authorization from the individual, body, or authorities of the Member providing the information.
Panels may seek information from any relevant source and may consult experts to obtain their opinion on certain aspects of the matter. With respect to a factual issue concerning a scientific or other technical matter raised by a party to a dispute, a panel may request an advisory report in writing from an expert review group. Rules for the establishment of such a group and its procedures are set forth in Appendix 4.
This clearly is not a “specific stipulation” of authority for discovery, as noted by Mani and as provided for in the Statute of the ICJ.
69 Indonesia - Certain Measures Affecting the Automobile Industry (Indonesia - Automobiles), Report of the Panel adopted on 23 July 1998, at para. 14.6.
70 Id.. at para. 14.7.
71 United States - Measure Affecting Imports of Woven Wool Shirts and Blouses from India (Shirts and Blouses), WT/DS33/AB/R, Report of the Appellate Body adopted on 23 May 1997, at 14.
WT/DS70/R
Page 24
“’burden shifts to the other party, who will fail unless it adduces sufficient evidence to rebut the presumption.’”72 In this regard, Canada quotes Dr. Mojtaba Kazazi:
“First, it is an established rule that if the claimant does not provide any evidence at all, there is no duty for the other party to do so. In Anglo-American Law, from which seemingly the rule of collaboration originates, if the claimant is not able to show at the early stages of the proceedings that it has a solid case, the judge will announce that there is no case to answer, and the respondent will not be required to provide any evidence”. [emphasis added by Canada]
“Second, as to the order of production of evidence, it is also an established rule in both municipal and international procedure that the respondent does not have to initiate the production of evidence before the claimant does”. [emphasis added by Canada]
“Third, not only should the claimant initiate the production of evidence but the evidence, it provides should be of some value. The minimum standard of proof known in many municipal jurisdictions is prima facie evidence”. [emphasis added by Canada]
Canada argues that the duty to cooperate in no way relieves the claimant of its burden of proof, nor does it allow a claimant to embark upon a broad and unfounded request for information. In this regard, Canada argues that the situation in the Argentina – Footwear case is not even remotely similar to the situation in the present case. Canada notes that in its report, that Panel considered evidence based on specific transactions that was adduced by the United States. In particular, the United States submitted copies of invoices of shipments to support its arguments. Argentina challenged the validity of these documents, and argued that the evidence submitted was not the best evidence. 73
Canada also notes that the Panel observed that the United States had tried to obtain the original copies that were in Argentina’s possession. The Panel then noted the duty to collaborate and stated that in the absence of the originals, and after careful examination and consideration of the evidence, it considered that the copies submitted by the United States constituted sufficient evidence to allow it to make conclusions.74 The Panel stated that the obligation to provide it with relevant documents which are in its sole possession does not arise until the claimant has actually produced some prima facie evidence in support of its case. In Canada’s view, Brazil has not done so. According to Canada, vague or ambiguous quotations presented out of context and jumbled to produce a misleading impression of what the person cited actually meant, does not constitute evidence. Nor, in Canada’s view, is there precedent or place for wide-ranging discovery in the WTO.
For Canada, it is difficult to see how the WTO dispute settlement mechanism could work if even before a complainant has shown that it has a prima facie case, it could force the responding party to furnish it with evidence in the course of a wide-ranging and apparently limitless fishing expedition. For Canada, this would open the door to endless intrusion in confidential records of WTO Members, such as tax files, banking and financial records, regulatory proceedings and a host of other governmental activities. Canada asserts that this is not the way the WTO functions.
Canada argues that it has nothing to hide, and is fully prepared to cooperate in the normal way of WTO proceedings, and acknowledges the duty of collaboration as discussed and applied by the panel in Argentina – Footwear:
72 Id..
73 Argentina – Footwear at paras. 6.52-6.53
74 Argentina – Footwear at para. 6.58
WT/DS70/R
Page 25
“Another incidental rule to the burden of proof is the requirement for collaboration of the parties in the presentation of facts and evidence to the panel, and especially the role of the respondent in that process. It is often said that the idea of peaceful settlement of disputes before international tribunals, is largely based on the premise of co-operation of the litigating parties. In this context, the most important result of the rule of collaboration appears to be that the adversary is obligated to provide the tribunal with relevant documents which are in its sole possession. This obligation does not arise until the claimant has done its best to secure evidence and has actually produced some prima facie evidence in support of its case. It should be stressed, however, that “’discovery’ of documents, in its common-law system sense, is not available in international procedures”. We shall, therefore, follow these general rules when addressing, for instance, the request of the United States to Argentina for production of documents and the fact that Argentina did not do so.”75 [emphasis added by Canada]
With respect to any attempt to import discovery into the WTO process, Canada notes and agrees with the observations of the panel in Argentina – Footwear,76 referring to the works of Dr. Mojtaba Kazazi, on the burden of proof before international tribunals:
It should be stressed, however, that “’discovery’ of documents, in its common-law system sense, is not available in international procedures”.77
Brazil’s request further supports Canada’s preliminary submissions
Canada asserts that Brazil’s request confirms the position of Canada set out in its preliminary submissions. First, Canada asserts, it will put before the Panel information, including commercial confidential information, necessary to rebut Brazil’s allegations when they are known; and that it is willing to provide the Panel with information, including commercial confidential information, in response to the Panel’s requests. Canada notes that to those ends it has made a preliminary submission regarding procedures governing confidential business information, which notes that evidence it may need to submit in its defence could contain confidential proprietary business information. For Canada therefore, it is imperative that adequate safeguards be developed for the protection of such information. Canada states that its proposed confidentiality procedure, once established by the Panel, would provide such a safeguard and permit Canada to adduce such evidence as may be necessary to defend its case.
Canada also takes issue with Brazil’s request for “’fact-finding’” in respect of “’all operations’” under the listed programmes. Canada recalls its argument (see paras. 4.38-4.58) that Brazil’s request for a panel is not consistent with Article 6.2 of the DSU as it fails to identify the “specific measures at issue.” Canada reiterates that the panel process is not a licence for a fishing expedition; that the responding party must be given notice of the case against it in the request for a panel, which forms the terms of reference of the Panel; and that, as a result, it is important that the request for a panel not contain a sweeping statement relating to programmes that could cover, for example in the case of “financing by the Export Development Corporation”, thousands of transactions in the context of a Can$10 billion portfolio.
According to Canada, Brazil wishes to engage the Panel and, by necessary implication, Canada, in this fishing expedition. In Canada’s view, having failed to inform Canada of the specific programmes, activities or transactions that are at issue as required by DSU Article 6.2, Brazil
75 Argentina – Footwear, at para. 6.40.
76 Argentina – Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items, WT/DS56/R, Report of the Panel adopted as modified by the Appellate Body on 22 April 1998.
77 Id.., at para 6.40.
WT/DS70/R
Page 26
compounds the difficulty by asking the Panel to engage in fact-finding with respect to “all operations” under the listed programmes.
Accordingly, for Canada, this request by Brazil confirms that Brazil has yet to state its case against Canada. In Canada’s view, the Panel should not assist a complaining party to find or develop a case not made.
Response of Brazil
Brazil counters by arguing that it has encountered significant difficulty in obtaining information about the various Canadian export subsidy programmes at issue. According to Brazil, citing confidentiality, Canada has declined to provide transaction-specific information concerning the operations of EDC, the Canada Account, Technology Partnerships Canada, the Canada-Québec Subsidiary Agreements on Industrial Development, and the Société de Développement Industriel du Québec, and has not even notified EDC or Canada Account funding, as required by Article 25 of the SCM Agreement, to enable other Members like Brazil “to evaluate the trade effects and to understand the operation of” support programmes provided thereunder. According to Brazil, Canada’s notifications pursuant to Article 25 of the SCM Agreement fail in some instances to provide any description of these programmes.
Brazil appreciates concerns regarding the confidentiality of proprietary business information, but states that in this case Canada’s posture seems to be, “Our officials do not mean what they say when testifying before Parliament or when they issue official reports or when they speak to industry groups. We cannot tell you why, but despite what these officials might say, we are acting in conformity with the terms of the SCM Agreement. Trust us.”
Brazil acknowledges that it bears the burden of proving its affirmative case, but states that at the same time Canada bears the burden of collaboration, a concept well-founded in public international law and in WTO law. In addition to the obligation of Members to act in good faith, Brazil recalls the Appellate Body’s statement in India – Patent Protection for Pharmaceutical and Agricultural Chemical Products, that all parties engaged in dispute settlement “’must be fully forthcoming from the very beginning,’” and that this duty includes the requirement that facts “’must be disclosed freely.’” In Brazil’s view, facts have not been disclosed freely by Canada in this case.
In Brazil’s view, particularly where, as here, concerns regarding confidentiality hinder access by a complaining Party to information peculiarly within the control of a responding Party, that responding Party’s burden to produce evidence in good faith to the Panel is considerable. Brazil argues that where a party fails to provide information peculiarly within its control, or where, in the words of the Appellate Body in India Pharmaceuticals, “in the aftermath of consultations, any party believes that all the pertinent facts relating to a claim are, for any reason, not before the panel, then that party should ask the panel . . . to engage in additional fact-finding.”
Regarding Canada’s contention that Brazil’s request is not appropriate until the Panel has received the first submissions and heard the first substantive arguments of the Parties, Brazil states that it is aware of no such limitation, noting that the Appellate Body, in India Pharmaceuticals stated only that such a request should be made in the “’aftermath’” of consultations, and stating that Brazil made its request in the aftermath of consultations.
Brazil also states that even if Canada is correct, as of receipt of the first submissions of the parties and the hearing of the first substantive arguments, it is time for Canada, even by its own terms, to fulfil its obligation to cooperate with the Panel and Brazil in producing all relevant information.
Regarding Canada’s argument that the Panel’s authority to seek information is dependent upon presentation by Brazil of its prima facie case, Brazil finds two problems. First, Brazil argues,
WT/DS70/R
Page 27
Canada confuses the prerequisites for the Panel to engage in additional fact-finding with the prerequisites for the Panel ultimately to decide that a complainant has satisfied its burden of proof, which are not the same thing. Second, Brazil asserts that in making its submission of approximately 1,600 pages of supporting documentation and two expert reports, it has presented a prima facie case.
As regards Canada’s concern over the breadth of Brazil’s request, Brazil does not view its request as a fishing expedition, but as a proper attempt to obtain evidence relevant to the issues before this Panel. Brazil submits that to the extent Canada does not produce, in a timely fashion, necessary information in its possession, the Panel should draw the appropriate adverse inferences.
Rebuttal of Canada
Canada submits that Brazil’s allegations of secrecy are unfounded, arguing that given Canada’s character as a flourishing democracy with a vigorous opposition, hyperactive and multifaceted media, an independent judiciary whose powers are enhanced by constitutionally entrenched Charter of Rights, as well as state-of-the-art access to information legislation, it is counter- intuitive to suppose that Canada’s public institutions are as secretive as Brazil claims them to be. According to Canada, Brazil’s much vaunted 1,600-plus pages of evidence from public sources by itself belies this assertion. Canada notes that Brazil has, in fact, relied on many sources of public information about the pertinent Canadian institutions, including annual reports of the EDC and the Department of Industry, the public accounts of the federal government and the provinces, the reports of the Auditor General and even the considerable volume of materials obtained by an Opposition Member of Parliament and handed over to Brazil’s Embassy in Ottawa. In spite of all these efforts, Canada argues that Brazil has failed to produce a single shred of material, credible evidence that will stand up to scrutiny, because there is simply nothing to find.
As to the alleged secrecy of EDC and the Canada Account, which Canada states are the principal targets of Brazil’s allegations of secrecy, Canada notes that these are financial institutions, essentially export banks, and questions what bank or financial institution willingly releases legally protected transaction-specific, confidential business information. According to Canada, the very nature of the EDC’s operations as a commercial lender means that this is what would be required to make out the negative proof that Brazil alleges is necessary to clear EDC of the general presumption that any state-owned financial institution active in export markets ipso facto constitutes an export subsidy measure.
CANADA’S REQUEST FOR A DEADLINE FOR THE SUBMISSION OF NEW EVIDENCE
Arguments of Canada
Canada observes that despite three volumes of annexes, Brazil’s first written submission suffers from a paucity of credible evidence, and states that this is not surprising. The impugned programmes, activities and transactions are consistent with the SCM Agreement; Brazil brought this dispute before the WTO in response to Canada’s case against Brazil’s export subsidy programmes. Nevertheless, in view of the vagueness of Brazil’s claims in its request for a panel, Canada is concerned that Brazil might advance allegations about programmes, activities or transactions other than those already mentioned in Brazil’s first submission, or adduce evidence at a late stage in these proceedings.
Canada submits that, given the accelerated procedure under Article 4, any such late submission of allegations or evidence would be prejudicial, as Canada would be effectively denied an adequate opportunity to respond. Canada requests, therefore, that the Panel rule that the complaining Party may not adduce new allegations or evidence after the end of the first substantive meeting of the Panel with the Parties.
WT/DS70/R
Page 28
Arguments of Brazil
Brazil counters that Canada’s request that Brazil, as the complaining Party, not be permitted to adduce new allegations or evidence after the end of the first substantive meeting of the Parties is without support in WTO law. According to Brazil, no provision of either the DSU or the SCM Agreement would require or even suggest such a rule. Brazil argues that in Argentine Footwear, the Appellate Body affirmed the Panel’s authority to permit the submission of new facts at any point in the proceedings, provided that the other party is given an adequate opportunity to respond.
Brazil maintains that it has not deliberately held back any evidence for later presentation in an effort to surprise Canada or to attempt in any way to limit its due process ability to respond. Brazil states that it cannot predict, however, whether particular arguments to be made by Canada will require a response from Brazil employing new arguments or additional evidence. Brazil asserts that any such evidence would not, of course, expand the scope or range of the dispute, or in any way increase the number of claims involved, but rather would be responsive to any evidence made available by Canada.
Brazil submits that it also has a due process right to react to what Canada argues and presents, particularly when, as here, Canada has been so unforthcoming about a number of highly non- transparent measures.
BRAZIL’S REQUEST REGARDING DEADLINE FOR SUBMISSION OF NEW EVIDENCE
Arguments of Brazil
At the first meeting of the Panel, Brazil requested that the Panel not permit Canada to introduce evidence in support of any affirmative defences after the first meeting of the parties. In this regard, Brazil notes a suggestion by Canada (para. 6.161) that Canada Account financing and loan guarantees are excluded from the prohibition in Article 3 of the Agreement because they comply with the interest rates provisions of the OECD Arrangement on Guidelines for Officially Supported Export Credits, or are not used to secure a material advantage in the field of export credit terms. This suggests to Brazil that Canada might intend to assert that the provisions of paragraph (k) of Annex I to the Agreement excuse its otherwise prohibited subsides. Brazil notes that a similar but slightly varied suggestion is made with regard to financing activities of the Export Development Corporation (para. 6.64).
Brazil argues that if this is Canada’s position, it is an affirmative defense for which Canada bears the burden of proof. Brazil believes that good faith implementation of the Dispute Settlement Understanding requires a party making an affirmative defense to set out the grounds for that defense in its first written submission. Brazil asserts that it did exactly that with regard to its claim under paragraph (k) in the other Panel proceeding.
For Brazil, Canada’s allusion to a paragraph (k) defense, combined with the absence of any relevant evidence in its first submission, suggests to Brazil that Canada may be withholding the evidence in this time-constrained proceeding so as to hinder Brazil’s ability to respond. Brazil does not believe there is room in WTO dispute settlement, between parties called upon to act in good faith, for conduct of this kind. Accordingly, Brazil requests that the Panel not accept any evidence or claims of defense in the nature of affirmative defences with regard to the Canada Account, the EDC or any other measure at issue which is not submitted to the Panel and to Brazil prior to the scheduled close of the first meeting of the Panel.
For Brazil, this is particularly important in a fast track proceeding, where the time constraints on the Panel could limit, or even preclude, the Panel’s giving Brazil a reasonable time to respond to
WT/DS70/R
Page 29
new information after the date for the final submission, or particularly after the second meeting of the Panel.
Arguments of Canada
Canada notes a Panel request during the first meeting that the parties consider whether they could provide the Panel with additional information. Canada states that in light of this request, and of Brazil’s concern about being ‘sandbagged’ by new evidence, it would endeavour to provide the Panel with additional contextual evidence in its second written submission.
Canada insists that this decision was not made lightly, as Canada continues to believe that Brazil has failed to present a prima facie case and that Canada is under no obligation to produce any evidence in response to a case that has not been made (see paras. 5.5- 5.12). Canada also recalls its statements (see paras. 4.173 - 4.183) that the amended confidentiality procedure does not adequately protect private sector interests and the interests of the Government of Canada in maintaining effective control over the dissemination of business confidential information.
Canada emphasizes that the introduction of any additional evidence pursuant to the Panel’s request would be without prejudice to Canada’s position on these two issues, and that the introduction of such evidence would not be done to ‘sandbag’ Brazil, an allegation that Canada rejects.
Actions of the Panel
In connection with the second meeting of the Panel, held on 12-13 December 1998, the Panel posed a number of questions to both parties, many of which were of a factual nature, and announced a deadline of one week after the meeting for the parties to respond, and one further week for the parties to comment on any new factual evidence or arguments submitted in response to questions. This was in addition to the Panel’s letter to Canada, dated 9 December 1998, in which the Panel stated that it “expects to receive clear and complete answers” to certain of the Panel’s original questions to which Canada had not “replied in full”. Because the one week period to prepare comments on any new factual evidence or arguments submitted in response to the Panel’s questions fell over the Christmas holidays, and because Brazil believed that Canada might submit voluminous new evidence in answer to questions, Brazil expressed concern over the shortness of time for commenting on Canada’s replies, and requested that the Panel withdraw all of the questions to Canada, and rely on the evidence before it at that time. Canada did not agree with Brazil’s suggestion. The Panel, in response, indicated that it could allow one additional week for comments on answers to questions, if the parties would agree to a one-week extension in the timetable for the Panel (in light of footnote 6 to Article 4 of the SCM Agreement that any time periods in that Article may be extended by mutual agreement). In response to this Panel proposal, Brazil responded as follows.
Response of Brazil
In a letter to the Panel dated 13 December 1998, Brazil argued that with the parties clearly satisfied with the cases they had put forward the panel would have been entitled to decide the case on the record as made to this point; neither party would have grounds to complain that it had not had fair opportunity to present evidence or advance argument.
Brazil acknowledges the panel's right to request information from any source at any time, but believes it was not necessary for the Panel to request this particular information at this particular time. More important for Brazil in this accelerated proceeding, the Panel's decisions have placed it in a position in which, depending upon the nature of Canada's response, it might not be able to comment at all on the new information that might be supplied by Canada.
WT/DS70/R
Page 30
Brazil recalls its concerns that Canada might be planning on presenting an affirmative defence to one or more of Brazil's complaints in its second written submission or even at the second meeting of the Panel, and its related request that the Panel not accept any such information after the first meeting of the Panel (paras. 4.119- 4.122)
Brazil notes a Panel statement that even before receiving Brazil's request it had decided to ask Canada, in questions to be presented to the parties, whether or not it intended to assert any defence based upon items (j) or (k) of the Illustrative List of Export Subsidies, and its further statement that it expected a clear and complete response to the questions, on which it also gave the other party a period to comment.
Brazil notes that Canada did not invoke any affirmative defences in its response to questions from the Panel, nor did Canada supply information that would support such defences. Brazil indicates that the Panel granted Canada an additional opportunity to submit answers responsive to certain of the Panel's original questions, and notified the parties of its intention to present further questions, and stated that except for responses to those questions, it would not accept further submissions from the parties after the deadline for those responses, except for good cause shown. In the same message, Brazil argues, despite the fact that Canada had not responded with affirmative defence information by the deadline for answers to previous questions, the Panel stated that "’Canada shall invoke any positive defences it intends to raise in this case during or before, the second meeting with the Panel on Saturday, 12 December 1988‘".
Brazil contends that although by this action the panel gave Canada an additional eight days to provide affirmative defence information, information that, in Brazil's view, properly belonged in its first submission and not in a response provided at the second meeting with the Panel, Canada again declined to submit the information by 12 December, and reaffirmed that it did not intend to use any kind of affirmative defence since, in its view, Brazil had not presented a prima facie case. Brazil states that in fact, the Panel orally confirmed at the second meeting that that meeting would be the last opportunity for Canada to invoke an affirmative line of defence.
However, in Brazil’s view, the additional questions provided by the Panel to Canada In connection with the second meeting effectively asked for the kind of information on which such a defence could be based, which could consist of a large quantity of transaction-specific, confidential business information that would take time to produce and time to analyse. The panel's questions therefore in Brazil's view effectively extend Canada's time to provide this kind of information by 17 days. Brazil notes that to analyse and prepare any comment it may wish to make on such response, the Panel has provided Brazil seven days, including the Christmas holiday, and that should Canada concur in a one-week extension of the Panel's deadline, the Panel would extend Brazil's time for comment by an additional week, encompassing the New Year's holiday.
Brazil believes this procedure is unnecessary and unfair to Brazil; unnecessary because Canada, as a Member of the WTO, was entitled to have its case decided on the basis on which it chose to present it. This is particularly important, in Brazil's view, in an accelerated procedure under the SCM Agreement.
Brazil believes it is unfair because, not only does it permit Canada to supply last-minute information, it effectively denies Brazil a fair opportunity to respond. In Brazil’s view, the occurrence of major religious and secular holidays in the midst of the response period would cause difficulty for any Member, and causes particular difficulty in this case because of Canada’s indication that the information Canada will furnish will include confidential business information that may be dealt with only under strict procedures to ensure its security.
Brazil states that as a developing country, it does not have the same resources or personnel as Canada to enable it to prepare and present its argumentation in the manner envisioned by the DSU,
WT/DS70/R
Page 31
noting that in this case, it chose to utilize the services of non-government specialists. Brazil argues that this presents a difficulty because these specialists are resident in neither Geneva nor Brasilia, where the Brazilian officials with whom they are working are located, and it had been learned that airline reservations for them to return to Geneva over this holiday period would be extremely difficult if not impossible to obtain.
Brazil argues that it thus was unable to choose between the options provided by the panel, as its needs would depend entirely on what Canada would supply in the way of answers to the Panel's questions.
Proposal by the Panel
In light of the concerns raised over the proposed schedule for answers to questions and comments thereon, and in light of an absence of consensus among the parties on extending the Panel’s overall timetable, the Panel proposed a new schedule that would allow the parties 17 days to comment on one another’s answers to questions, but would reduce the time for interim review. The Panel noted that 17 days was the same amount of time as between the deadlines for first and second submissions. That is, the parties would have exactly the same amount of time for commenting as if all of the information had been presented in the first submissions.
Joint letter from Brazil and Canada, and Panel action
In response to the Panel’s proposal, the parties submitted a joint letter to this Panel and to the Panel in the parallel case on Brazil –Export Financing Programme for Aircraft. In this letter, the parties note that the time available under the Panel's proposed new schedule for the review of the interim report and for the travel of delegations to Geneva for a possible additional meeting with that Panel was likely to be insufficient, and requested that the two Panels examine the possibility to readjust their respective timetables as follows:
the interim reports of both Panels shall be issued on the same dates, but no later than 17 February;
the final reports of the two Panels shall be issued on the same dates;
the interim review meetings should be held on dates as close as possible to each other; and
adequate time should be allowed for all stages of the interim review, providing sufficient time for internal consultations and travel arrangements.
The letter further states that Brazil and Canada realize that such request would call for some flexibility regarding the date of issuance of the final report, and indicate that they therefore would agree to a one-week postponement of the deadline for issuance of the final report.
The Panel, taking into consideration this letter, announced its revised timetable.
ARGUMENTS CONCERNING THE SUBMISSION OF EVIDENCE IN RESPONSE TO PANEL QUESTIONS POSED IN CONNECTION WITH THE SECOND MEETING
Comment of Canada
In replying to the Panel’s questions posed in connection with the second meeting, Canada indicates that its replies are provided in the light of certain concerns. First, Canada states, the Panel has requested the production of evidence in respect of defences Canada has not made. Canada recalls
WT/DS70/R
Page 32
that consistent with the principle of judicial economy, particularly in the context of the expedited proceedings under SCM Article 4, it is neither necessary nor desirable that the respondent make out a defense for both of the two prongs of the test for a prohibited export subsidy in Part II of the SCM Agreement. Canada states that Brazil must prove both that a subsidy exists within the meaning of Article 1, and that the subsidy is export contingent within the meaning of Article 3. In response, it is sufficient for Canada to demonstrate in relation to each impugned measure, the measure is either not a subsidy or not export contingent. Canada asserts that in its letter to the Chairman of the Panel of 13 December 1998, Brazil has agreed with this thesis.
Second, Canada states, the Panel has requested the production of evidence in respect of matters where in Canada’s view Brazil has clearly not made out a prima facie case. Canada notes that the Panel has not ruled on whether Brazil has made out a prima facie case on any of the impugned programmes, activities or transactions. Canada notes that the covering letter to the Panel’s questions of 10 December 1998 states that the questions are posed without prejudice to the Panel’s eventual findings with respect to any issue raised by either party.
Third, Canada objects that the Panel has requested the production of evidence in respect of transactions where Brazil has not even made an allegation. Canada does not consider it appropriate to adduce evidence in response to an allegation that has not been made and a case that has not been established.
Finally, Canada states that it has been placed in a very difficult position because the Panel’s questions ask Canada to produce business confidential information, recalling its position (see paras. 4.173- 4.183) that the Procedures Governing Business Confidential Information do not provide the requisite level of protection for such information.
Comment of Brazil
Regarding Canada’s comments about the Panel’s questions, and Canada’s answers thereto, Brazil raises concerns regarding what it views as Canada’s failure to uphold its obligation to disclose information peculiarly within its control. Brazil notes its view that the Panel could, consistent with the principle of judicial economy, have decided the issues raised in this case on the basis of the facts before it as of the second meeting, rather than offer Canada yet another chance to present information in its defense (para. 4.126). Brazils states that regardless, the Panel exercised its right, pursuant to Article 13 of the WTO Understanding on Rules and Procedures Governing the Settlement of Disputes (“DSU”), to request further information from Canada, and in Brazil’s view, this request triggered Canada’s obligation, again under the terms of Article 13, to “’respond promptly and fully‘” to the Panel’s request.
Brazil asserts that Canada has failed to meet this obligation, noting Canada’s repeated invocation in its answers to the Panel’s questions of “’the lack of adequate procedures to protect business confidential information‘” to justify its failure to provide information specifically requested by the Panel. Brazil argues that where documentary evidence has been provided by Canada, it has been redacted to such an extreme degree that it contributes nothing to the Panel’s understanding of the programmes and issues involved. Brazil states that in other instances, Canada has selectively provided particular pages, lines or figures from various documents, claiming that these extracts support its defense.
Brazil contends that Canada must bear the consequences of its decision to withhold information from the Panel. Where Canada has expressly refused to provide documentary information specifically requested, or where Canada has wholly or selectively redacted documentary information specifically requested, in Brazil’s view the Panel should adopt adverse inferences, presuming that the information withheld constitutes inculpatory evidence of Canada’s infringement of the SCM Agreement. In Brazil’s view, adoption of such adverse inferences is not prohibited by the
WT/DS70/R
Page 33
DSU, and is entirely consistent with the practice of international tribunals such as the Iran-US Claims Tribunal,78 the Inter-American Commission on Human Rights,79 and the Inter-American Court of Human Rights.80 Moreover, Brazil asserts, reliance on adverse inferences in instances where a party fails to produce information specifically requested and within the party’s exclusive possession and control provides a Panel with the only practical means at its disposal of upholding the duty of collaboration recognized in Argentina - Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items as central to the WTO dispute settlement process, and of upholding the obligation incumbent on a WTO Member under Article 13 of the DSU to “’respond promptly and fully to any request by a panel‘” for information.
Brazil argues that Canada’s claims concerning the “’inadequacy‘” of the Panel’s confidentiality procedures, if permitted, would make a sham of WTO dispute settlement proceedings concerning activities alleged to violate the SCM Agreement. Brazil submits that in requesting that private parties waive their confidentiality rights, Canada apparently represented that “’the Brazilian government (including Embraer)‘” would receive copies of confidential business information.81 Brazil states that this is simply untrue, and that it calls into question the sincerity of Canada’s effort to secure the agreement of private parties to the release of particular information.82
More importantly according to Brazil, activities reviewable under the SCM Agreement will nearly always involve contributions to a private party, and will therefore depend upon information which can be characterized by a private party as sensitive or business confidential. If dispute settlement proceedings under the SCM Agreement are to be at all meaningful, Brazil maintains, WTO Members must not be permitted to appeal to domestic confidentiality concerns as a way to impede a Panel’s right, under Article 13 of the DSU, to seek information, or its obligation, under Article 11 of the DSU, to undertake “’an objective assessment of the facts of the case and the applicability of and conformity with the relevant covered agreements . . .‘” Brazil states that it is for these reasons, and to ensure consistency with the obligation, under Article 3(10) of the DSU, of WTO Members to engage in dispute settlement in good faith, that Brazil has been responsive to questions requiring the submission of confidential business information to a separate Panel reviewing the consistency of Brazil’s PROEX programme with the terms of the SCM Agreement.
Moreover, Brazil states, in some instances, Canada’s decision to withhold documentation from the Panel is based not upon the refusal of private parties to waive their confidentiality rights, but instead upon the Canadian government’s own refusal to release documents prepared for itself by itself. In particular, Brazil directs the Panel’s attention to Canada’s refusal to provide Canadian government documents constituting TPC “’project assessments and funding decisions.‘”(para. 6.260.) Brazil submits that the decision to withhold these documents, which are exclusively within the possession and control of the Canadian government, must carry adverse consequences for Canada, or the good faith obligation contained in Article 3(10) of the DSU will become meaningless. Brazil
78 INA Corporation v. The Government of the Islamic Republic of Iran, Award No. 184-161-1 (13 August 1985), reprinted in 8 Iran-US CTR, 373-384, at Westlaw pg. 8 (extracted in Exh. BRA-100); William J. Levitt v. Islamic Republic of Iran, Ministry of Agriculture and Natural Resources of Iran, et al., Award No. 520-210-3 (29 August 1991), reprinted in 27 Iran-US CTR, 145-187, at paras. 56-66 (extracted in Exh. BRA-101).
79 Tom Farer, “Finding the Facts: The Procedures of the Inter-American Commission on Human Rights of the Organization of American States,” in Fact-Finding by International Tribunals (R. Lillich, Ed., 1991), pgs. 275-287, at 281 (extracted in Exh. BRA-102).
80 Thomas Buergenthal, “Judicial Fact-Finding: Inter-American Human Rights Court,” in Fact-Finding by International Tribunals (R. Lillich, Ed., 1991), pgs. 261-274, at 266 (extracted in Exh. BRA-103).
81 Exh. CDN-106 (letter to Blair Hankey, Associate General Counsel, Trade Law Division, Department of Foreign Affairs and International Trade, from Peter Keyser, Manager, Business Development Programmes,
Allied Signal, at pg. 2).
82 The Panel notes that Canada, in its 29 January 1999 cover letter transmitting Canada’s comments on the draft descriptive part of this report, denies as unfounded and untrue this allegation by Brazil.
WT/DS70/R
Page 34
argues that he Panel should presume that these documents contain information prejudicial to Canada’s position.
PROCEDURES FOR PROTECTION OF BUSINESS CONFIDENTIAL INFORMATION
Procedure Proposed by Canada
Canada requests that the Panel adopt as part of its working procedures, pursuant to DSU Article 12.1, a procedure for the protection of confidential proprietary information that may be submitted to the Panel. Canada requests that the proposed procedures be adopted prior to the deadline for Brazil’s first submission. Canada notes that such a procedure is necessary as, in making its defence, it may have to submit evidence to the Panel that contains confidential proprietary business information.
Canada acknowledges that some provisions of the Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU) address the need for confidentiality. Article 18.2 of the DSU provides:
“Written submissions to the panel or the Appellate Body shall be treated as confidential, but shall be made available to the parties to the dispute. Nothing in this Understanding shall preclude a party to a dispute from disclosing statements of its own positions to the public. Members shall treat as confidential information submitted by another Member to the panel or the Appellate Body which that Member has designated as confidential. A party to a dispute shall also, upon request of a Member, provide a non-confidential summary of the information contained in its written submissions that could be disclosed to the public.”
Canada cites paragraph 3 of the Working Procedures (DSU Appendix 3):
“The deliberations of the panel and the documents submitted to it shall be kept confidential. Nothing in this Understanding shall preclude a party to a dispute from disclosing statements of its own positions to the public. Members shall treat as confidential information submitted by another Member to the panel which that Member has designated as confidential. Where a party to a dispute submits a confidential version of its written submissions to the panel, it shall also, upon request of a Member, provide a non-confidential summary of the information contained in its submissions that could be disclosed to the public.”
For Canada, however, these DSU provisions do not provide adequate procedural protection for confidential proprietary business information.
Canada maintains that there is scant WTO practice with respect to the protection of such information in the context of a dispute, but that this problem has arisen in a previous dispute under the SCM Agreement (Indonesia – Certain Measures Affecting the Automobile Industry)83 in which the United States indicated in its first submission that it possessed proprietary business information relevant to its serious prejudice claims. Canada notes that the United States was reluctant to provide such information to the Panel unless the Panel first established adequate procedures to protect that
83 Indonesia – Certain Measures Affecting the Automobile Industry (Complaints by Japan, European Communities, United States) (1998), WTO Doc. WT/DS54,55,59,64/R (Panel Report) [hereinafter Indonesia – Automobiles].
WT/DS70/R
Page 35
information,84 and Indonesia requested the Panel to require the United States to submit this information before the first substantive meeting of the Panel with the parties.85
Canada recalls that the Indonesia Panel did not require the United States to submit the requested information, noting that there was no obligation on parties to present all factual information in its first submission, but nonetheless found:
“Finally, we would like to remind all parties that Article 18.2 of the DSU does allow parties to designate information as confidential. Such designation will be respected by this Panel, the WTO Secretariat and the other parties to the dispute. Accordingly, we encourage all parties to submit to the Panel such information as they consider may be helpful to the resolution of this dispute. In this respect, we note that the parties agree that the complainants alleging serious prejudice must demonstrate its existence by positive evidence. If the United States considers that the information in question is necessary in order to meet that burden, and if it believes that Article 18.2 is inadequate, the United States may propose to the Panel in writing, at the earliest possible moment, a procedure that it considers sufficient to protect the information in question.”86 [emphasis added]
Canada notes that the United States did not propose or request the Panel to adopt any such procedure, and that the Panel found that while complainants could not be required to submit confidential business information, neither could they “…invoke confidentiality as a basis for their failure to submit the positive evidence required, in the present case, to demonstrate serious prejudice under the SCM Agreement.”87 The Panel concluded that the United States had not demonstrated serious prejudice by positive evidence.88
Canada submits that as complainant in the Indonesia - Automobiles case, the United States had a choice as to whether it would submit evidence necessary to meet its evidentiary burden. Canada argues that in this case, if Brazil meets its prima facie burden, Canada cannot afford the luxury of not adducing the evidence necessary to defend its impugned programmes, and may have to rely upon confidential proprietary business information.
Canada submits that the protection of confidential proprietary business information is a new and significant challenge to the WTO dispute settlement process, in which the Panel must balance two competing interests, both deeply rooted in fairness and due process, neither in itself having a claim to better protection than the other: first, that reasonable access to such information, when introduced into evidence, must be provided to the Panel and the other disputing parties; second, that additional procedural safeguards are necessary to provide private business interests with adequate protection for their proprietary business information when a disputing party deems it necessary to present such evidence in support of its case. Canada believes that its proposed procedure strikes the necessary balance.
The procedure proposed by Canada would allow access to business confidential information only to persons having signed a declaration of non-disclosure and who were either panelists, assistants to panelists, WTO Secretariat staff, PGE members, or “representatives” of parties, the latter defined as employees of or agents for parties, excluding employees, officers or agents of private companies engaged in aircraft manufacturing. The procedures as proposed would require business confidential information to remain on the premises of the WTO Secretariat, which would be the only place where
84 Id.. at para 4.56.
85 Id.. at para. 4.51.
86 Id.. at para. 14.7.
87 Id.. at para. 14.235.
88 Id.. at para. 14.236.
WT/DS70/R
Page 36
it could be consulted. The information would need to be kept in a locked receptacle at the WTO Secretariat and the Secretariat staff would control and monitor access to and use of the information. Approved persons could take summary notes on the information but would not be permitted to copy or distribute it, or remove it from WTO premises. Only approved persons would be permitted to be present in panel meetings at which business confidential information was discussed and tapes of such meetings would be subject to the same procedures as the written material submitted. At the conclusion of the panel process, the BCI would be returned to the submitter and all tapes and transcripts would be destroyed.
Arguments of Brazil
Brazil, while receptive to Canada’s proposal, outlined a number of concerns, in light of which it proposed a modified procedure.
Advisors of a Party should be considered an approved person
Brazil notes that Canada’s definition of “representative” includes reference to employees, agents and legal counsel but omits reference to other types of representatives or advisors. Brazil states that the complexity of the aircraft financing market is such that much of the information that Brazil would want to present to the Panel in support of its position would of necessity rely upon the analysis of experts and others knowledgeable about the industry. For Brazil it is crucial to be able to share confidential business information with its advisors to rebut arguments presented by Canada. Brazil agrees with Canada that in no circumstances should the definition of an advisor or other representative include an employee, officer or agent of a private company engaged in aircraft manufacturing.
Review of Confidential Business Information should be permitted outside the WTO premises
Brazil notes that Canada’s proposal omits reference to copies submitted to any party, and that DSU Article 18.2 provides that “[w]ritten submissions . . . shall be made available to the parties to the dispute.” Brazil submits that Article 18.2 requires Canada to provide a complete copy of its submissions, and that forcing a party, a third party, a panel member or a PGE member to travel to the WTO to review submissions places enormous administrative burdens on approved parties. Especially given the short time-frames in this dispute, Brazil submits that an approved person – whether a party, a panel member or a PGE member – must have unrestricted access to the information. Brazil argues that a copy of such information therefore should be permitted to be kept in secure locations outside the premises of the WTO.
In Brazil’s view, by limiting the access to such information to the premises of the WTO, Canada would prohibit effective review of confidential business information, information that may be crucial to the case, and in so doing, would effectively bar any approved party from analyzing portions of submissions containing confidential business information. For Brazil, this would significantly hamper any substantive rebuttal of Canada’s submissions.
Brazil argues that providing a party copies of the confidential business information presents no risk to Canada, as each party is bound to ensure that its representatives comply with the non- disclosure obligation. Brazil notes that DSU Article 18.2 provides that “Members shall treat as confidential information submitted by another Member . . . which that Member has designated as confidential.”
Brazil also notes the Vienna Convention on the Law of Treaties, which states in Article 31 that “a treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to
WT/DS70/R
Page 37
the terms of the treaty in their context and in the light of its object and purpose.”89 In Brazil’s view, Canada’s original proposal is too restrictive in its limits regarding access to confidential business information; as written, it would prevent Canada’s submissions from being made effectively “available” to Brazil or to any other “approved person”, including a panel member, in direct contradiction to the spirit of the DSU and to the provisions of DSU Article 18.2.
Third Parties should be granted access to Confidential Business Information
Brazil notes that Canada’s definition of “approved persons” omits reference to third parties, although DSU Article 10.3 provides that “Third parties shall receive the submissions of the parties to the dispute to the first meeting of the panel,” and paragraph 6 of the Working Procedures (DSU Appendix 3) provides that “All third parties . . . shall be invited . . . to present their views during a session of the first substantive meeting of the panel . . . All such third parties may be present during the entirety of this session.” (emphasis added by Brazil). For Brazil, since third parties are entitled to receive written submissions to the first panel meeting and are entitled to be present during the entire first panel meeting, they should also be granted access to confidential business information disclosed by a party in its first submission or at the first panel meeting.
Brazil believes that the proposed provisions requiring that only approved persons be present at panel meetings where business confidential information is discussed are redundant. By adding third parties to the definition of approved parties, as required by DSU Article 10.3 and paragraph 6 of the Working Procedures (DSU Appendix 3), Brazil submits, no persons other than approved persons may be present at a panel meeting. Brazil notes Canada’s acknowledgement of paragraph 3 of the Working Procedures (DSU Appendix 3): “The deliberations of the panel . . . shall be kept confidential.” Brazil also does not object to the requirement that parties and third parties provide a non-confidential summary of the information contained in written submissions that could be disclosed to the public at the same time that they provide their submission to the panel.
A complete administrative record should be maintained for the Appellate Body
Brazil notes that Canada’s proposal omits reference to the Appellate Body, and recalls that in accordance with the provisions of DSU Article 17, the Appellate Body may hear appeals from panel cases and may uphold, modify or reverse the legal findings and conclusions of the panel. To do so, Brazil submits, the Appellate Body must have unrestricted access to a complete administrative record. Destroying confidential business information at the end of the panel process might seriously limit the Appellate Body’s ability to obtain a true understanding of the factual arguments made by the parties.
Action by the Panel
After considering the arguments of the parties, the Panel transmitted to the parties procedures for the protection of business confidential information and indicated that if either party identified concerns that the Panel considered sufficiently serious to call into question the type of information to be submitted by the parties, it would consider amending the procedures. Comments were received from Brazil, which requested that the procedures include a requirement that one copy of business confidential information be served on each party’s Geneva mission, and from Canada, which objected to Brazil’s proposal as in Canada’s view such a procedure would provide insufficient protection of private interests.
89 Vienna Convention on the Law of Treaties, Art. 31; see also, United States - Standards for Reformulated and Conventional Gasoline, WTO Doc. WT/DS2/R at para. 6.7.
WT/DS70/R
Page 38
The Panel, after considering these comments, adopted the procedures for the protection of business confidential information contained in Annex 1 to this report.90
Statements of Canada
After the Panel adopted the procedures, Canada sent a letter to the Panel indicating that it did not consider the procedures as adopted to confer sufficient protection for business confidential information. Canada indicates a serious concern regarding the protection of confidential business information, and states that it cannot divulge confidential business information in the absence of adequate procedures to protect that information. Canada notes that it proposed procedures that it considered sufficient to protect confidential business information.91 In Canada’s view, the original procedures adopted by the Panel governing confidential business information provided a sufficient level of protection for that information.
Canada notes that its position regarding the treatment of certain sensitive confidential information is not an indication of a lack of good faith. In Canada’s view, such information necessitates the strongest possible confidentiality procedure. The procedure suggested by Canada is such a procedure and it is clearly workable within the context of WTO dispute settlement proceedings. Insofar as it provides Members with the confidence to provide such information, such a procedure advances the goals and objectives of the WTO.
Canada submits that requiring a Party to hand over business confidential information to an adverse Party having a legal and commercial interest in that material and the absence of legal sanctions for the breach of these procedures, does not constitute effective procedures for the protection of such information. In putting in place such a procedure and subsequently requesting in its questions of 27 November and 10 December 1998 that Canada produce business confidential information, Canada argues that the Panel has placed it in a very difficult position.
Canada also stated that, while it would produce certain of the business confidential information requested by the Panel, it would do so only where it had been able to obtain releases from the private parties to the pertinent transactions, waiving their contractual rights to confidentiality.
Canada notes that in its view, the procedures governing confidential business information as modified at Brazil’s request do not provide the requisite level of protection for any confidential business information that Canada would adduce. Canada argues that this information is of extreme sensitivity and is well protected in domestic law, under which the release of such information is subject to civil and criminal sanctions. Canada submits that in contrast, the WTO system does not provide for such sanctions, so that if a breach were to occur, there would be no effective recourse except potentially under domestic law against the very government that produced the evidence in compliance with a request from a Panel, an impossible situation. As a result, Canada states, it is not in a position to provide confidential business information to the Panel. Canada notes that Brazil’s concerns over access could be overcome by special arrangements for 24-hour access to business confidential information at the WTO Secretariat and at any Canadian mission or embassy designated by Brazil.
The Panel asked Canada to explain how the level of protection under the procedures ultimately adopted by the Panel for the protection of confidential business information differs, in substance, from what would have been afforded by the procedures to which Canada indicates it could have agreed, given that under these latter procedures, only good faith would prevent a representative
90 The Panel notes that Canada, on 21 December 1998, sought clarification in the procedures to indicate that the “representative” referred to in Articles VI.2 and VIII.3 shall be an approved person. The procedures in Annex 1 reflect this change.
91 Indonesia –Automobiles, WT/DS54,55,59,64R (Panel Report) at para. 14.7.
WT/DS70/R
Page 39
of a party that had signed a non-disclosure form from taking verbatim notes on the confidential business information on the premises of either the WTO or the opposing party, removing those notes from those premises and then disclosing them to non-authorized persons. The Panel asked whether exactly the same degree of good faith would not be involved under either set of procedures.
In response to this question, Canada indicated that the Panel’s amended confidentiality procedures do not adequately protect private sector interests and the interests of the Government of Canada in maintaining effective control over the dissemination of business confidential information. In particular they may not be adequate to protect the Government of Canada from potential liability under domestic law, in the event the confidentiality provisions are not strictly observed.
In Canada’s view, the essential difference between the confidentiality procedures originally proposed by the panel and those ultimately adopted are that the original procedures entrusted the commercial confidential materials to the care of a neutral third party, i.e. the WTO Secretariat. The modified procedures oblige Canada to entrust the materials to the care of the opposing Party, which Party, or at least nationals of the Party on whose behalf the present case is being pursued, may have an immediate and commercial interest in these materials.
For Canada, under the original procedures, any breach of these procedures is likely to take the form of detailed or verbatim notes on a document, as is implied in the Panel’s question. Under the modified procedures, breach could take the form of a photocopy of the actual document. In Canada’s view, the difference between the commercial and legal effects of the two potential forms of breach are substantial.
As to commercial effects, Canada submits that the uncertain authenticity of notes on a document will diminish its utility and correspondingly decrease the potential commercial injury that might be caused by its disclosure to an interested party. For example, in negotiations for the sale of aircraft, a purchaser might pass a copy of the term sheet proposed by one manufacturer/seller to the agents of a competing manufacturer/seller, as a means of inducing the first manufacturer/seller to lower its price or otherwise improve the terms of its offer. Canada views a copy of a term sheet as a far more effective inducement for the manufacturer/seller to vary his proposal than the assertion by the purchaser that the competing manufacturer/seller has offered better terms, even if the specific content of the alleged better terms is detailed by the purchaser.
As to legal liability, in Canada’s view any claim, based on the production of detailed or verbatim notes, that business confidential information was improperly disclosed would constitute hearsay under Canadian law, and its evidentiary value would be limited. A photocopy of an original document, however, would constitute best evidence.
ARGUMENTS OF THE PARTIES REGARDING LEGAL ISSUES CONCERNING SCM ARTICLES 1 AND 392
GENERAL
Arguments of Canada
Principles of Treaty Interpretation
Canada observes that Article 3.2 of the Understanding on Rules and Procedures Governing Dispute Settlement (DSU) provides that the WTO dispute settlement system serves to clarify the existing provisions of the covered agreements in accordance with “customary rules of interpretation of
92 Except as otherwise noted, the footnotes and citations, and the emphasis in the text in this section are as contained in the parties’ submissions.
WT/DS70/R
Page 40
public international law.” Canada recalls the Appellate Body’s view that the rules of treaty interpretation set out in Articles 31 and 32 of the Vienna Convention on the Law of Treaties (Vienna Convention)93 have attained the status of customary or general international law.94
Canada states that according to Article 31 of the Vienna Convention, an international treaty must be interpreted “in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.”95 For Canada, Article 32 of the Vienna Convention provides that supplementary means of interpretation, including the negotiating history, may also be used to confirm an interpretation of the agreement or to resolve ambiguities in the text.96
Canada submits that several other interpretative principles are also applicable to the present case, recalling that the Appellate Body has noted that the principle of effectiveness flows from Article 31:
“One of the corollaries of the “general rule of interpretation” in the Vienna Convention is that interpretation must give meaning and effect to all terms of the treaty. An interpreter is not free to adopt a reading that would result in reducing whole clauses or paragraphs of a treaty to redundancy or inutility.”97
For Canada, similarly, good faith underlies the principle that the interpreter must avoid interpreting the treaty in a way that would lead to a manifestly absurd or unreasonable result.98
Burden of Proof
Canada observes that in its report in United States - Measure Affecting Imports of Woven Wool Shirts and Blouses from India,99 the Appellate Body dealt extensively with the issue of allocation of the burden of proof,100 stating:
“In addressing this issue, we find it difficult, indeed, to see how any system of judicial settlement could work if it incorporated the proposition that the mere assertion of a claim might amount to proof. It is, thus, hardly surprising that various international tribunals, including the International Court of Justice, have generally and consistently accepted and applied the rule that the party who asserts a fact, whether the claimant or the respondent, is responsible for providing proof thereof. Also, it is a generally-accepted canon of evidence in civil law, common law and, in fact, most jurisdictions, that the burden of proof rests upon the party, whether complaining or defending, who asserts the affirmative of a particular claim or
93Vienna Convention on the Law of Treaties (Vienna Convention), 1969, 1155 U.N.T.S. 332 (entered into force 1980) (Exh. CDN-4).
94United States - Standards for Reformulated and Conventional Gasoline (Reformulated Gasoline), W/DS2/AB/R, Report of the Appellate Body adopted on 20 May 1997, at 16-17; Japan - Taxes on Alcoholic
Beverages (Japan Liquor Tax), WT/[DS8/DS10/DS11]/AB/R, Report of the Appellate Body adopted on 11 November 1997 at 10; most recently reaffirmed in India - Patent Protection for Pharmaceutical and Agricultural Chemical Products (Pharmaceuticals), WT/DS50/AB/R, Report of the Appellate Body adopted on 16 January 1998, at paras. 43-48.
95Vienna Convention, at Article 31.
96Id. at Article 32.
97 Reformulated Gasoline, at 23.
98I. Sinclair, The Vienna Convention on the Law of Treaties, 2nd rev. ed. (Manchester U.K.: Manchester University Press: 1984) at 120 (Exh. CDN-5).
99WT/DS33/AB/R, Report of the Appellate Body adopted on May 23, 1997 (“Shirts and Blouses”).
100Id.., at 12-17.
WT/DS70/R
Page 41
defence. If that party adduces evidence sufficient to raise a presumption that what is claimed is true, the burden then shifts to the other party, who will fail unless it adduces sufficient evidence to rebut the presumption.”101 [emphasis added and footnotes omitted by Canada]
For Canada, a prima facie case of inconsistency means that the complainant must adduce sufficient evidence that would permit a tribunal to find for the complainant in the absence of evidence to the contrary.102 The respondent must then present enough evidence to cast doubt on the proposition of the complainant to prevent the tribunal from finding in favour of the complainant.
According to Canada, international tribunals, for obvious reasons, have refused to accept unsupported statements of the parties as evidence. Canada recalls the statement by the British- Mexican Claims Commission of 1926, “’if an international tribunal were to accept all ... allegations without evidence, it would expose itself to the not unjustifiable criticism of placing jurisdiction as between nations below the level prevailing in all civilized States for jurisdiction as between citizens.‘”103
A complainant must present a prima facie case in respect of each element of its claim.
For Canada, a prima facie case is one which, in the absence of effective refutation by the defending party, requires a panel, as a matter of law, to rule in favour of the complaining party presenting the prima facie case.104 Canada states that the Appellate Body has observed that a panel should begin its analysis of each legal provision by examining whether the complaining party has presented evidence and legal arguments sufficient to demonstrate that the challenged measures are inconsistent with respect to each relevant legal provision for which an inconsistency is alleged,105 and that for claims with multiple elements, all the elements of that claim must be established.106
Thus, in Canada’s view, Brazil must present a prima facie case with respect to each element of its claim, that is, Brazil must demonstrate that a programme, activity or transaction at issue is:
a subsidy within the meaning of Article 1 of the SCM Agreement, that is, there is a “financial contribution” within the meaning of Article 1.1(a) that imposes a net cost on the government making the contribution and that confers an advantage above and beyond the market; (paras. 5.28 - 5.38) and
contingent on export performance, in the sense that it would not be paid unless exports took place, that there would be rewards if exports took place or that there would be penalties if exports did not take place (paras. 5.55 -5.80).
The principles applicable to the evaluation of evidence
101Id.., at 14.
102See M. Kazazi, Burden of Proof and Related Issues: A Study on Evidence Before International Tribunals (The Hague: Kluwer Law International, 1996) at 326-343 (Exh. CDN-6).
103Id.., at 39-42 quoting Further Decisions and Opinions of Commissioners, Claim of W. Allen Odell, p.
63 (Exh. CDN-6).
104 European Communities - Measures Concerning Meat and Meat Products (European Communities - Hormones), WT/DS26/AB/R and WT/DS48/AB/R, Report of the Appellate Body adopted on 13 February 1998, at para. 104.
105 Id.. at para. 109.
106 Id.. at paras. 214-15.
WT/DS70/R
Page 42
Canada notes the Appellate Body’s observation that in evaluating the evidence presented by the complaining party to determine whether a prima facie case has been made with respect to each element of its claim, a panel must be satisfied that the adduced evidence is “sufficient to raise a presumption that what is claimed is true”,107 that mere assertions of fact do not amount to proof and cannot be relied upon by a complaining party to meet this standard,108 and that it is not enough for the complaining party to merely raise a reasonable doubt as to whether a measure might be inconsistent with a WTO provision.109
Canada submits that a panel should not accept that a prima facie case has been established where the evidence presented in support of an allegation:
does not support the proposition for which it is tendered;
is internally inconsistent and does not amount to proof;
consists of press reports which are uncorroborated or do not otherwise contain material with an independent title of credibility and persuasiveness;110 or
is so meagre that it can be ignored.111
In Canada’s view, submitting a large volume of documents does not demonstrate that a prima facie case has been made. (para. 4.110). Rather, it must be shown that the evidence adduced is credible and cogent, and that it establishes a prima facie case for each element of the claim. Canada submits that applying these principles, Brazil has not made out a prima facie case against any of the programmes, activities or transactions it has impugned.
Judicial Economy
Canada argues that the principle of judicial economy has guided panels since the inception of the GATT. Canada states that, relying on GATT panel practice and Article 11 of the DSU, the Appellate Body observed in Shirts and Blouses that:
“Nothing in this provision [Article 11] or in previous GATT practice requires a panel to examine all legal claims made by the complaining party. Previous GATT 1947 and WTO panels have frequently addressed only those issues that such panels considered necessary for the resolution of the matter between the parties, and have declined to decide other issues. … In recent WTO practice, panels likewise have refrained from examining each and every claim made by the complaining party and have made findings only on those claims that such panels concluded were necessary to resolve the particular matter. …
“A panel need only address those claims which must be addressed in order to resolve the matter in issue in the dispute.”112 [italics added by Canada]
107 United States – Measures Affecting Imports of Woven Wool Shirts and Blouses from India (Shirts and Blouses), WT/DS33/AB/R, Report of the Appellate Body adopted on 23 May 1997, at 14.
108 Id..
109 India – Patent Protection for Pharmaceutical and Agricultural Chemical Products, WT/DS50/AB/R, Report of the Appellate Body adopted on 16 January 1998, at paragraph 74.
110 Keith Highet, “Evidence and Proof of Facts” in The International Court of Justice at Crossroads,
(New York: Transactional Publishers, Inc., 1987) at 367 (Exh. CDN-54).
111 Kazazi, at 326
112 Shirts and Blouses, at 18-19.
WT/DS70/R
Page 43
Canada submits that given the accelerated time-frame of the SCM Agreement Article 4 process, the principle of judicial economy is particularly important; in Canada’s view the Panel should address only those issues which must be addressed in order to resolve the matter in issue in the dispute.
Canada observes that in the present case, Brazil makes one claim with respect to each impugned Canadian programme, transaction or activity: a violation of SCM Agreement Article 3. For each of these claims, Canada states, Brazil must demonstrate two distinct and necessary elements:
that as a result of that programme, transaction or activity, a subsidy exists (SCM Agreement Article 1); and
that the subsidy is contingent, in law or in fact, upon export performance (SCM Agreement Article 3).
Canada believes that in response it suffices for Canada to show that Brazil has failed to demonstrate either one of the two necessary elements. Indeed, Canada argues, to resolve each of the claims in the dispute, if the Panel finds that either one of the two necessary elements has not been shown, then the relevant claim fails and the aim of the dispute settlement system – to secure a positive solution to the dispute -- will have been met.
In accordance with the principle of judicial economy, Canada argues, it is not necessary for this Panel to determine whether impugned programmes, activities or transactions are “subsidies”, if it finds that they are not “contingent … on export performance”, and vice versa. Canada indicates that its arguments with respect to Articles 1 and 3 of the SCM Agreement focus, therefore, on the element that allows the Panel to dispose of each of Brazil’s claims in the most efficient manner possible.
Response of Brazil
Regarding the weight to be attached to the evidence it has submitted, Brazil notes that this evidence includes statements by current and past presidents of the EDC, other EDC officials, the Canadian Department of Industry Minister, other Industry Canada officials, and members of the Canadian Parliament. Brazil refers the Panel to an article provided by Canada113. Brazil argues that Canada considers this article, which concerns the evidentiary standards applied by the International Court of Justice, to provide the relevant framework for the Panel in its consideration of the evidence presented by Brazil. Brazil agrees.
In particular, Brazil directs the Panel’s attention to the ICJ’s treatment of public statements by high government officials in the Nicaragua case, discussed thoroughly in the article, and notes a finding by the Court in that case:
“The material before the Court also includes statements by representatives of States, sometimes at the highest political level. Some of these statements were made before official organs of the State or of an international or regional organization, and appear in the official records of those bodies. Others, made during press conferences or interviews, were reported by the local or international press. The Court takes the view that statements of this kind, emanating from high-ranking official political figures, sometimes indeed of the highest rank, are of particular probative value when they acknowledge facts or conduct unfavourable to the State represented by the person who made them. They may then be construed as a form of admission.”
113 Exh. CDN-54.
WT/DS70/R
Page 44
In Brazil’s view, the Court in fact considers statements by a party that are “’against its own interest‘” to be “’of superior credibility‘” and on equal footing with evidence offered by disinterested witnesses, and deems that such statements “’may . . . be construed as a form of admission.‘”
According to Brazil, the author of the article concludes that the Court “’arrived at a process of taking at face value the public statements of high officials when such statements were arguably made against the legal interest of the state in the government of which such officials served,‘” and of taking “’the unfavourable evidence as probative, but in general to reject the favorable evidence as not being disinterested or veracious‘”. Brazil states that the author concludes that these evidentiary practices are not restricted to the particular circumstances of the Nicaragua case, but have in fact “’long been employed by the Court in normal bilateral litigations as a way of winnowing through the vast quantities of documentary assertion and evidentiary pleading to which it is normally subjected . . .‘”
Brazil reiterates that it has faced tremendous difficulty in these proceedings, given the unreasonable secrecy of the Canadian government and given Canada’s failure to fulfill its duty of collaboration, recognized in Argentina – Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items, as central to the WTO dispute settlement process. For this reason, and because Canada has apparently agreed that the ICJ’s evidentiary rules are relevant to the Panel’s weighing of the evidence, Brazil asks that the Panel attach particular weight to the contemporaneous statements by Canadian government ministers and officials offered by Brazil as evidentiary proof of its claims. Brazil argues that these statements, made against Canada’s legal interest, are, in the ICJ’s words, “’of superior credibility.’”
Brazil argues that although Canada has secured a statement by a Bombardier officer attempting to explain away his recent comments confirming Bombardier’s use of Canada Account funds at non-commercial rates, no other such statements have been offered to correct or explain the dozens of comments by current and past presidents of the EDC, other EDC officials, the Canadian Department of Industry Minister, other Industry Canada officials, and members of the Canadian Parliament offered by Brazil as “’statements against interest‘” in support of Brazil’s claims.
According to Brazil, the many statements by many Canadian ministers and officials therefore retain, in the words of the ICJ, their “superior credibility” and persuasiveness as reliable indications of the accuracy of the facts asserted. For Brazil, post-hoc rationalizations and defences offered by Canada in the course of these proceedings, coming not from the ministers or officials themselves, and lacking any contemporaneous connection to the original statements, are not under the ICJ’s practice considered credible or persuasive.
“SUBSIDY” PER SCM ARTICLE 1
Arguments of Canada
Canada argues that based on the language of SCM Article 1, for a “subsidy” to exist, two elements must be present, and there must be a causal link between the two. First, there must be either one of the government “financial contribution(s)” enumerated in subparagraphs (i) to (iv) or a form of income or price support as defined in Article XVI of GATT 1994. Second, a “benefit” must thereby be conferred.
According to Canada, the customary rules of interpretation in international law require that interpretation reflect the ordinary meaning of the words used in context and in the light of the object and purpose of the agreement in question (paras. 5.1- 5.4). As well in Canada's view, in interpreting and analysing the relevant provisions of an international agreement, the interpreter must ensure that
WT/DS70/R
Page 45
the resolution of any ambiguity does not upset the “carefully drawn balance of rights and obligations of Members”114 set down in that agreement.
The meaning of “financial contribution”
Canada submits that under Article 1.1(a)(1), the term “financial contribution by a government or any public body” is defined exhaustively115 and is limited to the circumstances described in the four sub-paragraphs (i) to (iv). Accordingly, Canada states, if the measures at issue do not fall within any of the items in (i) to (iv), then the practices cannot be considered to be “financial contributions by government or any public body”.
The meaning of “benefit”
Canada notes that the term "benefit" is not defined in the SCM Agreement or elsewhere in the WTO Agreement, and argues that to determine under what circumstances a benefit is "conferred", one must turn to the customary rules of treaty interpretation, in particular Article 31 of the Vienna Convention.
Canada submits that the word "benefit" as it is found in Article 1.1(b) of the SCM Agreement has two components: a benefit is conferred when a financial contribution by a government a) imposes a cost on the government, and b) results in an advantage above and beyond what the market could provide. For Canada, this conclusion flows from the ordinary meaning of "benefit", the context in which it is found and the object and purpose of the SCM Agreement read as a whole.
Ordinary meaning
Canada submits that the relevant ordinary meaning of benefit is "advantage", but that the dictionary definition does not adequately narrow the term, i.e., that it contains nothing that would distinguish a subsidy from normal commercial activity. For Canada, any commercial contract could accord an advantage to a firm relative to its competitors, and therefore it is necessary also to consider context and object and purpose.
Context
Canada argues that “benefit” should not be interpreted simply to mean “the advantage given beyond commercial or market activity”, an interpretation that in Canada's view Brazil makes (paras. 6.60, 6.148, 6.179). According to Canada, viewed in the context of the SCM Agreement, including its annexes, “benefit” and hence “subsidy” requires a more nuanced and sophisticated approach.
Canada notes that under the SCM Agreement the calculation of subsidy may be done in one of two ways: Article 14 sets out the guidelines, based on a market-referenced “benefit to the recipient” test, that are to be used for determining the maximum value of countervailing duties to be imposed. Canada further notes that for the purpose of determining the amount of ad valorem subsidisation under Article 6.1, Annex IV sets out a “cost to government” test.
114United States – Restrictions on Imports of Cotton and Man-made Fibre Underwear (United States - Cotton Underwear) WT/DS24/AB/R, Report of the Appellate Body adopted 25 February 1997, at 15.
115That is, for Canada, the use of the term “i.e.” indicates that the items that follow are intended to be
an exhaustive, rather than an illustrative list. In this regard, Canada notes that the Uruguay Round negotiators agreed to replace the term “such as” in the Cartland I draft of the SCM Agreement (18 May 1989)(Exh. CDN-21), with “i.e.” in Cartland II (Exh. CDN-22), thus clearly indicating an intent to move from an illustrative definition to an exhaustive definition. Canada also notes that “e.g.” is also used within the subsidy definition in Article 1 to indicate an illustrative list.
WT/DS70/R
Page 46
According to Canada, neither test defines “benefit” or provides a definitive guide as to when a “benefit” exists under Article 1, but taken together, they provide important contextual guidance as to the types of governmental measures that could be considered “subsidies”, and therefore the measures that could be considered to have conferred a benefit.
Canada submits that this contextual element should be considered in the light of the consideration that the SCM Agreement should be interpreted so as to function logically and seamlessly. In Canada's view, taking one test -- as Brazil seems to have done -- without considering how it would work within the SCM Agreement as a whole would have the potential of rendering the SCM Agreement nonsensical, i.e., if “benefit” were based solely on a market-based “benefit to the recipient”, a “subsidy” found under Article 1 might be found under Article 6.1 to have no value. For that reason, Canada argues, “benefit” must be interpreted in such a way as to ensure that a subsidy found to exist under Article 1 would also be found to exist -- even if the values were different -- under both tests set out in the SCM Agreement.
The object and purpose of the SCM Agreement
Canada maintains that a pure “benefit to the recipient” approach is not consistent with the object and purpose of the SCM Agreement, as demonstrated by anomalous results set out in the “Finan Report” and relied upon by Brazil.116 In this regard, Canada states that Brazil argues using this approach (para. 6.184) that the “subsidy value” of a repayable contribution of $87 million was between $94 million and $211 million -- that is, potentially over twice the subsidy value of an equivalent outright grant.117
Canada submits that Brazil thus implicitly argues that there is more distortion in the international economy when a government requires that research and development contributions be repaid, than when such contributions are made with no expectation of return or repayment. In Canada's view, to the extent that the object and purpose of the SCM Agreement is to restrict the use of trade distorting subsidies, Brazil’s approach is not consistent with that object and purpose.
Canada submits that the “mischief” that the Agreement seeks to discipline are measures that distort the market by a) imposing a cost on the treasury of the providing Member, and b) an advantage to the recipient above and beyond the market.
Canada further argues with respect to credit terms that the first paragraph of Item (k) of the Illustrative List of Export Subsidies of Annex I of the SCM Agreement (Annex I) provides a specific contextual indication of what constitutes a subsidy. Canada notes that the first paragraph of Item (k) identifies two elements in determining whether particular credit terms are subsidies: first, where governments provide credit at rates below those which they have to pay for the funds so employed, and second, where such credit secures a material advantage in the field of export credit terms. For Canada, the test in determining whether government credit is a subsidy is therefore whether there is a net cost to the government, and whether as a result an advantage is granted above and beyond what the market would provide.
Arguments of Brazil
Brazil argues that Article 1.1 includes a three-part test: first, paragraph (a)(1) of Article 1.1 requires that, for a subsidy to occur, there must be a “financial contribution by a government or any
116 The “revenue foregone” is not to be confused with the “cost to government” method set out in Annex IV. The cost to the Government of Canada of a contribution cannot, by definition, exceed the value of that contribution as an outright grant. The “cost” to the Government, in the case of a TPC contribution, therefore would be calculated by subtracting repayments from the original contribution.
117 The Parties’ arguments regarding the Finan Report are set forth in Section VI.F of this report.
WT/DS70/R
Page 47
public body within the territory of a Member . . .”; second, under clauses (i) through (iv) of paragraph (a) of Article 1.1, the financial contribution may take one of several forms; third, whether a “benefit” is conferred, within the meaning of paragraph (b) of that Article. For Brazil, Canada’s interpretation of the word “benefit” would rob the term of its ordinary meaning, and indeed, of any recognizable meaning.
Brazil, considering the ordinary meaning of the phrase “a benefit is . . . conferred”, notes that Webster’s Third New International Dictionary defines the verb “confer” as to “grant,” “bestow” or “give.” The noun “benefit” means “advantage” or “something that guards, aids, or promotes well- being.” To “aid,” in turn, means “to give help or support to.” Alternatively, an “advantage” is “a more favorable or improved position or condition.” For Brazil, Article 1.1 therefore states that a subsidy exists where a government contributes something, and in so doing gives help or support, or improves the recipient’s condition. For Brazil, EDC and the Canada Account, in granting assistance, improve a recipient’s condition by granting the recipient something, in words taken directly from Canada’s first written submission, “above and beyond the market.”
Brazil submits that the structure of Article 1.1 requires this construction: paragraph (a) of Article 1.1 tells about the source of the financial contribution – the government; as structural matter, paragraph (b), in requiring that a “benefit” be conferred, must then speak to the effect this contribution has on its recipient.
Brazil notes that Canada appears to agree that demonstrating a “benefit to the recipient” is one part of the “benefit” test, but argues that Canada invents a second, additional requirement – that the government, in making its contribution, realize a “net cost.” Brazil rejects each of Canada’s reasons for this additional asserted requirement.
Brazil disputes Canada’s argument that applying the ordinary meaning of the term “benefit” – which in Brazil’s view means “advantage” or “aid” – would not “adequately narrow the term”, arguing that no provision of the Vienna Convention requires, that the ordinary meaning of a term be narrowed in order for it to be valid. Brazil notes Canada’s statement that a broad definition of “benefit” could mean that a “commercial contract” could possibly be considered a “subsidy,” but argues that such a contract is not, without something more, a subsidy prohibited by the terms of the SCM Agreement. Brazil asserts that Canada’s concern is with the scope of the prohibitions included in the SCM Agreement, and not the definition of the term “ subsidy,” and is therefore misplaced.
Brazil also disagrees that item (k) of Annex I to the SCM Agreement offers “context” supporting Canada’s proposed test. According to Brazil, Annex I does not speak to whether government activity constitutes a subsidy, but rather to whether government activity constitutes a prohibited export subsidy. Brazil submits that a measure may constitute a subsidy, but not be on the Illustrative List of Export Subsidies included in Annex I. For Brazil, the relevant reference to the Illustrative List of Export Subsidies included in Annex I falls in Article 3, which specifies, among those activities already identified as subsidies, those that are prohibited by the SCM Agreement. Brazil states that Article 1 of the Agreement contains no reference to Annex I relevant to the definition of a subsidy, and that the Canadian definition must therefore be rejected.
Brazil also takes issue with Canada’s view that paragraph (a) of Article 6.1 of the SCM Agreement provides contextual support for its net cost argument given that Article 6.1(a) requires that for purposes of that paragraph the value of a subsidy be calculated on the basis of cost to the granting government. Brazil disagrees with Canada that if “benefit” under Article 1.1 means “benefit to the recipient,” a government contribution identified under Article 1.1 as a “subsidy” could not be valued under Article 6.1 and Annex IV, stating that Canada ignores that valuation is only one of several ways to establish “serious prejudice” under Article 6.1 of the SCM Agreement. For Brazil, if the particular subsidy in question is defined based on the “benefit to the recipient,” “serious prejudice” could still be found based on the other alternative provisions of Article 6.1 which characterize the effect of the
WT/DS70/R
Page 48
subsidy according to the extent of the benefit realized by the recipient. Moreover, for Brazil, Article
6.1 is not relevant to an export subsidy, which need not be quantified or subject to specific valuation in order to trigger the prohibition of Article 3.
Regarding Canada's argument that using a “benefit to the recipient” test is against the object and purpose of the SCM Agreement, which for Canada is “'to restrict the use of trade distorting subsidies'” , Brazil finds incomprehensible Canada's implication (see para. 5.37) that the type of benefit estimated by Brazil under TPC's $87 million contribution to Bombardier is not, consistent with the object and purpose of the SCM Agreement, “’trade distortive.‘” Brazil also does not understand how Canada's argument demonstrates that the “benefit to recipient” test is inconsistent with the object and purpose of the Agreement.
Brazil notes that Revenue Canada’s Special Import Measures Act Handbook states that while the determination of “benefit” on a government loan is generally related to whether the government recovers its costs, such a test will not always capture the true effect of the subsidy. According to Brazil, the Handbook states that:
“[I]t is also possible that a benefit would accrue to an exporter or an importer as a result of a government guarantee which would not necessarily result in a cost to the government. The benefit could be a lower interest rate or a loan at a commercial rate which the company would otherwise not get without government involvement.”
Furthermore, according to Brazil, in defining a subsidy,” the Canadian Handbook states that a “benefit” can be direct or indirect:
“A direct financial or other commercial benefit is one which accrues directly to the person, firm or industry which is the intended recipient, such as an outright grant of funds to a producer of goods. An indirect benefit is one which does not accrue directly, but which alters the economic environment within which firms operate, and hence the level of their costs.”
For Brazil, therefore, Canada’s position concerning the definition of “subsidy” and “benefit” before this Panel is irreconcilable with that adopted in its own law, noting that in the Handbook, Canada agrees with a definition of the term “subsidy” which focuses on whether a benefit accrues either directly to the recipient, or somehow alters the recipient’s, and not the grantor’s, costs. Brazil also states that Canada agrees with Brazil that covering the government’s cost does not necessarily mean that the recipient is getting a rate it would otherwise get on the market, without government involvement.
For Brazil, neither the ordinary meaning of Article 1.1, nor its context, nor the object and purpose of the SCM Agreement, nor Canada itself, outside the confines of these proceedings, suggest, much less require, a “net cost to government” test. Brazil argues that the proper test is evident from the ordinary meaning of Article 1.1: a subsidy exists where a government contributes something, and in so doing grants an aid, which gives help or support, or an advantage, which improves the recipient’s condition above and beyond the market.
Response of Canada
In response, Canada notes Brazil’s own admission that under its definition, a purely commercial contract could be covered by Article 1 of the SCM Agreement. This conclusion, arrived at through a contextual interpretation of the word “benefit” is inescapable: according to economic theory, any contract involves at least one party that is better off – that benefits from the contract – and one party that, at least, does not lose. If “benefits” or advantage were considered without the broader
WT/DS70/R
Page 49
context of the SCM Agreement, including Article 14 and Annex IV, then all purely commercial contracts entered into by governments would be subject to the disciplines of the SCM Agreement. This, in Canada’s view, widens the scope of application of the SCM Agreement significantly beyond what the drafters intended. In this context Canada reiterates the anomalous results that would be obtained if the Article 14 approach – benefit to the recipient – to the determination of the value of a subsidy were used for the purposes of Article 1: an CAN$87 million repayable contribution was found by Brazil to have a subsidy value nearly three times higher than the subsidy value of the same amount as a cash grant.
Canada also observes that the Revenue Canada Handbook definitions relate to the valuation of subsidies for the purpose of imposing countervailing duties. In that respect, the definitions and methodologies set out in the Handbook are in Canada’s view fully consistent with the guidelines set out in Article 14, for the express purpose of the valuation of subsidies in countervailing duty cases. The Handbook, though fully consistent with Article 14, does not, however, predetermine Canada’s understanding of what types of governmental contributions or activities should come within the scope of the SCM Agreement.
“CONTINGENT, IN LAW OR IN FACT, … ON EXPORT PERFORMANCE”
Arguments of Canada
Brazil’s interpretation of Article 3.1(a)
Canada argues that Brazil’s first submission nowhere states how Brazil interprets “contingent
… upon export performance” in Article 3.1(a); and Canada infers that, according to Brazil, this criterion is met where a subsidy is granted in any one of the following conditions:
by an organisation that has the mandate to support export trade and the capacity to engage in that trade and to respond to international business opportunities; (para. 6.49)
because the industry is export-oriented or has significant export potential;(para. 6.221)
because an export-oriented company is required to stay in business; (para. 6.287)
an objective of a subsidy programme is promoting domestic and international sales; (para. 6.308) or
the sales of specific products of the recipient have all been export sales.(para. 6.221)
In Canada's view, these implicit interpretations of what constitutes “contingent … upon export performance” are incorrect under the customary rules of interpretation of public international law, as required by Article 3.2 of the DSU.
Article 3.1(a) prohibits subsidies that are conditional on or tied to export performance
Canada submits that in accordance with the customary rules of international law on the interpretation of treaties, Article 3.1(a) should be interpreted to apply to subsidies that are, in law or in fact, conditional on or tied to export performance.
More specifically, Canada submits that the following factors are useful in determining whether subsidies are in fact contingent upon export performance:
evidence that the subsidy would not have been paid but for the exports flowing from it;
WT/DS70/R
Page 50
whether there are penalties -- in the sense of reduction or withdrawal of payments -- if
exports do not take place; or
whether there are bonuses or additional payments if exports do take place.
For Canada, this interpretation accords with the ordinary meaning of Article 3.1(a), in the light of its context and the object and purpose of the SCM Agreement, and is supported by the negotiating history of the SCM Agreement. Canada submits that Brazil’s interpretations, if adopted, would lead to a manifestly absurd or unreasonable result.
The ordinary meaning of “contingent ... upon export performance”
For Canada, the key words to interpreting Article 3.1(a) are “contingent ... upon export performance”. Canada argues that the relevant ordinary meaning of “contingent upon” is “dependent for its existence on something else”, “conditional; dependent on, upon;”118 the ordinary meaning of “performance”, in pertinent part, is “execution or accomplishment of an action, operation or process undertaken or ordered.”119 Thus, for Canada, on its plain meaning, Article 3.1(a) applies only to subsidies that are conditional on exports being executed or accomplished.
Canada submits that this is further supported by footnote 4 to Article 3.1(a), which states the following regarding whether an ostensibly domestic subsidy is “in fact” contingent upon export performance:
“This standard is met when the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in fact tied to actual or anticipated exportation or export earnings. The mere fact that a subsidy is granted to enterprises which export shall not for that reason alone be considered to be an export subsidy within the meaning of this provision.” [emphasis added by Canada]
In Canada's view, the relevant ordinary meaning of “tied to” is “restrain or constrain to or from an action etc; limit or restrict as to behaviour, location, conditions, etc.”120 Thus, Canada argues, it must be shown that, exportation or export earnings must be a condition for the grant of a subsidy for it to be prohibited under Article 3.1(a). The fact that exports take place, or that increased exports were intended (without the subsidy having been made conditional on exportation), does not render a subsidy a prohibited subsidy.
For Canada, this interpretation finds additional support in the Panel Report in Indonesia – Autos, which noted that Article 3 “…prohibits subsidies which are conditional on export performance and on meeting local content requirements …”.121 Canada argues that the prohibition in Article 3 is not triggered by the effect of a subsidy,122 by the objective of a subsidy. Rather, Canada submits, Article 3 addresses subsidies that are granted or maintained only if certain conditions are met: Article 3.1(a) captures subsidies where the condition attached to the subsidy is that exports must be executed
118 The New Shorter Oxford English Dictionary on Historical Principles, Vol. 1, (Oxford: Clarendon Press, 1993) at 494 (Exh. CDN-8).
The French version uses “subordonné”, which translates as “subordinate to”; Robert & Collins French-English Dictionary, (Paris & London: Dictionnaires Le Robert and Collins Publishers, 1987) at 680 (Exh. CDN-9).
119 The New Shorter Oxford English Dictionary on Historical Principles, Id.. Vol. 2 at 2160 (Exh. CDN-10).
120 Id.. at 3307 (Exh. CDN-11).
121 Indonesia – Certain Measures Affecting the Automobile Industry (“Indonesia- Autos”) Report of the Panel adopted on 23 July 1998, at para. 14.33.
122 In contrast, actionable subsidies are triggered by effect: SCM Agreement Article 5 and 6.
WT/DS70/R
Page 51
or accomplished; likewise, Article 3.1(b) captures subsidies where the condition attached to the subsidy is that there must be domestic content.
Canada contends that based strictly on its ordinary meaning, Article 3 does not, therefore, prohibit programmes or subsidies that only have as a general objective the expansion of trade or the increase in international competitiveness that might lead to increased exports. For Canada, a subsidy may have, as a general objective, an increase in competitiveness and hence increased exports, and may meet its competitiveness objectives and lead to increased exports. If, however, such a subsidy is available to recipients whether or not they engage in exports, in Canada's view the subsidy is not contingent upon export performance and therefore not inconsistent with Article 3.1(a).
The context of Article 3.1(a)
Canada argues that three contextual elements support this analysis.
First, Canada observes, the SCM Agreement makes a fundamental distinction between prohibited and non-prohibited subsidies: export subsidies, together with domestic content requirement subsidies, are singled out for prohibition under Part II, whereas other subsidies may be actionable under Part III or non-actionable under Part IV. In Canada's view, this distinction is central to the legal structure of the SCM Agreement, under which there are three discrete categories of subsidies. Canada argues that a subsidy falling within Article 3 is prohibited, regardless of its actual or expected impact on international trade or on the interests of other WTO Members; and that the trade distorting impact of actionable subsidies is presumed only under certain conditions (set out in Article 6), and even then, unlike in the case of prohibited subsidies, the presumptions are rebuttable (Article 6.2); and that subsidies that fall within the category of “non-actionable” subsidies are subject to disciplines only in the rare circumstance that they “cause damage which would be difficult to repair” (Article 9.1).
For Canada, Brazil’s argument tries to obscure these essential distinctions. Canada notes the example of certain research and development subsidies that are inherently aimed at increasing “international competitiveness” and in that broad sense possibly lead to the expansion of exports. In Canada's view, to the extent that the subsidies are not paid only on condition that exports take place -- that is, the exports do not arise as a result of the desire to take advantage of a subsidy that can only be obtained if exports take place, but are incidental to the general increase in competitiveness that has resulted from the subsidies -- such subsidies must not be considered “contingent …upon export performance” and therefore prohibited by Article 3.
Canada refers to a second contextual element, Article 3.1(b), which prohibits subsidies “contingent… upon the use of domestic over imported goods.” According to Canada, Brazil’s interpretation of “contingency” as “propensity”, if pushed to its logical conclusion in Article 3.1(b), would result in the prohibition of subsidies made to domestic industries that have a domestic content “propensity” or “orientation”. For Canada, if Brazil’s argument with respect to Article 3.1 is correct, at a minimum, subsidies to the following sectors or concerns would be prohibited:
natural resources processing plants;
subcontractors for major manufacturing concerns;
enterprises in remote locations; and
medium- to small-sized enterprises in large economies.
Canada submits that these enterprises tend, by the nature of their operation, to source their inputs domestically, and are thus “oriented” to the use of domestic goods over imported goods. For
WT/DS70/R
Page 52
Canada, according to Brazil’s interpretation of “contingent upon”, subsidies to such enterprises would be prohibited.
A third contextual element referred to by Canada is the list of export subsidies in SCM Agreement Annex I. Canada states that all of the listed examples describe situations in which the subsidy is conditional on or tied to the export of a good. At the same time, Canada argues, the fact that Item (a) identifies "direct subsidies…contingent upon export performance" does not by a contrario implication, lead to the conclusions that indirect subsidies are thereby excluded from the scope of Article 3.
Further in this regard, Canada, in answer to a panel question, submits that its argument that no subsidy under Article 1 may be found where no net cost to the treasury of the granting country can be shown is not an a contrario conclusion based on Item (k), but rather a statement setting out the elements that have to be established under Article 1. Canada submits that there is a resemblance between the structure of Item (k) and Article 1, and states that this similarity is the reason why Canada referred to Item (k) as contextual guidance in interpreting Article 1 of the SCM Agreement in its First Written Submission.
Canada notes with respect to the example of direct and indirect subsidies (para. 5.68) that while no a contrario conclusion can be drawn, from Item (a), that indirect subsidies are permitted, a responding Party might show, under Article 3, that an impugned subsidy is not “contingent upon export performance”; or it might demonstrate that an impugned contribution is not a subsidy, and it would do so under Article 1. For Canada, neither argument would be drawing a contrario conclusions from an Item in the Illustrative List.
The object and purpose of the SCM Agreement
Canada notes that the object and purpose of the SCM Agreement is not set out in a purposive clause, or in a preamble, but can be determined nonetheless from the text and structure of the SCM Agreement itself, as well as from the historical context in which disciplines on export subsidies developed.
Canada submits that the SCM Agreement is based on the premise that some forms of government intervention distort international trade, some have the potential to distort, and still others do not distort at all, and that the disciplines imposed by the SCM Agreement reflect this accepted approach, commonly known as the “traffic light approach”:123 trade distorting subsidies are to be prohibited outright (red light); potentially trade distorting subsidies are to be disciplined if they cause distortions in the market (amber); non-trade distorting subsidies are not subject to disciplines (green). Hence the prohibition on export subsidies, the disciplines on actionable subsidies if they cause serious prejudice, and the absence of disciplines on certain types of research and development subsidies, according to Canada.
It is essential, Canada submits, that in construing the scope of Article 3.1 the Panel does not blur this distinction. For Canada, Article 3.1 does not capture domestic subsidies merely because an objective and effect of a programme is to increase export competitiveness.
123 See Communication from Switzerland, 1 February 1988, MTN.GNG/NG10/W/17 at 1-2 (Exh. CDN-12); Note by Secretariat, Negotiating Group on Subsidies and Countervail Measures, Meeting of 1- 2 June, 1998, MTN.GNG/NG10/7 (Exh. CDN-13); and Montreal Meeting of the Trade Negotiations Committee, MTN.TNC/7(MIN) 9 December 1988, at 18-20 (Exh. CDN-14).
WT/DS70/R
Page 53
The negotiating history of the SCM Agreement
Canada argues that its interpretation is supported by the negotiating history of the SCM Agreement, as provided for in Article 32 of the Vienna Convention. Canada submits that the question of how to determine whether a subsidy is “in law or in fact” an export subsidy was an issue that highlighted the key fissures in the negotiations on the SCM Agreement between the United States and almost all other countries, including Brazil. According to Canada, the United States, basing its position on its own “disproportionality” test for the imposition of countervailing duties on export subsidies, argued for a quantitative approach and an “export propensity” test to determine whether a given measure was a prohibited subsidy.124 Canada states that this position was rejected by other delegations,125 because it would be inequitable for small economies that were more dependent on export markets, quoting a "status report" from the Chairman of the Negotiating Group:
“Several participants found the proposal to prohibit subsidies granted to predominantly exporting firms unacceptable. They considered it biased against countries with small internal markets, where firms were forced to export most of their production to be economically viable. They also considered that the same disciplines should apply to all firms irrespective of whether their sales happened to take place in the domestic or a foreign market.”126
Similarly, Canada asserts, the evolution of the footnote to Article 3.1(a) demonstrates the negotiators’ rejection of an object or intent-based test for that of a conditions-based test. According to Canada, in the earliest draft, the footnote was cast as a test of intent:
“This standard is met whenever the granting authority knew or should have known that the subsidy, without having been expressly made contingent upon export performance, was intended to increase exports.”127
which was modified to:
“This standard is met whenever the facts which were known or should have been known to the government when granting the subsidy demonstrate that the subsidy, without having been made expressly contingent upon export performance, would operate as an export subsidy.”128
and subsequently to:
124 See: Communication from the United States, 15 June 1988, MTN.GNG/NG10/W/20 at 5 (Exh. CDN-15); Elements of the Framework for Negotiations – Submission by the United States, 22 November 1989, MTN.GNG/NG10/W/29 (Exh. CDN-16); Elements of the Negotiating Framework – Submission by the United States, 27 September 1990, MTN.GNG/NG10/W/39 at 3 (Exh. CDN-17).
125 See, for example: Elements of the Framework for Negotiations – Submission by Japan, 6 October 1989, MTN.GNG/NG10/W/27 at 1 (Exh. CDN-18); Elements of the Framework for Negotiations – Submission by India, 30 November 1989, MTN.GNG/NG10/W/33 at 1 (Exh. CDN-19).
126 Note by the Secretariat, Negotiating Group on Subsidies and Countervailing Measures, Meeting of 27-28 March 1990, MTN.GNG/NG10/17 at paragraph 3 (Exh. CDN-20), recording the response of some delegations to the US export propensity proposal.
127Status Report by the Chairman, 18 July 1990, MTN.GNG/NG10/W/38, containing a text distributed
by the Chairman on 18 May 1990 at 2 (“Cartland I”) (Exh. CDN-21)
128 Draft Text by the Chairman, 4 September 1990, MTN.GNG/NG10/W/38/Rev.1, at 2 (“Cartland II”) (Exh. CDN-22)
WT/DS70/R
Page 54
“This standard is met whenever the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in practice tied to actual or anticipated exportation.”129
and later to:
“This standard is met whenever the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in practice tied to actual or anticipated exportation or export earnings.”130
and finally to:
“This standard is met whenever the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in fact tied to actual or anticipated exportation or export earnings. The mere fact that a subsidy is accorded to enterprises which export shall not for that reason alone be considered to be an export subsidy within the meaning of this provision.”131
For Canada, the rejection of an “export propensity” test, along with the change from a footnote test based on “intent” to one based on conditions attached to the subsidy (i.e. whether a subsidy is “in fact tied to” exports) must give rise to the “commonplace inference”132 that neither export propensity nor “intent” to increase exports is a correct measure for determining whether a subsidy is “in fact” contingent upon exports. Rather, Canada asserts, the correct test is whether the subsidy in question is available only for exports, or only on condition that goods are exported.
The implications of Brazil’s interpretation
Canada submits that it is an axiom of treaty interpretation that the interpreter must avoid interpreting the treaty in a way that would lead to a manifestly absurd or unreasonable result.133 In Canada's view, Brazil proposes that “contingent upon” should be interpreted more expansively than its ordinary meaning, and argues that Article 3.1(a) applies not only to subsidies that are conditional upon export performance, but also those subsidies that have an “export propensity”, an interpretation that, in addition to ignoring the plain meaning of Article 3 and the negotiating context of the SCM Agreement, would lead to a manifestly absurd or unreasonable result in three ways.
First, Canada asserts, such a definition would create a serious imbalance between the obligations of smaller and larger economies under the SCM Agreement. In Canada's view, such an outcome was not only not intended, but would be manifestly unreasonable, for it would mean that
129 Draft Text by the Chairman, 2 November 1990, MTN.GNG/NG10/W/38/Rev.2 at 5 (“Cartland III”) (Exh. CDN-23)
130 Draft Text by the Chairman, 6 November 1990, MTN.GNG/NG10/W/38/Rev.3 (“Cartland IV”) (Exh. CDN-24).
131 Draft Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, 20 December 1991, MTN.TNC/W/FA (“Dunkel Draft”) at I.3 (Exh. CDN-25).
132 United States - Restrictions on Imports of Cotton and Man-made Fibre Underwear (Cotton
Underwear), WT/DS24/AB/R, Report of the Appellate Body adopted on 25 February 1997, at 17:
“We believe the disappearance in the ATC of the earlier MFA express provision for backdating the operative effect of a restraint measure, strongly reinforces the presumption that such retroactive application is no longer permissible. This is the commonplace inference that is properly drawn from such disappearance. We are not entitled to assume that that disappearance was merely accidental or an inadvertent oversight on the part of either harassed negotiators or inattentive draftsmen. That no official record may exist of discussions or statements of delegations on this particular point is, of course, no basis for making such an assumption.” [emphasis added]
133 Sinclair, (Exh. CDN-5).
WT/DS70/R
Page 55
there would be one law for larger economies that are not dependent on international trade and another for smaller economies that are.
Second, Canada states, if “export propensity” -- so clearly rejected as a benchmark throughout the negotiations -- were considered the appropriate standard for making a prohibited subsidy finding, Article 3.1 would be over-broad to the point of absurdity. For Canada, under such a definition, almost any subsidy that would help in the development of the competitive advantage of a state -- that would make its industries more efficient globally -- would then be prohibited as an export subsidy.
Third, in Canada's view, Brazil’s interpretation would render government planning for WTO- consistent subsidies impossible. Canada submits that any purely domestic subsidy could have the effect of increasing the general competitiveness of a sector or a company, and that a properly structured subsidy would have the achievement of this outcome as its prime directive. For Canada, even if the output of the sector or the company in question only serves the domestic market, greater efficiencies in one sector or company could have a ripple effect in the economy as a whole, and such greater efficiency could lead to greater exports. Canada maintains that if Brazil’s interpretation were acceptable, the resulting exports would render the otherwise WTO-consistent subsidy illegal, making it impossible for governments to exercise their right under the SCM Agreement to structure WTO- consistent subsidies.
The OECD Arrangement on Guidelines for Officially Supported Export Credits (OECD Consensus)
Finally, Canada notes, Article 3 is subject to exceptions, recalling that Footnote 5 to Article 3 provides that “[m]easures referred to in Annex I as not constituting export subsidies shall not be prohibited under this or any other provision of this Agreement.” Canada submits that one such “measure” is set out in Annex I, Item (k), which provides:
“The grant by governments (or special institutions controlled by and/or acting under the authority of governments) of export credits at rates below those which they actually have to pay for the funds so employed (or would have to pay if they borrowed on international capital markets in order to obtain funds of the same maturity and other credit terms and denominated in the same currency as the export credit), or the payment by them of all or part of the costs incurred by exporters or financial institutions in obtaining credits, in so far as they are used to secure a material advantage in the field of export credit terms.
Provided, however, that if a Member is a party to an international undertaking on official export credits to which at least twelve original Members to this Agreement are parties as of 1 January 1979 (or a successor undertaking which has been adopted by those original Members), or if in practice a Member applies the interest rates provisions of the relevant undertaking, an export credit practice which is in conformity with those provisions shall not be considered an export subsidy prohibited by this Agreement.” [emphasis added]
Canada observes that the OECD Consensus134 is such an “international undertaking”, of which Canada is a Member.
Canada argues that if, therefore, an export financing transaction is entered into consistently with the interest rate provisions of the OECD Consensus, then such export credit practice, according to the express terms of the SCM Agreement, “shall not be considered an export subsidy prohibited by this Agreement.”
134Exh. CDN-3.
WT/DS70/R
Page 56
Canada notes its view that item (j) and the first paragraph of item (k) of Annex I to the SCM Agreement are not exceptions; they merely set out what type of practice would be ipso facto an export subsidy. That is, an export credit provided by government that met the conditions of the first paragraph of Item (k) would be an export subsidy; a complainant need only prove the elements of Item (k) and does not need to further prove, for example, that the provision of the credit was “contingent on export performance”. Canada submits that the examples provided in the Illustrative List do not identify a contrario what would not constitute an export subsidy, and that such an interpretation would turn the Illustrative List into an exhaustive list.
The United States misunderstands Canada’s interpretation of Article 3
Canada disagrees with the argument in the United States' third party submission (Section VII) that Canada’s interpretation of Article 3 is premised on a flawed interpretative analysis because Canada is reading words into Article 3.1(a), is ignoring the word “anticipated” in footnote 4, and has presented an inaccurate and incomplete discussion of the negotiating history.
Canada argues that the US position in its third party submission is little different from the position it took throughout the Uruguay Round – a position that was specifically and unambiguously rejected in the SCM Agreement. In Canada's view, the United States also misunderstands Canada’s position and misinterprets footnote 4 by emphasising the wrong verb in that provision, making the example it provides (see paras. 7.41 - 7.42) both inconsequential and illogical.
For Canada, the prohibition in Article 3 is both forward looking and backward looking. A subsidy that is conditional on or tied to past exports as well as expected or anticipated exports is one that is “contingent upon … export performance.” In this sense, Canada asserts, “actual or anticipated exportation or export earnings” is a reformulation of “export performance”, just as “tied to” is a reformulation of “contingent upon”. In Canada's view, the introduction of “anticipated” in the footnote does not in any way change the conditionality explicit in “tied to” into a test based on general objectives or intent.
For Canada, an export subsidy that is in fact tied to anticipated exportation is found where, for example:
a subsidy programme is, in law, neutral as to destination; and
it is administered so that a subsidy:
is granted only if exports are anticipated (whether or not this is the only condition or there are other criteria), or
is not granted where exports are not expected to take place.
Accordingly, Canada argues in commenting on a hypothetical example posed by the Panel, a subsidy that is restricted in law only to enterprises that are, based on past performance, “export- oriented” is not, for that reason alone, contingent on export performance.
Canada submits that a specific example of a subsidy that is in fact tied to anticipated exportation would be a tax holiday granted for the construction of a plant that will in fact only produce products that do not meet domestic standards.
Canada additionally submits that, whether “export performance” is the only criterion, or one among many, does not change the conditionality of Article 3 and does not transform the test into an intent-based test: “tied to … anticipated exportation” cannot and does not translate into “granted with the general intent that exports somehow increase”; but rather means that “one of the conditions for
WT/DS70/R
Page 57
the grant of the subsidy is the expectation that exports will flow thereby.” For Canada, the distinction, though perhaps subtle, is nevertheless important.
In Canada's view, the United States fails to identify this distinction and therefore presents the Panel with a hypothetical that is at once inconsequential and illogical. Canada agrees that the US hypothetical is an export subsidy. For Canada, however, in the US example the grant is in law tied to anticipated exportation, i.e., it is tied to or conditional on export performance. In Canada's view, the US example adds nothing to the analysis of export contingency in fact.135
Canada also takes issue with the US suggestion that Canada’s discussion of the negotiating history is incomplete because it ignores the purported reason for adding footnote 4: a proposal by the European Communities.136 Canada submits that the United States has not explained that this proposal was made137 seven months before the first of the draft texts of the SCM Agreement – i.e. Cartland I.
138
Canada acknowledges that the first draft of the footnote reflected the proposal of the European Communities – almost word for word,139 but states that the first iteration of footnote 4 was part of the on-going negotiations for establishing the scope of Article 3; that the European Communities was not the only participant in the negotiations; and that the test it proposed was clearly rejected by the negotiators. Canada states that the United States acknowledges (see para. 7.39) that “the final text of footnote 4 does not contain a ‘knowledge’ test as originally articulated by the EC
…”. For Canada, the commonplace inference to be drawn from this is that knowledge and intent are not part of the conditionality test contained in Article 3.
Canada states that its view on intent is confirmed by the original proponent of the intent test, noting that the European Communities stated in its third party submission to this Panel:
“The EC supports Canada’s view that the effect of a subsidy or the objective of a subsidy cannot on their own be sufficient to establish de facto export contingency under Article 3.1(a). Exportation, or export earnings must be a condition for the grant of the subsidy.” (para. 7.3).
Canada asserts in response to a panel question that a subsidy the intent or objective of which is to increase exports can be distinguished from a subsidy that is tied to anticipated export performance, that a subsidy may have “increasing exports” as a general objective, but that because this general objective may be achieved in any number of ways, the existence of that intention or objective alone does not mean that a subsidy is tied to anticipated exports. For example, Canada states, a country may suffer from declining productivity, or declining competitiveness due to currency
135 The US hypothetical is also illogical. Governments do not grant export subsidies for planned exportation only. A government that sets itself on the course of providing export-contingent subsidies would be expected, in the US hypothetical, to adjust the eligibility criteria so as to ensure that subsidies are paid when exports actually take place. Or, if it is simply interested in granting subsidies contingent on plans and expectations whether or not exports actually do take place, one would expect such a government to structure a domestic subsidy programme rather than one that is prohibited under the SCM Agreement. In any event, as a practical matter WTO Members have little incentive to challenge an ostensibly export-contingent subsidy that does not result in any exports.
136 Id.. at para. 14.
137 Elements of the Negotiating Framework, Submission by the European Community, MTN.GNG/NG10/W/31, 27 November 1989 (Exh. CDN-56).
138Status Report by the Chairman, 18 July 1990, MTN.GNG/NG10/W/38, containing a text distributed by the Chairman on 18 May 1990 at 2 (“Cartland I”) (Exh. CDN-21)
139 The draft footnote in Cartland I states:
“This standard is met whenever the granting authority knew or should have known that the benefit is conferred on certain enterprises.”
WT/DS70/R
Page 58
realignments or changes in the pattern of trade in resources. It may, to spur its economy -- and with a general objective that exports might increase – establish a system of domestic subsidisation, to encourage productivity, increase competitiveness or effect a change from reliance on one set of resources or industries to another.
According to Canada, such subsidies may take many forms and focus on many different things. For example, subsidies may be made to change factories over from textiles to newer industries, or indeed services. Shipyards may be helped to move from oil tankers to cruise-ships. Canada maintains that in all of this, any improvement in productivity or competitiveness, or indeed any restructuring or change over to other sectors, may well result in increased exports. For Canada, however, to argue that these essentially domestic subsidies are tied to anticipated exports would be to stretch the term beyond recognition.
In response to a question of the Panel regarding whether “tied to” could also be interpreted to mean that one of the “reasons” for the grant of the subsidy is the expectation that exports will flow thereby, or that the subsidy is granted because of the expectation that exports will flow thereby, Canada refers to an article in the Economist of 31 October 1998, concerning Magnetic Levitation trains,140 and argues that the country or company that finds the right solution to land transport is guaranteed world market dominance for the foreseeable future. The article notes that NASA, the National Aeronautics and Space Agency of the United States, is “paying for an experimental version” of the trains. Canada submits that the ostensible reason is NASA’s own requirements, but that since it is clear that if the technology takes off, world market dominance would follow, it might be argued that the reason may well be increased exports.
Canada argues that this payment is not a subsidy tied to anticipated exports, because Article 3 prohibits subsidies that are contingent on or tied to export performance, that is, the subsidy would not be paid but for the expectation that exports would ensue. For Canada, the reason why a subsidy is granted, even if it could be adequately determined, is only relevant to the extent that it establishes the condition; that reason in itself – that is, the intention in itself – is not enough to turn an otherwise domestic subsidy into a prohibited subsidy.
Canada further argues that determination of de facto export contingency cannot be made on the basis of a single example, and submits that Article 2.1(c) of the SCM Agreement, which deals with de facto specificity, is helpful in this regard, particularly its provision that the length of time a subsidy programme has been in operation is one of the factors that should be taken into account in such a determination.
This also applies to de facto contingency, according to Canada. Therefore, while establishing conditionality for anticipated exportation may not be easy in respect of a single transaction, it could be indeed done over time, in Canada's view where for example, parts of a stream of payment are not paid because export expectations are not fulfilled; or where future eligibility is curtailed because such expectations are not fulfilled; or, where additional subsidies are paid as exports increase.
Arguments of Brazil
Brazil notes Canada’s argument that contingency on export can be determined on the basis of three factors: that the subsidy “’would not have been paid but for the export flowing from it‘”; that “’penalties . . . in the sense of reduction or withdrawal of payments” occur “if exports do not take place‘”; or that “’bonuses or additional payments‘” occur “’if exports do take place.‘”
For Brazil, none of these factors finds textual support in the SCM Agreement, and they share one common assumption, which is that Article 3.1(a) applies only to those subsidies made conditional
140 Exh. BRA- 89. “Maglev trains: A permanent solution”, The Economist, 31 October 1998, at 88.
WT/DS70/R
Page 59
on exports actually being executed or export earnings actually being realized. With this presumption, Brazil argues Canada has failed to give effect to the ordinary meaning of Article 3.1(a) of the SCM Agreement, which states at footnote 4 that a subsidy is considered contingent “in fact” upon export not only if tied to actual exports, as the Canadian definition requires, but also if tied to anticipated export or export earnings. In Brazil’s view, Canada’s interpretation thus lacks support in the relevant rules of treaty interpretation, identified by Canada itself as derived from Articles 31 and 32 of the Vienna Convention on the Law of Treaties.
Brazil submits that the ordinary meaning of the word “anticipate” is to “look forward to,” or to “expect.” In Brazil’s view, under the ordinary meaning of Article 3.1(a), therefore, a subsidy is considered contingent in fact upon export if it is tied to expected exportation or export earnings. If in granting a subsidy a Member “anticipates,” or “expects,” that exports or export earnings will occur, it has granted a subsidy contingent in fact upon export. In Brazil’s view, the ordinary meaning of the word “anticipated” defeats Canada’s interpretation, and the very fact that the word “anticipated” survived the various drafts of footnote 4 and emerged in the final text belies Canada’s claim that a Member’s intent is irrelevant to determine whether a subsidy is “in fact” contingent upon export. For Brazil, nothing in Article 3.1(a) requires that to be considered contingent upon export, post hoc penalties be levied or bonuses paid depending on whether exports are actually made.
Brazil asserts that Canada also fails to recognize that footnote 4 to Article 3.1 states that export performance may be either the sole contingency for the subsidy or merely “’one of several other conditions.’” Brazil notes that while Article 3.1(a) provides that the mere granting of a subsidy to an exporting entity will not “’for that reason alone’” make it a prohibited export subsidy, the Panel is not here faced with such a case.
Brazil disagrees with Canada’s view that Brazil’s entire claim turns on what Canada terms the “export propensity” of the Canadian regional aircraft industry alone. For Brazil, there are many factors contributing to Brazil’s conclusion that various support to the Canadian regional aircraft industry (including through TPC) is “in fact tied to actual or anticipated exportation or export earnings.” Brazil cites information concerning TPC, and submits that the Panel is faced with a situation in which several factors converge, together illustrating that the Canadian Government and the provinces have supported the Canadian regional aircraft industry precisely because it is a total export industry, and precisely because they anticipate that this performance will continue (see also para. 6.227):
TPC funding statistics (see para. 6.220), which for Brazil demonstrate that TPC is captive to the Canadian aerospace industry, and more specifically, the regional aircraft industry, and TPC’s own recognition that its main beneficiary is “highly export oriented.”141
The historical high level of support (asserted to be approximately $2 billion) through TPC’s predecessor, DIPP to the aerospace sector.142
The statements of the Canadian Government in announcing its $267 million in grants the regional aircraft industry demonstrate that those grants are tied to the Canadian Government’s anticipation, based on ample historical evidence, that the industry will maintain its 100 per cent export orientation (see para. 6.227).
141 TPC Annual Report 1996-1997, pg. 5 (Exh. BRA-5).
142 Canadian Taxpayers Federation study, “Corporate Welfare – A Report on Sixteen Years of Industry Canada Financial Assistance,” 16 April 1998, Tab 4 (Exh. BRA-31). This statement of total DIPP funding to the aerospace sector represents aggregate DIPP funds released to four Canadian companies, i.e., Bombardier, de Havilland, Pratt & Whitney Canada Inc., and CAE Electronics Ltd.
WT/DS70/R
Page 60
According to Brazil, neither these statements by senior Canadian Government officials nor the statements about the industry in TPC’s promotional materials were crafted in a vacuum. Brazil maintains that at the time the Canadian Government was making these statements, it was eminently aware of the fact that every single sale of Dash 8 series aircraft made since 1992, and every single sale of the CRJ since its development and commercialization, was for export. For Brazil, the industry is devoted to exports, and the Canadian Government has therefore made it very clear that it maintains massive amounts of support to the Canadian regional aircraft industry precisely because it is an export industry and precisely because it anticipates that the Canadian regional aircraft industry will continue to be an export industry.
Regarding Canada’s arguments (para. 5.86) concerning anticipated exportation or export earnings, Brazil indicates first that it agrees with Canada’s statement that “’[t]he prohibition in Article 3 is both forward looking and backward looking’”, noting that it is for this reason that Brazil cites the overwhelming export orientation of the Canadian regional aircraft industry as one reason supporting the conclusion that assistance granted under or through TPC, the Subsidiary Agreements, SDI/IQ and the sale of de Havilland constitute subsidies contingent upon export. Brazil argues that in disbursing funds to the Canadian regional aircraft industry under the auspices of these programmes, the federal and provincial governments, based on the past export orientation of the industry, anticipate or expect that the this orientation will continue in the future.
Second, Brazil also agrees with Canada’s statement that “’tied to . . . anticipated exportation’” means that “‘one of the conditions for the grant of the subsidy is the expectation that exports will flow thereby’” (para. 5.90). Brazil contends that the evidence before the Panel demonstrates that assistance under or through TPC, the Subsidiary Agreements, SDI/IQ and the sale of de Havilland would not have been granted were it not for the virtually total export orientation of the Canadian regional aircraft industry. In this sense, exportation is a condition of the Canadian subsidies.
According to Brazil, there is no requirement in Article 3.1 that for a subsidy to be contingent “in fact” upon export, the condition be express, and were that the case, there would be no distinction between a subsidy contingent “in law” and a subsidy contingent “in fact” upon export. Members thus would be able to avoid the prohibition of Article 3.1 by simply leaving the word “export” out of their laws, regulations and dealings.
Third, Brazil emphasizes that it is challenging the application of TPC, the Subsidiary Agreements and SDI/IQ to the Canadian regional aircraft industry, which is overwhelmingly export oriented. As a result, Brazil submits that it is not relevant to the Panel’s decision in this case that TPC, the Subsidiary Agreements and SDI/IQ may also grant assistance to other industries with sales in export markets, or to other industries with sales in domestic markets. Brazil argues that when assistance under these programmes is granted to the regional aircraft industry, it is granted because the industry is totally export oriented, and because the Canadian federal and provincial governments anticipate or expect that it will remain so. Brazil states that it has identified many factors apart from export performance or orientation as contributing to its conclusion that that assistance granted to the regional aircraft industry under these programmes is therefore contingent “’in fact’” upon export (e.g., paras. 6.227, 6.231). Thus, Brazil disagrees with Canada’s argument that the export performance of the regional aircraft industry is the only factor called upon by Brazil in support of its case.
Fourth, Brazil disagrees with Canada’s statement that the terms “’actual’” and “’anticipated’” from footnote 4 to the SCM Agreement are nothing more than a “’reformulation’” of the term “’export performance’” in the body of Article 3.1(a). According to Brazil, the implication of Canada’s statement is that the terms “’actual’” and “’anticipated’” should not be accorded their ordinary meaning. Brazil submits that footnote 4 discusses the meaning of the term contingent “’in fact,’” and therefore the ordinary meaning of the terms included in that footnote are crucial to an understanding of contingent “’in fact.’” Brazil notes that footnote 4 provides that subsidies are considered contingent in fact on export if they are “tied to actual or anticipated exportation or export
WT/DS70/R
Page 61
earnings”. Brazil recalls its argument regarding the “’ordinary meaning’” of the word “’anticipate’” and Brazil contends that funding to the Canadian regional aircraft industry under or through TPC, DIPP, the Subsidiary Agreements, SDI/IQ, and the de Havilland sale was granted because the Canadian federal and provincial governments anticipated, or expected, based on ample historical evidence, that the industry would remain virtually totally export oriented.
Regarding Canada’s argument that Brazil’s view of Article 3.1(a) as applying to situations of “’anticipated’” exportation or export earnings would lead to “’a manifestly absurd or unreasonable result’” (para. 5.4), namely that “’that there would be one law for larger economies that are not dependent on international trade and another for smaller economies that are’”, Brazil states that Canada is effectively arguing that despite the terms of Article 3.1(a), Canada’s alleged position as a “’small economy’” warrants that it be accorded special treatment. Brazil doubts Canada’s claim of “’small economy’” status – as a founding Member of the WTO, a member, with the United States, Japan and the European Communities, of the WTO “Quad,” and a member of the OECD. For Brazil, moreover, Canada’s argument must fail as a matter of law, because there is only one law, applying equally to large and small economies alike, and it is found in the ordinary meaning of Article 3.1(a).
Brazil argues that this is not a case of a small economy of necessity having a small share of a global market, as Canada claims, or of support being given to an industry by a government indifferent to whether that industry sells in domestic or export markets, but is a case in which, by the deliberate action of the Canadian Government, whatever demand existed in the domestic market was satisfied by what were legally export sales, not domestic sales.
In this regard, Brazil recalls the EDC President’s comments to Parliament about EDC’s support, with export funds, of a sale to a foreign company of planes developed with TPC funds and ultimately leased to a domestic carrier, in order, according to the EDC President, to “‘launch an aircraft that has a world market.’”
For Brazil, these facts belie Canada’s “small market” claim, and demonstrate that the Panel is not faced with a situation in which domestic sales are a small percentage of total sales because the domestic market is small, but is rather faced with a situation in which domestic demand, however large or small, was satisfied by the Export Development Corporation with export development funds for the purpose of launching an export product which had benefited from TPC development funds. To Brazil, these facts establish that Canada’s use of TPC support for the aircraft industry was, de facto if not de jure, an export subsidy.
Response of Canada
In response to Brazil’s argument that Canada’s proposed tests for de facto export contingency do not find textual support in the SCM Agreement (paras. 5.101-5.102), Canada notes that it proposed three tests for determining whether a subsidy is in fact tied to exports. Canada submits that the tests Canada proposed find their support in the text of the SCM Agreement, in the words “contingent upon” and “tied to”, and that though anchored in legal provisions, tests, by definition, attempt to elaborate and clarify those provisions. Thus, Canada argues, tests are used as an aid to the application of laws. Canada notes that Brazil has not shown what other criteria would be used, or useful, in determining whether a subsidy is in fact contingent on export performance. Canada submits that it is because of that, and to help the Panel in its determination, that Canada has put forward these suggested tests. Canada acknowledged that regardless of the test used, the legal requirement is still the same: one of the conditions for granting a subsidy must be that exports take place. Regarding the United States's (paras. 7.35-7.37) and Brazil's (paras. 5.101-5.102) objection that Canada’s argument on Article 3.1(a) applies only to actual exports, rather than anticipated or expected exports, Canada asserts that the objection is based on a reading that the test proposed in footnote 4 is one of intent rather than conditionality. Canada submits that this reading of Canada’s arguments is not accurate, in that
WT/DS70/R
Page 62
Canada's view is that footnote 4 prohibits subsidies that are conditional on, or tied to, exportation,
whether anticipated or actual.
For Canada, Article 3.1 refers to both actual and anticipated exports, but whether exports are taking place or are expected to take place, the key verbs in Article 3.1, are contingent upon in the body and tied to in footnote 4. Canada submits that each of these denotes the existence of a condition; each connotes a “but for” test. Canada notes that exports that were expected might not happen, but that this does not mean that a subsidy that had been contingent upon such expectations is saved. Canada observes, however, that if a government wants to grant export subsidies, and exports do not result, the government would change the programme to ensure exports do follow. For Canada, that is how a de facto export subsidy can be detected, even if initially no exports take place.
Regarding Brazil’s arguments that Canada’s interpretation “lacks support” in the relevant rules of treaty interpretation, Canada recalls that the principles of treaty interpretation require that the ordinary meaning of the words of a treaty be viewed in the light of its context and object and purpose. For Canada, the phrase “anticipated exportation or export earnings” should therefore be read in context – that is to say, in the context of the condition explicitly provided for in the verb tied to. In Canada's view, nothing in footnote 4 or in Article 3 changes the nature of the condition: whether it is the only one or one of many, and therefore whether anticipated or actual, exportation must be a requirement of a subsidy for it to fit within Article 3.
In this sense, Canada asserts, “actual or anticipated exportation” is simply a reformulation – an elaboration – of “export performance”, just as “tied to” is a reformulation of “contingent upon.” That is, the first sentence of footnote 4 does not purport to create a new obligation, but rather, elaborates the existing obligation in Article 3.
Regarding Brazil's views on footnote 4, Canada argues that Brazil ignores the second sentence, which according to Canada was the sentence that finally made compromise on this provision possible. Canada submits that Brazil’s claims should be examined in the light of the interpretation that Canada has put forward for Article 3, an interpretation that respects the second sentence of footnote 4. In Canada's view, this sentence renders Brazil’s case null and void.
WT/DS70/R
Page 63
ARGUMENTS OF PARTIES REGARDING CANADIAN MEASURES ALLEGED BY BRAZIL TO BE PROHIBITED EXPORT SUBSIDIES IN THE SENSE OF SCM ARTICLES 1 AND 3143
EXPORT DEVELOPMENT CORPORATION (“EDC”)
General arguments of the parties
Alleged concessionary terms of EDC financing; EDC risk level and performance
Arguments of Brazil
Brazil states that the Export Development Corporation (“EDC”) is an agency of the Government of Canada which was established by the Export Development Act “’for the purposes of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.’”144 In response to a question from the Panel concerning the scope of Brazil’s claim, Brazil indicates that it challenges the EDC as de jure export subsidies prohibited by SCM Article 3, and thus as a programme per se. As a result, Brazil states that it also challenges the measure as applied in the context of the regional aircraft industry.
According to Brazil, “’EDC’s mandate is to help Canadian business compete and succeed in the global marketplace’”145 attempting to satisfy “’the seemingly endless appetite of Canadian exporters for financial support’”146 through a variety of financial and risk absorption services, including export trade insurance, sales financing, loan guarantees, and equity investments. Brazil states that these benefits are not available to Canadian exporters from private financial institutions, arguing that EDC itself acknowledges that “’EDC complements the banks and other financial intermediaries, but cannot substitute for them.’”147 Thus, according to Brazil, EDC’s “’goal is to help absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries.’”148
Brazil argues that EDC provides high levels of financing assistance to Canadian regional aircraft exporters. Brazil cites EDC Vice President Henri Souquières as indicating that of Can$1.7 billion in funding provided to the Canadian aerospace sector in 1995, fully 62 per cent, or
$1 billion, constituted support for the Canadair Regional Jet alone.
According to Brazil, this is equivalent to US$18.3 million per airplane delivered by Bombardier in that year alone, representing virtually 100 per cent of the price of the airplane.
For Brazil, EDC’s programmes involve the direct transfer of government funds through grants, loans, and equity infusions, and the potential direct transfer of funds or liabilities through loan guarantees, and confer a clear benefit to the recipient. Brazil argues that any aircraft manufacturer that receives government largesse that approximates 100 per cent of the selling price of its product is benefiting handsomely, and that any party receiving any direct or indirect benefit of the kind described in Article 1 of the Agreement and conferred by EDC is being benefited, albeit perhaps not
143 Except as otherwise noted, the footnotes and citations, and the emphasis in the text are as contained in the parties’ submissions.
144 Export Development Act, Section 10(1). (The text of the Export Development Act, as amended, is attached as Exh. BRA-6).
145 Export Development Corporation, 1995 Chairman and President’s Message (“Message”), p. 2 (Exh.
BRA-7).
146 Id. at 4.
147 Id. at 4.
148 Id. at 2.
WT/DS70/R
Page 64
as handsomely as Bombardier was in 1995. According to Brazil, EDC’s former President, Paul Labbé, acknowledged this in stating that EDC’s programmes give “’Canadian exporters an edge when they bid on overseas projects.’”149
Brazil states that EDC’s financing of up to 90 per cent (or more) of an aircraft’s cost constitutes a direct transfer of funds by grant or loan, within the meaning of Article 1.1 of the Agreement. Brazil argues that similarly, EDC’s investment in SPCs established to support the export of aircraft constitutes a direct transfer of funds by equity infusion. Moreover, according to Brazil, EDC’s provision of loan guarantees, and the grant of residual value guarantees, constitute the potential direct transfer of funds or liabilities, within the meaning of Article 1.1 of the SCM Agreement. Thus, for Brazil, all of these transfers (or potential transfers) of funds confer a benefit, within the meaning of Article 1.1.
Brazil asserts that regardless of the form in which export financing is offered, EDC has itself acknowledged the concessionary nature of its subsidies, noting that in order to avoid “’los[ing] money, [EDC] should be making at least the rate of inflation on [its] capital base which is [its] aim. That goal is a long cry from the 15 per cent or 20 per cent return on equity that would be required to survive in the private sector.’”150 According to Brazil, EDC’s annual reports underscore its shortcoming in this regard, as EDC’s net interest margin was a mere 2.82 per cent in 1997, and 3.03 per cent in 1996,151 and even the Canadian Parliament has expressed its concerns with EDC’s poor returns. Brazil indicates that the Standing Senate Committee on Banking, Trade and Commerce in April 1996 observed that despite the appearance of positive returns, “’only looking at the bottom line can give a misleading impression of how EDC is faring on its loan portfolio.’”152 Brazil states that the Committee noted that EDC was in actuality masking poor returns with two types of “government debt relief”: first, with Can$151 million in direct government relief to EDC for two of its “‘problem’ accounts,” and; second, with earnings realized from EDC’s practice of trading in financial instruments (facilitated by its ability, as a Crown institution, to borrow at privileged rates and lend at slightly higher rates, “’thus generating profits by essentially acting as an intermediary between borrowers and capital markets.’”).153
In response to a question from the Panel regarding what net interest margin Brazil believes private sector investors would expect for financing in the civil aircraft sector, Brazil discusses the risk level of EDC’s loan portfolio. According to Brazil, impaired loans, for which the EDC no longer has reasonable assurance that principal and interest will be collected on a timely basis, comprise 14.4 per cent of gross loans receivable154, and 57.6 per cent of EDC’s performing loan portfolio (total loans less impaired loans) has been classified by EDC itself, in its 1997 Annual Report, as “’below investment grade’” or “’speculative grade.’”155 Brazil states that it cited EDC’s low net interest margin156 to establish that EDC is not being compensated for this extra risk and,
149 CanadExport On-Line, Focus on Export Development Corporation, p.2 (Exh. BRA-67).
150 Committee on Banking, Trade and Commerce at 48:22 (Exh. BRA-10).
151 EDC 1997 Annual Report at 22 (Exh. BRA-21); EDC 1996 Annual Report at 33 (Exh. BRA-22).
152 Report of the Standing Senate Committee on Banking, Trade and Commerce, April 1996, at p. 43 (Exh. BRA-23).
153 Id. at 42-43.
154 Exh. BRA-79.
155 EDC 1997 Annual Report, pgs. 32-33 (Exh. BRA-21). Combined sovereign and commercial ratios are as follows: for below investment grade (Ba1 to Ba3) gross loans receivable net of non-accrued capitalized interest, as a per cent of total performing gross loans receivable net of non-accrued capitalized interest, 30 per cent; and for speculative grade (<Ba3) loans, 27.6 per cent.
156 Net interest margin is an expression of “net interest income expressed as a percentage of average
total assets.” Net interest income, in turn, “is interest and dividends earned on assets, less interest paid on liabilities.” Canadian Imperial Bank of Commerce Annual Report, dated 31 October 1997, pg. 30 (Exh. BRA-80).
WT/DS70/R
Page 65
conversely, that recipients of EDC funding are paying less for that funding than they would for funding from commercial sources.
Brazil also submits that Canada’s comparison of EDC’s net interest margin with that of commercial banks (para. 6.27), fails to make the proper adjustment for risk -- a factor key to the analysis of EDC’s loan portfolio. Brazil states that a fundamental tenet operating in financial markets is that “investors increase their required rates of return as perceived risk (uncertainty) increases”157, and that it is therefore impossible to assess return in isolation, without knowledge of the underlying risk involved.
Brazil submits a comparison both by risk and tenor (term) of the market return demanded by bondholders lending to the different risk classes to illustrate this risk/return principle158. Brazil states that for maturities of three to four years, as the riskiness of the borrower increases (no risk at AAA to very speculative at CCC+ or lower), the return demanded by lenders increases by 599 basis points.159 For loans extended for 15 or more years, Brazil argues that the comparison demonstrates that the spread demanded by bondholders lending to the most speculative class of borrowers increases to 1,242 basis points.
According to Brazil, EDC lends to speculative borrowers for regional aircraft financing for terms of 15 or more years160, and the comparison demonstrates that private sector independent bondholders would demand a spread of 1,242 basis points over riskless US Treasuries of 15+ years maturity, or 17.73 per cent, for this risk class. The comparison therefore, in Brazil’s view, supports a conclusion that EDC’s 1996 and 1997 reported net interest margins (para. 6.7) do not reflect returns commensurate with the risk borne by EDC.
Brazil submits that to understand why Canada’s comparison of EDC’s net interest margin to that of commercial banks is not valid, it is necessary to examine two key ratios which amplify the riskiness of EDC’s loan portfolio relative to those of three major Canadian banks and the US banking industry averages: the ratio of gross impaired loans to gross loans receivable, and the ratio of the allowance for losses on loans to gross loans receivable. Brazil submits that concerning this ratio, the Panel should take account of EDC’s own statements in its 1997 Annual Report:
Loans are classified as impaired when circumstances indicate that [EDC] no longer has reasonable assurance that the full amount of principal and interest will be collected on a timely basis in accordance with the terms of the loan agreement.161
Brazil submits a comparison162 to show that that EDC’s ratio of gross impaired loans to gross loans receivable is 14.4 per cent, compared with a ratio of approximately one per cent for the Royal Bank of Canada, Canadian Imperial Bank of Commerce, and the industry average for all US banks. This ratio is less than one per cent for the Bank of Montreal. Brazil cites Standard and Poor’s as saying that “’when [the ratio of nonperforming loans to total loans] exceeds 3 per cent . . . it can cause concern. For a bank with a very high level of nonperforming loans – approaching 7 per cent or more
the future may be doubtful.’”163
157 Frank K. Reilly, Investment Analysis and Portfolio Management (Dryden Press), Fourth Edition, pg.
25 (Exh. BRA-81).
158 Exh. BRA-82
159 A basis point is 1/100th of a percentage point.
160 “Optimism over Regional Jets Hit by Wrangling over Export Support,” Commuter/Regional Airline News, 21 July 1997 (Exh. BRA-83); “Export Credit,” Airfinance Journal, October 1995, pg. 30 (Exh. BRA-84).
161 EDC 1997 Annual Report, pg. 47 (Exh. BRA-21).
162 Exh. BRA-79
163 Standard & Poor’s Banking Industry Survey, dated 11 June 1998, pg. 23 (Exh. BRA-86).
WT/DS70/R
Page 66
Concerning the ratio of allowance (or reserve) for losses on loans to gross loans receivable, Brazil notes that “’[t]o cover possible future loan losses, banks are required to maintain a reserve for loan losses.’”164 The level of reserve maintained by a bank therefore “’reflects management’s judgment regarding the quality of its loan portfolio.’”165 Brazil submits a table166 to show that EDC’s allowance for loan losses as a per cent of total loans is 13.2 per cent, compared with amounts between one and two per cent for each of three major Canadian banks and the US industry average. Brazil recalls as well that approximately 57.6 per cent of EDC’s performing loan portfolio has been classified by EDC as “below investment grade” or “speculative grade.”167
For Brazil, there is therefore a clear and marked differential in risk between EDC’s loan portfolio and the loan portfolios of commercial banks. Brazil submits that commercial banks would not be able to sustain the risk levels inherent in EDC’s loan portfolio, and that while other independent investors might be willing to take on this level of risk, they would demand a return significantly higher than that shown in Canada’s comparison168 as compensation. Brazil recalls its view that financing of EDC’s risk class would demand a spread of 1,242 basis points over riskless US Treasuries of 15+ years maturity, or 17.73 per cent.
Brazil asserts that Canada’s statement that EDC’s exemption from corporate income tax and the payment of dividends is not relevant to the calculation of net interest margin (para. 6.32) misses the point, which is that comparing EDC’s net interest margin with that of commercial financial institutions of similar rating is not valid. Brazil submits that unlike commercial financial institutions, the EDC does not, in forecasting an acceptable rate of return, need to account for payment of corporate taxes or dividends, and as such can obviously accept a lower rate of return than can commercial financial institutions, putting EDC at a decided advantage in the marketplace. Thus, the risks undertaken by private, commercial, tax- and dividend- paying institutions are in no way comparable to those engaged in by the EDC and thus a comparison of their returns is meaningless absent accounting for the relative underlying riskiness of their portfolios.
Arguments of Canada
Evidence submitted by Brazil on alleged concessionary terms of EDC financing
Canada argues in the first instance that Brazil’s allegations about EDC financing are not supported by the evidence cited. Canada objects to what it terms Brazil’s selective quotation of certain materials adduced as evidence, to the point of changing the very sense of the material itself.
Canada states that at the beginning of its discussion on the Export Development Corporation (EDC), Brazil begins with what appears to be a descriptive paragraph (para. 6.2.), and that the description and “’quotations’” of this paragraph underlie Brazil’s claims about the “’benefits’” allegedly conferred by the EDC in the course of its operations. Canada asserts that this paragraph is thus the central element of Brazil’s final conclusion that “EDC is precisely what Article 3 of the SCM Agreement was intended to prohibit.” ( para. 6.49). Canada indicates that Brazil, apparently quoting a message from the Chairman and the President of the EDC (the Message), states that:
“[EDC] attempts to satisfy ‘the seemingly endless appetite of Canadian exporters for financial support’ through a variety of financial and risk absorption services, including export trade insurance, sales financing, loan guarantees, and equity investments. These benefits are not available to Canadian exporters from private
164 Id. at pg. 21.
165 Id.
166 Exh. BRA-79
167 EDC 1997 Annual Report, pgs. 32-33 (Exh. BRA-21).
168 Exh. CDN-38
WT/DS70/R
Page 67
financial institutions. As EDC itself acknowledges, ‘EDC complements the banks and other financial intermediaries, but cannot substitute for them.’ For this reason, EDC’s ‘goal is to help absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries.’” [emphasis added by Canada]
According to Canada, these quotations are taken out of context, in that the first two are actually from the concluding part of the Message, and that Brazil’s submission omits a connecting sentence that substantially qualifies the passage. Canada submits that the full paragraph, as set out in the fourth page of Exhibit BRA-7, reads:
“It will be hard to maintain the pace of 1995 and earlier, but EDC has a lot of growing to do before it begins to satisfy the seemingly endless appetite of Canadian exporters for financial support and advice. However, EDC cannot nor should not strive to be the solution for all the challenges faced by Canadian exporters. EDC complements the banks and other financial intermediaries, but cannot substitute for them.” [emphasis added by Canada]
For Canada, by saying that EDC cannot substitute for banks and other financial institutions, the Chairman and the President of the EDC were acknowledging that EDC should not try to satisfy the “’endless appetite of Canadian exporters.’” Canada submits that contrary to Brazil’s conclusion from the misquoted passages, this is manifestly not what the EDC attempts to do. In addition, Canada submits that Brazil’s argument that “…export trade insurance, sales financing, loan guarantees, and equity investments…” are “benefits” that “…are not available to Canadian exporters from private financial institutions” is not true, and notes that it has provided documents from a variety of private financial institutions operating in Canada, demonstrating that they provide these services to Canadian exporters.169
Canada notes that the third quotation, concerning the absorption of risk by the EDC, is repeated later by Brazil (para. 6.60), to support the allegation that “[n]o private financial institution or investor would provide this degree of financing on concessionary terms” [emphasis added by Canada], and to allege (para. 6.49), that “every move [EDC] makes” is in support of this “risk absorption” goal. According to Canada, the quotation is the basis for Brazil’s claim that “EDC is precisely what Article 3 of the SCM Agreement was intended to prohibit.”
Canada asserts that Brazil has omitted the qualifying subordinate clause of the quoted sentence, and that the full sentence, placed in context, is:170
“In addition to the shift from sovereign to commercial loans, the complexity, scale and duration of financing are changing, and thereby changing the risks associated with insuring and financing Canadian exports.
To reinforce its capacity to manage these changing risks, EDC has established a new Financial Services Office and procedures for evaluating loan portfolios on an industry, geographic, and individual transaction basis. Our goal is to help absorb risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries, by diversifying the Corporation’s business both on a country and sectoral basis. We are determined to achieve this goal through growth in both emerging and established markets.” [emphasis added by Canada]
Canada states that, contrary to Brazil’s assertion, the passage when quoted in full does not
support the proposition that EDC enters into “’financing on concessionary terms,’” nor the assertion
169 Exh. CDN-52
170 Exh. BRA-7, p. 2
WT/DS70/R
Page 68
that “’every move’” the EDC makes is to absorb risks on behalf of Canadian businesses. According to Canada, the sentence quoted by Brazil is concerned with portfolio diversification by the EDC, an elementary and prudent market activity, and has nothing to do with whether EDC financing confers a benefit.
Canada also takes issue with Brazil’s allegation regarding EDC’s “’concessionary’” rates or terms, arguing that Brazil does not define this term, and adduces no evidence in support of the allegation. Canada maintains that there is no such evidence. According to Canada, the EDC does not provide financing at concessionary rates. Canada states that apparently as support for its assertion, Brazil relies on statements that do not in any way support its allegation.
Canada notes Brazil’s allegation (para. 6.56) that “’EDC funding may be structured as direct financing at concessionary rates for up to 90 per cent of the cost of an aircraft.’” [emphasis added by Canada], noting that Brazil cites two sources as support for this assertion, one of which is a newspaper article, dated 5 April 1995,171 which quotes an Industry Canada official as saying:
“the government is risking less than it once did. The Air Canada guarantee may look similar to those provided for years on sales of such aircraft as the Dash 8 …. But in those deals, Ottawa put up as much as 90 per cent of the purchase price, Mr. Dixon said. Here, as little as 20 per cent is at risk.”
Canada asserts that Mr. Dixon is identified as an official at Industry Canada, that he was not speaking on behalf of the EDC, and that his statement did not refer to the EDC and in any event did not mention “concessionary financing”. Canada notes that the other source cited by Brazil is the testimony, on 11 May 1995, of Mr. Paul Labbé, then-President of the EDC172. Canada indicates that Mr. Labbé notes that “[h]istorically, we would have financed 85 per cent of the cost of an aircraft. Our capacity won’t do that any more.” Canada maintains that there is no support in this statement for the proposition that the EDC provides financing at concessionary rates.
EDC’s risk level and performance
According to Canada, EDC operates on commercial principles and is self-sustaining, as attested to by the fact that it earns a net interest margin that is equal to or better than most commercial financial institutions of similar credit rating. Canada states that net interest margin represents the difference between gross interest income and gross interest expense on all interest bearing assets, divided by the value of all interest bearing assets, and thus is a useful measure of how well a financial institution’s assets are performing. Canada submits that EDC ranks just behind Chase Manhattan Bank and Citibank, and well ahead of other banks such as Barclay’s Bank, Deutsche Bank, and the Union Bank of Switzerland.173
In response to a Panel question in light of Brazil’s arguments concerning the risk level of EDC’s portfolio, regarding whether EDC’s credit rating is attributable to the fact that it is a Crown Corporation of Canada, Canada replied that EDC’s credit rating of Aa2/AA+ is based upon the rating accorded to the Government of Canada, and that the EDC has – on more than one occasion – requested Moody’s and Standard & Poor’s to rate the EDC separately from the Government of Canada, based upon the EDC’s portfolio of assets, but that these agencies have indicated that they see no benefit in providing such a rating.
Canada asserts that in making an assessment of a financial institution, the rating agencies would assess, among other things, the sufficiency of the financial institution’s provisioning levels.
171 Exh. BRA-8, p.2
172 Exh. BRA-9
173 Bankstat, Net Interest Margin Chart, 2 October 1998 (Exh. CDN-38).
WT/DS70/R
Page 69
Based on regular discussions with the rating agencies, EDC provisions so that it maintains a notional rating for its total portfolio of business at a targetted level of Aa2/AA+. According to Canada, the balance sheet of the EDC remains at the Aa2/AA+ level due to the overall level of its provisions and the EDC’s conservative accounting treatment of non-performing loans.174
In turn, Canada states, the EDC’s provisions have been funded by the risk margins charged by the EDC to its clients. According to Canada, the EDC maintains an overall Net Interest Margin (NIM)
which takes into account the carrying costs of the non-performing loans on which no income is earned -- which compares favourably with a number of private financial institutions. 175 The net interest margin is a useful measure as it allows an apples to apples comparison of financial institutions.
Canada argues that EDC employs one of the largest pools of trade financing experts in Canada, and has experience in over 200 markets and expertise in many highly specialised fields. According to Canada, private financial institutions such as Citibank or Bank of America employ a similar pool of talent,176 and like Citibank and Bank of America, the “competitive edge” that EDC gives Canadian exporters is EDC’s experience and expertise (para. 6.49).
Canada notes, in reply to a Panel question, that EDC pays business taxes, but not corporate income tax, and does not normally pay a dividend. For Canada, these facts are not relevant for the net interest margin calculation because the net interest margin is calculated before tax and before dividends are paid. That is, whether or not a financial institution pays corporate income taxes or dividends does not affect its cost of funds, or the risk margin charged.
Regarding Brazil's assertion that EDC's portfolio is riskier than the portfolios of other commercial financial institutions, Canada disagrees with Brazil's conclusion that the EDC’s net interest margin is insufficient given the level of the EDC’s non-performing loans in its portfolio, because net interest margin already takes non-performing loans into account (para.6.12 - 6.15).
According to Canada, when a loan becomes non-performing, all financial institutions, including the EDC, cease to recognize interest income on that non-performing loan, but continue to carry the interest expense of that non-performing loan. That is, the gross interest income is the interest income on performing loans, but interest expense is the cost of funding all loans in the portfolio, performing and non-performing. Therefore, Canada maintains that it is incorrect for Brazil to assert that a further deduction should be made from the net interest margin to cover the cost of funding these non-performing loans, as this would be double-counting.
Regarding Brazil's arguments about EDC’s non-performing loan portfolio, Canada argues that EDC recognizes three classes of loans: 1) investment grade - which corresponds to an S&P credit rating of AAA to BBB; 2) below investment grade – which corresponds to the range BB to B and 3) speculative grade, which is CCC. Canada states that EDC’s civil aviation portfolio is as follows: 65 per cent investment grade; 13 per cent below investment grade and 22 per cent speculative grade, before security has been recognized. According to Canada, every commercial loan in EDC’s aircraft portfolio is secured against the aircraft in the transaction, much like a mortgage on a house. Canada asserts that credit rating agencies suggest that a credit can improve by up to two categories if
174 (EDC 1997 Annual Report at p. 47, “Net Loans Receivable” – Exh. BRA-21).
175 Exh. CDN-101.
176 Bank of America, “Industry Expertise” on line: Bank of America Homepage
http://www.bankamerica.com (date accessed: 5 November 1998) (Exh. CDN-39); Citibank, “Small Business & Professional” and “Global Banking Services” online: Citibank homepage
http://www.citibank.com/US/businessbanking/main.htm and … /global.htm (date accessed: 3 November 1998) (Exh. CDN-40).
WT/DS70/R
Page 70
secured.177 178 Canada maintains that securing the loan against the asset – the aircraft – improves the credit rating of the credit typically by one to two rating designations, and that where EDC is over- secured, the credit enhancement is even greater. Canada argues that after considering the value of the security, 91 per cent of EDC’s aircraft portfolio is investment grade or higher - in other words BBB or better.
Canada characterizes Brazil’s comparison of spreads between corporate bonds, on the one hand, and secured lending, on the other hand, as fundamentally flawed. Canada states that corporate bonds are unsecured and generally have “bullet” payments, in which the full amount of the principal is paid in a lump sum at the end, while EDC’s lending is secured, and the principal is amortized over the life of the loan. For Canada, there is obviously a considerable difference between the two, which has an impact on the credit risk. According to Canada, EDC is not, therefore, engaged in the high risk lending in the civil aircraft sector that Brazil implies.
Regarding Brazil's assertion that “'EDC’s allowances for loan losses as a percentage of total loans is 13.2 per cent, compared with amounts between one and two per cent for each of three major Canadian banks and the US industry average'" (para. 6.14) [emphasis added by Canada], Canada argues that EDC allowances for loan losses are at this level to ensure that EDC’s total portfolio maintains an equivalent credit rating of at least AA. In Canada's view, the only thing that matters is how much money is lost on a portfolio and whether there is adequate provisioning. Canada states that the EDC’s write-offs against provisions have never exceeded 0.15 per cent of gross loans receivable over the past 8 years, in comparison to the Canadian commercial banks, none of which in that same period have had write-offs of less than 0.21 per cent and which in one case has been as high as
1.52 per cent. Canada notes in addition, quoting from Brazil’s Exhibit 86, that the US Federal Deposit Insurance Corporation stated that the net write-offs of the US banking industry amounted to
0.63 per cent of average loans in 1997, down from a peak of 1.59 per cent in 1991. Thus, in Canada's view, EDC’s write-offs against provisions are more conservative than any of these other institutions.
Rebuttal of Brazil on EDC’s risk level and performance
Regarding Canada’s arguments concerning EDC’s credit rating (para. 6.28) Brazil argues that the fact that Moody’s and Standard & Poor’s have “’indicated that they see no benefit’” in rating EDC separate from the Government of Canada does not mean that EDC has earned an Aa2/AA+ rating by having reserves for loan losses commensurate with the risk of its portfolio; rather, it means that Moody’s and Standard & Poor’s recognize that the EDC is backed by the Government of Canada. Based on the evidence of the riskiness of EDC’s portfolio (paras. 6.12- 6.15), Brazil submits that were Moody’s or Standard & Poor’s to consider it necessary or beneficial to rate EDC independently, EDC would be accorded a rating significantly lower than the Aa2/AA+ rating accorded the Government of Canada.
Concerning Canada’s rebuttal arguments on net interest margin, the ratio of gross impaired loans/gross loans receivable and the allowance for losses on loans/gross loans receivable, Brazil argues that on net interest margin Canada’s response ignores the fundamental premise of Brazil’s argument, which is essentially the risk/return principle: “’Investors increase their required rates of return as perceived risk (uncertainty) increases.’”179 For Brazil, simply asserting that EDC’s net interest margin is roughly in line with those achieved by commercial banks ignores the fact that EDC should, in fact, earn a much higher return as compensation for the higher risk of the portfolio. Brazil
177 Exh. CDN-98 and 99
178 Standard & Poors, S&P’s Corporate Finance Criteria, (McGraw-Hill, 1994) at 53-54 (Exh. CDN-98); Moody’s, A Sense of Security – Moody’s Approach to Evaluating Bank Loan Structure and
Collateral, (New York, Moody’s: 1997) (Exh. CDN-99).
179 Frank K. Reilly, Investment Analysis and Portfolio Management, Fourth Edition (Dryden Press),
pg. 25.
WT/DS70/R
Page 71
notes Canada’s statement, with respect to “Analysis of EDC’s Financial Performance,”180 that EDC’s “’average portfolio of business is of a poorer risk quality’”, and disputes Canada’s statement that Brazil argues that a further deduction should be made from the net interest margin to cover the cost of funding these non-performing loans. Brazil states that it made no such assertion, but instead asserts investments with higher risks demand higher returns, which higher return compensates the investor for taking on additional risk, not for the cost of funding the investments.
Regarding the ratio of gross impaired loans/gross loans receivable, Brazil notes Canada’s acknowledgement (para. 6.47) that EDC’s non-performing loans comprise 14.4 per cent of gross loans receivable, and recalls that this same ratio is approximately one per cent for The Royal Bank of Canada, The Bank of Montreal, Canadian Imperial Bank of Commerce, and the average of all US FDIC-insured commercial banks.181 Brazil also notes that non-performing (also called “impaired”) loans are defined by the EDC as follows:
Loans are classified as impaired when circumstances indicate that [EDC] no longer has reasonable assurance that the full amount of principal and interest will be collected on a timely basis in accordance with the terms of the loan agreement.182
Brazil notes that Canada provides a breakdown of sovereign versus commercial non- performing loans and notes that the EDC’s level of non-performing loans has been declining in recent years183. For Brazil, Canada fails to respond to Brazil’s assertion that the overall level of impaired loans provides strong evidence that the riskiness of the EDC’s overall portfolio far exceeds those of commercial banks and should be accompanied by a much higher return on the portfolio.
Brazil argues that Canada’s argument actually focuses on the EDC’s performing loans, omitting a discussion of impaired loans, and notes Canada’s acknowledgement (para. 6.45) that the EDC’s civil aviation performing portfolio is classified by the EDC itself as 65 per cent investment grade, 13 per cent below investment grade, and 22 per cent speculative grade. Brazil argues that in the same paragraph and without citation to any source, Canada asserts that the EDC has apparently misclassified its own loans, with the result that 91 per cent of EDC’s aircraft-related loans should be considered, according to Canada, of investment grade or higher. Brazil finds the EDC’s own classification, with 13 per cent below investment grade and 22 per cent speculative grade, to be more credible than the speculation offered by Canada.
Regarding the allowance for losses on loans/gross loans receivable, Brazil states that Canada acknowledges that the allowance for losses on loans/gross loans receivable reflects the “’poorer risk quality’” of the EDC portfolio184, which is precisely the point made by Brazil; that is, that the allowance ratio of 13 per cent (versus one to two per cent for commercial banks) provides strong evidence of a much riskier portfolio than those of commercial banks.
Brazil notes Canada’s conclusion (para. 6.37) that “’the only thing that matters is how much money you lose on your portfolio and whether you have adequately provisioned against this loss’”, and the information on EDC’s write-off history provided by Canada as evidence that the EDC has not lost much money on its portfolio. Brazil notes that Standard and Poor’s Banking Industry Survey states that “’[w]hen management deems a loan uncollectible, it’s written off the books and a deduction is made from the reserve for loan losses.’”185 Brazil states that managers exercise a great
180 Exh. CDN-101
181 Exh. BRA-79.
182 EDC 1997 Annual Report, pg. 47 (Exh. BRA-21).
183 Exh. CDN-101
184 Exh. CDN-101, Section I(b) (“Due to the nature of the business, EDC’s provision charge is higher than that for a commercial bank because the average portfolio of business is of a poorer risk quality.”).
185 Standard and Poor’s Banking Industry Survey, dated 11 June 1998, pg. 21 (Exh. BRA-86).
WT/DS70/R
Page 72
deal of discretion in deciding when to write off a loan, and argues that the EDC’s 1997 net write-offs of 0.15 per cent of gross loans receivable is absurdly low, particularly given the very high level of impaired loans. For Brazil, the level of write-offs does not provide evidence that the EDC “’has not lost much money’”, but rather begs the question of what criteria are applied by the EDC in determining when loans are written off, and whether or not these criteria are similar to those employed by commercial banks.
Response of Canada on EDC’s risk level and performance
Regarding Brazil’s response to Canada’s discussion of EDC’s risk level, Canada comments that Brazil ignores that the cost of carrying non-performing loans lowers the EDC’s NIM, recalling that the EDC’s non-performing loans comprise 14.4 per cent of gross loans receivable, in contrast to one to two per cent for commercial banks, and arguing that if it were not for the fact that the EDC’s portfolio contains a relatively higher level of non-performing loans, the EDC’s NIM would be higher. In spite of these non-performing loans, Canada argues, the EDC maintains a NIM that is equal to or better than many commercial banks with a comparable credit rating, that according to Brazil have less risky portfolios and lower levels of impaired loans. Thus, contrary to Brazil’s assertions, the EDC’s NIM does not indicate that the EDC does not earn a return commensurate with the risk of its portfolio.
Canada also denies that it asserts that the EDC has misclassified its own loans, stating that Brazil ignores the explanation Canada provided (para. 6.35) regarding the effect of taking security on these loans. Canada notes that it focussed on the EDC’s aircraft portfolio and noted that these loans are secured against the aircraft in the transaction, which according to Canada is common in aircraft financing186, and effectively raises the percentage of investment grade assets from 65 to 91 per cent187.
Regarding allowance for losses on loans/gross loan receivable, Canada maintains that a more complete quotation of its exhibit188 than that cited by Brazil reads “Due to the nature of the business; EDC’s provision charge is higher than that for a commercial bank because the average portfolio of business is of a poorer risk quality…” [emphasis added by Canada]. According to Canada, this higher provision charge has resulted in higher allowances for losses on loans189, and these allowances effectively offset the risk in EDC’s portfolio.
Canada rejects Brazil’s suggestion that EDC employs different criteria for loan write-offs than do commercial banks, and refers the Panel to a letter from Canada’s Auditor General reproduced at page 43 of the EDC’s 1997 Annual Report190. According to Canada, this letter confirms that the EDC’s financial statements are presented in accordance with Generally Accepted Accounting Principles, which cover financial disclosure by all entities that are subject to them, including both the EDC and commercial banks. According to Canada, essential elements of Generally Accepted Accounting Principles are the criteria applied and the consistent application of those criteria when valuing and writing off loans, and an auditor must be assured that management applies these principles on a consistent basis so that the value of assets would not be overstated or manipulated, and thus the Auditor can opine that the financial statements present fairly in all material respects the financial position of the entity. Canada argues that Brazil provides no evidence to support its assertion that “’managers exercise a great deal of discretion in deciding to write off a loan’”, or that EDC uses criteria different from those employed by commercial banks.
186 Exh. CDN-48
187 Exh. CDN-98 and 99
188 Exh. CDN-101
189 Exh. BRA-79.
190 Exh. BRA-21
WT/DS70/R
Page 73
Contingency on export performance
Arguments of Brazil
Brazil submits that the EDC provides prohibited export subsidies. First, Brazil argues, the EDC exists solely to support export transactions. Brazil cites to testimony by EDC’s President before the Canadian Parliament that “[t]he goal for Canada is to make sure that we have a competitive advantage for Canadian exporters, not just a level playing field.”191 According to Brazil, every subsidy EDC grants, every benefit it confers, every move it makes, is intended to “’absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries.’”192 Brazil asserts that EDC bills itself “not [as] a profit maximizer, but an export maximizer.”193 Brazil argues that subsidies granted Canadian enterprises by EDC are, as a matter of legal directive, “’for the purposes of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.’”194 Brazil states that these subsidies are, within the meaning of Article 3, contingent in law and in fact upon export performance, and are therefore prohibited; and that EDC is precisely what Article 3 of the SCM Agreement was intended to prohibit.
Brazil contends that Section 10(1) of the Canadian Export Development Act states that EDC was established “'for the purposes of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.'”195 Brazil asserts that EDC is required to fund exports, as opposed to domestic sales, and that as a result, EDC itself and assistance to the regional aircraft industry granted thereunder, is de jure contingent upon export, within the meaning of Article 3 of the SCM Agreement.
Arguments of Canada
In answer to Panel questions whether Canada concedes that EDC activities are contingent upon export performance, and whether given EDC’s mandate it is reasonable to assume that any transaction financed with EDC assistance is necessarily an export transaction, Canada replies in the negative. Canada states that the EDC has the mandate, under section 10 of the Export Development Act, to engage in activities with, in essence, objectives of supporting and developing Canada’s export trade; and Canadian capacity to engage in that trade, and respond to international business opportunities.
According to Canada, EDC’s mandate allows it to offer a full range of risk management services and financing products “for the purpose of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.” Canada states that EDC offers, therefore, a variety of services and products, some of which are contingent on export, and others – such as foreign investment insurance, domestic credit insurance, funding investments overseas, and various equity investments – that are not contingent on export.
Canada states that an article submitted by Brazil196 supports this, as it refers to the “’the new legislative mandate adopted by the EDC in June this year [1994]’” and notes that the new regulations
191 Testimony of EDC President Mr. Ian Gillespie before the Standing Committee on Foreign Affairs and International Trade, 6 November 1997, p. 15 (Exh. BRA-66).
192 Message at 2 (Exh. BRA-7).
193 Testimony of EDC President Mr. Ian Gillespie before the Standing Committee on Foreign Affairs and International Trade, 6 November 1997, p. 3 (Exh. BRA-66).
194 Export Development Act, Section 10(1) (Exh. BRA-6).
195 Export Development Act (Exh. BRA-6).
196 Exh. BRA-13
WT/DS70/R
Page 74
“’allow EDC to, among other things, provide comprehensive cover for domestic as well as export receivables …’”. While Canada does not vouch for the accuracy of the rest of the article, it states that this part is accurate.
Further, for Canada, Mr. Labbé’s comments relied upon by Brazil establish simply the following: that, although the focus of the EDC is exports and international business (as indeed it cannot be otherwise), it may engage in certain domestic transactions that help develop Canada’s capacity to engage in export trade and to respond to international business opportunities. That is to say, for Canada, the fact that the EDC is involved in a transaction does not mean that the transaction itself is necessarily an export one.
Response of Brazil
Regarding Canada’s argument concerning EDC’s “’new legislative mandate’”, Brazil questions Canada’s reliance upon the article submitted by Brazil, in view of Canada’s characterization of that article as a press report which is “’uncorroborated or does not otherwise contain material with an independent title of credibility and persuasiveness’” (paras. 6.97). In addition, Brazil argues that Canada surely could have provided the Panel with the actual legislative or regulatory language evidencing this change in Canadian law, rather than relying on a press article. Brazil argues that under similar circumstances, the Panel in Indonesia - Certain Measures Affecting the Automobile Industry considered that reliance on press reports constituted “’failure to submit the positive evidence required’” to support a claim
. . . given that the affected companies certainly had at their disposal copious evidence in support of the claims of the complainants, such as the actual business plans relating to the new models, government documentation indicating approval for such plans (assuming the "approval" referred to by the complainants with respect to the Optima means approval by the Indonesian government), and corporate minutes or internal decision memoranda relating both to the initial approval, and the subsequent abandonment, of the plans in question.197
EDC Debt Financing
Arguments of Brazil
Brazil states that EDC export financing takes various forms, and may be structured as direct financing at concessionary rates for up to 90 per cent of the cost of an aircraft.198 Brazil asserts that EDC itself acknowledges the concessionary nature of its export assistance, referring to testimony before Parliament by Paul Labbé, President and Chief Executive Officer of EDC, that EDC strives to “mak[e] at least the rate of inflation,” and that such a return is well below the return “that would be required to survive in the private sector.”199 Brazil cites as an example of EDC’s direct financing an April 1997 agreement by Bombardier for the sale of 30 CRJs (with options for an additional 60 aircraft), valued at approximately US$600 million, to ASA Holdings, Inc. and its subsidiary, Atlantic
197 Indonesia - Certain Measures Affecting the Automobile Industry, WT/DS54/R, WT/DS55/R, WT/DS59/R, WT/DS64/R, adopted 23 July 1998, at paras. 14.234-14.235.
198 “Ottawa Joins Bombardier in jet-leasing venture,” The Globe and Mail, 5 April 1995 (Exh. BRA-8) (To support Dash 8 sales, “Ottawa put up as much as 90 per cent of the purchase price.”). See also House of Commons of Canada, 35th Parliament, 1st Sess., Evidence, Standing Committee on Foreign Affairs and
International Trade, Meeting No. 43, 11 May 1995, 43:10 (“Committee on Foreign Affairs and International Trade”) (Exh. BRA-9).
199 House of Commons of Canada, 35th Parliament, 1st Sess., Standing Committee on Banking, Trade and Commerce, Issue No. 48, 28 November 1995, 48:22 (“Committee on Banking, Trade and Commerce”) (Exh. BRA-10).
WT/DS70/R
Page 75
Southeast Airlines.200 According to Brazil, the transaction included a commitment from EDC to finance up to 85 per cent of the lease or purchase price for all 30 CRJs, an option exercised by ASA for those aircraft delivered to date201 to finance 16.5-year leases of the CRJs.202
According to Brazil, EDC President Paul Labbé, in information published by EDC, expressly acknowledged the benefit granted to Canadian exporters by EDC financing:
EDC’s financing support gives Canadian exporters an edge when they bid on overseas projects. . . . Trade deals increasingly depend on complex and tightly negotiated financing arrangements where a few basis points in interest rates can make or break the deal. Exporters are having to bid not just on the basis of quality and price, but also on the basis of the financing package supporting the sale.203
For Brazil, EDC’s help – to the tune of a “’few basis points’” – must be better than that which would otherwise be commercially available, or an EDC financing package would not, in EDC’s former President’s words, “’give Canadian exporters an edge.’” For Brazil, this “’edge’” is a “benefit” conferred upon exporters, within the ordinary meaning of Article 1.1 of the SCM Agreement.
Similarly, according to Brazil, one of the benefits of EDC financing is that a Canadian exporter can “'advis[e] potential foreign buyers that Canadian financing may be available for their purchase,'” thereby “’enhanc[ing] the competitiveness of [the exporter’s] sales proposal.'”204 Brazil asserts that the Canadian regional aircraft exporter’s product is more attractive to a purchaser, in turn, for the simple reason that it costs less than it would without the Canadian Government’s help.
According to Brazil, therefore, EDC’s provision of financing of up to 90 per cent (or more) of an aircraft’s cost over a 15-year or 15-year-plus period at concessionary rates confers the obvious benefit, within the meaning of Article 1.1, of lowering the price of an exported aircraft for the purchaser. For Brazil, no private financial institution or investor would provide this degree of financing on concessionary terms, which is why EDC steps in to “’help absorb the risk . . . beyond what is possible by other financial intermediaries.‘”205
Brazil argues that even if it is relevant, under Article 1.1, to inquire whether the government in making a financial contribution has covered its costs, pure analysis of the cost to the lender is not appropriate in the case of EDC, which boasts of its ability to “’absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries.’”206 As compensation for this risk, which is a cost according to Brazil, EDC should be expected to collect a risk premium. Brazil notes that it is a fundamental tenet operating in financial markets that “’investors increase their required rates of return as perceived risk (uncertainty) increases.’”207 Yet, Brazil argues, EDC’s former president, in testimony to Parliament, has spoken of EDC’s goal of avoiding losing money by “’making at least the rate of inflation,’” which he recognized is well below the return “’that would be
200 ASA Holdings, Inc., Form 10-Q, filed with the US Securities and Exchange Commission for the quarterly period ended 31 March 1997 (File No. 333-13071), pgs. 8-9 (Exh. BRA-11).
201 ASA Holdings, Inc. Annual Report 1997, “Liquidity and Capital Resources,” Lexis Disclosure Report, pgs. 15-16 (Exh. BRA-12).
202 Id.
203 CanadExport On-Line, Focus on Export Development Corporation, at 2 (Exh. BRA-67).
204 Infoentrepreneurs website, pg. 13 (Exh. BRA-68).
205 Message at 2 (Exh. BRA-7).
206 Export Development Corporation, 1995 Chairman and President’s Message, pg. 2 (Exh. BRA-7).
207 Frank K. Reilly, Investment Analysis and Portfolio Management (Dryden Press), Fourth Edition, pg.
25 (Exh. BRA-81).
WT/DS70/R
Page 76
required to survive in the private sector.’”208 Brazil questions how, if EDC is absorbing risk beyond that which the private sector is willing to accept, merely making the rate of inflation allows it to meet its costs?
Brazil recalls its reference to EDC’s low net interest margin to establish that EDC is not being compensated for this extra risk and, conversely, that recipients of EDC funding are paying less for that funding than they would from commercial sources (paras. 6.7-6.16, 6.38-6.44).
Brazil takes issue with Canada's response referring to EDC’s 1996 net interest margin of 3.03 per cent – which Canada claims is higher than that achieved by some commercial banks -- as evidence that it is covering costs incurred on its lending activities. (para. 6.73). Brazil maintains that for a commercial bank to accept EDC’s loan portfolio a much higher return would be required, and submits an analysis of the differential in risk between EDC’s loan portfolio and the loan portfolios of commercial banks209. For Brazil, this analysis shows that financing of EDC’s risk class by private investors would demand a spread of 1,242 basis points over riskless US Treasuries of 15+ years maturity, or 17.73 per cent, a conclusion which demonstrates the degree to which EDC’s lending activities, yielding a net interest margin in 1996 of only 3.03 per cent, do not effectively cover the costs associated with the riskiness of EDC’s loan portfolio, and grant borrowers access to rates not otherwise available from lenders disciplined by market forces. Brazil further argues that even if all of EDC’s loans are secured, private lenders still demand a spread of at least 150 basis points above the riskless US Treasuries of identical tenor. In response to a Panel question, Brazil states that this assertion is based on a comparison of the yield for riskless US Government 15-year securities and the yield for investment grade-rated unsecured non-rail transportation bonds.210
Arguments of Canada
Substantive issues
Canada submits that financing activity by the EDC under its corporate account (EDC financing) is not a subsidy within the meaning of Article 1, and that it is therefore not necessary for the Panel to determine whether such activity is contingent upon export performance. Canada further submits that Canada Account activity is not inconsistent with Article 3 as it benefits from the exception contained in Item (k) of Annex I of the SCM Agreement. Canada notes, finally, that Brazil’s other allegations about the EDC are factually incorrect and must therefore fail.
Canada states that as an official export credit agency and agent in all respects for the Crown, the EDC has a public policy function that it carries out applying commercial disciplines and principles, and that it may also conduct financial activities that the Government may deem to be in the national interest, any obligations under which would be funded by the Government of Canada. Canada notes that these latter activities of EDC are called the “Canada Account”, and that EDC’s activities on its own account as a commercial operator are referred to as “corporate account” or simply EDC financing or loan guarantees. According to Canada, the EDC engages in such activities in accordance with its broad mandate, prudent risk management and international export credit disciplines.
According to Canada, the Government of Canada may, in view of its international commitments through the Paris Club, on occasion require that the EDC forgive sovereign debts under its corporate account. In these circumstances, Canada states, the Government of Canada indemnifies the EDC. Canada, noting Brazil’s reference to “’Can$151 million in direct government relief to EDC
208 House of Commons of Canada, 35th Parliament, 1st Sess., Standing Committee on Banking, Trade and Commerce, Issue No. 48, 28 November 1995, pg. 48:22 (Exh. BRA-10).
209 Exh. BRA-85
210 Exh. BRA-85
WT/DS70/R
Page 77
for two of its ‘problem' accounts‘” (para. 6.7), maintains that this amount was paid to the EDC as indemnification for the forgiveness of the sovereign debts of Poland and Egypt by the EDC in accordance with instructions by the Government of Canada and Canada’s international commitments, and that these were unrelated to the civil aircraft sector.
Canada maintains that Canada Account and EDC activities are not prohibited under Article 3 of the SCM Agreement. Canada recalls Brazil’s argument that EDC financing may be structured at “’concessionary rates,‘” asserting that Brazil alleges seven times that the EDC provides loans at concessionary rates, but provides no evidence for these allegations, nor defines what it means by “’concessionary rates‘”. According to Canada, Brazil extrapolates the concessionary “’nature‘” on the basis of an unfounded allegation that the EDC’s profits are lower than those expected in the private sector.
Canada denies that the EDC provides subsidies, stating that the EDC does not finance at concessionary rates, and characterizing Brazil’s argument as based on unfounded speculation. Canada notes that Brazil does not define what it means by “’concessionary rates‘”; Canada submits that Article 38(c) of the OECD Consensus provides a useful point of reference for such practice, quoting this Article as follows:
“(c) For the purpose of calculating the overall concessionality level of an associated financing package, the concessionality level of the following credits, funds and payments are considered to be zero:
export credits that are in conformity with the Arrangement;
other funds at or near market rates; ….” [emphasis added]
Canada asserts that the EDC does not provide loans at “concessionary rates”, as the EDC’s financing activities are based on commercial pricing (“’at or near market rates‘”). That is, according to Canada, rates for EDC financings reflect commercial benchmarks211 and “’spreads‘”212 that are in accordance with commercial credit ratings213 -- and, where this is not available, internal credit ratings in accordance with prudent commercial practices. As well, Canada argues, EDC’s financing terms and structures are consistent with market trends and practices. In response to a question from the Panel seeking elaboration of these statements, Canada replied that it does not seek shelter in the ‘safe haven’ provided in the second paragraph of item k, with respect to EDC financing under its corporate account, and that its reference to Article 38(c) of the OECD consensus was to provide contextual guidance as to what Brazil might have meant by ‘concessionary rates’, and to make the point that EDC’s corporate account does not provide concessionary financing.
211 In a floating rate transaction the lender sets an interest rate that will be moved up or down in relation to general movements in interest rates in the wider economy. In floating-rate aircraft financing transactions, the benchmark used (to reflect the “general movements in interest rates”) is the three-month or the six-month LIBOR (London Inter-Bank Offer Rate).
In fixed-rate transactions, the borrower’s interest payments are set at the outset of the transaction and are not subject to variation in the underlying interest rate. In aircraft financing transactions (which are mostly US dollar-denominated) the benchmark used is US Treasury.
212 The charge added to a benchmark that reflects the credit risk of the transaction.
213 “Credit rating”, whether done by a rating agency or by a financial institution, is a measure of the
degree of risk of default by an enterprise of a loan or an investment, determined on the basis of the revenue, profitability and the assets of the enterprise, as well as collateral guarantees underwriting the transactions of that enterprise and the likelihood of change in any of these variables. Such ratings may also be given to bond issues covering, for example, the purchase of capital assets by an enterprise. The market perception of the asset, its value and durability may factor in determining the rating to be given to such an issue.
WT/DS70/R
Page 78
According to Canada, like several other international financial institutions, the EDC’s internal credit ratings are based upon the results of analyses using a sophisticated computer programme: LA Encore.214 Canada maintains that this programme is employed for the same purpose by other major financial institutions, such as Lloyd’s Bank and Barclay’s Bank in the United Kingdom.
In terms of the pricing process, according to Canada, EDC’s transportation group has a committee that reviews and approves the pricing on all transactions in the civil aircraft sector. Canada argues that in setting this pricing, EDC compares what the relevant borrower has recently paid in the market for similar terms and with similar security. The EDC then prices according to that benchmark. In the absence of this benchmark, the EDC compares the relevant borrower to borrowers of comparable credit standing in the civil aviation sector for whom a similar credit history exists; the EDC then prices according to this alternative benchmark.
Canada maintains that to ensure that the EDC’s pricing reflects the proposed risks being assumed and before approving the pricing, the EDC does one final test: based upon the credit rating and security of the transaction, the EDC ensures that the pricing will add value to the EDC’s capital base, using the EDC’s portfolio analysis tool for the civil aviation sector. For Canada, the rationale for this test is that market pricing is usually efficient and, if a transaction is priced to market, then generally it should always add value to the EDC’s capital base. Canada asserts that the EDC’s civil aviation portfolio is performing well and for any new deal, the EDC incorporates that deal into the current portfolio to ensure that it at least maintains the value of the portfolio; and that if not, the EDC will increase the interest rate such that the value of the portfolio is maintained. Canada responds to Brazil’s allegation, based on a quotation of Mr. Paul Labbé from an article that appeared in
CanadExport On-Line (para. 6.57)215, that “EDC’s help – to the tune of a ‘few basis points’ – must be
better than that which would otherwise be commercially available, or an EDC financing package would not, in EDC’s former President’s words, ‘give Canadian exporters an edge’.” Canada argues that Mr. Labbé’s statement is not evidence that EDC finances below market and refers to Mr. Labbé’s
quote earlier in the article describing the role of EDC:216
“What we bring to the table is a wide variety of financial solutions and insurance support, as well as extensive market and sectoral expertise… We have teams dedicated to different market sectors such as information technology and industrial equipment so that we understand your business as well as you do.”
In Canada’s view, taken in context, then, the edge that EDC’s financing support gives Canadian exporters derives from their knowledge of the various export markets, from their expertise in “complex and tightly negotiated financing arrangements” and from their awareness that a “few basis points in interest rates can make or break the deal”. Canada reiterated that this sort of knowledge and expertise is something that almost all financial institutions trumpet in their promotional literature.
Canada provided provided examples from Barclays, Warburg Dillon Read and UBS.217 Canada
submitted that in this context, Mr. Labbé’s statement shows that EDC recognises the factors that are important in making a deal in the international market. This is not, however, according to Canada a prima facie case that EDC’s finances below market.
214 Famas Inc. “Products: Famas Financial Statement Analysis Software” online: Famas homepage http://www.famas.com (date accessed: 3 December 1998) (Exh. CDN-57).
215 Brazil Exh. BRA- 67.
216 Id.
217 See Barclays, online: What Barclay’s Can Offer homepage
http://www.business.barclays.co.uk/corp/cb11.htm (date accessed: 08 December 1998). (Exh. CDN-95); Warburg Dillon Read, online: Banking Finance homepage http://www.wdr.com/public/products/finance (date accessed: 08 December 1998). (Exh. CDN-96); UBS, online: Data Mining homepage
http://www.ubs.com/e/index/about/ubilab/ext/projects/dm/e_index.htm (date accessed: 08 December 1998). (Exh. CDN-97)
WT/DS70/R
Page 79
Canada maintains that the commercial viability of EDC’s financing activity should be viewed in the context of the fact that the core lending business of many major banks is “increasingly unrewarding”,218 asserting that at 3.03 per cent, EDC’s net interest margin – which for Canada is a better measure of performance than “return on equity” -- is better than most commercial banks of similar or better credit rating.219 In commenting on Brazil’s response to the Panel’s question concerning risk spreads demanded by private lenders (para. 6.63), Canada observes that Brazil cites as its source the yield for investment grade-rated unsecured non-rail transportation bonds, purportedly in support of Brazil’s assertion that “[e]ven if we assume that all of the EDC’s loans for aircraft are secured, private lenders still demand a spread of at least 150 basis points…” Canada submits that the yield on unsecured bonds does not support an argument regarding secured lending.
Canada submits that Item (k) of Annex I provides a specific contextual guide as to the conditions under which EDC’s financing activity -- that is, export credits -- could be considered subsidies. For Canada, the structure of Item (k) parallels that of Article 1; as such, a strong inference may be drawn that the definition of subsidy in Article 1 should be informed by that set out in Item (k) as far as government credit is concerned.
Canada notes that Item (k) sets out a test with two elements. It provides, in relevant part, that: “The grant by governments (or special institutions controlled by and/or acting under
the authority of governments) of export credits at rates below those which they
actually have to pay for the funds so employed (or would have to pay if they borrowed on international capital markets in order to obtain funds of the same maturity and other credit terms and denominated in the same currency as the export credit), or the payment by them of all or part of the costs incurred by exporters or financial institutions in obtaining credits, in so far as they are used to secure a material advantage in the field of export credit terms.”
Thus, according to Canada, export credit constitutes a prohibited subsidy, and government credit constitutes a subsidy, where the rate of interest charged by a government in granting such a subsidy is lower than the rate it has paid, or would pay, to raise such amount, and where such credit secures a material advantage in the field of credit terms. That is, in the field of government credit, a subsidy exists where (a) a net cost to the treasury has been identified, that (b) results in an advantage above and beyond the market.
Canada argues that as the EDC always lends above its cost of funds, and does not incur a net cost on its financing activities, and that as it operates on the basis of commercial principles, it does not provide an advantage above and beyond the market. For Canada, EDC financing does not therefore constitute a subsidy, and Brazil’s claims in this respect must therefore fail.
In response to a Panel question seeking documentary support for its statement that EDC always lends above its cost of funds, Canada provides, as Business Confidential Information, a certified copy of the EDC’s Standing Board Resolution of 17 June 1992220. According to Canada, this Resolution applies to all business conducted by EDC under its corporate account, including the civil aircraft sector, and EDC has, in accordance with this resolution, lent above its cost of funds under its corporate account with respect to the civil aircraft sector since 1 January 1995.
Also in response to a Panel request for information and documentation relating to the terms of and conditions of the EDC financing allegedly provided to ASA, Canada states that the ASA deal is mentioned by Brazil at the end of a paragraph describing purported “concessionary” financing
218 “The trials of megabanks”, The Economist, 31 October 1998, 23 at 23. (Exh. CDN-41)
219 Bankstat, Net Interest Margin Chart, 2 October 1998 (Exh. CDN-38).
220 see BCI Tab 4.
WT/DS70/R
Page 80
provided by EDC, but that Brazil has made no allegation concerning ASA. Rather, Canada states, according to Brazil, ASA is merely an example of how EDC debt financing works (para. 6.56). Canada does not dispute the relevant quotes referred to by Brazil from the Form 10-Q filed with the US Securities and Exchange Commission by ASA Holdings Inc.
Canada states that the information requested by the Panel is sensitive business confidential information, and that Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. In this context, Canada does not consider it appropriate to adduce evidence in response to allegations not made and a case that has not been established.
Canada disputes Brazil's assertion that EDC funding for the CRJ “'is equivalent to US$18.3 million per airplane…virtually 100 per cent of the price of the airplane.'” (para. 6.4.) Canada asserts that the reference to $1.7 billion was to the volume of export insurance and export financing undertaken in the transportation sector, and that the reference to 62 per cent represents the percentage of EDC’s total aircraft portfolio that was accounted for by Canadair Regional Jets as of June 1996, rather than to the volume of business for 1995.
Arguments regarding whether Brazil has presented a prima facie case against EDC financing.
In Canada's view, Brazil has provided no evidence for the allegation of EDC loans at concessionary rates, and therefore has not made a prima facie case. According to Canada, Brazil’s allegations about EDC financing are without foundation and false, and Brazil has adduced no evidence to support its allegation that the EDC offers financing at “concessionary rates” (para.6.67 -
6.73 ); Canada asserts as well that the evidence that it has adduced to support its general claim that the EDC gives a benefit to Canadian companies does not stand up to scrutiny. (paras. 6.17- 6.26.)
For Canada, Brazil's reliance on a quote by then-EDC president Paul Labbé in support of its argument that “’EDC’s help – to the tune of a ‘few basis points’ – must be better than that which would otherwise be commercially available, or an EDC financing package would not, in EDC’s former President’s words, ‘give Canadian exporters an edge‘” (para. 6.58) is misplaced. Canada argues that earlier in the same article Mr. Labbé refers to EDC's expertise in “complex and tightly negotiated financing arrangements” and from their awareness that a “few basis points in interest rates can make or break the deal”. For Canada, the edge that EDC’s financing support gives Canadian exporters derives from this knowledge and expertise. Thus, Canada argues, Mr. Labbé's statement shows that EDC recognises the factors that are important in making a deal in the international market; it is not, however, a prima facie case that EDC’s finances below market.
In Canada’s view, to raise a prima facie case that EDC financing is a subsidy, Brazil must demonstrate that there is a “financial contribution” within the meaning of Article 1.1(a) that imposes a net cost on the government making the contribution and that confers an advantage above and beyond the market. (paras. 6.76)
Canada notes Brazil's disagreement with this interpretation, and states that Brazil’s interpretation of Article 1 appears to be that “’Article 1.1 therefore states that a subsidy exists where a government contributes something, and in so doing gives an advantage.‘” In Canada’s view, Brazil seems to suggest that Canada has the burden of presenting a prima facie case when there is no prima facie case raised by the complainant, as reflected in Brazil’s statement (paras. ) that:
“Apart from the fact that Canada has not as a factual matter offered any evidence establishing that EDC’s debt financing meets this test, Brazil considers that the plain meaning of Article
1.1 does not support Canada’s test.”
WT/DS70/R
Page 81
For Canada, whether Brazil agrees with “’Canada’s test’” is immaterial to the question of which party bears the burden of raising a prima facie case under Article 1. Canada asserts that it is the complainant Brazil, not Canada, and that it is not for Canada, “’as a factual matter’”, to offer evidence that EDC’s financing meets the test of Article 1 until Brazil has raised a prima facie case. According to Canada, regardless of whether the Panel adopts the Article 1 interpretation proposed by Canada or Brazil, Brazil has not raised a prima facie case.
In answer to a Panel question to Canada regarding what evidence a complaining party would need to adduce to establish a prima facie case of subsidization in respect of a credit at an interest rate below prevailing commercial rates, Canada provided the following examples of evidence that would constitute, in Canada’s view, prima facie evidence of subsidization in aircraft financing:
Statements by airlines (e.g., in annual report or securities filings) citing interest rates on debt resulting from government support that are:
Below that obtainable in the market and
Below the cost of funds of the supporting government (as evidenced by the yield on the government’s bonds at the time of the commitment);
Offering memoranda, investment prospectuses, or other documents published by a reliable source that would be privy to the financing terms of a transaction referring to subsidies provided by a government that reduce the costs (e.g., interest costs) of the aircraft purchaser;
Documents pertaining to a transaction that identify the payment of a subsidy to reduce financing costs; or
Legislation or regulations that mandate the payment of subsidies to reduce interest rates on export financing.
Canada submits that other types of evidence may also form prima facie evidence, and reasonable inferences may be drawn from each piece of evidence to establish a prima facie case in respect of the broader application of the programme or measure in question.
Response of Brazil
Regarding Canada’s response to the Panel’s request for information on EDC financing allegedly provided to ASA, Brazil argues that Canada has refused to provide the documents requested by the Panel. As a result, the Panel should adopt adverse inferences, presuming that the information withheld is prejudicial to Canada’s position (paras. 4.146-4.151).
Regarding the evidence submitted by Canada in support of its statement that EDC always lends above its cost of funds (para. 6.78), Brazil submits that Canada’s reply is not responsive to the Panel’s question. In Brazil’s view Canada does not, as requested by the Panel, provide documentary support for this statement, with respect to the aircraft sector. Rather, Brazil notes, Canada provides an EDC Standing Board Resolution covering all EDC activities. Therefore, in Brazil’s view, the Resolution identifying the general goal leaves unanswered whether “EDC’s lending to the civil aircraft sector” is above EDC’s cost of funds. Brazil argues that the Resolution also makes it clear that EDC’s general goal of covering costs may be overridden in particular circumstances, which could include EDC’s activities in the civil aircraft sector.
Brazil disagrees with what it characterizes as Canada’s “test” regarding whether EDC financing constitutes a subsidy, according to which a complainant must establish two factors: first,
WT/DS70/R
Page 82
that in lending money a government fails to recover its costs; and second, that the terms extended by the government are better than those available on the market. Brazil states that apart from the fact that Canada has not as a factual matter offered any evidence establishing that EDC’s debt financing meets this test, Brazil recalls that it considers that the plain meaning of Article 1.1 does not support Canada’s proposed test (para. 5.43)
For Brazil, Canada’s reliance on item (k) of Annex I to the SCM Agreement as support for its proposed test is misplaced, as Annex I does not speak to whether government activity constitutes a subsidy, but rather to whether government activity constitutes a prohibited export subsidy. Brazil argues that a measure may constitute a subsidy, but not be on the Illustrative List of Export Subsidies included in Annex I, and notes that Article 1 of the Agreement, entitled “Definition of a Subsidy,” contains no relevant reference to Annex I (para. 5.44). Rather, Brazil states, the relevant reference to the Illustrative List of Export Subsidies included in Annex I falls in Article 3, which specifies, among those activities already identified as subsidies, those that are prohibited by the SCM Agreement. For Brazil, the Canadian definition must therefore be rejected.
Brazil recalls its argument that Canada’s proposed test ignores the ordinary meaning and structure of Article 1.1. For Brazil, Article 1.1(a) indicates that to have a subsidy, the government must be the source of the financial contribution; Article 1.1(b), in turn, states that these government financial contributions are considered subsidies if, in giving them, “a benefit is thereby conferred.” Recalling its position that the verb “confer” means to “grant,” “bestow,” “give” or “endow”221 and the noun “benefit” means “advantage” or “something that guards, aids, or promotes well-being”222, Brazil submits that read in its entirety, Article 1.1 therefore states that a subsidy exists where a government contributes something, and in so doing gives an advantage (para.5.40).
It is not evident to Brazil what specific part of Article 1.1 imposes the requirement that a complainant demonstrate a “’net cost’” (para. 6.77) to the government by virtue of a financial contribution. Brazil believes that the ordinary meaning of Article 1.1(b) states that a subsidy exists where a government contribution gives an advantage. In Brazil’s view, lending above cost does not demonstrate that no advantage is given; governments may lend above their cost and give an advantage to a recipient relative to the terms a borrower could receive elsewhere.
Loan guarantees
Arguments of Brazil
Brazil argues that EDC supplements its direct financing by offering long-term loan guarantees to purchasers or lessors of Canadian regional aircraft. Brazil quotes a Canadian regional jet (“CRJ”) purchaser as stating that “’Comair expects to finance the aircraft . . . through a combination of working capital and lease, equity and debt financing, utilizing manufacturers’ assistance and government guarantees to the extent possible.’”223 Brazil cites a news report that EDC has provided “’up to $45 million in new loan guarantees’” for recent CRJ purchases which Canadian government officials reportedly stated would allow Bombardier “’to shave as much as 20 per cent off the selling price of a Canadair jet.’”224 For Brazil, this represents a substantial benefit, which has the effect of
221 Webster’s Third New International Dictionary of the English Language.
222 Id.
223 Comair Holdings Inc. Form 10-K, filed with the US Securities and Exchange Commission for the
fiscal year ending 31 March 1998, p. 27 (emphasis added) (Exh. BRA-14); Comair Holdings Inc. Form 10-K, filed with the US Securities and Exchange Commission for the fiscal year ending 31 March 1997, pgs. 18, 39-40 (Exh. BRA-15).
224 “Bombardier first to gain from equity investments by EDC,” Financial Post, 5 April 1995 (Exh. BRA-16).
WT/DS70/R
Page 83
increasing the amount of EDC direct financing well beyond 85 per cent, and which may be passed along to purchasers or lessors.
In response to questions from the Panel, Brazil clarifies that its claim regarding EDC loan guarantees covers such guarantees per se as de jure export subsidies, and thus in all instances.
Arguments of Canada
Canada rejects as incorrect Brazil’s assertions, arguing that Brazil implies that the very fact of loan guarantees indicates the existence of a subsidy, using an example of an airline that “’expect[ed]’” to finance its aircraft through a combination of instruments, including government guarantees, “’to the extent possible’”. Canada maintains that Brazil has failed to provide any evidence that the EDC has provided subsidies in the form of loan guarantees for any transaction concerning the civil aircraft sector, arguing that Brazil’s “example” is not evidence but innuendo, and that Brazil thus has not met the burden of adducing a prima facie case.
According to Canada, the loan guarantee referred to in the article cited by Brazil was in fact a guarantee by the federal government, through Industry Canada, to assist in the financing of the sales of CRJs to Air Canada, not an EDC loan guarantee. In addition, Canada states, this transaction was a domestic sale. In any event, Canada argues, press reports quoting unnamed officials which are uncorroborated or do not otherwise contain material with an independent title of credibility and persuasiveness simply do not constitute a prima facie case. Contrary to Brazil’s allegation, Canada argues, the EDC does not use any loan guarantees to supplement its financing activity, and Brazil’s argument in this respect must therefore fail.
Canada asserts that the EDC issues conventional financial guarantees of third party payment obligations, in which cases the financial guarantor is supporting the third party’s credit and will require that upon payment it will have recourse to the third party, either directly or by subrogation. Canada maintains that EDC loan guarantees are paid for by commercial fees from the recipient of the guarantee, and that there is no subsidy in this.
In response to a Panel question seeking clarification of Canada’s statement that it does not use any loan guarantees to supplement its financing activity, in light of other statements suggesting that EDC does undertake loan guarantees, Canada responds that these statements are consistent. That is, Canada states, EDC provides loan guarantees on its corporate account, and EDC provides loan guarantees on the Canada Account, but EDC has not supplemented its corporate account financing (i.e., direct lending) with loan guarantees on the same transaction, which Canada states was Brazil’s allegation.
In response to a Panel request for the details of any loan guarantees issued by EDC for transactions concerning the civil aircraft sector since 1 January 1995, Canada states its position that Brazil has not made a prima facie case against EDC loan guarantees, and notes the lack of adequate procedures to protect business confidential information. In this context, Canada provides to the Panel non-business confidential details of the two loan guarantees for export transactions provided by the EDC in the civil aircraft sector since 1 January 1995. The first of these was in support of the sale of two used de Havilland Twin Otters and two used de Havilland Dash 8-102s to an airline operating in the South Pacific. Canada states that the guarantee was provided at commercial rates. The second was in support of pre-shipment financing of a flight inspection system sold to a sovereign Latin American buyer, again at commercial rates.225
225 The Panel notes that Brazil, in its comments on this response of Canada to the Panel’s question, stated that Canada had not provided documents that in Brazil’s view Canada indicated were being provided. The Panel understands Canada’s reply to the question, recounted above, to constitute the entirety of the
WT/DS70/R
Page 84
Equity infusions into corporations established to facilitate the export of civil aircraft
Arguments of Brazil
Brazil argues that because of the high cost of maintaining direct financing at concessionary rates, and the toll such financing takes on its resources, EDC has developed alternative funding schemes.226 Brazil cites to a 4 April 1995 EDC press release regarding the formation, as one of “’several strategic initiatives aimed at further improving service and support to Canadian exporters’” of a new subsidiary -- Structured Finance Inc. (“SFI”) -- which would “’increase the capital available to Canadian exporters by leveraging private-sector investment and structuring risk-sharing partnerships.’”227 According to Brazil, the statement indicates that “’SFI will be the vehicle through which several special purpose companies, capitalized jointly by EDC and private-sector partners, are envisaged to support key export sectors . . . .,’”228 among the first of which was Canadair RJ Capital Inc. (“CRJ Capital”), which “’will be formed on a risk-shared basis primarily between EDC and Bombardier, with each partner providing equal amounts of equity.’”229 Brazil notes that SFI’s name subsequently was changed to Exinvest Inc. (“Exinvest”).
Brazil argues that according to the 1995 Message of the Chairman and President of EDC, Exinvest permits EDC “’to support its customers with a unique combination of debt, equity and insurance instruments.’”230 Brazil argues that this Message announced the realization of the hopes expressed in the April 1995 statement by confirming that “’[Exinvest’s] first investment is Canadair RJ Capital Inc., a 50-50 joint venture with Bombardier used specifically to finance the export of regional jets.’”231
Brazil asserts that although aircraft may initially be sold to airline operators, they are most frequently then immediately re-sold to corporations, referred to in the industry as Special Purpose Companies or “SPCs,”232 which are specifically established to own and lease aircraft. According to Brazil, the SPCs then lease the aircraft back to the airlines, a structure that is favoured for a variety of reasons, including protection against default and, in some jurisdictions, the realization of certain tax benefits. According to Brazil, these benefits result in reduced costs which may be passed along to enable the airline to achieve the lowest payments possible for the aircraft.
Brazil argues that CRJ Capital facilitates 233 export transactions by providing equity capital to the SPC, which supplements debt capital obtained from lenders, and asserts that in consultations, Canada declined to disclose to Brazil any pertinent information concerning the full scope of CRJ Capital’s operations, asserting that the information was confidential. Brazil states that it nevertheless has been able to determine from publicly-available information that the SPC will receive 20 per cent of its capital in the form of equity from CRJ Capital and the remaining 80 per cent from lenders.234 Thus, according to Brazil, the only charge on the SPC’s total capital is the charge required to service
information referred to by Canada in that reply, and in this regard notes that Canada’s reply makes no reference to documents.
226 Committee on Foreign Affairs and International Trade, 43:10 (Exh. BRA-9).
227 EDC Press Release No. 8, 4 April 1995 (Exh. BRA-17).
228 Id.
229 Id.
230 Message at 3 (Exh. BRA-7).
231 Id.
232 Transport Finance No. 74, dated 3 October 1996, at 3 (Exh. BRA-18).
233 The Panel notes that in its second submission, Brazil indicates that the references to CRJ Capital reflected in this and the following paragraph should have been to Exinvest, the provider of equity capital to CRJ
Capital, the SPC. The references as initially made by Brazil are retained in the text of this report, to ensure that Canada’s response arguments in the next section are comprehensible.
234 Id. at 3 (EDC chart of CRJ financing scheme).
WT/DS70/R
Page 85
the debt, since equity owners are paid only from profits. Moreover, Brazil argues, EDC typically guarantees a portion of the debt capital supplied by the lenders 235 (see section VI.A.3.)
Brazil states that after initially purchasing the aircraft from the manufacturer, the airline re- sells the aircraft to the SPC, which leases it, typically for periods of 15 or more years, back to the airline. According to Brazil, the stream of lease payments from the airline to the SPC is used to service only the debt obligations of the SPC; since the object of utilizing the SPC device is to minimize the cost to the customer, the stream of payments is not sufficient to provide a return to the equity investors in the SPC. Thus, Brazil asserts, at the end of the lease the SPC owns a 15-year-old used airplane, the residual value of which is the sole source of whatever return will accrue to CRJ Capital, and in turn to EDC, as a supplier of equity capital.
For Brazil, this residual value alone cannot provide a commercial return to EDC as an equity investor; a 30-per cent residual value, for example, would not begin to compensate a non-subsidizing equity investor. Brazil asserts that the EDC contribution of 10 per cent of total capital236 represents half of the equity capital in the SPCs established to support the export of civilian aircraft, which means that EDC’s share of the assumed 30-per cent residual value is half of that, or 15 per cent. According to Brazil, this would yield an annual return of a mere 2.74 per cent, which for Brazil demonstrates quite clearly and quite graphically that EDC equity infusions are little more than outright grants.237 Brazil further states that even a 50-per cent residual value produces an annual return of only 6.3 per cent, hardly the level of return sought by risk-taking equity investors.238
In response to a question from the Panel seeking Brazil’s view on what an appropriate level of return would be for such “risk-taking equity investors”, Brazil notes that in Parliamentary testimony, then-EDC President Paul Labbé identified 15 to 20 per cent as the “’return on equity that would be required to survive in the private sector‘”239, and refers to an article from Airfinance Journal240 that cites after-tax yields on private aircraft leasing deals of between seven and 14 per cent (or 15 to 20 per cent on a pre-tax basis). Brazil also submits a chart241 providing annual US returns on a range of low- risk to high-risk investments over the period 1991-1997, which Brazil states shows similar returns for investments involving any even nominal level of risk. Finally, according to Brazil, this rate accords with another appropriate benchmark – Bombardier’s pre-tax cost of capital, which factors at 16.9 per cent.242
Brazil notes, however, that a return on EDC’s equity investment of 6.3 per cent is highly speculative, presuming a 50-per cent residual value for the underlying aircraft, and states that a residual value of 30 per cent is more in line with the value projected by industry analysts,243 and in the case of a 16-year lease would yield a 2.57 per cent rate of return,244 a rate which fails in all but three of the last 20 years to outpace the US inflation rate.245 EDC’s acceptance of this rate of return for its
235 Id.
236 Since the SPC receives 20 per cent of its capital in the form of equity from CRJ Capital, and CRJ Capital is itself 50-per cent owned by EDC, EDC's equity contribution to the SPC amounts to 10 per cent.
237 See Exh. BRA-19.
238 Id.
239 Committee on Banking, Trade and Commerce, pg. 48:22 (Exh. BRA-10).
240 Exh. BRA-74
241 Exh. BRA-75
242 JP Morgan Securities, “Bombardier,” 1 September 1998, pg. 10 (adjusting 11 per cent to pre-tax cost of capital) (Exh. BRA-76).
243 “Consider Residual Values When Computing RJ Costs, Say Analysts,” Commuter/Regional Airline News, 7 April 1997 (Exh. BRA-77).
244 "Evaluation of Export Subsidies Received by Bombardier Inc. and Certain Suppliers", Expert
Report of Horst, Frisch, Clowery and Finan, 29 October 1998 ("Finan Report") (included as a separate attachment to the first written submission of Brazil), at Exhibit B, Table B.4.
245 Stocks, Bonds, Bills and Inflation, 1998 Yearbook (Exh. BRA-78).
WT/DS70/R
Page 86
equity financing vehicle serves a purpose; according to Industry Canada’s Director of Airframe Systems, Richard Dixon, it helps meet “[t]he challenge . . . to provide low cost financing to airlines with lower credit ratings,” and permits “airlines with double-B credit ratings . . . to lease new planes at interest rates normally offered to double-A credit risks.”246
Thus, Brazil argues, EDC’s equity infusions into SPCs confer a benefit by lowering the lease payment required from the ultimate customer, the airline, thereby effectively lowering the price of the aircraft and making it more competitive than would otherwise be the case.
Brazil notes that the foregoing analysis assumes a 15-year lease, and that to the extent a lease exceeds 15 years, the residual value of the aircraft would be less because it is older, which would in turn affect the equity investor's ultimate return. That is, according to Brazil, the equity investors’ return would be less because its pay-back would occur at a later date; for example, a 16-year lease, with an anticipated 30-per cent residual value yields an annual return of 2.57 per cent.247
Brazil states that in consultations, Canada declined to disclose specific information concerning the number of regional aircraft sales benefiting from EDC’s equity financing vehicle. According to Brazil, EDC officials have stated that CRJ Capital “could be involved in purchasing and leasing as many as 75 jets.”248
According to Brazil, publicly-available information indicates that CRJ Capital has been used to finance sales both to Comair and Air Canada.249 Brazil asserts, citing to the expert report of Grey, Clark, Shih and Associates, Ltd. (the “Clark Report”) that the sale to Air Canada, which at first impression would seem to be a domestic sale, is in fact an export sale entered into with the explicit aim of promoting export sales.250 According to Brazil, when questioned by a Committee of Parliament concerning EDC support for sales of civilian aircraft to Air Canada -- seemingly a domestic sale and not, therefore, within EDC’s purview -- EDC President Paul Labbé explained that these in fact were export sales, having been made through an SPC (a “tax vehicle”) established in the United States,251 which in turn leased the aircraft to Air Canada, thereby qualifying for treatment as an export transaction within EDC’s authority. Brazil quotes Mr. Labbé as saying:
Our focus is export financing. What we’re trying to do in this particular case -- this is an exceptional case -- is launch an aircraft that has a world market. The Canadian market for this aircraft is minimal compared with what this world market is going to be, but it has to be launched. You have to get a good customer base established so that you have a good market for that thing.252
According to Brazil, EDC’s equity infusion into the SPC amounts to approximately 10 per cent of the value of each aircraft, or approximately $1,500,000 to $2,000,000 for every airplane sold. Brazil argues that the present value in 1998 of benefits to Bombardier associated with these equity infusions ranges from a low estimate of US$894 million to a high estimate of US$979 million.253
Brazil argues that SPCs are designed not to earn a profit during the term of the lease; therefore, only the debt portion of their capital needs to be serviced during the 15-year or 15 year-plus
BRA-8). 8).
246 “Ottawa Joins Bombardier in jet-leasing venture,” The Globe and Mail, 5 April 1995 (Exh.
247 Finan Report at Exhibit B, Table B.4.
248 “Ottawa Joins Bombardier in jet-leasing venture,” The Globe and Mail, 5 April 1995 (Exh. BRA-
249 “Closing the Stable Door,” Airfinance Journal, July/August 1995, p. 25 (Exh. BRA- 20).
250 Clark Report at pgs. 13, 21-24.
251 Committee on Foreign Affairs and International Trade, 43:30 (Exh. BRA- 9).
252 Id.
253 Exhibit B, Table B.5 of Finan Report.
WT/DS70/R
Page 87
period of the lease, and no payment is made to the equity investors. Thus, for Brazil, the greater the percentage of equity capital in the SPC, the lower the percentage of debt capital that actually must be serviced by the lease payments. In effect, Brazil argues, when EDC utilizes the SPC mode of support, it purchases 10 per cent of a new aircraft, with no prospect of return on this investment during the entire period of the lease, in exchange for a 50-per cent interest, 15 or more years later, in a 15-year- old or more used aircraft.
Brazil states that this equity infusion is a benefit because no private financial institution or investor would contribute equity on such unremunerative terms; and argues that were non-subsidizing private investors willing to provide funds on such terms, there would be no need for EDC, since its mission is to go “beyond what is possible by other financial intermediaries” by, in this instance, providing equity infusions to feed “the seemingly endless appetite of Canadian exporters for financial support.”254 According to Brazil, these equity infusions go far beyond “what is possible by other financial intermediaries,” and in so doing, confer benefits of the type envisioned by Article 1.1 of the SCM Agreement. Thus, for Brazil, EDC, through CRJ Capital, therefore confers a clear benefit on Canadian producers and exporters of regional aircraft by assuming an equity investor’s risk in its contribution of capital, with no prospect of obtaining an equity investor’s reward.
In addition, Brazil states that using Canada’s words, EDC incurs a “net cost” in that no private financial institution or investor would contribute equity on such unremunerative terms, as evidenced by the apparent fact that no one other than Bombardier, the final beneficiary, has done so. Brazil submits that EDC’s only return depends upon the residual value of a 15-year-old (or older) airplane, which the Finan Report255 and Exhibit Bra-19 establish cannot provide anything remotely resembling a commercial return to EDC as an equity investor.
Brazil further argues that because all EDC funding is de jure contingent upon export, all sales using the equity financing vehicle, including those sales to Air Canada, have been structured as export transactions, and that as a result, extension of this type of funding constitutes an export subsidy within the meaning of Article 3 of the Agreement.
Brazil states that it is aware of one other apparent SPC, Canadian Regional Aircraft Finance Transaction, or “CRAFT,” which may be a successor to or replacement of sorts for Exinvest. According to Brazil, CRAFT was launched in April 1998 as an aircraft securitization structure to provide lease and loan financing for customers buying Bombardier’s CRJ and Dash 8 aircraft.256 Brazil, referring to a chart of CRAFT’s structure and operations that it submits257, states its understanding that EDC and the Government of Québec provide equity capital and are preferred shareholders in CRAFT.
Brazil indicates that CRAFT came to the market with an initial portfolio of 26 aircraft and a total appraised value of $458 million, which can be expanded over the next 18 months to reach 70 aircraft worth $1.1 billion.258 In Brazil’s view, CRAFT diversifies the risk to debt holders by packaging together deals and offering portions of the total package to the debt holders, thereby lowering the cost of financing. Brazil asserts that EDC and the Government of Québec provided preferred capital to the CRAFT structure. Brazil states that it is estimated by Standard & Poor’s that
254 Message at 2, 4 (Exh. BRA- 7).
255 See Exhibit B, Table B.4 of Finan Report.
256 Standard & Poor’s Presale Report, “CRAFT No. 1 Trust 1998 – A,” 17 April 1998 (Exh. BRA-70); “Turning finance into a fine craft,” Airfinance Journal, June 1998 (Exh. BRA-71); “CRAFT takes off for
Bombardier,” Euroweek, 8 May 1998 (Exh. BRA-72).
257 Exh. BRA-73
258 Id.
WT/DS70/R
Page 88
EDC and the Government of Québec’s combined participation could exceed US$300 million.259 Brazil presumes that investment by the Canadian federal and provincial governments was required because the return to the other shareholders would be below the return demanded by private investors given the level of risk they would have to assume in the financing structure. In Brazil’s view, by accepting below market returns on their investment, EDC and the Québec Government allow the CRAFT SPC to reduce the monthly lease cost to the aircraft lessee.
Brazil notes that it has encountered significant difficulty obtaining access to information on EDC’s equity financing activities, as it has for other EDC activities. Brazil believes that in its defense of EDC’s equity activities (see paras. 6.128-6.129), Canada is relying on sophistic and misleading parsing of Brazil’s claim (as in Brazil’s view it did during the Panel meeting on 26 November with regard to Brazil’s claims concerning EDC’s extension of loan guarantees). Brazil recalls here Canada’s emphasis on the word “supplement” in that defence (see para. 6.99). Brazil reiterates that its factual understanding of EDC’s equity financing activities is derived directly from public information issued by EDC itself. For Brazil, Canada’s mere conclusory denial of facts which it has often and repeatedly held out to be true in public should not be considered by the Panel to be a sufficient rebuttal of Brazil’s prima facie case.
Arguments of Canada
Canada argues that CRJ Capital, Inc. (“CRJCI”) does not do what Brazil argues it does, i.e., that Brazil’s submission is based on a misunderstanding of the operation of the CRJCI. Canada further submits that Brazil’s allegation also casts in bold relief the difficulties that could arise as a result of a failure to conform to Article 6.2 of the DSU in fashioning a request for a panel (para. 4.38
- 4.58).
Canada states that its understanding of Brazil’s request for a panel has been that Brazil wished to challenge the EDC’s part ownership of the CRJCI, and that it was only upon receiving Brazil’s first submission that Canada understood that by “’corporations established to facilitate the export of civil aircraft‘” Brazil did not mean the CRJCI, but some “’special purpose corporations‘” that have never been the subject of discussion between Canada and Brazil.
For Canada, the short answer to Brazil’s claim (and the answer that Brazil would have received had the claim been put to Canada over the previous two years) is that neither the EDC nor the CRJCI have invested equity in the “special purpose corporations” mentioned by Brazil.
Canada states that the CRJCI was created as a result of a joint investment by Exinvest, a wholly-owned subsidiary of the EDC and Bombardier Inc. Each shareholder owns 50 per cent of CRJCI. Canada indicates that to date, in addition to the initial equity investment, the two shareholders have in equal proportion provided interest-bearing loans to CRJCI at commercial rates. According to Canada, CRJCI offers financing services for the CRJ through either direct lending at commercial rates or loan guarantees (paid for by fees from the borrower). To reduce the EDC’s risk exposure, Bombardier’s investment is in a first loss position, i.e., that in the event of a loss, Bombardier takes the loss first. Canada asserts that the only financing transactions that the CRJCI has participated in were for the sale of 17 CRJs to Air Canada. Canada indicates that these are domestic transactions and that therefore the financing provided did not constitute export credits, and that this is the only activity undertaken by CRJCI to date.
259 Standard & Poor’s Presale Report, “CRAFT No. 1 Trust 1998 – A,” pg. 3 (Exh. BRA-70) (Notes that the debt portion of the CRAFT SPC will be up to $770 million in amortizing notes, with a total portfolio of aircraft at an appraised value of $1.1 billion. The difference, $330 million, has therefore to come from equity financing).
WT/DS70/R
Page 89
Canada states that neither the EDC nor the CRJCI has made any investments in special purpose corporations, and that Brazil’s challenge is therefore without foundation.
Canada argues that Brazil’s allegations with respect to “equity infusions” by CRJ Capital into Special Purpose Companies rest solely on an article in a transportation financing journal (para. 6.104) which Canada asserts is inaccurate, as CRJ Capital has provided no such equity infusions. For Canada, this is a clear example of why press reports that are uncorroborated or do not otherwise contain material with an independent title of credibility and persuasiveness are insufficient to support a prima facie case.260
The Panel asked Canada to comment on a number of exhibits submitted by Brazil in support of the claim of EDC equity infusions: 2611995 EDC Chairman and President’s Message, which indicates, regarding Exinvest, that EDC is able to support its customers with a unique combination of debt, equity and insurance instruments, and states that Exinvest’s first investment is Canadair RJ Capital, a “50-50 joint venture with Bombardier used specifically to finance the export of regional jets”; a newspaper article262 stating that by forming CRJ Capital EDC “is getting into the aircraft- leasing business”, that CRJ Capital “marks the EDC’s first foray into equity financing”, and that “CRJ Capital will buy up to 30 per cent of each new Regional Jet, syndicate the rest to private-sector lenders and lease the aircraft back to the airlines”; a newspaper article263 containing a diagram, attributed to EDC, that shows RJ Capital Corp. providing 20 per cent of the equity to a special purpose company which owns the aircraft and leases it back to the airline; a newspaper article264 referring to CRJ Capital as “a leasing company designed to support sales of the Canadair RJ”, and stating that “CRJC’s initial mandate was to support sales of 75 aircraft”
Canada responded regarding the 1995 EDC Message that EDC has equity powers, but that these have not been used to invest in special purpose corporations. Regarding the first article, Canada states that it is incorrect in that: a) CRJ Capital is not an aircraft leasing company; b) CRJ Capital is not an equity financing vehicle; and c) CRJ Capital does not buy equity positions in aircraft. Regarding the second article, Canada states that the diagram is incorrect. Regarding the third article, Canada states that CRJ Capital is not a leasing company and has only participated in the commercial financing for the sale of 17 CRJs for Air Canada. Canada’s response further indicates that neither the Canadian government nor EDC has provided through CRJ Capital or, by any other means, equity participation in any sales of Canadian produced civilian aircraft. Canada also states that that neither Exinvest, which is 100 per cent owned by EDC, nor CRJ Capital, which is 50 per cent owned by Exinvest, are special purpose corporations. Canada submits that in the civil aircraft sector, a special purpose corporation is generally utilized for the sole purpose of owning a single asset, and is predominantly used in US leveraged leases and other types of tax leveraged leases; in Canada’s view, this appears to be consistent with Brazil’s definition of an SPC.
Canada also disputes as incorrect Brazil’s allegations that CRJ Capital acts as an aircraft leasing company, stating that CRJ Capital has not purchased aircraft and then leased them to airlines. Canada provides an officer’s certificate from CRJCI265, and an officer’s certificate for EDC’s subsidiary, Exinvest, to this effect,266 and notes that the officer’s certificate of CRJ Capital also states that CRJ Capital has not made any such equity investment, and that the officer’s certificate for Exinvest also states that Exinvest has not made any equity investment in any corporation aside from its equity investment in CRJ Capital.
260 Keith Highet, (Exh. CDN-54).
261 Exh. BRA-7
262 Exh. BRA-8 (also included in Exh. BRA- 17)
263 Exh. BRA-18
264 Exh. BRA-20
265 See Officer’s Certificate, CRJCI, 3 December 1998 (Exh. CDN-94).
266 Officer’s Certificate, Exinvest, 10 December 1998 (Exh. CDN-93).
WT/DS70/R
Page 90
Canada also challenges Brazil’s arguments with respect to Canadian Regional Aircraft Finance Transaction No. 1 Limited, (or CRAFT), as without foundation. Canada notes Brazil’s understanding that “EDC and the Government of Québec provide equity capital and are preferred shareholders in CRAFT”, which understanding Brazil appears to attribute to Standard & Poor’s: “It is estimated by Standard & Poor’s that EDC and the Government of Québec’s combined participation could exceed US $300 million.” (para. ) Canada argues that the source referenced by Brazil -- a Standard & Poor’s Presale Report on CRAFT267 - makes no reference to participation by the EDC or the Government of Québec, and that there was no reference in Exhibits BRA-71 or 72, either. Canada submits that the figure set out in Exhibit BRA-73, however, has a prominent, bolded box listing EDC and the Government of Québec as preferred shareholders in CRAFT and that the source of this information, according to that figure, is Embraer. Canada maintains that CRAFT is a special purpose, independent company without any ties to the Government of Canada or the Government of Québec. Canada submits a letter from the Director of CRAFT Ltd. attesting to the fact that neither the Government of Canada, nor the Government of Québec, nor any agency of the Government of Canada (in particular EDC) nor any agency of the Government of Québec, have participated in the financing of CRAFT.268
Response of Brazil
Brazil contests Canada’s denial of the facts underlying its equity activities. Although Canada claims that CRJ Capital does not lease aircraft, but rather helps finance aircraft, and that it is not an SPC, Brazil argues that public statements from EDC itself state flatly that CRJ Capital is both a leasing company and an SPC.
In this regard, Brazil cites an EDC announcement of the creation of “’Structured Finance, Inc.,’” later called “’Exinvest,’” which states that Exinvest is to be “’the vehicle through which several special purpose companies,’” or SPCs, were to be created, the first of which was CRJ Capital.269 Brazil notes that the announcement states that CRJ Capital receives equity infusions from EDC, through Exinvest. Brazil also cites EDC’s 1995 Message of the Chairman and President of EDC270 which states that Exinvest’s first investment was CRJ Capital, as well as a quote from a former EDC President271 that CRJ Capital was to be the model for other SPCs. For Brazil, the obvious implication of this is that CRJ Capital is an SPC itself.
Brazil also cites a quote from an EDC Vice-President272 characterizing CRJ Capital as a leasing company, and stating that the plan is for it to be used in the lease or sale of as many as 75 CRJs. Brazil asserts that in interviews273, EDC officials, including EDC’s president, stated that CRJ Capital purchases part of a plane, syndicates the rest to private-sector lenders, and leases the aircraft to an airline.
Brazil asserts that its factual understanding of EDC’s equity financing activities is, therefore, derived directly from information issued by EDC itself, and not, as Canada argues, based “‘solely on an article in a transportation financing journal.’” (para. 6.126) For Brazil, Canada’s denial of facts which EDC and its officials have often and repeatedly held out to be true in public should not be considered credible.
267 Exh. BRA-70
268 Letter from Timothy R. Myers, Director, Canadian Regional Aircraft Finance Transaction No. 1 Limited to Robert Greenhill, Bombardier Inc., dated 10 December 1998; letter from Charles Hyatt, Director, Greenwich NatWest Limited to Richard Sloan, Bombardier, dated 10 December 1998 (Exh. CDN-91).
269 Exh. BRA-17
270 Exh. BRA-7
271 Exh. BRA-8
272 Exh. BRA-20
273 Exh. BRA-8
WT/DS70/R
Page 91
In Brazil’s view, Canada’s arguments simply fail to respond to Brazil’s claims. Canada states, in response to a question from the Panel, that EDC, Exinvest and CRJ Capital have not “provided . . . equity participation in . . . sales of Canadian produced civilian aircraft.” According to Brazil, this is not, and has never been, Brazil’s claim. Rather, Brazil’s claim is that EDC, directly or indirectly, has made equity infusions into CRJ Capital which have facilitated CRJ Capital’s ability to lease or sell Canadian regional aircraft at a reduced price.
Brazil, it is irrelevant whether or not CRJ Capital leases, or helps finance, aircraft, and whether or not CRJ Capital is, or is not, an SPC. In Brazil’s view, the distinction is not relevant to the question whether EDC, through its direct or indirect equity investments in CRJ Capital, provides a “benefit.” The point is that EDC’s direct or indirect equity investment in CRJ Capital frees CRJ Capital up to accept a lower lease or loan payment from a lessor or purchaser of a Canadian regional aircraft than it would be able to accept in the absence of that equity investment, or to facilitate the ability of another SPC to do so. CRJ Capital is designed not to earn a profit during the term of the lease; therefore, only the debt portion of the capital used to finance the lease needs to be serviced during this period. According to Brazil, no payment is made to equity investors. Thus, the greater the percentage of equity capital in CRJ Capital, the lower the percentage of debt capital that must be serviced. The benefit, in short, is lower lease or debt payments for the airlines. As support, Brazil cites an Industry Canada official’s statement that EDC’s equity vehicle meets “[t]he challenge . . . to provide low cost financing to airlines with lower credit ratings,” by permitting airlines with double-B credit ratings to lease new planes at interest rates normally offered to double-A credit risks.
Brazil argues that all of the information submitted by it regarding CRJ Capital is based upon undisputed and undenied public statements by EDC officials and public announcements issued by EDC itself. The statements regarding the benefits of EDC’s equity vehicle remain unanswered “‘statements against interest,’”274 to which Brazil argues the Panel should attach “‘superior credibility.’” For Brazil, these statements illustrate both the nature and the degree of the “’benefit’” provided by EDC’s equity investments, and demonstrate why those efforts constitute a subsidy under Article 1.1 of the Agreement.
Residual value guarantees
Arguments of Brazil
Brazil asserts that at the end of an aircraft lease, in certain instances EDC also offers a “residual value guarantee,” protecting against the risk that the residual value of the used aircraft will be lower than anticipated, with the guarantee further reducing the cost of financing.275 According to Brazil, such guarantees generate savings due to the reduced risk of having to absorb a loss from a lower-than-expected residual value, which savings can be passed along to the airline customer in the form of lower lease payments, essentially increasing the amount of direct financing beyond 85 per cent.
Brazil argues that although a higher residual value would normally mean a higher return for the SPC’s equity investors, that benefit does not accrue to EDC, since EDC underwrites the cost of maintaining the higher residual value in the first place. According to Brazil, an inflated residual value does enhance Bombardier’s return on its equity investment in CRJ Capital, and the residual value guarantee also benefits the airline customer -- the SPC is relieved of the burden of absorbing a loss from a lower-than-expected residual value, and can pass any savings along to the airline customer in the form of lower lease payments. According to Brazil, residual value guarantees thereby confer a clear benefit under Article 1.1 of the SCM Agreement.
274 Exh. CDN-54
275 “A Very Tactical Regional Response,” Airfinance Journal, November 1994, at pgs. 18-20 (Exh. BRA- 13).
WT/DS70/R
Page 92
In response to Panel questions regarding the coverage of Brazil claim in respect of residual value guarantees, Brazil states that its claim includes the grant by EDC of residual value guarantees per se as de jure export subsidies, and thus in all instances.
Arguments of Canada
According to Canada, a residual value guarantee, which may take the form of an insurance policy, will guarantee to the lessor that, if the equipment is returned at the location and in the condition required by the lease or the guarantee, the equipment will be worth not less than a stated sum. Canada states that the residual value guarantor’s sole recourse is to the equipment and its risk is entirely dependent on equipment value, not the credit of another person.276
Canada denies Brazil’s allegations regarding residual value guarantees, arguing that Brazil’s “’evidence‘” with respect to “’residual value guarantees‘” is based on an article that notes a “’suggestion‘” that a deal completed in 1992 “’may have‘” involved a residual value guarantee. The suggestion of a possibility in respect of a deal made prior to the entry into force of the WTO Agreement does not, in Canada’s submission, amount to a prima facie case. In Canada’s view, this is an example of a press report which is uncorroborated or does not otherwise contain material with an independent title of credibility and persuasiveness. Canada argues that Neither the EDC nor the Government of Canada has provided residual value guarantees through CRJ Capital277 or any other means in support of civil aircraft. Canada provides, in support of its argument, an officer’s certificate from CRJCI 278 and an officer’s certificate for EDC’s subsidiary, Exinvest, to this effect.279
Response of Brazil
Brazil argues that although Canada has submitted “officer’s certificates” claiming that CRJ Capital and Exinvest have not issued a residual value guarantee, no such document has been submitted with regard to EDC itself. Brazil quotes from Exhibit CDN-54 to argue that Canada’s “‘failure to deny news stories published on the subject’” (paras. 5.18-5.21) – such as, in Brazil’s view, the statements in an article included in Exhibit BRA-13 regarding EDC’s provision of residual loan guarantees – constitutes a valid reason to infer that Canada’s failure to include a similar certificate for EDC itself constitutes an admission of the practice.
Brazil believes that the circumstances of this case make this rule all the more applicable. Brazil states that when questioned by the Panel regarding its statements that “‘EDC does not use any loan guarantees to supplement its financing activities,’” Canada defended itself by stating that it does not consider EDC loan guarantees to supplement EDC’s other financing activities (para. 6.99). In light of Canada’s tendency, in Brazil’s view, to employ semantics to reveal less than the full and complete truth about EDC’s operations, Brazil believes that the Panel should find the omission of a sworn statement regarding EDC’s issuance of residual value guarantees to constitute an admission of the practice. Brazil states that it stands by its claim, therefore, that EDC has issued a residual loan guarantee on one or more occasions.
276 Milbank, Tweed, Hadley and McCloy, Re Residual Value Guarantees, 17 November 1998 (Exh. CDN-59); also, H. Ruda, Asset-Based Financing: A Transactional Guide, Vol. 4 (New York: Matthew Bender, 1998) at 38-20 to 38-21 (Exh. CDN-60).
277 Officer’s Certificate, CRJCI, 3 December 1998 (Exh. CDN-58).
278 Officer’s Certificate, CRJCI, 3 December 1998 (Exh. CDN-58).
279 Officer’s Certificate, Exinvest, 10 December 1998 (Exh. CDN-93)
WT/DS70/R
Page 93
Canada Account
Arguments of Brazil
Brazil asserts that the operations of the Canada Account are marked by an extreme degree of secrecy. Brazil states that EDC’s 1996 and 1997 Annual Reports omit any meaningful description of the Canada Account, and that the two most recent Canada Account reports have not in fact been tabled in the Canadian Parliament,280 and quotes EDC’s 1995 Annual Report as follows regarding the Canada Account:
Canada Account funds are used to support export transactions which the federal government deems to be in the national interest but which, for reasons of size or risk, the Export Development Corporation (EDC) cannot support through regular export credits. Transactions are negotiated, executed and administered by EDC on behalf of the government, and are accounted for separately on the books of the Department of Foreign Affairs and International Trade (DFAIT). These activities are known collectively as the Canada Account.281
Therefore, according to Brazil, before a transaction can be supported by the Canada Account, it first must be considered and rejected by the EDC: “The Canada Account provides the Government of Canada with the authority and means to support transactions which, on the basis of prudent risk management as defined by the Board of Directors of EDC, cannot be supported by the [Export Development] Corporation.”282 Brazil asserts that the transaction is then referred to the Minister of International Trade who, together with the Minister of Finance, may approve transactions up to Can$50 million,283 contingent upon a finding by the Ministers that the transaction is in the national interest. Brazil argues that among the grounds for finding a transaction to be in the national interest is the “’importance of the transaction to the exporter‘”284. Brazil cites the EDC as indicating that Canada Account financing may be extended at below-market or concessional rates.285
Brazil indicates that there are no publicly-available statements of the level of funding for regional aircraft development and sales provided under the auspices of the Canada Account, and notes that during consultations, Canada stated that details concerning all Canada Account transactions were confidential, and refused to disclose any details of the operations of the Canada Account.
Brazil submits that in response to a March 1996 inquiry from a Member of Parliament concerning assistance provided during the prior 15 years to the Canadian aerospace sector, EDC, which administers the Canada Account, stated that during that period, Bombardier received Can$450 million and US$61.172 million, while de Havilland received Can$131.04 million and US$279.656 million for loans and US$14.95 million for loan guarantees,286 that US$11.0 million was provided for a single project for Canadair, and that a loan to a foreign buyer in the amount of US$14.2 million was provided in support of a joint sale by Bombardier and Canadair.287 Brazil notes
280 EDC’s Summary Reports to Treasury Board on Canada Account Operations for the periods 1994- 95, 1993-94 and 1992-93, respectively, Exhs. BRA-24, 25 and 26.
281 EDC 1995 Annual Report, “Canada Account Profile” (Exh. BRA-27).
282 Summary of the Report to Treasury Board on Canada Account Operations for the Fiscal Year 1994- 1995 by the Export Development Corporation, p. 1 (Exh. BRA-24).
283 EDC 1995 Annual Report, “Canada Account Profile” (Exh. BRA-27). Larger transactions must be approved by the Cabinet.
284 Id. (emphasis added).
285 Id. See also Summary of the Report to Treasury Board on Canada Account Operations for the Fiscal Year 1994-1995 by the Export Development Corporation, p. 3 (Exh. BRA-24).
286 EDC Response to Inquiry of Ministry, Question No. Q-30 by Mr. Schmidt, dated 19 March 1996 (Exh. BRA-28).
287 Id.
WT/DS70/R
Page 94
that the spreadsheets attached to the official response appear to include only those benefits granted through 1991.288
According to Brazil, the official response makes clear that the amounts listed represent only those transactions for which press releases were issued, and do not include transactions which were not the subject of press releases, as such information is considered commercially confidential.289 Brazil argues that because EDC administers the Canada Account, it is possible that these amounts include Canada Account transactions, and that without further disclosure from Canada, the Panel should so presume, and should presume as well that the funding is extended on concessional terms.
Brazil maintains that the major difference between the EDC and the Canada Account is that projects financed through the Canada Account do not even meet EDC’s criteria because of extremely high risk, and therefore require ministerial approval. One exception according to Brazil is that the Canada Account has not been used for equity infusions to support exports of civilian aircraft, but instead has supported exports of civilian aircraft over the years with millions of dollars in grants, low- interest loans, interest-free loans, loan guarantees and similar direct or potentially-direct transfers of funds from the Canadian government. For Brazil, the Canada Account, unlike the EDC, does not even pretend to be a “self-supporting” operation, but is a direct charge to the books of the Department of Foreign Affairs and International Trade.290
Brazil asserts that like the EDC, the Canada Account exists for one purpose only: to support exports -- risky exports, recalling the EDC President’s statement that “’Canada Account funds are used to support export transactions which the federal government deems to be in the national interest but which, for reasons of size or risk, the Export Development Corporation (EDC) cannot support through regular export credits.‘”291 According to Brazil, the Canada Account not only “’absorb[s] the risk on behalf of Canadian exporters beyond what is possible by other financial intermediaries,‘”292 it absorbs risk beyond what is possible from the EDC, and is an integral, vital part of Canada’s effort to satisfy “’the seemingly endless appetite of Canadian exporters for financial support.‘”293 Brazil claims that as an alternative to EDC and the financier of last resort for export transactions, the Canada Account grants funds contingent in law or in fact upon export performance, and is therefore prohibited by Article 3 of the Agreement. Brazil, in response to a Panel question, states that its claim regarding Canada Account is that Canada Account per se is de jure contingent on export, and that Brazil also challenges the Canada Account as applied.
Brazil argues that Canada does not challenge its assertion that Canada Account assistance is contingent on export, within the meaning of Article 3 of the SCM Agreement, and that Canada could not do so, in light of the above language of the EDC’s 1995 Annual Report.
Brazil maintains that the grants, interest-free loans, low-interest loans, guarantees and other give-aways and support supplied through the Canada Account confer a benefit on the exporters of civilian aircraft in Canada by artificially facilitating their sale. For Brazil, in the absence of subsidies, high-risk buyers would pay higher interest rates and would be required to make higher down- payments. According to Brazil, the Canada Account eliminates these requirements, to the benefit of civilian aircraft exporters, and therefore is a subsidy and confers a benefit within the meaning of Article 1.1 of the SCM Agreement.
288 Id.
289 Id.
290 EDC 1995 Annual Report, “Canada Account Profile” (Exh. BRA- 27).
291 Id.
292 Message at 2 (Exh. BRA- 7).
293 Id. at 4.
WT/DS70/R
Page 95
Brazil also argues that the most recent WTO Trade Policy Review of Canada reports that export financing through the Canada Account soared from $100 million in 1996 to $1.6 billion in 1997, an increase of 1600 per cent in a single year.
According to Brazil, EDC’s 1995 Annual Report states that a portion of Canada Account financing is extended on what it refers to as non-concessional terms, while the remainder is extended at admittedly below-market or concessional rates. Brazil notes that a nominally non-concessional rate would amount to a concessional rate if extended to a borrower with a poor credit rating, i.e., that whether a rate is or is not concessional depends in large part upon the risk associated with a particular borrower. Brazil asserts that the Canada Account funds the riskiest of borrowers.
Brazil argues that from the perspective of the recipient, financing extended at rates that are concessional by definition constitutes a clear “benefit” to that recipient within the meaning of Article 1.1. For Brazil, support at concessional rates confers a benefit on the exporters of Canadian regional aircraft by artificially facilitating their sales.
Brazil submits a press article294 which quotes Bombardier’s executive vice-president as stating that Bombardier has used the Canada Account for some transactions, under terms of financing which he describes as “’close to commercial‘” (emphasis added by Brazil). Brazil requests the Panel to ask Canada, in light of this statement, to provide the Panel and Brazil with all of the details regarding all of the support given to the regional aircraft industry in Canada through the Canada Account.
Brazil argues that Canada has refused to reveal the existence, much less the terms, of any support from the Canada Account to the Canadian regional aircraft industry or its customers, while Bombardier has acknowledged using the Canada Account for some transactions on “close to commercial”295 terms. Brazil states that it utilized the services of an expert Canadian economic consulting firm and of a highly-reputable Canadian law firm to obtain information regarding the Canada Account. Brazil asserts that it has produced all the evidence about Canada Account support to the Canadian regional aircraft industry it could locate from public sources, which evidence in Brazil’s opinion establishes that the Canada Account has been used to support the sale of Canadian regional aircraft on non-commercial terms. For Brazil, Canada is obligated as a legal matter under the rule of collaboration recognized in Argentina – Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items, “to provide the tribunal with relevant documents which are in its sole possession”. Brazil argues that Canada has refused to comply with this obligation, and submits that the Panel accordingly should adopt inferences adverse to Canada with regard to the Canada Account.
Brazil further submits that Canada has suggested an appeal to item (k) of Annex I to the SCM Agreement, which Canada contends exempts Canada Account financing and loan guarantees from the prohibition in Article 3 of the Agreement (para. 6.161). According to Brazil, Canada bears the burden of proving its entitlement to this exemption, and has not done so.
Arguments of Canada
Canada denies Brazil’s allegations, maintaining that Canada Account operations are not inconsistent with Article 3. Canada characterizes “Canada Account” operations as financing and loan guarantee activities undertaken by the EDC on the accounts of the Government of Canada; that is, any obligations under the Canada Account are funded by the Government of Canada. The Canada Account is therefore not a pool of money used to subsidise export sales, Canada argues.
294 “Ottawa slams Brazil’s Bombardier claims,” The Globe and Mail, 23 November 1998 (Exh. BRA-87).
295 Id.
WT/DS70/R
Page 96
Canada submits that the EDC has been, throughout its history, self-sufficient and operates on the basis of prudent risk management, and that as a result it would not, on that basis, accept all requests for financing or loan guarantees. According to Canada, among the factors that would dissuade the EDC from extending financing or guarantees on its own account are that the transaction could:
exceed the EDC’s exposure guidelines for a particular country (that is, the maximum amount of business the EDC has decided it can prudently undertake in a specific market, rather than the riskiness of the venture itself);
involve countries where the EDC is “off-cover” (markets where, for reasons of commercial or political risk, EDC is unwilling to support Canadian export business);
involve an amount or a term in excess of that which the EDC would normally undertake for a single borrower or transaction; or
require below-market financing to match subsidised foreign financing.296
Canada states that in exceptional circumstances the EDC may provide financing or loan guarantees under the Canada Account, and that requests that meet the EDC’s basic objectives are authorised by Canada’s Minister for International Trade with the concurrence of the Minister of Finance after taking into consideration the following:
EDC’s usual criteria, such as the number and quality of other financial parties involved, country risk considerations, and legal and regulatory structures;
the Government’s general willingness to consider the country risk and/or the creditworthiness of non-sovereign borrowers;
an assessment of the economic costs and benefits to Canada, including the employment generated by the transaction in question;
the importance of the transaction to the exporter;
foreign policy implications, including Canada’s bilateral relationship with the country of the purchaser; and
the importance of the purchasing market to Canada.297
According to Canada, Canada Account financing and loan guarantees for exports committed since the entry into force of the SCM Agreement have been consistent with the interest rate provisions of the OECD Consensus, as required by Item (k) of Annex I, and such financing and loan guarantees are therefore not prohibited, as provided in footnote 5 to Article 3. In response to a Panel question, Canada indicates that it is not invoking the second paragraph of item (k) as a positive defense to any of the claims made by Brazil, in view of its position that Brazil has not made out a prima facie case. Canada argues that in the context of government credit, a subsidy exists where: first, there is a net charge to the treasury of the Member providing the loan (for example, where a government provides credit at rates below those at which it borrows); and second, there is an advantage above and beyond the market (for example, where a government lends at rates below the market). Therefore, according to Canada, to make out a case on Article 1 that Canada Account constitutes a subsidy, Brazil, as the
296 Exh. BRA-27.
297 Id..
WT/DS70/R
Page 97
complainant, must establish a prima facie case on both elements of the test of Article 1. In Canada’s view, Brazil has not done so.
Canada asserts that Brazil relies on two pieces of evidence in support of its allegation that the Canada Account is a subsidy, and that Brazil invites the Panel to presume that a subsidy exists because of an alleged “blanket of confidentiality” or a “veil of secrecy” that surrounds the Canada Account. For Canada, this evidence does not stand for the proposition in support of which it is adduced, and does not stand up to closer scrutiny.
Canada objects to the conclusion that Brazil draws a statement in the EDC’s Annual Report -- that Canada Account funds are “’used to support export transactions which … for reasons of size or risk, the … EDC cannot support through regular export credits‘” -- that the Canada Account provides subsidies. According to Canada, the EDC may determine not to engage in a given financing transaction on the basis of prudent portfolio management. A transaction may be rejected by the EDC because further exposure to a particular country or credit is judged not to be prudent for the corporation, given its existing portfolio. At the same time, the EDC under the Canada Account may not have the same exposure to the country or credit in question, and if the transaction is deemed by the Canadian Government to be in the national interest, such transaction may be considered under the Canada Account. Canada states that like the corporate account, the Canada Account is managed in accordance with prudent commercial risk practices, and therefore, the fact that the EDC under the Canada Account may advance credit where the EDC, for the reasons of portfolio management, would not do so under its corporate account, does not support the proposition that the Canada Account imposes a net charge against the treasury of Canada or that it provides an advantage above and beyond the market.
In response to Brazil’s arguments on the alleged “non-transparency” of the Canada Account and Brazil’s statement that an expert Canadian economic consulting firm and highly reputable Canadian law firm have produced all the evidence about Canada Account they could find from public sources, Canada provided a copy of the Summary of the Report to the Treasury Board on EDC’s Canada Account Operations for the Fiscal Year 1995-96 that was posted on EDC’s website.298 Canada acknowledges that, as argued by Brazil (para. 6.153) Canada Account loans may be provided on either concessional (that is, below market) or non-concessional (that is, commercial) terms. Canada notes that the Canada Account report299 contains two tables – one detailing concessional loans, and the other non-concessional loans. According to Canada, the “concessionary” financing is part of Canada’s Official Development Aid, and is provided in accordance with the Helsinki tied-aid rules of the OECD Consensus. Canada maintains that since 1995, tied-aid has not been used in respect of any country other than China, and has not been used in respect of civil aircraft, which in any event is prohibited under the OECD Consensus. Since this concessionary financing is not used in respect of civil aircraft, it is Canada’s view that it falls outside the Panel’s jurisdiction.
Regarding the newspaper quote from Bombardier that Canada Account financing is at “close to commercial rates” (para. 6.155), Canada asserts that the comment was made in reference to the Commercial Interest Reference Rates established by the OECD Consensus.300 Canada notes its argument that Canada Account transactions are consistent with the interest rates provisions of the OECD Consensus.
298 Summary of the Report to the Treasury Board on EDC’s Canada Account Operations for the Fiscal Year 1995-96, EDC online: homepage http://www.edc.ca/edcweb/about/summary.html (date accessed: 10 December 1998). (Exh. CDN-92)
299 Exh. CDN-92, pp. 8-9
300 Letter to whom it may concern from Dr. Yvan Allaire, dated 27 November 1998. (Exh. CDN-64); Arrangement on Guidelines for Officially Supported Export Credits (OECD Consensus) TD/Consensus(97)70 (Exh. CDN-3).
WT/DS70/R
Page 98
Canada submits that according to the OECD Consensus, the Commercial Interest Reference Rates “represent commercial lending interest rates in the domestic market of the currency concerned.”301 These commercial reference rates are, therefore, by definition, “’close to commercial rates‘”, and indeed , may be above or below the market for a particular credit, Canada states. Accordingly, in Canada’s view, Brazil’s reference to this comment does not amount to proof that Canada Account financing confers an advantage above and beyond the market.
More important for Canada, the quoted statement says nothing at all about the “’net cost‘” element of the test in Article 1, in Canada’s view because the relevant Commercial Interest Reference Rate for the currency in which the Canada Account transactions for civil aircraft have been done (the commercial reference rate for the US dollar is equivalent to the yield on US Treasuries plus 100 basis points) was well above the cost of funds for the Government of Canada in both instances.
According to Canada, Brazil therefore has not raised a prima facie case with respect to either of the two prongs of the test for establishing that government export credit is a subsidy.
Canada further submits that since Brazil has not made a prima facie case that Canada Account lending constitutes an export subsidy, Canada is not invoking the second paragraph of Item (k) of Annex I of the SCM Agreement as a defence to any of the claims made by Brazil in respect of Canada Account financing.
Canada submits that in the regional aircraft sector, Canada Account plays a minor role in financing of exports. According to Canada, out of a total of 281 aircraft financed and delivered in the period 1 January 1995 to 29 October 1998,302 Canada Account was used to finance only six deliveries, all of them Dash 8-300’s303 split equally between LIAT of Antigua and South African Express of South Africa.304
In response to a Panel request for the details of these six deliveries, including terms, conditions and copies of the relevant finance arrangements, Canada replies that the information requested by the Panel is sensitive business confidential information, and that Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute.
Comments of Brazil
Regarding Canada’s response to the Panel’s request for information on Canada Account financing of the six deliveries of regional jets, Brazil argues that Canada has refused to provide the documents requested by the Panel. As a result, the Panel should adopt adverse inferences, presuming that the information withheld is prejudicial to Canada’s position (paras. 4.146-4.151). Such treatment is particularly compelling here, where Canada has acknowledged that the Canada Account has been used to finance particular regional aircraft transactions (para. 6.170), and where Bombardier has acknowledged that it has used Canada Account financing under terms described as “’close to commercial.’”305
301 OECD Consensus, at 12 (Exh. CDN-3).
302 Chart showing Canadian Regional Aircraft Exports - Financed and Delivered: 1 January 1995 – 29 October 1998 (Exh. CDN-61).
303 Chart showing Canadian Regional Jet Exports - Financed and Delivered: 1 January 1995 – 29 October 1998 (Exh. CDN-62).
304 Chart showing Canadian Regional Turboprop Exports - Financed and Delivered: 1 January 1995 – 29 October 1998 (Exh. CDN-63). See also, table provided in answer to Question 17 of the Panel.
305 Exh. BRA-87.
WT/DS70/R
Page 99
TECHNOLOGY PARTNERSHIPS CANADA (“TPC”) AND PREDECESSOR DEFENCE INDUSTRY PRODUCTIVITY PROGRAMME (“DIPP”)
General arguments of the parties
Arguments of Brazil
Brazil states that in 1996, Technology Partnerships Canada (“TPC”) “’was created to address the need by established companies in specific industrial segments to ensure that near-market products
-- those with a high potential to stimulate economic growth and job creation -- actually reach the marketplace.‘”306 According to Brazil, sectors eligible for assistance from TPC are “’environmental technologies, enabling technologies, and aerospace and defence industries‘”307, although Brazil maintains that in reality TPC represents a captive financing opportunity for the aerospace industry generally and the regional aircraft industry specifically.
Brazil indicates that TPC’s predecessor was the Defense Industry Productivity Programme (“DIPP”), under which subsidies to the Canadian aerospace sector totaled approximately Can$2 billion.308 Brazil quotes Canadian Industry Minister John Manley as emphasizing that despite the reference to “Defence,” “’virtually all funding under the DIPP fund has been earmarked for civilian applications,‘”309 and argues that DIPP in fact “’had been used primarily by aerospace companies . . .‘”310 Brazil argues that in launching TPC, the Canadian government created a programme which “’is really nothing more than a redesigned and renamed Defence Industry Productivity Programme,‘”311 noting that Industry Canada announced that “’no new commitments are being made under DIPP,‘” but that “’[a]ny future funding will be considered under the auspices of the new programme, Technology Partnerships Canada,‘”312 and quoting TPC’s annual report that TPC funding “’is intended to cover outstanding commitments under the Defence Industry Productivity Programme (DIPP) . . .‘”313
According to Brazil, TPC explicitly targets “’conditionally repayable investments‘”314 to projects that result in a high technology product for sale in “’export markets.‘”315 For Brazil, this explains the programme’s extensive support for the Canadian aerospace industry, which according to Brazil is described by TPC itself as “’highly export oriented.‘”316 Brazil asserts that the overwhelming export orientation of the Canadian regional aircraft industry has been confirmed in the
306 Industry Canada website, “Technology Partnerships Canada” (emphasis added) (Exh. BRA-29).
307 Industry Canada website, “Technology Partnerships Canada -- How It Works” (Exh. BRA-30).
308 Canadian Taxpayers Federation study, “Corporate Welfare - A Report on Sixteen Years of Industry Canada Financial Assistance,” 16 April 1998, at Tab 4 (Exh. BRA-31). This statement of total DIPP funding to the aerospace sector represents aggregate DIPP funds released to four Canadian companies, i.e., Bombardier, de Havilland, Pratt & Whitney Canada Inc., and CAE Electronics Ltd.
309 Hansard, House of Commons Debates, 14 March 1996, Oral Question Period, at 709 (Exh.
BRA-32).
310 “Superfund to boost science, technology: Federal budget to unveil $200-million pot for development of new products for export,” The Globe and Mail, 5 March 1996 (Exh. BRA-33).
311 Id.
312 Industry Canada Response to Inquiry of Minister, Question No Q-30 by Mr. Schmidt, dated 19 March 1996 (Exh. BRA-34).
313 TPC Annual Report 1996-1997, p. 20 (Exh. BRA-5).
314 Id. at pgs. 5, 15.
315 Industry Canada website, “Technology Partnerships Canada -- How It Works” (emphasis added) (Exh. BRA- 30). See also “Superfund to boost science, technology: Federal budget to unveil $200-million pot
for development of new products for export,” The Globe and Mail, March 5, 1996 (TPC contributions are to cover “the development costs of new products, particularly those destined for export.” (Exh. BRA-33).
316 TPC Annual Report 1996-1997, pgs. 5, 15 (Exh. BRA-5). TPC’s Advisory Board, which includes among its members Bombardier Aerospace Group’s President and Chief Operating Officer, maintains this support network by “provid[ing] expert advice on the vision and goals of TPC’s programme.” Id. at 23, 26.
WT/DS70/R
Page 100
Clark Report, a study submitted separately by Brazil, which concludes that all CRJ sales, and all Dash 8 sales since 1992, have been for export.
In answer to a question from the Panel concerning the coverage of its claim regarding TPC, Brazil indicates that it is challenging all TPC and DIPP funding extended to the Canadian regional aircraft industry after 1 January 1995.
Arguments of Canada
Canada argues Brazil’s case against Technology Partnerships (TPC) appears to be based on an “export propensity” and “intent” interpretation of Article 3. Brazil argues that TPC contributions are subsidies that are “contingent, in law or in fact, on export performance” because the Canadian aerospace industry is an “export-oriented” sector, because exports in the sector are growing at 10 percent a year and because the “building of exports” was, along with job creation, one of the objectives of TPC.
Canada asserts that Brazil’s arguments are based on an incorrect reading of the law and faulty methodology in analyzing the evidence. Canada argues that TPC is not contingent, in law or in fact, upon export performance.
Subsidy
Arguments of Brazil
Brazil asserts that TPC transfers funds in the form of grants and loans on concessional terms to specified industries, including the civilian aircraft industry, specifically on a “’royalty basis,‘” which for Brazil means that repayment will only occur if the underlying project achieves a certain degree of success as described by Brazil. In Brazil’s view, as with any government-provided low (or interest-free) loan or grant, these funds confer an obvious benefit on the recipient; namely, the recipient has no down-side risk-- if the project is unsuccessful, TPC loans need not be repaid. Brazil argues that TPC is not compensated for the risk that it will not be repaid, or for the extended repayment period during which neither principal nor interest on TPC contributions to the Canadian regional aircraft industry are due. Brazil further argues that even if the Canadian government recovers its investment – which, according to Brazil, recent audits show that it does not do for most conditionally-repayable assistance granted to Bombardier, de Havilland and Pratt & Whitney -- its anticipated rate of return is approximately 1.76 per cent for the CRJ loan from TPC, and 3.02 and 3.31 per cent for the Dash 8 loans from TPC, figures well below that expected by a “’reasonable market investor’” (paras. 6.188-6.192)317.
In response to a question from the Panel, Brazil stated its view that a loan that is conditionally-repayable on a royalty basis constitutes a subsidy under Article 1.1 of the SCM Agreement where, if the loan is repaid, the rate of return is such that the lender is not compensated for either the risk that it would have received no repayment or the extended repayment period during which neither principal nor interest are due. If compensated for these two factors, such a loan could be considered commercial. If the lender is not so compensated, the recipient is realizing a benefit by receiving access to funds at a rate not available on the market. For Brazil, TPC is not compensated for that risk, but rather takes an equity investor’s risk in exchange for a secured creditor’s rate of return.
Brazil argues that on 21 October 1996, TPC announced that it was providing a Can$87 million loan to assist the development of Bombardier’s 70-seat Canadair Regional Jet project,
317 See Tables in Exhibit B of Finan Report.
WT/DS70/R
Page 101
known as the CRJ-700.318 According to Brazil, the loan is conditionally repayable on a “royalty basis,” which for Brazil means that the loan will be repaid only if the project generates profits.319 Brazil asserts that Bombardier has as an historical matter failed to repay its conditionally-repayable loans from the Government of Canada, citing an April, 1996 Industry Canada audit of conditionally- repayable Canadian Government assistance to Bombardier which concluded that repayments amounted to a mere five per cent of funds received by the company.320
Brazil states that repayment, if any, of the TPC loan for the CRJ-700 project will therefore begin only after Bombardier realizes profit on the project, which it is not projected to do until it has sold 250 of the CRJ-700s, the first of which is scheduled for delivery in 2000.321 Furthermore, Brazil argues, Bombardier CEO Laurent Beaudoin has estimated that the company must sell 400 of the CRJ- 700s to enable it to pay back the principal amount of the loan, without interest.322 Thus, according to Brazil, any return on TPC’s investment would only begin to accrue with the 401st plane sold, and Brazil asserts that industry estimates predict that only approximately 25 of the CRJ-700s will be produced annually.323 Brazil also submits an article which, it indicates, reflects “industry forecasts” projecting that Bombardier will produce between 127 and 178 CRJs by 2005.324
Brazil asserts that assuming a “royalty rate” of Can$580,000 per aircraft,325 with 25 aircraft sold annually, TPC’s expected rate of return on this loan is 1.76 per cent,326 a return which illustrates that TPC “investments” are nothing more than outright gifts. For Brazil, as a result of this rate of return, the Government of Canada on an annual basis is foregoing an extremely conservative minimum interest amount of Can$4,054,200 (based on the difference between the 1.76 per cent expected rate of return and an extremely conservative benchmark rate of 6.42 per cent for the 1997 Canadian 10-year bond).327 Brazil states that assuming a more realistic benchmark rate -- the rate a commercial investor would expect to achieve for such an investment, i.e., 16.91 to 21.92 per cent328 -- Canada is foregoing between Can$13,180,500329 and Can$17,539,200330 annually on this TPC loan.
Brazil emphasizes that these figures do not account for the fact that repayment of this loan will not even begin until the year 2011, after Bombardier has sold 250 of the CRJ-700s.331 In the meantime, according to Brazil, Canada is foregoing altogether a return on its TPC “investments” in the CRJ-700, resulting in a delay, and therefore increased risk, that would generally translate into a demand for a higher rate of return once such return is actually paid. Brazil asserts that with this considerable delay, no reasonable market investor could be expected to accept TPC’s expected rate of return of 1.76 per cent, noted above. Brazil argues that assuming the benchmark rate a commercial
318 Industry Canada News Release, October 21, 1996 (Exh. BRA- 35).
319 Id. See also Industry Canada definition of “conditionally repayable contribution” (Exh. BRA- 36).
320 Industry Canada audit report, included in Canadian Taxpayers Federation study, “Corporate Welfare Volume Two,” 4 June 1998, at Tab 6, pgs. 57-61 (Exh. BRA- 37). Although the Canadian government, to
preserve confidential information, heavily edited the report of the Bombardier audit before releasing it to the Canadian Taxpayers Federation, Bombardier’s identity and its repayment level were not expunged.
321 See Exhibit A, pgs. A.3-A.4 of Finan Report.
322 Id.
323 Id.
324 Exh. BRA-92
325 Calculated as Can$87 million/(400-250), reflecting the fact that repayment will not begin until 250 jets have been sold, and will be paid in full when 400 jets have been sold. Exhibit A, p. A.4 of Finan Report.
326 Id.
327 Using the 1997 rate of 6.42 per cent for Canada’s “10 Year and Over Federal Government Bond Yield to Maturity.” (Exh. BRA- 38).
328 Exhibit B, Table B.6 of Finan Report.
329 ($87,000,000)(.1691-.0176).
330 ($87,000,000)(.2192-.0176).
331 Exhibit B, Table B.6 and Exhibit A, pgs. A.3-A.4 of Finan Report.
WT/DS70/R
Page 102
borrower would expect to pay for such an investment, i.e., 16.91 to 21.92 per cent,332 the present value in 1998 of the benefit received by Bombardier from this TPC “investment” is between Can$94 million and Can$211 million.333
In response to a question from the Panel concerning why Brazil considers that TPC will only be repaid if the CRJ-700 project is profitable, Brazil argued that under the repayment plan described above, before it has to begin repaying the principal amount of the contribution, Bombardier must sell
250 planes, which, at approximately US $23 million334 per plane, would gross Bombardier
$5.75 billion. To begin to pay TPC a return on the contribution, Bombardier must sell 400 planes, gross income on which would reach $9.2 billion. Bombardier has stated that developing the CRJ-700 cost it $645 million.335 The conclusion to draw from this information, according to Brazil, is that TPC will not even recover the principal amount of its contribution, let alone realize any return, until the CRJ-700 has achieved for Bombardier a very significant degree of success which one can reasonably deem “profitable.”
Brazil states that on 17 December 1996, TPC announced that it was providing a further Can$57 million to Bombardier’s de Havilland subsidiary to develop a 70-seat “stretch” version of its Dash 8 turbo-prop airplane.336 According to Brazil, this loan, too, is repayable only when and if the venture is ever profitable.337 Using calculations similar to those used with regard to the $87 million loan, Brazil estimates that TPC's expected rate of return (if any) is 3.02 per cent, which Brazil states is well below that which would be expected by a market investor.338 Moreover, citing to an Industry Canada list, Brazil asserts that de Havilland has a rather poor repayment history having repaid a mere one per cent of the Can$424 million in DIPP funds provided to the company.339 Assuming the asserted benchmark rate a commercial borrower would expect to pay for such an investment, i.e.,
16.91 to 21.92 per cent,340 Brazil estimates that the present value in 1998 of the benefit received by de Havilland from this TPC “investment” is between Can$43 million and Can$77 million.341
Brazil further argues that in January 1997, TPC announced a total investment of Can$147 million in Pratt & Whitney Canada, Can$100 million of which was targeted for work on the firm’s 6,500 SHP PW150 turboprop engine, used in Bombardier’s Dash 8 aircraft, all of which according to Brazil are destined for export.342 Brazil characterizes Industry Minister Manley as emphasizing, in granting this funding, the crucial role played in the Canadian economy by export- oriented sectors such as the Canadian aerospace industry, quoting Minister Manley as stating that “[a]erospace is a crucial sector for Canada’s economy, with exports growing at 10 per cent per year.”343 Using calculations similar to those discussed with regard to the $87 million and $57 million loans, Brazil estimates TPC's expected rate of return at 3.31 per cent, which according to Brazil is
332 Exhibit B, Table B.6 of Finan Report.
333 id.
334 "Orders soar for Canadair jet", The Globe and Mail, February 20, 1997 (Exh. Bra-90).
335 Id. See also “Le Quitte où double de Bombardier,” Commerce, December 12, 1996, pg. 3 (Exh. BRA-92).
336 Industry Canada, News Release, 17 December 1996 (Exh. BRA- 4).
337 Id.
338 Exhibit A, p. A.5 and Exhibit B, Table B.7 of Finan Report.
339 “Billions in Federal loans unpaid: Biggest aerospace companies return as little as 2 per cent of money advanced for technology development,” The Globe and Mail, 13 June 1998 (Exh. BRA- 39).
340 Exhibit B, Table B.6 of Finan Report.
341 Exhibit B, Table B.7 of Finan Report.
342 See Clark Report.
343 Industry Canada, News Release, 10 January 1997 (Exh. BRA- 40).
WT/DS70/R
Page 103
well below that which would be expected by a market investor.344 As with Bombardier and de Havilland, Brazil states, Pratt & Whitney’s repayment history is poor, indicating that of nearly Can$900 million in “direct cash subsidies” from the Canadian government since 1980, less than Can$100 million has been repaid.345 Assuming the benchmark rate which Brazil argues a commercial borrower would expect to pay for such an investment, i.e., 16.91 to 21.92 per cent,346 Brazil estimates that the present value in 1998 of the benefit received by Pratt & Whitney from this TPC “investment” is between Can$97 million and Can$182 million.347
In response to a Panel question as to the rate of return that a “reasonable market investor” in the civil aviation sector would expect, Brazil submits that its argument in the context of EDC regarding risk and return (para. 6.11) frames the issue clearly: investors in speculative grade bonds expect a return greater than 17 per cent.
For Brazil, another relevant benchmark is the particular return expected by Bombardier on its own investments. In Bombardier’s case, Brazil states, the required pre-tax rate of return is Bombardier’s weighted average pre-tax cost of capital, which Bombardier expects to achieve across all risk classes of its investments; that is to say, this rate represents the average return on Bombardier’s investment in new aircraft development, capital equipment and structures, and other investments. Brazil states that Bombardier’s after-tax cost of capital is 11 per cent.348 Applying a 35 per cent effective tax rate, Bombardier’s pre-tax cost of capital is determined to be 16.9 per cent. According to Brazil, if Bombardier’s investments fail to achieve this rate of return, then Bombardier fails to provide a return to its debt and equity holders sufficient to compensate them for the risk they bear in investing in Bombardier.
According to Brazil, the required cost of capital return for Bombardier is not constant across all investments. Certain investments are inherently more risky and, hence, should have a higher expected return to compensate for the additional risk. Development of a new airframe, such as the CRJ-700, certainly falls into this riskier class of investment.
Brazil states that economic studies report higher costs of capital for these types of research and development projects. For example, a Federal Reserve Bank of New York study of the cost of capital in the United States and other countries found that the required cost of capital return for R&D projects averaged 19.2 per cent, while the required cost of capital return for equipment and machinery with a physical life of 20 years yielded an average return of 10.6 per cent.349 Brazil notes that other researchers have found R&D to have an even higher rate of return, in the range of 25 per cent to 35 per cent.350
Brazil states that based on this analysis, the Finan Report concluded that the benchmark rate a commercial borrower would expect to pay for airframe development expenses is between 16.91 per cent, which is Bombardier’s pre-tax cost of capital, and 21.92 per cent, which includes a five per cent risk premium to compensate for the risk normally associated with R&D investment projects, as discussed in the previous paragraph.
344 Exhibit A, pgs. A.5-A.6 and Exhibit B, Table B.13 of Finan Report.
345 “As Bombardier takes off, Pratt descends: Diverging fortunes raise questions about subsidies,” The Globe and Mail, 19 October 1998 (Exh. BRA- 41).
346 Exhibit B, Table B.6 of Finan Report.
347 Exhibit B, Table B.13 of Finan Report
348 J.P. Morgan Securities report, “Bombardier,” 1 September 1998, pg. 10 (Exh. BRA- 76).
349 Federal Reserve Bank of New York, “Explaining International Differences in the Cost of Capital,” Quarterly Review, Summer 1989, Table 2 (average for the years 1984 – 1988) (Exh. BRA-95).
350 See US Congressional Budget Office report on “Federal Investment in Intangible Assets”
(July 1991) (Exh. BRA-96). See also Edwin Mansfield, “Appropriating the Returns from Investment in R&D Capital,” in K. Cool, D. Neven, and I. Walters (eds.), European Industrial Restructuring in the 1990s (1992) (Exh. BRA-97).
WT/DS70/R
Page 104
According to Brazil, other TPC “repayable investments” for the development of Canadian regional aircraft, all of which are destined for export, have been recently announced, although without enough detail to permit analysis of the precise rates of return to be expected by TPC. In April 1997, Brazil states, TPC announced a Can$12.7 million “repayable investment” in Allied Signal Aerospace Canada, a portion of which is targeted for development of the power management generating system for the Dash 8-400.351 Brazil also states that in March 1998, TPC announced a Can$9.9 million “repayable investment” in Sextant Avionique Canada Inc., to be used for development of the avionics system for the Dash 8-400 and the flight control system for the CRJ-700.352
Brazil asserts that the most recent WTO Trade Policy Review of Canada lists TPC as one of a number of “’Selected federal subsidy programmes.‘” Brazil considers it significant that the Report has identified TPC as a “’subsidy.‘” Moreover, according to Brazil, the WTO identified the form of subsidy offered by TPC as “’grants,‘” as opposed to certain other programmes identified as “’repayable contributions‘” or “’conditionally-repayable‘” contributions.
Arguments of Canada353
Canada confirms that there have been commitments, under the TPC programme, for repayable contributions of $87 million to Bombardier for the development of the CRJ-700;
$57 million to de Havilland for the development of the Dash 8-400; $100 million to Pratt & Whitney for the development of the PW150 turboprop engine; $12.7 million to Allied Signal Aerospace Canada, a portion of which is for the development of the power management generating system; and
$9.9 million to Sextant Avionique Canada inc., for the development of avionics systems. Canada states that in every case, these amounts have not been completely disbursed to the recipients, but that a contribution is disbursed in predetermined portions as milestones set out in the project’s work programme are met.
Canada challenges Brazil’s description of the repayment obligations under the TPC.354 According to Canada, royalty-based repayment is not tied to profits, but to sales: in the case of some contributions, such as the contribution for the CRJ-700, the royalties are related to the sales of the specific product that are to be developed under the project; with respect to other projects, although the contribution may be in respect of a specific product, royalties would be paid on the sales of a family of products that would be developed following the original project. Canada notes as an example that the contribution towards the Pratt and Whitney 150 engine, under which royalties are payable on the sales of the family of engines that may be developed based on the original engine. Finally, Canada states, if it is a process or a technology that is being developed – such as one related to fuel efficiency
then royalties would be due on the basis of the total sales of the recipient enterprise. Thus, Canada states, in no circumstance is profitability a criterion of repayment.
Regarding Brazil’s reference to Canada’s recent WTO Trade Policy Review, Canada recalls that this Trade Policy Review, dated 19 November 1998, was conducted pursuant to the Agreement Establishing the Trade Policy Review Mechanism (TPRM Agreement) contained in Annex 3 of the WTO Agreement. Section A(i) of the TPRM Agreement specifically provides that:
“… the review mechanism … is not … intended to serve as a basis for the enforcement of specific obligations under the Agreements or for dispute settlement procedures …”.
351 Industry Canada, News Release, April 25, 1997 (Exh. BRA- 42).
352 Industry Canada, News Release, March 6, 1998 (Exh. BRA- 43).
353 See also comments of Canada and responses of Brazil on the Finan Report regarding TPC in Section VI.F.
354 This paragraph responds to Question 18 of the Panel to Canada.
WT/DS70/R
Page 105
Canada notes that the information contained in the report concerning Canada’s alleged subsidy practices was obtained from Canada’s notification under Article 25 of the SCM Agreement.355 Article 25.7 specifically provides that:
“Members recognize that notification of a measure does not prejudge either its legal status under GATT 1994 and this Agreement, the effects under this Agreement, or the nature of the measure itself.”
For Canada, the Secretariat’s description of a Canadian programme in a TPR review therefore has no relevance whatever for the purposes of dispute settlement.356 Canada further states that the notification referenced by Brazil is Canada’s Notification of the Technology Partnership Programme, and not Technology Partnerships Canada. Canada states that according to the Notification (at page 15), the objective of the Technology Partnership Programme is “to provide support to enable Canadian small- and medium-sized enterprises to enter into partnerships with university laboratories to develop university research to the point where it can be exploited by industry.” Canada notes that the TPR Review correctly describes Technology Partnerships Canada, the programme at issue in this dispute, in its paragraph 117.
Canada contends that TPC contributions usually constitute a portion – on average less than 30 per cent – of the eligible costs incurred by the recipient, that eligible costs are limited to research and development expenses, and that TPC contributions may not be made for costs associated with facilities or capital equipment. Further, Canada argues, these contributions are repayable, generally on a royalty basis tied to the market success of a project: the more successful a product, the higher its returns. In that sense, according to Canada, TPC contributions are investments rather than loans: the Government of Canada shares in the rewards of successful projects. Unlike debts, TPC returns are not limited to a fixed amount.
In response to a Panel request for full details on the terms and conditions for the provision and repayment of the funds provided by TPC for the five transactions identified (para. 6.195) Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Accordingly, Canada indicates, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to support Canada’s submissions that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
With regard to the Panel’s request for “commercial borrowing rates” for debt of comparable size, risk level and maturity, Canada submits that the relevant consideration for determining the rate of return for the TPC contributions in question was the Government of Canada’s cost of funds, and states that at the time these transactions were entered into, Canada’s long term bonds were yielding approximately 6 to 7 per cent357 According to Canada, TPC repayment terms are negotiated so that on a net present value basis they are cost-revenue neutral or better. Canada provides as business confidential information in response to a Panel question regarding whether it agreed with any of the rate of return calculations of the Finan Report a calculation of TPC’s rate of return on this transaction, and stated that this was clearly above Canada’s cost of funds of 6-7 per cent.
In response to a Panel request for full details and documentation on TPC contributions to Bombardier, CAE Electronics and Pratt & Whitney listed in the 1996-97 Public Accounts of Canada, Canada states that the figures in the Public Accounts detail disbursements actually made in respect of
355 WT/TPR/S/53, at 71.
356 Indeed, the description of TPC in the TPR itself is internally inconsistent: see WT/TPR/S/53. The description of TPC at page 69, para. 117 correctly describes TPC investments and repayment conditions. The reference to “Grants” in the Table on page 71 is to the Technology Partnerships Programme.
357 Exh. CDN-108
WT/DS70/R
Page 106
existing commitments, and that information requested by the Panel concerning the TPC contributions to Bombardier for the CRJ-700 was provided in Canada’s answer to the Panel’s question concerning this funding. With respect to the CAE Electronique contribution, Canada states that Brazil has made no allegation and adduced no evidence in respect of this contribution, and that Canada does not consider it appropriate to adduce evidence in response to an allegation that has not been made and a case that has not been established.
Comments of Brazil
Brazil observes that in response to a question from the Panel, Canada has provided documents for only one of the five TPC grants challenged by Brazil, and that with regard to that grant, Canada has wholly or selectively redacted any and all information which could be at all informative to the Panel. Brazil submits that the Panel should adopt adverse inferences, presuming that all information withheld by Canada is inculpatory, and that each of the five TPC contributions to the Canadian regional aircraft industry challenged by Brazil constitute export subsidies under Article 3 of the SCM Agreement. In this connection, Brazil argues that Canada has failed to provide documents specifically requested by the Panel on the TPC contributions identified in the Public Accounts of Canada (para. 6.203), and Brazil therefore requests the Panel to adopt adverse inferences regarding these transactions.
In Brazil’s view, the Panel should presume, from Bombardier’s decision to permit disclosure of information regarding the $87 million TPC grant for the development of the CRJ-700, that permission to disclose information regarding the $57 million grant to its subsidiary, de Havilland, for the development of the Dash 8-400 was withheld because such information would prove incriminating.
Moreover, Brazil argues, Canada has selectively provided in BCI Tab 1 various pages - largely redacted - from certain market assessments, presumably to show that the potential market for the CRJ-700 is large enough to give some hope that TPC will one day recover its “’investment‘” in the development of the aircraft. Brazil notes Canada’s reply to a Panel question (para. 6.367) that Brazil’s conclusions in the Finan Report concerning the market prospects of the CRJ-700 are “’unduly pessimistic,‘” citing to what Brazil terms Canada’s selective and incomplete excerpts from the market forecasts included in BCI Tab 1. Brazil states that Canada has left out a critical portion of these studies - projections regarding Bombardier’s production capacity for the CRJ-700 - which make Brazil’s and the Finan Report’s conclusions regarding the near impossibility of Canada’s recovery of its “investment” crystal clear, noting that Brazil submitted358 a report of the Forecast International study extracted in Canada’s BCI Tab 1, and indicating that that forecast concludes that Bombardier is able to manufacture only 178 CRJ-700s over the next ten years, or by 2006.359 For Brazil, this is a perfect illustration of why the Panel should adopt adverse inferences, presuming that the information withheld by Canada is prejudicial to its position.
Additionally Brazil states, the December 1997 and July 1998 programme updates and the July 1998 programme forecast included in BCI Tab 1, along with the progress reports included in BCI Tab 2, are redacted to such an extreme degree as to make them utterly worthless. Brazil states that all schedules and reports contained therein are redacted in their entirety, and that Canada’s claim regarding whether destinations are listed in the order book redacted from the July 1998 programme update, for example, is without any support, as the information which would permit the Panel to reach this conclusion has been redacted.
358 Exh. BRA-92
359 Another forecast cited in Exh. BRA-92, by the Teal Group, concludes Bombardier will manufacture only 127 CRJ-700s by 2005.
WT/DS70/R
Page 107
TPC’s anticipated or expected return is below market
Brazil recalls the documentation it submitted to demonstrate that commercial lenders to speculative borrowers (bonds rated CCC+ to C) as of the fourth quarter 1998 were requiring returns of at least 17 per cent (para. 6.110), and its argument that development of a new airframe, such as the CRJ-700, certainly falls into this risk class, and that investment therein therefore should yield a return commensurate with this risk (paras. 6.188-6.192) Brazil states that economic studies report higher costs of capital for these types of research and development projects. For example, a study by the Federal Reserve Bank of New York concluded that the required cost of capital return for R&D projects averaged 19.2 per cent, while the required cost of capital return for equipment and machinery with a physical life of 20 years yielded an average return of 10.6 per cent360, and other researchers have found R&D to have an even higher rate of return, in the range of 25 per cent to 35 per cent.361 As a result, Brazil argues, the Finan Report demonstrated that the benchmark rate a commercial borrower would expect to pay for airframe development expenses is between 16.91 per cent, which is Bombardier’s pre-tax cost of capital, and 21.92 per cent, which includes a five per cent risk premium to compensate for the risk normally associated with R&D investment projects.
Brazil argues that as a result of the structure of the TPC contribution, it is impossible for TPC to achieve this rate of return, or any rate of return even remotely compensating TPC for the substantial risk undertaken in making this contribution: first, if Bombardier were to deliver only 200 CRJ-700s, for example, TPC’s return would be zero, and it would fail to even recover the principal amount of its contribution; second, only if CRJ-700 deliveries meet the most optimistic of market forecasts - which, as discussed above (para. 6.206), have been misrepresented by Canada in its reply - does TPC’s return reach the figure offered by Canada in its reply to question 33 as the maximum rate of return; third, were Bombardier to sell these aircraft over a longer time horizon than ten years, TPC would fail to earn the return heralded by Canada, which is, according to Canada, the maximum TPC can earn; fourth, while TPC could suffer a negative return, its return on the upside is capped, no matter what Bombardier’s actual success in the market is.
Therefore, Brazil submits, it is only under the most optimistic of forecasts that TPC gets back both the principal amount of its contribution, and earns the maximum return permitted by the contribution agreement concluded with Bombardier, and even this return fails to compensate TPC adequately for the risk involved in its “investment.”
TPC’s expected return - “expected” at the time the loan commitment was made - was insufficient, even under Canada’s “cost of funds” test
According to Brazil, the risk inherent in TPC’s $87 million CRJ-700 grant is not with the design or development stage of the project, noting Canada’s statements (para ) that “’project risk was considered manageable in light of Bombardier’s proven track record and because the technical risk was mitigated by the fact that this aircraft was a stretch version of an existing platform - the CRJ50‘”, and that the “’market risk was also manageable.‘”
Brazil states that the risk inherent in TPC’s $87 million CRJ-700 grant is attributable to two factors, the first of which (see para. 6.184) involves the lengthy delay before repayment of principal, let alone interest, even begins. According to Brazil, the documents provided by Canada in response to
360 Federal Reserve Bank of New York, “Explaining International Differences in the Cost of Capital,” Quarterly Review, Summer 1989, Table 2 (average for the years 1984 – 1988) (Exh. BRA-95).
361 See US Congressional Budget Office report on “Federal Investment in Intangible Assets” (July 1991) (Exh. BRA-96). See also Edwin Mansfield, “Appropriating the Returns from Investment in R&D Capital,” in K. Cool, D. Neven, and I. Walters (eds.), European Industrial Restructuring in the 1990s (1992) (Exh. BRA-97).
WT/DS70/R
Page 108
Panel questions do not in any way rebut this factor, which would motivate any reasonable market investor to seek returns well beyond those anticipated by TPC for this type of loan.
For Brazil the second factor concerns uncertainty surrounding the number of CRJ-700 aircraft that will be delivered. Brazil argues that Canada cites to selectively quoted market forecasts to conclude that the size of the potential market for 60-90 seat regional jets will support the projections memorialized in the contribution agreement with Bombardier, and that TPC’s anticipated return is very sensitive to the number of CRJ-700 aircraft delivered per year, and states that there is wide variation in the projections for the total number of units of the CRJ-700 that will be delivered (see para. 6.206).
Brazil does not ask to engage the Panel or Canada over which forecast is better or more plausible, but makes the point that for TPC to earn its anticipated return, and to meet even Canada’s own estimate of its cost of funds (six to seven per cent) everything must go exactly according to plan. According to Brazil, the rate of return cited by Canada as TPC’s anticipated return is actually based on a market projection which lies at the upper end of a very wide range of forecasts. Brazil argues that if the number of CRJ-700s delivered per year slips below that projection, TPC would fail to achieve the six to seven per cent minimum return required to meet its cost of funds.
Brazil argues that therefore there is a critical shortcoming with Canada’s justification or support for the rate of return it calculates. In Brazil’s view, Canada simply shows that it may be feasible to achieve that return under ideal circumstances, but this is not what an investor would focus on in deciding what its anticipated return on an investment should be. For Brazil, the question is not which forecast of future CRJ-700 deliveries is more credible; rather, the question is what an investor would do when faced with uncertainty about the potential market size for the CRJ-700. In Brazil’s view, the investor would not simply rely upon the highest possible value for unit deliveries and use that to justify the expected rate of return, but rather, faced with uncertainty over the number of future aircraft deliveries, would is compute an expected value for the future deliveries of the CRJ-700, using this information to determine the expected royalty stream and, in turn, the expected rate of return.
Brazil submits that given the range of equally credible forecasts of CRJ-700 annual deliveries, the expected value is the mid-point of the range. At that midpoint, Brazil argues, TPC’s expected return on annual deliveries of CRJ-700s fails to meet the cost of funds threshold identified by Canada itself.
Regarding Canada’s assertion that TPC does not grant lump-sum contributions, Brazil argues that whether the TPC contributions are disbursed as a lump-sum or disbursed over time, the result remains the same - a benefit is provided to the recipient, who gains access to a below-market source of funds. In this context, Brazil submitted a table (BCI Exhibit Bra-1) showing the cash flows associated with the $87 million TPC grant for development of the CRJ-700, calculated with the grant being disbursed both as a lump sum, and over time.
Brazil further argues that business confidential information submitted by Canada in connection with the $87 million contribution to Bombardier suggests that the repayment terms do not appear to be consistent with that set out in TPC’s repayment policy. In this connection, Brazil discusses, in business confidential comments on Canada’s 21 December 1998 replies to the Panel’s questions, a number of examples of what Brazil considers to be departures from TPC’s repayment policy in the case of this particular transaction. For Brazil, this begs the question whether the Canadian government departs from this policy in the case of grants to the Canadian regional aircraft industry.
WT/DS70/R
Page 109
Export contingency
Arguments of Brazil
Brazil discusses a number of factors which in its view demonstrate that TPC and DIPP support to the Canadian regional aircraft industry is “in fact tied to actual or anticipated exportation or export earnings”. Brazil submits that the comments of Industry Minister John Manley in announcing its “’partnership‘” with de Havilland, that “’Aerospace is a crucial sector for Canada’s economy, with exports growing at 10 per cent per year‘” apply equally to the TPC’s interest-free $87 million loan to Bombardier, the $147 million loan to Pratt & Whitney, the loans to Allied Signal and Sextant Avionique, and the DIPP loan for initial development of the CRJ. Brazil also states that Mr. Herb Gray, Leader of the Government in the House of Commons and Solicitor General of Canada, added, in remarks that also apply equally to the other loans, that “’[t]hese two outputs of the Dash 8-400 project -- the creation of jobs and the building of exports -- are just what the government had in mind when we established Technology Partnerships Canada earlier this year.‘”362 Brazil emphasizes that each Dash 8 series aircraft sold since 1992, and each CRJ sold since its development, has been sold for export.
According to Brazil, in June 1998, after only two years of operation, TPC investments in the Canadian aerospace industry totaled Can$407.6 million,363 representing 72 per cent of total TPC funding to all sectors. According to Brazil, of that total Can$267 million (paras. 6.181, 6.186, 6.187, 6.193), or 66 percent, was invested in the regional aircraft industry. Brazil notes a press report indicating that the Canadian aerospace industry has indicated that an “’additional $300 million in annual TPC funding is required to build on current technology and secure the future for Canadian aerospace.‘”364
For Brazil, TPC and DIPP assistance to the civilian regional aircraft industry is contingent in law or in fact upon export, in that, according to Brazil (citing to the Clark Report), all CRJ aircraft, and all Dash 8 aircraft sold since 1992, have been sold solely for export. For Brazil, although under the terms of the SCM Agreement “’[t]he mere fact that a subsidy is granted to enterprises which export shall not for that reason alone be considered to be an export subsidy,‘”365 funding mechanisms meet the definition of subsidy when they are “’in fact tied to actual or anticipated exportation or export earnings.‘”366 Brazil asserts that the Canadian Government supports the Canadian aerospace industry (and Bombardier) through TPC precisely because it is an “’export-oriented‘” sector the “’well-being‘” of which “’is vital to Canada’s economy.‘”367 Most significantly for Brazil, every Dash 8-400 and every CRJ-700 sale benefiting from TPC and DIPP funding has been for export.368 Accordingly, the TPC and DIPP “investments” advanced to support these aircraft are export subsidies in law or in fact and are, accordingly, prohibited by Article 3 of the SCM Agreement.
That is, according to Brazil, Canada maintains massive amounts of support to its regional aircraft industry precisely because it is 100-per cent export oriented and precisely because they anticipate that this trend will continue. For Brazil, the available evidence supports only one conclusion: were it not for the total export orientation of the Canadian regional aircraft industry, Canada would not continue to give it billions of dollars in subsidies.
362 Exh. BRA-4 (emphasis added by Brazil).
363 TPC investment summary, from TPC website, dated 22 July 1998 (Exh. BRA- 47).
364 “As Bombardier takes off, Pratt descends: Diverging fortunes raise questions about subsidies,” The Globe and Mail, 19 October 1998 (Exh. BRA- 41).
365 SCM Agreement, Article 3.1(a), note 4.
366 TPC Annual Report 1996-1997, p. 5 (Exh. BRA- 5); Industry Canada, News Release, 17 Dec. 1996
(Exh. BRA- 4).
367 Id.
368 Clark Report, at pgs. 7-8, 13, 50-51, and Tables 1 and 2.
WT/DS70/R
Page 110
In Brazil’s view, this is particularly important in determining whether TPC and DIPP grants constitute, within the meaning of Article 3 of the SCM Agreement, export subsidies “in fact.” Brazil maintains that the drafters of Article 3 appreciated the fact that a subsidy need not be “legally contingent on export performance,” if it is instead “in fact tied to actual or anticipated exportation or export earnings.”
Brazil submits that the significant support offered to the Canadian regional aircraft industry is both a reward for past export performance, and a grant which various Canadian statements make eminently clear is given in anticipation of continued, exceptional export growth and total, 100 per cent export orientation.
For Brazil, Canadian practices embody precisely the meaning of the term “export subsidies in fact.” To fail to apply the de facto export subsidy provision in these circumstances would be to render the prohibition of de facto export subsidies empty, and to give Members a license to provide export subsidies with abandon, subject only to the limitation that they avoid using the word “export” in their laws and regulations and in their dealings with export-oriented companies.
Brazil acknowledges that some TPC and DIPP funds may have gone to other industries with sales in export markets, or to other industries with sales in domestic markets, but states that Brazil’s claim is unrelated to these possible instances of TPC and DIPP funding. For Brazil, when TPC funds were disbursed to the Canadian regional aircraft industry, they were given to an industry which is 100-per cent export oriented, precisely because it is an export industry and was anticipated to remain so.
Brazil contends that while Article 3.1(a) provides that the mere granting of a subsidy to an exporting entity will not “for that reason alone” make it a prohibited export subsidy, the Panel is not here faced with such a case. In Brazil’s view, although Canada would have the Panel believe that Brazil’s entire claim turns on what Canada calls the “export propensity” of the Canadian regional aircraft industry alone, there are many convergent factors contributing to Brazil’s conclusion that support to the Canadian regional aircraft industry through TPC and DIPP is “in fact tied to actual or anticipated exportation or export earnings,” i.e., that the Canadian Government and the provinces have supported the Canadian regional aircraft industry precisely because it is a total export industry, and precisely because they anticipate that this performance will continue:
That TPC funding statistics (para. 6.220) demonstrate that TPC is captive to the Canadian aerospace industry, and more specifically, the regional aircraft industry; that TPC itself recognizes its main beneficiary as “highly export oriented,”369 and evidently believes that fact is noteworthy in justifying TPC’s own operation.
That such support is, as an historical matter, par for the course: that TPC’s predecessor, DIPP, channelled subsidies to the aerospace sector totalling approximately $2 billion,370 giving TPC ample experience testing whether its anticipations or expectations regarding the industry’s export performance are founded.
That statements made by the Canadian Government in announcing its $267 million in grants to the regional aircraft industry demonstrate that those grants are tied to the
369 TPC Annual Report 1996-1997, pg. 5 (Exh. BRA-5).
370 Canadian Taxpayers Federation study, “Corporate Welfare – A Report on Sixteen Years of Industry Canada Financial Assistance,” 16 April 1998, Tab 4 (Exh. BRA-31). This statement of total DIPP funding to the aerospace sector represents aggregate DIPP funds released to four Canadian companies, i.e., Bombardier, de Havilland, Pratt & Whitney Canada Inc., and CAE Electronics Ltd.
WT/DS70/R
Page 111
Canadian Government’s anticipation, based on ample historical evidence, that the industry will maintain its 100 per cent export orientation:
Statement of Industry Minister John Manley that “[a]erospace is a crucial sector for Canada’s economy, with exports growing at 10 per cent per year.”
Statement of Mr. Herb Gray, Leader of the Government in the House of Commons and then-Solicitor General of Canada, that “[t]hese two outputs of the Dash 8-400 project -- the creation of jobs and the building of exports -- are just what the government had in mind when we established Technology Partnerships Canada earlier this year.”
Statement of Mr. Paul Labbé, then-President of the EDC explaining EDC’s decision to support the launch of the CRJ:
Our focus is export financing. What we’re trying to do in this particular case – this is an exceptional case – is launch an aircraft that has a world market. The Canadian market for this aircraft is minimal compared with what this world market is going to be, but it has to be launched. You have to get a good customer base established so that you have a good market for that thing.371
That when these statements were made, the Canadian Government was aware of the fact that, according to Brazil, every single sale of Dash 8 series aircraft made since 1992, and every single sale of the CRJ since its development and commercialization, had been for export. For Brazil, the Canadian Government has therefore made it very clear that it maintains massive amounts of support to the Canadian regional aircraft industry precisely because it is an export industry and precisely because they anticipate that it will continue to be an export industry.
Brazil disagrees with Canada’s argument concerning small economies (para. 5.74). In Brazil’s view, Canada is effectively arguing that despite the terms of Article 3.1(a), its alleged position as a “small economy” warrants that it be accorded special treatment. Brazil questions Canada’s claim of “small economy” status for itself, noting that Canada is a founding Member of the WTO, a member, with the United States, Japan and the European Communities, of the WTO “Quad,” and a member of the OECD. For Brazil, Canada’s argument also must fail as a matter of law, because there is only one law, applying equally to large and small economies alike, and it is found in the ordinary meaning of Article 3.1(a).
Brazil further argues that this is not a case of a small economy of necessity having a small share of a global market, as Canada claims, or of support being given to an industry by a government indifferent to whether that industry sells in domestic or export markets, but is a case in which, by the deliberate action of the Canadian Government, whatever demand existed in the domestic market was satisfied by what were legally export sales, not domestic sales (paras. 5.113-5.115).
Arguments of Canada
Canada argues that the evidence cited by Brazil regarding the alleged export contingency of TPC’s funding for civil aircraft does not support Brazil’s argument. In Canada’s view, Brazil’s argument (para. 6.175) that “’TPC explicitly targets ‘conditionally repayable investment’ to projects
371 Committee on Foreign Affairs and International Trade, pg. 43:30 (emphasis added) (Exh. BRA-9).
WT/DS70/R
Page 112
that result in a high technology product for sale in ‘export markets,’ which explains the programme’s extensive support for the Canadian aerospace industry …‘” [emphasis in original] is a complete distortion of the document on which Brazil relies. According to Canada, quoted in full, the cited passage states that the TPC targets “’projects that result in a high technology product or process for sale in domestic and export markets.‘” [emphasis added by Canada] Thus, in Canada’s view, the passage stands for precisely the opposite of the proposition for which Brazil’s selective and misleading quotation stands. Thus Canada argues, to the extent that Brazil alleges that TPC is in law contingent upon export performance, its allegation is based on a misquotation of TPC’s objectives.
Canada notes Brazil’s argument that Brazil’s claim is unrelated to possible instances of TPC funds to other industries with sales in export markets, or to other industries with sales in domestic markets. In Canada’s view, here Brazil is arguing that the mere fact that companies in the aircraft sector engage in exports turns a programme that is also available, as agreed by Brazil, for domestic markets into an export contingent subsidy. For Canada, this directly contradicts the second sentence of footnote 4.
Canada further argues that Brazil’s case against Technology Partnerships Canada (TPC) appears to be based on an “’export propensity‘” and an “’intent‘” interpretation of Article 3. In Canada’s view, Brazil argues that TPC contributions are subsidies that are “’contingent, in law or in fact, on export performance‘” because the Canadian aerospace industry is an “’export-oriented‘” sector, because exports in the sector are growing at 10 per cent a year and because the “’building of exports‘” was, along with job creation, one of the objectives of TPC. Canada maintains that Brazil’s arguments are based on an incorrect reading of the law and faulty methodology in analysing the evidence, and submits that TPC is not contingent, in law or in fact, upon export performance.
Canada states that TPC is a programme administered by the Department of Industry of the Government of Canada (Industry Canada). It was established in 1996 under the authority of the Department of Industry Act,372 with a funding in the amount of $150 million in 1996/97, $200 million in 1997/98 and $250 million thereafter, with repayments and revenues recycled to sustain and increase the fund.373
According to Canada, TPC provides support to a broad base of sectors and technologies that touch on virtually all industrial sectors of Canada. Specifically, eligible sectors and technologies include the Aerospace and Defence sector (including defence conversion), environmental technologies and “’enabling‘” technologies, which include biotechnologies, information and communication technologies, and advanced materials and advanced manufacturing technologies. Canada indicates that as of September 30, 1998 TPC had approved 65 projects representing a total of
$582 million in multi-year investments; and that of these, 48 projects ($174.5 million) involved environmental and enabling technologies, with the balance going to the Aerospace and Defence sector.
Canada states that the basic objectives of the TPC programme, set out in its Charter374, include “’to maintain and build the industrial technology and skill base essential for internationally competitive products and services‘”375. Canada asserts that TPC does not have “’export performance‘” as a condition, in law or in fact, of project support. Canada submits that nothing in the
372Statutes of Canada 1995, c. 1, sections 4(1)(a), 5(c), 5(d), 6(c) and 14(1)(c) (Exh. CDN-26).
373 Technology Partnerships Canada, 1996-97 Annual Report (Ottawa: Industry Canada, 1996) at 26-
28 (Exh. CDN-27).
374 Technology Partnerships Canada, Charter, 2 December 1996 (Ottawa: Industry Canada, 1996) at 3. (Exh. CDN-28).
375 Id..
WT/DS70/R
Page 113
application documents or the contribution agreements of TPC identifies export performance as a condition for eligibility for or approval of contributions.376
Canada states that the basic objectives of TPC, approved by the Government of Canada following its 1995 Budget, are to:
be an investment approach supportive of the government’s priorities for jobs, growth and sustainable development with all repayments of TPC contributions being recycled to help fund the programme;
be strongly market driven and results-oriented;
focus on activities in environmental technologies, strategic enabling technologies (e.g., advanced manufacturing and processing, advanced material processes and applications, applications of biotechnology and applications of selected information technologies) and aerospace and defence (including defence conversion); and
adhere to the twin principles of international competitiveness and national access by putting in place the necessary programme machinery, rules and processes to ensure that competitive and capable high technology small and medium-sized enterprises from all regions of the country are encouraged to participate and have fair access to the programme.
Canada argues that TPC’s eligibility criteria reflect these objectives and are set out in the application documents.377
Canada provides Volumes I and II of the Interim Reference Binder (Binder) of the TPC programme,378 including the following documents379:
“Terms and conditions: Technology Partnerships Canada”, which sets out the eligibility criteria for the programme;
“Technology Partnerships Canada: Repayment of Contributions”, which sets out the policy guidelines for determining repayment obligations;
The “Project Summary Form” and criteria (environmental assessment forms are excluded); and
A “Statement of Work”, to be filled out by an applicant.
Canada argues that none of these documents in any way indicates or intimates that “’export performance‘” is a criterion of eligibility for the TPC programme. Canada states that Ministerial assessment criteria under TPC are not related to export performance.
For Canada, the “export propensity” of the aerospace sector is a fact of the market rather than a condition or requirement of the programme. The world aircraft industry is one of the most
376 A TPC applications kit and a TPC model contribution agreement are set out in Exhs. CDN-29 and
30.
377 Technology Partnerships Canada, Terms and Conditions, at 1 (Exh. CDN-65).
378 Exh. CDN-65
379 Canada initially provided the documents listed in this paragraph, and subsequently, in response to a
request from the Panel, provided Volumes I and II in their entirety.
WT/DS70/R
Page 114
globalised industries, with few countries producing all the necessary technology domestically and all relying on economies of scale for profitability, and the Canadian aerospace sector is no different.
Canada submits that in view of the small size of the Canadian market, the significant dependence of the Canadian economy in general and the manufacturing sector in particular on exports,380 it would not be unusual if a large proportion of the sales of Canadian manufacturers are made in markets other than Canada.381
For Canada, however, there is no requirement, in law or in fact, that the products resulting from the research and development investment by the Government of Canada be exported; there are no penalties if exports do not take place; royalties are not reduced if exports increase; repayment obligations are not in anyway affected by whether the sales are made in Canada or outside Canada. Accordingly, Canada argues, TPC contributions are not, in law or in fact, conditional on or tied to export performance.
Canada submits that Brazil has not presented a prima facie case with respect to export contingency. Canada challenges the evidence adduced by Brazil in support of its “export propensity”- and “intent”-based argument that TPC funding to the civil aircraft industry is contingent on export performance.
For Canada, evidence adduced by Brazil from Industry Canada’s annual report, acknowledging the export orientation of the aerospace sector (para. 6.175) is not relevant to the question of export contingency.
Canada also disputes as incorrect the conclusion of the “Clark Report” that all CRJ and Dash 8 sales since 1992 have been for export. (See section VI.G for detailed arguments concerning the Clark Report.) Thus, for Canada, this evidence has been rebutted, and does not establish a prima facie case.
Canada also maintains that the statements cited by Brazil of Canadian Ministers acknowledging the importance of the aerospace sector to the Canadian economy and that the creation of jobs and building of exports was an objective of TPC is not relevant to the question of export contingency.
That is, Canada argues, Brazil has not adduced any evidence to show that TPC contributions are contingent on export performance, in the sense that contributions would not be paid unless exports took place, that there would be rewards if exports took place or that there would be penalties if exports did not take place. Canada maintains that there is no such evidence because this is not how TPC contributions are made.
According to Canada, TPC’s area of activity is defined very broadly, and TPC has invested in projects ranging from closed loop, zero-effluent environmental solutions for the pulp and paper industry, research into a vaccine for cancer, precision laser welding technology to advanced flight simulators. Canada states that the TPC programme is thus aimed at building and improving Canada’s
380In 1997, the total value of Canadian exports was Can$344 bn, constituting 40 per cent of Canada's GDP. See Statistics Canada, “Gross domestic product (expenditure-based)”, (1998) online: Statistics Canada Homepage http://www.statcan.ca/english/Pgdb/Economy/Economic/Econ04.htm (date accessed: 14 November 1998) (Exh. CDN-31).
381 Exports constitute 70 per cent of the total sectoral production. 53.5 per cent of manufacturing shipments were exported in 1996. Canadian Industry Statistics Development Team, “The Canadian Economy”
(1998) Strategis, online: Industry Canada homepage http://strategis.ic.gc.ca/SSC/ci00034e.html (date accessed: 14 November 1998) (Exh. CDN-32).
WT/DS70/R
Page 115
technological and international competitiveness. TPC invests in products and technologies that advance this general objective.
Canada argues that the TPC is not contingent on export performance because it is neutral as to the destination of a successful product or the fruits of a successful technology. That is, a contribution is made where a product or a technology has the potential of achieving market success, either domestically or internationally.
According to Canada, eligibility for the programme does not depend on export performance: export performance does not increase TPC’s share of the costs that are eligible under the programme
that is, there will not be more TPC contributions if there are exports; there are no penalties if there are no export sales; there are no rewards if products are exported. Rather, practically the opposite is true: higher sales – export or domestic – mean that a producer has to pay more in royalty payments to the Government of Canada.
Canada submits that at no point in the process of qualifying for a TPC investment is there a requirement that exports take place: the eligibility criteria for the TPC do not include export performance; in applying for a TPC contribution, an enterprise is not required to show that it makes any exports; and in entering into an agreement to receive TPC contributions, an enterprise is not required to show or promise that it will export. Canada notes the example of a contribution for the treatment of tar sands in Athabasca in Northern Canada, indicating that there is no requirement there that the process that is developed is then exported. For Canada, this is how TPC is applied in the aerospace sector; thus the TPC programme is not contingent on export performance.
According to Canada, TPC contributions are conditional on success – and on presenting a successful business plan. Canada argues that regardless of the destination of sales in a business plan, the questions that the administrators of the programme will put to the proponent of the project will not concentrate on the export aspects of the project, but rather on whether the project is going to be viable, i.e., whether the business plan makes sense so that the contribution is not wasted. In answer to a Panel question regarding assessment criteria, Canada indicates that “other assessment criteria” that may apply include regional distribution; sectoral balance (both within and across eligible sectors); the need for an appropriate mix of projects between large firms and small to medium enterprises (SME); availability of funds; and portfolio management considerations (for example, balanced risk profile, matching cash flows to requirements). Also in response to a Panel question regarding performance indicators for projects in the aerospace and defense sector, Canada states that there are no performance indicators unique to projects in the Aerospace and Defence sector, but that rather the following key indicators (and as reported in the 1997/98 TPC Annual Report) are tracked for the Programme as a whole: progress on SME delivery: the distribution of the number of projects and dollars of investment by size of recipient; regional balance: the distribution of the number of projects and dollars of investment by region; sectoral balance: the distribution of the number of projects and dollars of investment by component (TPC’s target being that the Environmental and Enabling Technologies component of the programme utilizes one-third or more of the programme’s budget by 1998-99; leverage: dollars of private sector investment in innovation per dollar of TPC investment; sharing ratio: the weighted average sharing ratio (TPC’s target being to average below 33 per cent; job creation: the number of forecasted jobs created or maintained over the life of the investments; and economic activity: the value of forecasted sales over the life of the investments.
Canada states that where a recipient fails to make any sales – including export sales – the investment of the Government of Canada in the project will have failed. According to Canada, the recipient does not suffer additional penalties because it did not sell into export markets, and if additional contributions by the government are due, they are not terminated because exports do not take place, nor are royalties payable to the government increased on domestic sales if export sales are not made.
WT/DS70/R
Page 116
Canada also argues that no distinction is made between export and domestic sales for the purpose of TPC royalty payments. That is, Canada states, export royalties are not lower than domestic ones.
Regarding the phrase “maintain and build upon the technological capabilities and production, employment and export base” of Canada in the TPC eligibility criteria for the aerospace and defence sector, which was the subject of a Panel question, Canada states that subsidies that develop a country’s global competitiveness and thus maintain and build upon its export base are not inconsistent with the SCM Agreement for that reason alone. To so argue in Canada’s view would be to suggest that the SCM Agreement permits only subsidies that are, at best, neutral as to competitiveness and productivity, which would simply render illegal just about every industrial and labour adjustment programme the world over.
Thus, for Canada, the fact that an increase in global competitiveness was an objective of the TPC does not make the TPC, or contributions given under it to the aerospace sector, illegal under Article 3, as governments do not generally make investments to make their economies less competitive.
Responding to a Panel question regarding advice from the Department of Foreign Affairs and International Trade to TPC, Canada states that the Government of Canada through TPC tries to spur investment by the private sector in technologies and sectors that are considered to be important for Canada’s competitiveness and for increased productivity. In the process, Canada states, the Government wishes to ensure that it shares in the upside benefits of its investments, so that new investments can be made, and in doing so has to ensure that a proposed project has a market. According to Canada, in determining the strength of this market, the Government takes advantage of all the information available to it, including the expertise of the Department of Foreign Affairs and International Trade.
In response to a Panel request for full details and documentation relating to the evaluation and decisions relating to the five TPC contributions acknowledged by Canada, Canada states that most of the information requested by the Panel is highly sensitive business confidential information, and that Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. According to Canada, these private parties, and others in the process of submitting applications under the TPC programme, have already expressed reluctance in sharing information, or additional information, concerning their business plans, and such reluctance, if it were to continue, would have a serious deleterious impact on the functioning of the TPC programme.
Canada states that it has requested the interested private parties to release Canada from obligations arising under the business confidentiality clauses of Canada’s arrangements with them. With the exception of Bombardier, these interested parties have indicated that they are not prepared to allow Canada to release business confidential information, or have not responded to the request for release382. Canada indicates that Bombardier agreed to release specific business confidential documents relating to the CRJ-700 programme that illustrate the operation of TPC, and that exceptionally sensitive confidential information in those documents was redacted to protect the commercial interests of Bombardier.383
Canada states that it is unable to comply with the Panel’s request for all project assessments and funding decisions documents. Since the level of contribution was in excess of $20 million, Cabinet approval was required for this investment, and therefore according to Canada, the recommendation, options, communications strategy, supporting rationale and analysis are contained in
382 Exh. CDN-106
383 Exh. CDN-107
WT/DS70/R
Page 117
a Memorandum to Cabinet (MC) and as such constitutes a cabinet confidence and cannot be divulged. Similarly, the Project Summary Form (PSF) in this case was required to be presented to the Minister of Industry Canada for signature, and therefore constitutes Ministerial advice that Canada states cannot be released.
Canada summarizes the basis for its investment in the CRJ-700 project, indicating that the key considerations were: that the timing of the project coincided with planned Canadair workforce reductions and would create or maintain 1,000 jobs during the development phase alone; that the CRJ- 700 was planned to advance the state-of-the-art in regional jet operating performance and would provide tangible benefits to operators in Canada and abroad; that the project would offer real opportunities for participation by domestic aerospace sub-contractors to expand their sales base and enhance their technical capabilities; that project risk was considered manageable in light of Bombardier’s proven track record and because the technical risk was mitigated by the fact this aircraft was a stretch version of an existing platform - the CRJ50; that market risk was also manageable given the CRJ50's strong position in the regional jet market, and the preference by airlines to utilize families of aircraft to reduce part inventories and training requirements; and that independent market forecasts and Industry Canada’s sectoral experts confirmed that the size of the potential market for a 60-90 seat regional jet to be in excess of 1,000 units through 2010.
Canada also indicates that there are no Memoranda of Understanding between TPC and relevant companies. The rights and obligations of the Government of Canada and the applicants are set out in the contribution agreements.
(c) Comments of Brazil
Brazil notes several documents included in the TPC reference binders which in its view demonstrate that TPC’s funding decisions for the aerospace and defense industry are tied to export:
Section 3.2.3 of TPC’s “Terms and Conditions” states that “[c]ontributions under the Aerospace and Defence component will be directed to projects that will maintain and build upon the technological capabilities and production, employment and export base extant in the aerospace and defence sector” (emphasis added by Brazil).
Section 3.3 of the TPC Charter, entitled “Aerospace and Defence (including Defence Conversion)” and included in TPC Reference Binder 2 states that “[i]nvestments will be directed to projects that build on and maintain technological capabilities and the production, employment and export base of the sector” (emphasis added by Brazil).
Part B of Schedule B of the TPC Aerospace and Defense Generic Model Agreement specifically calls for any representations by an applicant regarding “export markets penetration through marketing partnership agreements with foreign companies.”
Schedule C from the same Model Agreement, representing a form for “Report[s] on Estimated & Actual Sales and Royalties” requires the reporting of export sales revenues.
Page 10 of the 1996-1997 TPC Business Plan notes that TPC’s “approach” in the aerospace and defense sector is to “[d]irectly support the near market R & D projects with high export market potential.”
Page 12 of the 1996-1997 TPC Business Plan records the proportion of the aerospace and defense industry’s revenue allocable to exports.
According to Brazil, export orientation and performance are central, therefore, to TPC’s decision-making processes in the aerospace and defense sector, since maintaining exports is identified
WT/DS70/R
Page 118
as a key factor to consider in awarding TPC grants and evaluating TPC projects, and since information about a project’s export performance is requested and maintained by TPC. For Brazil this represents further evidence demonstrating that TPC grants to the industry are “’in fact tied to actual or anticipated exportation or export earnings‘” within the meaning of Article 3 of the SCM Agreement.
In Brazil’s view, the information provided by Canada regarding the $87 million CRJ-700 grant is so incomplete as to make it wholly unreliable, noting that information regarding financing terms, repayment plans, operations summaries, the timing of government disbursements, estimated development costs and projected programme revenues has been redacted from the first ten documents included in BCI Tab 1. Brazil asserts that much of the redacted information may have given the Panel information about whether the grant was contingent on export.
Finally, Brazil states, despite the Panel’s specific request, Canada has refused to provide project assessments and funding decision documents, including a Cabinet Memorandum detailing a “’recommendation, options, communications strategy, supporting rationale and analysis,‘” and a Project Summary Form. Brazil observes that Canada’s decision to withhold this information is based not upon the refusal of private parties to waive their confidentiality rights, but instead upon the Canadian government’s own refusal to release documents prepared for itself by itself. Brazil submits that these documents likely provide insight into the very issue central to the dispute between Canada and Brazil regarding TPC grants to the Canadian regional aircraft industry - whether those grants are “’in fact tied to actual or anticipated exportation or export earnings.‘” As a result of Canada’s strategic decision to withhold these documents, Brazil contends, the Panel should adopt adverse inferences, presuming that the information contained therein demonstrates that TPC grants to the Canadian regional aircraft industry are in fact tied to actual or anticipated export (paras. 4.146-4.151). Brazil recalls the evidence it has submitted supporting this conclusion (paras. 6.219-6.229), and argues that Canada’s decision to withhold documentation that could cast more light on the matter requires the Panel to conclude that Canada is hiding incriminating information.
SALE BY ONTARIO OF OWNERSHIP SHARE IN DE HAVILLAND
Subsidy
Arguments of Brazil
Brazil asserts that in January 1992, Bombardier and the Government of Ontario, through the Ontario Aerospace Corporation (“OAC”), purchased 51 and 49 per cent, respectively, of the shares in Boeing’s de Havilland Division, for a total purchase price of Can$100 million,384 and that at the same time, Bombardier reserved the option to purchase from the Government of Ontario the latter’s remaining 49-per cent share in de Havilland for a total of Can$49 million. Brazil states that in January 1997, Bombardier exercised this option, issuing to the Government of Ontario and the Ontario Aerospace Corporation (“OAC”) a 15-year promissory note bearing interest at seven per cent with annual principal repayments of Can$4.9 million in years six through 15.385 As a result, according to Brazil, increases in de Havilland’s equity did not accrue to OAC, whose return was strictly limited to its initial investment in de Havilland. Brazil further argues that OAC does not even appear to anticipate payment of Bombardier’s $49 million note, as according to Brazil, the Government of Ontario forgave the obligation effective 31 March 1996.386 Brazil argues that Bombardier therefore has at its disposal OAC’s equity capital, without the prospect of ever having to pass along an equity investor’s reward to OAC.
384 Bombardier press release, dated 29 January 1997 (Exh. BRA- 48).
385 Id. See also Bombardier 1996-1997 Annual Report, pgs. 49 (note 2), 55-56 (note 9) (Exh. BRA-
49).
386 Ontario Public Accounts, 1995-96, re. Ontario Aerospace Corporation, note 4 (Exh. BRA- 50). See
also Exhibit B, Table B.8 of Finan Report.
WT/DS70/R
Page 119
Brazil also draws attention to a statement in Bombardier’s Annual Report387 that Bombardier’s payments on the $49 million note to OAC are “‘reimbursable in anticipation under certain conditions‘”, and submits that the Panel should request that Canada explain completely what these conditions are, and why this does not mean that Bombardier will be reimbursed by some party for its payments on the $49 million loan.
Brazil further argues that during the period 1992-1997, OAC provided to de Havilland four distinct types of subsidies which made the purchase of OAC’s 49-per cent interest even more attractive to Bombardier: interest-free shareholder loans in the amount of Can$200 million; cash grants (under DIPP) of up to Can$100 million; sales financing support for de Havilland’s Dash 8 aircraft, including coverage of losses; and reimbursement of restructuring costs of up to Can$370 million,388 and OAC’s offer to forgive all loans to de Havilland, absent certain undefined events of default.389
For Brazil, OAC’s purchase of a 49-per cent interest in de Havilland and its simultaneous granting of an option to Bombardier to purchase that interest, within five years, for the same price initially paid by OAC, amounted to a cost-free option to Bombardier; Bombardier, in effect, got to play with OAC’s money. According to Brazil, if de Havilland did not prosper, then OAC was the loser, while if de Havilland did prosper, then Bombardier was the winner. Ultimately, Brazil argues, Bombardier exercised the option and took over OAC’s interest, and Bombardier’s note for payment of OAC’s remaining 49-per cent interest was subsequently forgiven. For Brazil, the Government of Ontario, through this action alone, conferred a benefit on Bombardier.
Moreover, the asserted extensive list of interest-free, non-repayable loans and grants provided by OAC to de Havilland during the period 1992-1997, totalling in excess of Can$300 million, confers according to Brazil a significant and obvious benefit to Bombardier. According to Brazil, the present value of these contribution is $874.7 million. For Brazil, it is inconceivable that OAC’s 49-per cent share in de Havilland was worth the same in 1997, when purchased by Bombardier, as it was when initially acquired in 1992, given the extensive array of subsidies provided to de Havilland by OAC in the interim period. Brazil states that as a result of these subsidies, Bombardier received a massive windfall in 1997 when purchasing OAC’s shares in de Havilland at a price fixed in 1992, before those subsidies made de Havilland a cash-rich venture. Brazil states that increases in de Havilland’s equity resulting from these government contributions did not accrue to Ontario, whose return was strictly limited to its initial investment in de Havilland. Rather, the entire benefit flowed to Bombardier.
Brazil maintains that the accrual of this subsidy benefit occurred in January 1997 when Bombardier purchased OAC’s remaining share of de Havilland. In response to a question from the Panel, Brazil clarified that it does not consider that the various contributions in the previous paragraph were themselves subsidies contingent upon export performance. Rather, they amplified the “benefit,” within the meaning of Article 1.1 of the Subsidies Agreement, accruing to Bombardier when it purchased the remaining share of de Havilland in January 1997. Brazil maintains that contributions, the present value of which constitute $874.7 million, were made by the Canadian and Ontario Governments to de Havilland between 1992, when the $49 million price was set, and 1997, when Bombardier completed the purchase of de Havilland. Yet, the purchase price remained unchanged from that set in 1992. As a result, according to Brazil, it did not reflect the value added by these contributions.
387 Exh. BRA-49
388 Ontario Public Accounts, 1992-93, re. Ontario Aerospace Corporation (Exh. BRA- 51). See also Exhibit B, Table B.8 of Finan Report.
389 Ontario Public Accounts, 1992-93, re. Ontario Aerospace Corporation, note 4 (Exh. BRA- 51).
WT/DS70/R
Page 120
Brazil estimates the present value in 1998 of benefits to Bombardier related to the acquisition and restructuring of de Havilland to be $874.7 million.390 For Brazil the amounts granted constitute both a transfer of funds and a massive benefit to Bombardier, within the meaning of Article 1.1 of the SCM Agreement.
Arguments of Canada
In answer to a request from the Panel for documentation regarding the determination of the interest rate on the sale, Canada states that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Canada notes that Brazil is in agreement with Canada regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Canada does not consider it appropriate to adduce evidence in response to what it views as baseless and false allegations and a case that has not been established, and in support of a defence Canada has not made. Accordingly, to the extent that any documents are produced by Canada in response to this question of the Panel, Canada states, they are provided to support Canada’s submission that the contributions at issue are not “’contingent on export performance‘” within the meaning of Article 3 of the SCM Agreement.
As a factual matter, Canada denies Brazil’s allegation that “’Bombardier’s note for payment of OAC’s remaining 49-per cent interest was subsequently forgiven’”. According to Canada, the Government of Ontario, forgave its loan to the Ontario Aerospace Corporation -- Ontario’s own wholly-owned corporation -- on 31 March 1996. According to Canada, the Finan Report concludes, without any evidence whatever, that “’[i]t is clear that OAC does not expect to receive payment from Bombardier for its 49 per cent share in de Havilland.‘”391
Canada states additionally that the promissory note of Bombardier was not forgiven. Canada confirms that the sale price of $49 million, agreed to in 1992, was paid in the form of a promissory note by Bombardier to the Government of Ontario at the rate of 7 per cent; a copy of this promissory note was adduced into evidence by Canada.392 Canada states that the interest rate was determined by the Ontario Financing Authority to be the appropriate market rate of interest to provide a commercial basis to the transaction. Canada submits a memorandum of January 1997 detailing the interest rate determination, in response to the Panel’s request. Canada maintains that interest payments owing to date have been paid by Bombardier, indicating that the first payment of $3,430,000 was received by the Government of Ontario in January 1998. A copy of the cheque for this amount was adduced by Canada into evidence.393
Canada further indicates that that on January 28, 1997, Ontario announced the sale of its 49 per cent interest in de Havilland Holdings Inc. to Bombardier. The sale took place on the basis of a “put/call” option on the purchase of de Havilland Holdings Inc. by Bombardier and Ontario in 1992. The exercise period for this option was 1 February 1996, to 31 January 1997. Contrary to Brazil’s assertion,394 Canada asserts, the option of selling its interest in de Havilland was exercised by Ontario.395
390 Exhibit B, Tables B.9 and B.10 of Finan Report.
391 See also Section VI.F for the parties’ comments on the Finan Report regarding the de Havilland sale. TPC Annual Report 1996-1997, p. 5 (Exh. BRA- 5); Industry Canada, News Release, 17 Dec. 1996 (Exh. BRA- 4).
392 Exh. CDN-49
393 Exh. CDN-51
394 Brazil asserts that:
“At the same time, Bombardier reserved the option to purchase from the Government of Ontario the latter’s remaining 49-per cent share in de Havilland for a total of Can$49 million. In January 1997, Bombardier exercised this option, issuing to the Government of Ontario and the Ontario Aerospace Corporation (“OAC”) a 15-year promissory note bearing interest at seven per cent with
WT/DS70/R
Page 121
Regarding the $49 million sale price of de Havilland determined in 1992, Canada argues that whether or not de Havilland was solvent, whether or not its equity had increased, whether or not the restructuring loans would add value to de Havilland, the sale price in 1997 would be what had been agreed to in 1992. Canada explains that the Government of Ontario agreed to the sale price in 1992 because Boeing had poured $1.3 billion in de Havilland, only to sell it for $100 million after 6 years of operation. According to Canada, Ontario’s participation in the project to revive de Havilland in 1992 was conditional on ensuring that at the very least Ontario would get its purchase money back.
Regarding the equity value of de Havilland at the time of the 1997 sale to Bombardier, in request to a question from the Panel, Canada submits a letter from the Vice President Legal Services of Bombardier. According to the letter, at the point of its sale to Bombardier by Ontario, de Havilland’s audited statements showed that it had a negative equity value. The Panel requested copies of the audited statements and accompanying notes, and in response Canada states that the information requested by the Panel – detailed audited statements of de Havilland along with all accompanying notes and any other relevant documentation - is highly sensitive business confidential information that is not in the possession of the Government of Canada or the Government of Ontario, and that Canada’s desire to present to the Panel such information as may help it to arrive at a decision must be balanced against the concerns of private parties not Party to this dispute.
Canada further asserts that Brazil has adduced no evidence in support of its assertions about the increase in de Havilland’s value as a result of alleged subsidies between 1992 and 1997. In Canada’s view, Brazil’s unfounded and incorrect allegations about the alleged forgiveness of the
$49 million promissory note show that Brazil’s assertions about de Havilland suffer from a fundamental lack of credibility, and that Brazil’s allegations about the rise in value of de Havilland should be viewed in that light.
Response of Brazil
Regarding Canada’s citation to Brazil’s belief in the principle of “judicial economy”, Brazil notes that in a letter to the Panel dated 13 December, reacting to the list of questions presented to Canada by the Panel, Brazil stated that in the course of these proceedings, Canada had made a deliberate decision not to present evidence rebutting Brazil’s claims. In light of that decision, Brazil stated, the Panel could, consistent with the principle of judicial economy, have decided the issues raised in this case on the basis of the facts before it, rather than offer Canada yet another chance to present information in its defense.
Brazil disagrees with Canada’s argument that “‘Brazil’s central allegation’” is that Ontario forgave Bombardier’s $49 million loan to Bombardier for the purchase in 1997 of Ontario’s remaining share in de Havilland. According to Brazil, paying this $49 million debt will not bring Bombardier anywhere close to offsetting the tremendous benefits that accrued to Bombardier when it purchased the remaining shares of de Havilland in 1997, given the contributions with a present value of $874.7 million made by the Canadian and Ontario Governments to de Havilland between 1992, when the $49 million price was set, and 1997, when Bombardier completed the purchase of de Havilland (paras. 6.271)396. Yet, the purchase price remained unchanged from that set in 1992, accruing a massive benefit to Bombardier when it completed the transaction in 1997.
annual principal repayments of Can$4.9 million in years six through 15.” [emphasis added by Canada ] (para. 6.254).
395 Government of Ontario, News Release, “Ontario Recoups Taxpayer Investment in De Havilland” (28 January 1997) (Exh. CDN-34).
396 Finan Report, pg. A.9-A.10.
WT/DS70/R
Page 122
Regarding the letter from Bombardier concerning de Havilland’s equity value at the time of the 1997 sale to Bombardier, Brazil comments first that even if de Havilland’s net equity value was negative at the time of its sale in 1997, this does not change the fact that it would have been even more negative if government contributions, the present value of which constitute $875 million, had not been made to de Havilland between 1992 and 1997. According to Brazil, these contributions increased the value of Ontario’s equity beyond the price set in 1992, but Ontario was given no share in this increase.
Second, Brazil states, a negative net equity value does not give the complete picture of the value of a shareholder’s equity. For Brazil, it does not mean that de Havilland was worth nothing – or less than nothing – to Ontario and Bombardier when purchased in 1997, noting for example that de Havilland’s book of orders at the time of the sale397, although not factored into net equity value, represented value which an equity investor selling its shares could reasonably expect to be captured in the selling price of its shares.
Regarding Canada’s response to the Panel’s request for the audited financial statements of de Havilland, Brazil argues that Canada has failed to supply the background documents supporting the statements in the letter submitted as evidence in support of its defense, and that Canada has failed effectively to rebut evidence submitted by Brazil from the Public Accounts of Ontario detailing contributions, the present value of which constitute $875 million, made by the Canadian federal and Ontario governments to de Havilland between 1992, when the $49 million purchase price for Ontario’s share in de Havilland was set, and 1997, when Bombardier completed the purchase of Ontario’s share. For Brazil, Canada’s argument that this injection of capital did not affect the value of Ontario’s equity lacks credibility.
Comments of Canada
Canada comments regarding Brazil’s argument concerning de Havilland’s order book that Brazil has adduced no evidence as to what the value of the de Havilland order book was on the date of sale, nor demonstrated that the value would have been sufficient to compensate the purchaser for the negative net equity value of de Havilland at the time. Thus, Canada submits, Brazil has not demonstrated that a benefit was conferred by the sale. Second, for Canada, any such valuation would have to reflect the possibility that orders might not result in actual sales. Third, according to Canada an order book does not indicate the viability of an aircraft manufacturer: Fokker’s order book just prior to its bankruptcy in early 1996 contained firm orders for 81 aircraft, according to a table submitted by Canada on “Fokker Orderbook – As of December 1995”, source: Lundkvist.
Export contingency
Arguments of Brazil
According to Brazil, as part of the transaction Bombardier is required “’to operate the de Havilland business with a view to continuing aircraft final assembly . . .‘”, which for Brazil constitutes a covenant making the transaction contingent upon maintenance of, among other things, de Havilland’s current export-oriented sales strategies. Brazil asserts that at the time this language was drafted, de Havilland’s order book stood at 61 aircraft, all but one of which was sold for export. In other words, Brazil states, Bombardier received a substantially discounted price for the purchase of de Havilland, in exchange for what the Government of Ontario anticipated would be the Dash 8’s continued export orientation. Brazil argues that Bombardier has held to this commitment; as established by the Clark Report, all of de Havilland’s Dash 8 aircraft sold since 1992 have been sold
397 The Clark Report notes that at the time of the sale, de Havilland had firm orders for the sale of 61 Dash 8 series aircraft. Clark Report, pgs. 39-40.
WT/DS70/R
Page 123
solely for export.398 Therefore, Brazil argues, the benefits accruing to Bombardier of OAC’s investment in de Havilland were tied to anticipated, if not to actual exportation or export earnings and thus are prohibited export subsidies within the meaning of Article 3 of the SCM Agreement.
Arguments of Canada
Canada describes the factors forming the background to the 1992 commitments by the Province of Ontario and the Government of Canada in support of the restructuring of de Havilland,399 noting that Boeing had owned de Havilland since 1986, and while the Dash 8 continued to be viewed as a superior product in the market, de Havilland’s losses throughout the late 1980s – amounting to
$1 billion by 1992 – persuaded Boeing to end its foray into the turboprop market. Canada states that sharply decreased orders following the recession of the early 1990s and aggressive sales financing and pricing practices by subsidised competitors had a considerable impact on the company’s finances, such that by the end of 1991, Boeing indicated that it intended to close de Havilland if a suitable purchaser could not be found.
Canada submits that in view of the contribution of de Havilland to the economy of the province, both in absolute monetary terms and in terms of strategic depth, the Government of Ontario and Bombardier examined the potential for long-term viability of de Havilland, and came to the conclusion that an appropriate business plan, a revitalized Dash 8 programme, continuing productivity improvements and diversification of business operations through involvement in other Bombardier aircraft programmes could help de Havilland in attaining long-term commercial strength. Canada states that it was also clear that restructuring would take several years and that the risk of failure was significant.
According to Canada, on 22 January 1992 Ontario and Bombardier entered into an asset purchase agreement with Boeing to acquire the assets of its de Havilland division, and the deal closed on 9 March 1992. At that time, de Havilland Holdings Inc. (DHI) acquired the assets of de Havilland. Canada states that DHI was capitalized at Can$100 million, and the equity was fully invested in de Havilland. For Canada, Ontario’s equity investment ensured that the Province would be a partner in decisions on the future direction of de Havilland. Ontario appointed three of seven members of the board of directors.
Canada argues that the Shareholders’ Agreement, concluded in 1992 and binding on both parties, included a “put” option: Ontario had the right to require Bombardier to purchase its equity for Can$49 million. In opting for a fixed price, Canada submits, Ontario had decided to forego any upside potential in the value of de Havilland following restructuring. At the same time, Canada states, given the significant possibility of the failure of restructuring, Ontario’s decision on the fixed price eliminated any downside potential as well. Canada notes that a five-year restructuring time-frame was established; the exercise period was set for 1 February 1996 to 31 January 1997.
According to Canada, in June 1995 the Government of Ontario changed, and the new Government was committed to withdrawing from certain functions in the marketplace and to selling certain government assets. Canada argues that by February 1996, Ontario determined that de Havilland had emerged as a viable entity, that continued involvement of the Province in de Havilland was no longer necessary, and that it could exercise its “put” option which right Ontario exercised
398 Id.
399 See Minister of Industry, Science and Technology and Minister for International Trade, News Release, “Wilson welcomes purchase of de Havilland and provides federal assistance”, 22 January 1992, (Exh.
CDN-66) and Ministry of Industry, Trade and Technology, News Release, “Ontario and Bombardier purchase de Havilland”, 22 January 1992 (Exh. CDN-67) listing the commitments made by the two governments at the time of the purchase of this company by the Ontario Government and Bombardier from Boeing in 1992.
WT/DS70/R
Page 124
within the permitted period, with the sale of Ontario’s equity position in de Havilland Holdings Inc. concluded on 28 January 1997.400
Regarding the covenant,401 which it argues is directly relevant to the question of de jure export contingency, Canada submits that the covenant required Bombardier to keep de Havilland running as an aircraft manufacturer if it made commercial sense, nothing more. Canada argues that if it makes commercial sense to sell aircraft domestically, then de Havilland will sell aircraft domestically and stay in business, and states that indeed, given that 20 per cent of all Dash 8s -- the most popular product manufactured by de Havilland -- in operation are in the Canadian market, de Havilland can expect to make more sales as required by market expansion and fleet renewal in Canada. Canada argues that this fact is well recognised by Brazil, which notes that the price of the sale had been agreed to long before -- in 1992. Canada maintains that the sale price was not conditional on or tied to export performance by de Havilland. It could not have been raised or reduced, and would not be raised or reduced, by virtue of the market into which de Havilland makes its sales. Accordingly, Canada argues, the sale of de Havilland by the Government of Ontario was not contingent, in law or in fact, upon export performance, and Brazil has not presented a prima facie case in this regard.
Canada also argues that Brazil overlooks the fact that, in line with the covenants it entered into, Bombardier has invested hundreds of millions of dollars in diversifying de Havilland’s operations. According to Canada, De Havilland now produces the wing for Bombardier’s new Learjet 45 and does the final assembly on the new intercontinental Global Express business jet, and the Learjet 45 was certified in late 1997, and the Global Express in mid 1998. Canada notes that there are firm orders from Canadian customers for both these aircraft.
For Canada, the important point is that Bombardier will not get a reduction in the sale price if it sells more into export markets, and will not pay a penalty -- through, for example, a prospective increase in the sale price -- if all of its sales are in the Canadian market.
CANADA-QUÉBEC SUBSIDIARY AGREEMENT ON INDUSTRIAL DEVELOPMENT
Subsidy
Arguments of Brazil
Brazil argues that pursuant to the 1984 Economic and Regional Development Agreement, the governments of Canada and Québec entered into two Canada-Québec Subsidiary Agreements on Industrial Development (the “Subsidiary Agreements”), signed in 1985 and 1992, which together pledged Can$743 million in assistance to industrial projects in Québec.402 According to Brazil, although the Subsidiary Agreements were to have expired effective 31 March 1998,403 programmes receiving funding through the Subsidiary Agreements were announced in June 1998.404
Brazil submits a study prepared by the Canadian Reform Party405, which it cites to illustrate that the Canadian regional aircraft industry has received “conditionally-repayable” contributions under the auspices of the Subsidiary Agreements. Brazil notes that the spreadsheet included in the
400 See Ministry of Economic Development, Trade and Tourism, News Release, “Ontario recoups taxpayer investment in de Havilland” 28 January 1997 (Exh. CDN-33).
401 Exh. CDN- 68
402 Report of the Auditor General of Canada to the House of Commons, November 1995, p. 21-8 (relevant chapter of Report – Exh. BRA-52).
403 Amendment No. 2 to the 1992 Subsidiary Agreement (Exh.BRA-59).
404 “Industry Canada - Contribution to Aeterna Laboratories Inc. under Canada-Québec Subsidiary Agreement on Industrial Development,” Canada NewsWire, 15 June 1998 (Exh. BRA-60).
405 Exh. BRA-58
WT/DS70/R
Page 125
study is limited to funding granted before 22 October 1996, the date of the study. According to Brazil, Canada has not been forthcoming concerning Subsidiary Agreement assistance to the regional aircraft industry after this date (or concerning details of the terms of assistance granted prior to 22 October 1996 and included on the Reform Party spreadsheet). In response to a Panel question regarding which of the transactions identified in that spreadsheet Brazil alleged constitute subsidies within the meaning of SCM Article 1, Brazil replies that it offered the list of transactions to illustrate that the Canadian regional aircraft industry has received “conditionally-repayable” contributions under the auspices of the Subsidiary Agreements, and that Brazil does not consider that any of the specific contributions listed themselves constitute export subsidies to the Canadian regional aircraft industry prohibited by Article 3 of the SCM Agreement.
Brazil submits406 an extract from the 1996-97 Public Accounts of Canada for the Department of Industry, which shows that Bombardier’s Canadair division received funds under the Subsidiary Agreements during that period. According to Brazil, Canada has not been forthcoming with information within its sole control about Subsidiary Agreement assistance to the Canadian regional aircraft industry. Brazil believes that the evidence it has presented is sufficient to warrant further questions from the Panel to Canada concerning the existence and terms of repayment for Subsidiary Agreement assistance to the regional aircraft industry.
In response to a question from the Panel asking Brazil to cite instances of Subsidiary Agreement assistance in the form of “non-repayable contributions” provided to the civil aircraft industry, Brazil replies that Schedule B of the 1992 Subsidiary Agreement407 states that assistance under the Subsidiary Agreement may be granted in the form of “a repayable or non-repayable contribution.” Brazil argues that the Canadian Reform Party study that it submits as an exhibit408 demonstrates that as an historical matter certain assistance granted Bombardier under the auspices of the Subsidiary Agreements has been characterized as “’non-repayable‘” or simply as a “’contribution.‘” Brazil argues that Canada refuses to release information which would permit Brazil, and the Panel, to determine whether non-repayable contributions under the Subsidiary Agreements subject to the SCM Agreement have in fact been made. Brazil considers that it has presented evidence sufficient to establish its prima facie case, and that the burden of production has shifted to Canada to rebut that case.
Brazil also argues that, according to Canada’s notification to the WTO’s SCM Committee, certain funds provided under the auspices of the Subsidiary Agreements are in the form of “non- repayable contributions,”409 and as such confer significant benefits on recipients within Article 1.1 of the SCM Agreement.
Brazil further argues that even those contributions classified as “repayable” should be considered to confer benefits within the meaning of Article 1.1, noting its arguments (para. 6.323) as to why it considers conditionally-repayable contributions failing to compensate for various factors to constitute subsidies within the meaning of Article 1.1 of the Agreement.
406 Exh. BRA-89
407 Exh. BRA-57
408 Exh. BRA-58
409 Communication at para. XXIX:5 (Exh. BRA- 56). See also 1992 Canada-Québec Subsidiary
Agreement on Industrial Development, Schedule B, p. 31 (Exh. BRA- 57). Canada has not disclosed sufficient information concerning the Subsidiary Agreement to determine whether even those contributions deemed repayable were in fact granted at concessionary rates.
WT/DS70/R
Page 126
Arguments of Canada410
In answer to a Panel request for information concerning alleged loans made under the 1992 Subsidiary Agreement to Rolls-Royce and Lamines CTEK, Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Canada notes that Brazil agrees with Canada regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Accordingly, Canada states, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to further support Canada’s submission that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
Canada also states that Brazil has made no specific allegation about contributions to Rolls Royce or Lamines CTEK, nor adduced any evidence in support of its allegations concerning “benefits” under the Subsidiary Agreement. Canada does not consider it appropriate to adduce evidence in response to allegations not made and a case that has not been established. In addition, Canada states that the contribution to Lamines CTEK was in respect of certain electronic materials not related to the civil aircraft sector, and as such is not within the Panel’s jurisdiction.
Finally, Canada indicates, most of the information requested by the Panel is sensitive business confidential information, and Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. Canada notes that with the consent of the Government of Québec, it provides the Rolls Royce contribution agreement, as Business Confidential Information, with clauses relating to repayment terms blocked. 411
In response to a question from the Panel, Canada indicates that a $2.25 million contribution identified in the report by the Auditor General involved pharmaceuticals, and further, that the Auditor General’s use of the term “repayable” includes conditionally repayable. According to Canada, all contributions under the Subsidiary Agreement are repayable, over time, in fixed annual payments. Canada states that there are three conditionally repayable contributions, based on sales, and that all other contributions are unconditionally repayable.
Comments of Brazil
Regarding Canada’s citation to Brazil’s belief in judicial economy, Brazil recalls its reference to judicial economy in its letter to the Panel dated 13 December 1998 (see para. 6.281).
Brazil states that Canada appears not to have provided all of the information requested by the Panel concerning Subsidiary Agreement assistance. First, Brazil notes, the repayment terms are redacted from the agreement setting out the terms of the $2.5 million contribution to Rolls Royce. According to Brazil, the Panel should therefore adopt adverse inferences, presuming that these terms tie repayment to export performance (paras. 4.146-4.151). Furthermore, Brazil notes that while Canada has provided the Rolls Royce contribution agreement, it appears to have provided as “Annex A” thereto the project description for the Lamines CTEK contribution, rather than the project description associated with the Rolls Royce contribution. Brazil submits that the Panel should adopt adverse inferences, presuming from Canada’s failure to produce the annex containing the project description for the Rolls Royce contribution that this description contains inculpatory information suggesting that the contribution is tied in fact to actual or anticipated export.
410 See also Section VI.F for the parties’ comments on the Finan Report regarding the Subsidiary Agreement.
411 See BCI Tab 4.
WT/DS70/R
Page 127
Export contingency
Arguments of Brazil
Under the Subsidiary Agreements, Brazil states, the governments of Canada and Québec jointly fund major industrial projects aimed at improving the competitiveness of Québec’s economy. The purpose of the Subsidiary Agreements is explicitly to foster development of export markets for Québec products by “[p]romot[ing] access to new foreign as well as domestic markets,”412 and for this reason, Brazil maintains, financing provided under the auspices of the Subsidiary Agreements is often explicitly directed at export-oriented projects and industries.413 In answer to a Panel question as to how the cited press articles support its “explicitly directed” allegation, Brazil states that the projects described are concrete examples establishing Brazil’s point that Subsidiary Agreement funding “explicitly” – meaning in actual fact – goes to export-oriented projects, as one describes Subsidiary Agreement funding to a company with 92 per cent of its revenue dedicated to export sales, and the other describes Subsidiary Agreement funding to a company “[m]ost of” the production of which is destined for export to the United States. In answer to a Panel question regarding why Brazil states that the purpose of the Subsidiary Agreements is to foster development of export markets, but cites only the 1985 Agreement, Brazil replied that the same quote does not appear in the text of the 1992 Subsidiary Agreement. Brazil reiterates that it does not claim that assistance granted to the Canadian regional aircraft industry under the Subsidiary Agreements is contingent in law on export, but rather that assistance granted to the Canadian regional aircraft industry under the auspices of the Subsidiary Agreements is contingent in fact on export. As a result, in Brazil’s view, it is not necessary that both, or either, of the Subsidiary Agreements include provisions speaking expressly to export contingency.
Brazil cites a study prepared by the Canadian Reform Party which reports that the regional aircraft sector has been a significant recipient of benefits under the Subsidiary Agreements,414 and notes that the Canadian regional aircraft industry is 100 per cent export oriented according to the Clark Report. Brazil also notes that 1992 Subsidiary Agreement identifies aerospace is one of the “’key sectors‘” the support of which is central to the development of the Québec economy.415
According to Brazil, assistance under the Subsidiary Agreements may be “in the form of repayable or non-repayable contributions.”416 In answer to a question from the Panel as to why Brazil considers that “repayable loans” under the Subsidiary Agreement, referred to in the articles submitted by Brazil417, constitutes subsidies within the meaning of SCM Article 1, Brazil states that it cited to that assistance for the purposes of illustrating, with the help of other evidence cited that funding under the Subsidiary Agreements is contingent in fact upon export performance within the meaning of Article 3 of the SCM Agreement. Brazil states that it has not claimed that the specific assistance
412 1985 Canada-Québec Subsidiary Agreement on Industrial Development, Art. 2.2(d). Id. at Schedule A, para. (d) (emphasis added) (Exh. BRA- 53).
413 “Rolls-Royce swells: Ottawa, Québec jointly lend $2.5 million to help finance Lachine plant expansion,” The Gazette (Montreal), 23 April 1997 (92 per cent of $2.5 million loan recipient’s sales were for export) (Exh. BRA- 54); “Federal and Provincial Governments Contribute to the Lamines CTEK Project in
Montreal,” Canada NewsWire, 17 October 1997 (Deputy Premier and Minister of State for Economy and Finance Bernard Landry states that “Most of the [funded] plant’s production will be exported to the United States.”) (Exh. BRA- 55).
414 Canadian Reform Party Research, “Bombardier and the Federal Government - 15 Years of Corporate Handouts,” dated 22 October 1996 (Exh. BRA- 58).
415 1992 Subsidiary Agreement, Schedule A, p. 27 (Exh. BRA- 57).
416 Canadian Communication Updating Notification Pursuant to Article XVI:1 of the GATT 1994 and Article 25 of the SCM Agreement, G/SCM/N/16/CAN, 25 June 1997, para. XXIX:5 (“Communication”) (Exh.
BRA- 56). See also 1992 Canada-Québec Subsidiary Agreement on Industrial Development, Schedule B, p. 31 (Exh. BRA- 57).
417 Exhs. BRA-54 and BRA-55
WT/DS70/R
Page 128
described in these exhibits constitutes a subsidy within the meaning of Article 1.1 of the SCM Agreement, among other reasons because this assistance is not to the regional aircraft industry.
According to Brazil, loans provided under the Subsidiary Agreements grant benefits to businesses with an established export trade or significant export potential, consistent with the goal of the Subsidiary Agreements to “’[p]romote access to new foreign as well as domestic markets.‘”418 For Brazil, together the above factors suggest that Subsidiary Agreement funds are in fact tied to actual or anticipated exportation or export earnings.
Arguments of Canada
Canada argues that the Subsidiary Agreement is not in law or in fact contingent upon export performance. Canada observes Brazil’s acknowledgement that contributions under the Subsidiary Agreement promote access “’to new foreign as well as domestic markets‘” [emphasis added by Canada] (para. 6.308)
Canada submits that the evidence that Brazil has adduced does not address the correct interpretation of export contingency. That is, Canada argues, Brazil has not adduced any evidence to show that Subsidiary Agreement contributions are contingent on export performance, in the sense that contributions would not be paid unless exports took place, that there would be rewards if exports took place or that there would be penalties if exports did not take place. Canada submits that there is no such evidence because this is not how contributions under the Subsidiary Agreement are made. In Canada’s view, therefore, Brazil does not put before the Panel a prima facie case that contributions under the Subsidiary Agreement are contingent upon export performance.
Canada states that the Subsidiary Agreement was signed by the Government Canada and the Government of Québec (Québec) on 27 March 1992 for a five year period,419 and that it was terminated on 31 March 1998.420 According to Canada, no new applications are entertained, but applications submitted prior to the date of termination are still in the process of being considered. Canada argues that the purpose of the Subsidiary Agreement was to improve the competitiveness and vitality of the Québec economy by providing financial support for major industrial projects in Québec.
Canada indicates that contributions under the Subsidiary Agreement have been made in 111 major industrial projects, including 7 feasibility studies, and that only nine projects -- 8 per cent of the total -- were in the aerospace sector. According to Canada, the involvement of the Government of Québec in support provided under the Subsidiary Agreement touched virtually all industrial sectors in Québec, including hygiene products, electric power systems, aircraft and aircraft parts, glass, textiles, plastics, ferro-alloys, major appliances, chemicals, waterworks and sewage systems, and cosmetics. According to Canada, if alleged subsidies under the Subsidiary Agreement are, as Brazil points out, available for both domestic and export markets, by definition they are not contingent on export performance. Canada argues that Brazil’s de facto argument rests on its “export propensity” interpretation, which in Canada’s view is incorrect (paras. 5.53- 5.94). In Canada’s view, Brazil’s argument must therefore fail.
In response to a Panel question as to the starting and ending dates for the 1985, 1988 and 1992 Subsidiary Agreements, and the dates on which funding began and ended under each, Canada states that the 1985 Subsidiary Agreement was entered into on 23 January 1985 and was terminated in
418 1985 Canada-Québec Subsidiary Agreement on Industrial Development, Art. 2.2(d). Id. at Schedule A, para. (d) (Exh. BRA- 53).
419 Canada-Québec Subsidiary Agreement on Industrial Development (1991), 27 March 1992 (Exh. CDN-34).
420 Amendment No. 2 to the Canada-Québec Subsidiary Agreement on Industrial Development (1991), 8 August 1996 at 2 (Exh. CDN-35).
WT/DS70/R
Page 129
June 1991, and contributions committed under it were terminated in 1996; and that the 1992 Subsidiary Agreement was entered into on 27 March 1992 and was terminated on 31 March 1998, and contributions committed under it will terminate in 2003. Canada further indicates that these agreements were entered into under the authority of a framework Memorandum of Understanding (MOU) on economic development between the Government of Québec and the Government of Canada. According to Canada, the first MOU was signed on 14 December 1984; the 1985 Subsidiary Agreement was entered into under the auspices of the 1984 MOU. The second framework MOU was signed on 9 June 1988, and the 1992 Subsidiary Agreement was entered into under this MOU.
In response to a Panel question asking why Canada in its arguments does not refer to Brazil’s arguments concerning the 1985 Subsidiary Agreement, Canada states that Brazil makes no claims in respect of that Agreement.
Response of Brazil
Brazil notes Canada’s argument in defending Subsidiary Agreement funding from characterization as “’contingent on export,‘” that because some Subsidiary Agreement funding supports domestic products or markets, or because such funding has been directed to projects outside of the Canadian regional aircraft industry, it cannot be considered to be contingent upon export performance.
Brazil does not contest that some Subsidiary Agreement funds may well have gone to other industries with sales in export markets, or to other industries with sales in domestic markets. Brazil submits that its claim that Subsidiary Agreement assistance constitutes a prohibited export subsidy under Article 3 of the SCM Agreement is unrelated to these incidences of Subsidiary Agreement funding. Rather, Brazil argues, when Subsidiary Agreement funds were disbursed to the Canadian regional aircraft industry, they were given to an industry that is totally export oriented according to the Clark Report, precisely because it is an export industry and was anticipated to remain so.
Brazil considers that the evidence it has submitted, regarding conditional repayability of Subsidiary Agreement funds, the Agreement’s objectives, the assistance to export-oriented companies and assistance to the regional aircraft industry (paras.6.297-6.298) constitutes a prima facie case that Subsidiary Agreement assistance to the Canadian regional aircraft industry constitutes a prohibited export subsidy.
SOCIÉTÉ DE DÉVÉLOPPEMENT INDUSTRIEL DU QUÉBEC (“SDI”)
Subsidy
Arguments of Brazil
According to Brazil, the Société de Dévéloppement Industriel du Québec (“SDI”) provides prohibited export subsidies, in the form of loans and guarantees, to the civil aircraft industry. Brazil states that SDI has recently been “’regrouped‘” for administrative purposes into a new corporation, known as Investissement-Québec (“IQ”).421 Brazil indicates that IQ maintains the resources, including the staff, of SDI.422 Brazil states that IQ administers the Industrial Development Fund and the Private Investment and Job Creation Promotion Fund (“FAIRE”),423 which provides various forms of “’financial assistance‘” and “’government support‘” to certain focus industries, foremost among which is the aerospace industry.424 Brazil draws particular attention to what Brazil terms FAIRE’s
421 Materials from SDI/Investissement-Québec website, at 1 (Exh. BRA- 65).
422 Id. at 1, 3.
423 Id. at 2, 4-6.
424 Id. at 7-9, 10-11.
WT/DS70/R
Page 130
pledge to extend “’refundable or non-refundable contributions,‘” loan guarantees, and the “’assumption of interest charges‘” for eligible projects.425
Brazil argues that SDI funding is extended on a conditionally-repayable basis, meaning that repayment will only occur if the underlying project achieves the degree of success outlined by Brazil. Brazil asserts that because the recipient has no down-side risk – i.e., if the project is unsuccessful, SDI funds need not be repaid -- a clear benefit is conferred, within the meaning of Article 1.1 of the SCM Agreement.
Brazil draws attention to the finding in the most recent WTO Trade Policy Review of Canada that IQ “provides export guarantees for projects considered too risky by private financial institutions,” and that IQ assistance “is available for export development or expansion, or purchases of foreign companies, or for contract finance.” For Brazil, by offering guarantees for projects which are, in the words of the Trade Policy Review “considered too risky by private financial institutions,” IQ makes a potential direct transfer of funds and confers the obvious benefit, within the meaning of Article 1.1 of the SCM Agreement, of providing resources which would otherwise simply not be available. In light of this, Brazil states, SDI funding need not be extended on a conditionally repayable basis to constitute a subsidy. Brazil also states that although the joint DIPP/SDI contribution for the development of Bombardier’s 50-seat regional jet is not subject to the Panel’s jurisdiction (paras. 4.74-4.78), it still considers the “’conditionally-repayable‘” terms attached to it to be persuasive as evidence that funding on such terms is as a general matter available and extended to the Canadian regional aircraft industry.
Brazil maintains that it has demonstrated that funds are being and have been provided to this industry, and that they are being and have been provided under conditions constituting a subsidy within the meaning of Article 1.1 of the SCM Agreement. Brazil believes that, having come forward with this evidence, Canada should be asked by the Panel to produce the evidence placed at the disposal of the WTO Secretariat in the process of conducting the Trade Policy Review.
Arguments of Canada426
In response to a Panel request for an explanation of, and the internal assessment documents and loan/share agreements under, Garantie Québec “ad hoc programmes” and “the financing of major projects”, in particular the three “financing of major project” activities totalling $7.1 million under the Aerospace Industry Development Fund, and five FAIRE activities, during the 1997- 1998 period Canada states that Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Canada notes that Brazil agrees with Canada regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Accordingly, Canada states, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to further support Canada’s submissions that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
Canada also states that Brazil has made no specific allegation about contributions made under the SDI, nor adduced any evidence in support of its vague and unspecific allegations concerning “benefits” under the SDI. Canada further states that Brazil has made no allegations and has adduced no evidence in respect of the Aerospace Industry Development Fund. Canada does not consider it appropriate to adduce evidence in response to allegations not made and a case that has not been established.
425 Id. at 5.
426 See also Section VI.F for the parties’ comments on the Finan Report regarding SDI.
WT/DS70/R
Page 131
Finally, Canada states, most of the information requested by the Panel is sensitive business confidential information, and Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. Accordingly, Canada indicates, with the express permission of the Government of Québec, it has provided as business confidential information the contribution agreements with les Industries Aerospatiales Mecair Inc., Héroux Inc., and Finition de Métal National
N.M.F. (Canada) Ltée. relating to the three projects under the Aerospace Industry Development Fund, with the clauses relating to repayment terms blocked. Canada states that each agreement details the terms of the project, the contribution, and reimbursement.427
Canada also notes that the “ad hoc programmes” listed under the SDI relating to the civil aircraft sector are, in fact, activities under the Subsidiary Agreement that are administered (for Québec) through SDI, and that as a result, these programmes have already been accounted for.
In addition, Canada states, the five FAIRE activities in the 1997-98 period were not in the civil aircraft sector: one project was in the wood products sector, one in the heavy vehicles (trucks) sector, one in the tourist industry sector, and two in the computer sector. Accordingly, Canada states, these FAIRE activities do not fall within the jurisdiction of the Panel.
Comments of Brazil
Regarding Canada’s citation to Brazil’s belief in judicial economy, Brazil recalls its reference to judicial economy in its letter to the Panel dated 13 December 1998 (see para. 6.281).
Brazil also argues that although Canada has provided the contribution agreements for the three Aerospace Industry Development Fund projects identified by the Panel, it has redacted the sections regarding repayment terms, as well as certain of the “conditions préalables.” Brazil submits that these sections likely provide insight into whether the contributions are “in fact tied to actual or anticipated exportation or export earnings.” Brazil argues that as a result of Canada’s strategic decision to withhold this information, the Panel should adopt adverse inferences, presuming that the information suggests that the contributions are in fact tied to actual or anticipated export (paras. 4.146-4.151).
Export contingency
Arguments of Brazil
For Brazil, SDI loans and guarantees (as well as those of Investissement-Québec) confer direct benefits on Québec businesses for the express purpose of assisting those businesses with “’[l]a conquète de marchés à l’exportation,‘”428 and to support transactions “’à l’étranger.‘”429
Brazil asserts that under SDI, the Government of Québec sponsors export programmes “’designed to promote the export of Québec-produced goods and services.‘”430 According to Brazil, SDI was formed with the objective “’to promote economic development in Québec, particularly by encouraging the development of businesses, the growth of exports, research and the development of new techniques.‘”431 Brazil states that SDI provides Québec businesses with “’guarantee[s] of
427 See BCI Tab 4.
428 SDI Website at 2 (Exh. BRA- 63).
429 Gouvernement du Québec, Ministère de L’Industrie, du Commerce, de la Science et de la Technologie, Répertoire des services offerts à l’exportation, p. 53 (Exh. BRA- 64).
430 Government Assistance Programmes - A Practical Handbook (15th ed., 1995), para. 135.230 (emphasis added) (Exh. BRA- 61).
431 An Act Respecting the Société de Développement Industriel du Québec, para. 2 (Exh. BRA- 62).
WT/DS70/R
Page 132
payment or repayment of a financial obligation‘” and “’loan[s] at the current market rate‘”432 for the purpose of assisting those businesses with “’[l]a conquète de marchés à l’exportation,‘”433 and to assist with “’tout projet d’exportation‘” or “’de démarrage à l’exportation.‘”434 According to Brazil, SDI does not fund only exports to other of the Canadian Provinces, but specifically supports transactions “’à l’étranger.‘”435 Brazil asserts that every CRJ sale benefiting from the Can$43 million SDI loan for the development of the 50-seat CRJ has been for export,436 and that accordingly, SDI funding to the civilian aircraft industry constitutes an export subsidy in law or in fact prohibited by Article 3 of the SCM Agreement.
Brazil does not contest that some SDI/IQ funds may well have gone to other industries with sales in domestic markets. Brazil states that its claim that SDI/IQ assistance constitutes a prohibited export subsidy under Article 3 of the SCM Agreement is unrelated to these incidences of SDI/IQ funding. For Brazil, when SDI/IQ assistance goes to the Canadian regional aircraft industry, it goes to an industry that is totally export oriented, precisely because it is an export industry and the Government of Québec anticipates that it will remain so. For Brazil, the ordinary meaning of Article 3.1(a) of the SCM Agreement prohibits such assistance.
Arguments of Canada
Canada maintains that the SDI is not contingent upon export performance. Canada observes that Brazil admits that the programme is available for goods destined for domestic as well as export market.
According to Canada, the SDI was established by the Loi sur la société de développement industriel du Québec437 (the SDI Act). In June 1996, the Garantie Québec programme replaced most operations under the SDI. Canada states that this programme is designed to guarantee loans, lines of credit or other forms of financing offered through a financial institution, and indicates that the programme supports projects dealing with start-up, expansion or modernisation, in the manufacturing or tourist sectors and in certain service industries; research, development or design, as well as marketing and financing of tax credits for scientific research or experimental development; marketing, acquisition of a company in another country, and financing of a line of credit; and business mergers, regrouping of common activities and take-over of one firm by another in the manufacturing and tourist sectors, and in certain service industries.
Canada states that according to Article 2 of the SDI Act, the objective of the SDI is to ‘«favoriser le développement économique du Québec, notamment en encouragent le développement des entreprises, la croissance des exportations et les activitiés de recherche et d’innovation.»’ Canada notes that Brazil acknowledges that ‘«exportation»’ in this context means exports outside Québec, including the other provinces of Canada. Therefore, for Canada, SDI has as a general objective the enhancement of Québec’s competitiveness and, as a desirable but not necessary result, an increase in Québec’s exports. Canada maintains that export performance is not a criterion of success for either the programme as a whole or its activities;438 and that export performance is manifestly not a condition for receiving contributions.
432 Id. at para. 135.240.
433 SDI Website at 2 (Exh. BRA- 63).
434 Gouvernement du Québec, Ministère de L’Industrie, du Commerce, de la Science et de la Technologie, Répertoire des services offerts à l’exportation, p. 53 (Exh. BRA- 64).
435 Id.
436 Clark Report, at pgs. 7-8, 13, 50-51, and Tables 1, 5 and 6.
437Lois Révisées du Quebéc, c. S-11.01 (Codification administrative 94-09-16) (Exh. CDN-36).
438 Société de développement industriel du Québec, 1997-1998 Annual Report (Québec: Gouvernement du Québec, 1998) at 36-37, 40 (Exh. CDN-37).
WT/DS70/R
Page 133
Canada maintains that the SDI comprises four sets of eligibility criteria, one of which is for ‘export development’, which includes credit insurance, financing for foreign direct investment by Québec companies, and inventory financing. Canada argues that 458 activities were authorised in 1997-1998, of which 53 per cent involved loans or loan guarantees of less than Can$200,000. One out of 39 financing activities and 96 out of 419 guarantees authorised was for exportation.439 According to Canada, the activities covered a wide range of industrial, transportation, resources and agricultural sectors (278 authorisations) and services (180 authorisations),440 but none were related to the civil aircraft sector.
Canada submits that in individual projects there are no rewards if exports take place and no penalties if they do not. More important, in Canada’s view, is that SDI is not in law or in fact contingent upon export performance.
According to Canada, the sole basis for Brazil’s allegation that SDI is inconsistent with SCM Agreement Article 3 appears to be that one of the objectives of the programme is “the growth of exports,” although Brazil admits that Québec counts sales to Canada’s other provinces as export sales. (para. 6.333) Accordingly, in Canada’s view, there is no prima facie case for Canada to answer.
Canada also objects, for the reasons identified in its arguments concerning TPC (paras. 6.197- 6.199) to what it views as the inappropriate use by Brazil of Canada’s trade policy review. Canada asks the panel to ignore Brazil’s arguments about SDI that are based on that document.
ARGUMENTS REGARDING THE FINAN REPORT SUBMITTED BY BRAZIL
Arguments of Canada
Canada raises a number of general criticisms of what it views as flawed analysis and assumptions, and inaccurate facts in the “Finan Report” submitted by Brazil in support of its allegations of subsidization of the Canadian aircraft industry. Canada questions 441 the timeframes over which the analysis is conducting, arguing that in general the rationale for the choice of any timeframe should be explicit, and that a consistent approach should be used in determining timeframes to be used. Canada takes issue with the Finan Report's use of different timeframes for different subsidies, with no rationale provided for any timeframe used, and no explanation of why different timeframes are used. That is, for some calculations the analysis extends through 2015, while for others it ends in 2000. For Canada, these concerns over timeframes are important because of their effect on the results of the analysis, in particular, the magnitude of the alleged subsidies. Canada argues that had different timeframes been used, the aggregate subsidy amounts calculated could have been greatly reduced.
Canada further argues that the Finan Report calculates a higher value on investments that require repayment than on the equivalent amount of funds as an unencumbered grant. For Canada, such a result is nonsensical, as the repayable funds clearly are less attractive to the recipient, given their repayment obligation, than outright grants with no strings attached.
Canada also criticises the Finan Report's comparative treatment of actual and anticipated benefits. According to Canada, there should be a clear distinction between those alleged subsidies that have already been received, and those that will only be received in the future if certain other events occur as well. Canada states that one-half of the benefits alleged in the Report have already
439Id.. at 22, Table 1 (Exh. CDN-37).
440 Id.. at 24, Table 2 (Exh. CDN-37).
441 The comments in paras. 6.343 and 6.346 are contained in a document prepared by KPMG LLP submitted as an attachment to Canada’s first submission.
WT/DS70/R
Page 134
been received, and the other half may or may not occur in the future, and that the lack of distinction between those categories of alleged benefits results in a potentially misleading statement of results.
Canada argues, with respect to the CRJCI, that the “Finan Report” assumes that “EDC’s return is generated by Exinvest’s share in the residual value of the aircraft”442, then assumes that “EDC participates in all of Bombardier’s sale of regional jets”443, and thirdly assumes implicitly that the EDC participates in all of Bombardier’s sales through equity infusions. Canada asserts that no evidence is adduced in support any of these “assumptions”, in spite of which the “Finan Report” declares that “EDC is clearly providing a subsidy to Bombardier.”
Canada argues the “Finan Report” observes that the Canada-Québec Subsidiary Agreement has been “previously determined to confer subsidy benefits to other companies.” [emphasis added] The “Finan Report” does not explain how such a finding is relevant to this dispute. It does not cite any evidence or authority for the assertion. It does not say when, by whom or under what circumstances any such determinations were made. Indeed, no support is offered and no evidence adduced in respect of an assertion without relevance. This is neither evidence nor argument; it is simply innuendo.
Canada states, with respect to Québec’s SDI, that the Report makes no “assumptions”; but rather an unsupported assertion of “fact”: “Projects must be aimed at markets outside Québec.”444 [emphasis added by Canada] Canada denies this (paras. 6.335-6.340). According to Canada, the SDI includes a number of programmes, only one of which is related to “exportation”, and in any event, Canada states, by “exportation” SDI refers to “exportation out of Québec”, including to the other provinces of Canada, as Brazil itself admits.
Canada recalls its argument that the conclusion of the “Finan Report” with respect to Ontario’s forgiveness of its loan to Bombardier is without foundation and false (see para. 6.275- 6.276). Nevertheless, Canada notes what it terms the methodological paradox created as a result of the assertion by the “Finan Report” that the alleged forgiveness of the loan concerning the sale in 1997 should be viewed as a cash grant in 1992: if the “subsidy” element of the sale was granted in 1992, then the sale is outside the jurisdiction of this Panel. If, however, the sale is to be the subject of WTO dispute settlement, any “subsidy” would have to be found to have been granted in 1997. If this is the case, Canada submits, the “present value in 1998” of the subsidy as calculated by the “Finan Report” is inflated and, in any event, irrelevant.
Canada notes that the Finan Report in Table B4 estimates the alleged benefits received by Bombardier as a result of “equity infusions” from the EDC into the CRJCI. According to Canada, these alleged benefits are estimated on the presumption that private sector investors require a 15 per cent to 20 per cent pre-tax return on similar aircraft leasing deals, reflecting a misunderstanding of the nature of US leveraged leasing transactions. According to Canada, US leveraged leases are driven by tax considerations: because of the tax benefits received, private equity investors are prepared to earn very low pre-tax returns; indeed, pre-tax returns may be zero or even negative. This is because, Canada submits, their after-tax returns, incorporating the tax benefits associated with leveraged leasing, reach acceptable levels. 445 Accordingly, in Canada's view, the “Finan Report” grossly over- estimates the value of the alleged benefit from these transactions. In these circumstances, Canada maintains, even assuming that all other assumptions are correct, EDC’s alleged pre-tax return on equity of 2.75 per cent would, in view of the after-tax returns that EDC would expect in a leveraged lease transaction, be consistent with what private sector investors would expect.
442 Finan Report, at A.1.
443 Id.. at A.2.
444 Id.. at A.3.
445 Andrew Littlejohns and Stephen McGairl, eds., Aircraft Financing, 3rd ed. (New Jersey, US: Euromoney Books, 1997) at Chapter 8 (Exh. CDN-48) and Appendix 1 (Exh. CDN-75).
WT/DS70/R
Page 135
Response of Brazil
In response to Canada’s general argument that the Finan Report overstates or improperly calculates the present value in 1998 of the benefits received by the Canadian regional aircraft industry from subsidies, Brazil emphasizes that whether the benefit is $1 or $1 billion is in large part irrelevant. According to Brazil, the fundamental point established by Brazil and the Finan Report is that whatever the value, and whether the test is “benefit to recipient” or “cost to government,” the Canadian programs confer a benefit and constitute a subsidy within the meaning of Article 1 of the SCM Agreement.
In responding to Canada's specific comments, Brazil notes generally that the Finan Report performed two separate calculations: first, the aggregate present value in 1998 of benefits received by Bombardier in the past as well as those expected to be received in the future.446 second, benefits per 50-seat jet delivered in 1998.447 Brazil states that the per-aircraft benefits relate only to benefits which have already been received and do not include benefits relating to the development of 70-seat aircraft.
The Time Frames Over Which Subsidies are Quantified and Aggregated
Brazil states that it does not disagree with KPMG's proposed tests for evaluating the timeframes used in the Finan Report, but emphasizes the importance of applying the correct time- frame consistent with the facts and circumstances of the case. Brazil argues that since each of the different Canadian Government subsidies analysed in the Finan Report involve different facts and circumstances, it correctly applied different time frames in calculating the benefits to Bombardier appropriate to the facts:
EDC Equity Infusions
Time-frame: Benefits calculated for each of 20 years from fiscal 1996 through 2015.
Rationale: The EDC is presumed to participate in the financing of every jet sold by Bombardier, beginning in January 1996 and continuing for the estimated 20-year economic life of the product.448 Brazil notes that extending the analysis out 20 years does not significantly alter the analysis of aggregate benefits, because benefits received in the future are discounted back to the present at 16.9 per cent. Thus, 75 per cent of the net present value of benefits relating to EDC equity infusions are generated in the first 10 years of the analysis, i.e. through 2005.
Can$87 Million “Investment” in the 70-seat CRJ-700
Time-frame: For the purpose of calculating an aggregate benefit to Bombardier, the analysis extends to 2019. However, no per-aircraft benefit relating to this subsidy is calculated, since deliveries of the 70-seat jet have not yet commenced.
Rationale: Brazil notes that the government expects repayment, plus receipt of a modest profit, through per-aircraft royalties commencing with the 250th jet sold. Therefore, the analysis is extended through the estimated life of the 70-seat jet, 20 years.
446 The aggregate amounts appear in Table B.1 of The Finan Report.
447 The per-aircraft amounts are summarized in Table B.16 of The Finan Report.
448 See R. Baldwin and P. Krugman, "Industrial Policy and International Competition in Wide-Bodied Jet Aircraft", in R. Baldwin, ed., Trade Policy Issues and Empirical Analysis (Chicago: University of Chicago Press, 1988). Baldwin and Krugman state: "… the product cycles in aircraft are quite long … (T)he life of a successful airframe is something like twenty years."
WT/DS70/R
Page 136
Can$57 Million “’Investment‘” in the Dash 8-400
Time-frame: For the purpose of calculating an aggregate benefit to Bombardier, the analysis extends to 2019. However, no per-aircraft benefit relating to this subsidy is calculated, since deliveries of the Dash 8-400 have not yet commenced.
Rationale: As in the case of the Can$87 million “investment”, Brazil argues that the government expects repayment, plus receipt of a modest profit, through per-aircraft royalties. Therefore, again the analysis extends through the estimated life of the product, 20 years.
Can$100 Million “Investment” in Turboprop Engines
The time-frame and rationale are identical to those relating to the Can$57 million "'investment'".
The Acquisition and Restructuring of de Havilland
Time-frame: Benefits are amortized over 13½ years, beginning in the year in which funds were received.
Rationale: Amortization is based on the depreciable life of Bombardier fixed assets for the period 1994-1998.449 This is appropriate since the subsidies relate to the acquisition of the assets of the de Havilland business. Amortizing the benefits based on the average depreciable life of fixed assets anchors the duration of benefit to the life of the underlying assets.
The Comparative Treatment of Repayable and Non-Repayable Contributions
Brazil concedes that the value of a grant should exceed the value of a repayable contribution, and that any statement to the contrary would be silly. Brazil states that its “low estimate” correctly computes the value of the repayable contributions, as noted by KPMG, but acknowledges a spreadsheet error made in calculating the “high estimate” for benefits relating to the Technology Partnerships Canada subsidies. Correcting this technical error does not change the low estimate of the present value of benefits received by Bombardier, and reduces the high estimate from US$2.4 billion to US$2.3 billion.
The Comparative Treatment of Actual and Anticipated Benefits
Brazil does not object to making the distinction between benefits already received and those that may be received in the future. Brazil asserts that its calculation of the benefits per 50-seat jet delivered in 1998 relate entirely to benefits received in 1998 and prior years, and that the 1998 per- aircraft benefit excludes EDC benefits generated in later years and ignores benefits relating to the development of 70-seat aircraft.
In separating past vs. future benefits in the calculation of the aggregate present value of benefits, Brazil states that the only “anticipated” benefits are EDC benefits for 1999 and beyond. Brazil disagrees with the percentages calculated of the percentage of EDC benefits which may be considered “anticipated”. Brazil argues that the number of planes actually sold is not the relevant benchmark to be used in distinguishing between actual and anticipated benefits, but rather the present value of benefits relating to deals consummated in or before 1998 vs. deals which are expected to take
449 The method used to determine the depreciable life of Bombardier's fixed assets is discussed in
White, A. Sondhi, D. Fried, The Analysis and Use of Financial Statements, (John Wiley, New York) 1994, page 453.
WT/DS70/R
Page 137
place in the future. Brazil calculates the present value of benefits relating to deals taking place in or before 1998 at between US$215 million and US$236 million. Brazil's estimate of the present value of benefits relating to future deals is between US$679 million and US$743 million. Thus, Brazil states, the portion of the total present value of benefits relating to anticipated benefits is roughly 30 per cent.
3. Rebuttal by Canada
Canada submitted a report by KPMG in support of Canada's criticisms of the Finan Report. In the report, Canada notes, first, that for EDC equity infusions, extending the timeframe of analysis increases the amount of the aggregate alleged benefits, because most of the alleged benefits identified were to be received in the future, as aircraft are sold. For the other funds for the development of airplanes and engines, Canada asserts, the alleged benefit has already been entirely received up front; extending the timeframe, be increasing the estimated repayments or royalties, diminished the magnitude of the alleged benefit.
Canada notes that Brazil acknowledges a spreadsheet error with respect to only one calculation. Canada submits that the same error is equally applicable to the calculations pertaining to TPC.
In response to Brazil’s assertion that the Can$87 million TPC contribution in 1996 would have a net present value, in 1998, of Can$119 million, Canada notes first, that the Finan Report assumes that the TPC contributions are all paid in lump sum, which according to Canada is not the case. Canada states that as of the date of its First Written Submission, only about Can$22 million had been disbursed. Canada also indicates that the rate of inflation in Canada over the previous two years was only about 1.5 to 2 percent. Thus, for Canada, even if the Can$87 million had been disbursed in lump sum in 1996, a 25 percent inflation in two years in the “net present value” of the contribution was unjustified and again highlighted the difficulties in the methodology of the Finan Report. Canada also argues that the Finan Report is unduly pessimistic about the market prospects of the CRJ700, in view of the market forecasts by independent and highly respected forecasters that Canada has submitted in response to a Panel question.
ARGUMENTS REGARDING THE CLARK REPORT
Arguments of Canada
Saturation of the Canadian regional turboprop market
Canada criticises the “Clark Report” submitted by Brazil in support of its arguments concern the proportion of regional aircraft exported by Canada on a number of grounds. Canada asserts, first, that the report chooses a period of analysis that ignores the reality that the Canadian regional turboprop aircraft market has been saturated since 1992. Canada states that because of this saturation, since 1992, there have been only two firm sales of any new regional turboprop aircraft to Canadian customers450, and one import of a used Dash-8, in 1998.
450 Two 2 Saab 340Bs were ordered in 1994; ten British Aerospace J41 were ordered in 1995, but all were subsequently cancelled or returned to British Aerospace. The remaining additions to the regional aircraft fleets of Canadian airlines post 1992 have been 26 Canadair CRJs delivered to Air Canada: see Orders, Canada, 20-90 Seat Market, 1984-1998 (Lundkvist Aviation Research, 1998) (Exh. CDN-42).
WT/DS70/R
Page 138
According to Canada, in the period 1984-90, Canadian airlines invested heavily in turboprop regional aircraft and the Canadian domestic regional aircraft fleet increased by 776 per cent451, thereafter stabilizing, as the Canadian market was saturated.
Canada argues that as of 1 January 1998, Canadian carriers operated almost 100 Dash 8’s in regular service452, and states that from its introduction in 1983/84 to 31 December 1997, 481 Dash 8’s had been delivered. According to Canada, domestic Canadian carrier fleets therefore represent about 20 per cent of the Dash 8 deliveries.
For Canada, it thus is not surprising that Dash 8 sales post 1992 have been for export. Canada argues, however, that when the domestic demand for turboprop regional aircraft returns as older aircraft need replacing, it is likely that Canadian airlines will order new Dash 8s, given the economies of training on and maintaining new aircraft that have the same lineage as their established turboprop fleets.
The 26 CRJs delivered to Air Canada
According to Canada, the Clark Report's “determination” that the 26 CRJs delivered to Air Canada are “export sales”453 is untenable. Canada submits that these 26 CRJs were purchased by a Canadian airline (Air Canada) from a Canadian manufacturer (Bombardier), and were never intended for export: each aircraft was required by contract to be manufactured in compliance with Transport Canada requirements for a Canadian Certificate of Airworthiness and a Canadian Certificate of Registration. The Standard Certificate of Airworthiness should be contrasted with a Certificate of Airworthiness for Export: CRJs acquired by Air Canada require the former, whereas export customers (such as Slovenia and the United States) require the latter. Air Canada took delivery of the aircraft directly from Bombardier. Canada states that the aircraft are based and registered in Canada,454 and fly regular routes between Canadian centres, and between Canadian and US centres.455
Canada also takes issue with the Report's statement that that Air Canada ordered 26 of the 50-passenger CRJs in several separate transactions and that Air Canada was not interested in owning these aircraft.456 According to Canada, this assertion is not supported by the evidence. Canada states that to finance its acquisitions, Air Canada chose to use US leveraged leases for 24 CRJs, with the remaining two CRJs purchased for cash.457 Thus, Canada maintains, Air Canada has demonstrated a patent interest in owning CRJs. Canada asserted that the latter two purchases are established beyond any doubt by the Bills of Sale and Certificates of Acceptance for these aircraft.
According to Canada, airlines purchasing aircraft look to a variety of financing mechanisms in order to reduce financing costs, and at present, the US leveraged lease is one of the most popular financing vehicles, as it allows an airline to pass US tax benefits to an equity participant, in return for
451 Regional Market History: Orders, Deliveries, and Total Fleet, Canadian Market (Lundkvist Aviation Research, 1998) (Exh. CDN-43).
452 Ninety-seven Dash 8-100 and 300 series aircraft were operated by Canadian carriers as of 31 December 1997: see Summary of Dash 8 Aircraft in Service with Operators of Canadian Airlines as at 1 January 1998 and World Airline Directory, the Americas, Flight International 25-31 March 1998 at pp 37-60 (Exh. CDN-44).
453 “Clark Report” at 10, 13, 21-24.
454 Transport Canada Registry of Civilian Aircraft, online: Transport Canada homepage
http://www.tc.gc.ca/cgi-bin/webdbc.dll/Av (date accessed: 10 November 1998) (Exh. CDN-45).
455 Air Canada, Press Release, “Air Canada to Add Two New Canadair Jets to Fleet” (31 January 1997) (Exh. CDN-46).
456 “Clark Report” at 13.
457 Air Canada, 1997 Annual Report (Exh. CDN-47).
WT/DS70/R
Page 139
lower financing costs.458 Canada states that similar advantages can be found in other jurisdictions: Japan, Germany and the United Kingdom.
In Canada's view, given the globalisation of financial services and capital movements, these transactions cannot be characterised as exports simply because they are financed or underwritten in another country. That is, Air Canada’s choice of financing vehicle cannot change the fundamental nature of these transactions: they are domestic and not export sales.
Thus, Canada argues, the “Clark Report” incorrectly concludes that 100 per cent of all of Bombardier’s regional aircraft sales have been for export.
Canada also argues that the Clark Report's definition of “sale”, which counts firm orders as sales, is incorrect.459 Canada maintains that sales are properly marked when the plane is delivered; even firm orders may not turn into sales, as was evident during the downturn in the aircraft industry in the early 1990’s.460 Canada noted that Air BC acquired two Dash 8 aircraft in April and May 1992 on rolling one-month interim finance leases and subsequently took title to the aircraft in July 1992. Thus, using the proper definition of sale, de Havilland made domestic sales in the "review period" chosen by the Clark Report.
Response by Brazil
Brazil takes issue with Canada's criticism that the Clark Report does not include sales of Dash 8 series aircraft sold to domestic airlines prior to 1992, and that had the Clark Report taken account of the period 1984-1990, it would have found evidence of domestic sales (paras. 6.368-6.371).
Brazil notes that it has conceded, pursuant to arguments by Canada, that a 1989 alleged subsidy by DIPP and SDI was not within the Panel's jurisdiction because it was made before the entry into force of the WTO Agreement (para. 4.77), and argues that Canada cannot “have its cake and eat it too.” For Brazil, the period before 1 January 1995 is either relevant and subject to the Panel’s jurisdiction or it is not. Canada cannot have it both ways.
Brazil also disagrees with Canada's argument that because 26 CRJs are operated by Air Canada and registered to operate in Canada, they represent domestic rather than export sales (paras. 6.372-6.376).
Brazil notes, first that 10 of the 26 CRJs operated by Air Canada were sold in 1993, before the effective date of the SCM Agreement.461 Brazil states that according to Canada, events occurring before the effective date of the SCM Agreement are not subject to the Panel’s jurisdiction.
Second, Brazil argues, Canada admits that for at least 24 of the 26 CRJs, the aircraft were not sold to and are not currently owned by Air Canada, but that rather because they are operated by Air Canada, they constitute domestic sales. Brazil does not dispute that these aircraft are, for the moment, part of the Air Canada fleet, but submits that they are not owned by Air Canada and were in fact sold
458 Andrew Littlejohns and Stephen McGairl, eds., Aircraft Financing, 3rd ed. (New Jersey, US: Euromoney Books, 1997) at Chapter 8 (Exh. CDN-48).
459 “Clark Report” at 4: “Our analysis is based on the numbers of aircraft sold, i.e., those which are the subject of firm orders or commitments, not on value. As deliveries may lag firm orders by several months or several years, whenever possible we selected the date of the reported firm order, not subject to any conditions, as the most appropriate basis for determining the date of sale.”
460 For example, a cancellation of a near-term order by GPA Jetprop caused de Havilland to reduce its production of Dash 8’s and to shutdown the Downsview, Ontario plant for an extended period in the summer of
1993: See “GPA Jetprop Cancels Backlog, DHC Forced to Extend Summer Shutdown”, Commuter Regional Airline News, 26 April 1993 at 3. (Exh. CDN-72).
461 Clark Report, Tables 5D and 5F.
WT/DS70/R
Page 140
via EDC’s equity financing vehicle to a US SPC, from which Air Canada merely leases the aircraft, a fact which Brazil states that Canada acknowledges. In response to a Panel question concerning the basis for this alleged acknowledgement by Canada, Brazil notes Canada’s statement (para. 6.373) that “’US leveraged leases‘” were used to finance the sale of these aircraft, and Canada’s statement (para. 6.124) that CRJ Capital, which according to Brazil is the “’EDC equity financing vehicle‘”, was used for the Air Canada transaction. Brazil also states that this transaction is the focus of former EDC President Paul Labbé’s comments to the Canadian Parliament’s Committee on Foreign Affairs and International Trade (see para. 6.112). Brazil states that its arguments in this regard do not speak to the remaining two CRJs, which Canada alleges were purchased by Air Canada.
Brazil notes that even if the two remaining CRJs operated by Air Canada and sold in 1997462 constitute domestic sales, the result would be that during the period from 1 January 1995 to the present, 99.64 per cent of sales by the Canadian regional aircraft industry would still have been for export.463
Third, regarding Canada's argument (para. 6.372) that Air Canada’s compliance with Transport Canada requirements regarding a Canadian Certificate of Airworthiness and a Canadian Certificate of Registration demonstrates that these CRJs “were never intended for export . . .”, Brazil states that as a matter of Canadian law, it is illegal to operate aircraft in Canada without a Canadian Certificate of Airworthiness.464 Brazil notes that because the CRJs are in fact operating in the Air Canada fleet, this would presumably subject them to this domestic regulatory requirement. For Brazil, however, the fact that Air Canada would secure a Canadian Certificate of Airworthiness does not diminish the fact that the aircraft were sold to a foreign buyer.
Brazil also maintains that it is illegal under Canadian law for Air Canada to operate foreign- registered aircraft without obtaining a Canadian Certificate of Registration,465 and for Brazil the fact that Air Canada needed to secure this certificate simply makes it all the more evident that the CRJs at issue had been sold to and are owned by a US SPC, and are therefore “foreign-registered” within the meaning of Transport Canada regulations.
Brazil characterizes Canada's argument (para. 6.375) that although these aircraft were sold to foreign buyers and structured as export sales precisely so that they would fall within EDC’s export financing mandate, the “'choice of financing vehicle cannot change the fundamental nature of these transactions‘” as another instance of Canada's trying to “’to have its cake and eat it too.‘” Brazil recalls a statement by EDC’s former president in testimony to Parliament that the “reality” of this sale was that it was to a US party,466 in order to take advantage of EDC financing to launch an export product. For Brazil, Canada cannot credibly now be heard to claim that these were not export sales.
Rebuttal of Canada
In response to Brazil's argument that 10 of the 26 CRJs were sold prior to 1995, Canada argues that counting a sale as when a delivery is made, Air Canada has acquired 22 CRJs since 1 January 1995467 two of which were paid for in cash in January 1997 (para. 6.373), and the remaining 20 were financed through US leveraged leases. With respect to Dash 8 sales and Brazil's argument
462 Clark Report, Table 5H.
463 Clark Report, Table 8 (during the period 1 January 1995 to the present, there have been 563 sales of Canadian regional aircraft, two (or 0.36 per cent) of which are operated by Air Canada).
464 Section 700.05(1)(a) of the Transport Canada regulations (attached in relevant part in Exh. BRA-98).
465 Section 202.42(1) of the Transport Canada regulations (attached in relevant part in Exh. BRA-98).
466 House of Commons of Canada, 35th Parliament, 1st Sess., Evidence, Standing Committee on Foreign Affairs and International Trade, Meeting No. 43, 11 May 1995, pg. 43:30 (emphasis added) (“Committee on Foreign Affairs and International Trade”) (Exh. BRA-9).
467 Bombardier Inc, list of CRJ delivery dates to Air Canada (Exh. CDN-90)
WT/DS70/R
Page 141
that Canada cannot "have its cake and eat it too", Canada submits that its comments on the Clark Report were simply responding to the assertions and analysis of that document and that it was the Clark Report that chose to review the period before 1995. In Canada's view, its comments on the Clark Report took no position on whether the transactions prior to 1995 are within the scope of the Panel's jurisdiction.
Canada’s denies that it acknowledges that any of the 26 Air Canada CRJs were sold via EDC’s equity financing vehicle to a US SPC (para. 6.381). Canada asserts that the statements of Canada cited by Brazil in this regard make no mention of EDC involvement, nor of CRJ Capital, and argues that Brazil continues to mischaracterize CRJ Capital as “’EDC’s equity financing vehicle‘”. Canada recalls its statements and the officers’ certificate that it submitted,468 to this effect.
Regarding Brazil’s argument that the aircraft acquired by Air Canada through leveraged leases are export sales, Canada submits that the fact that the owners of these US leveraged lease financed aircraft are owner-trustees located in the United States does not make these transactions into export sales. For Canada, US leveraged leases are just a means of financing an acquisition which are used to finance a variety of capital goods. Canada states that, for example, a Canadian electrical utility could finance a new coal-fired electrical power plant built and located wholly within Canada with a US leveraged lease, and that the power plant never leaves Canadian soil and thus is not exported.469
For Canada, regarding the term “export” for the purposes of Article 3 of the SCM Agreement, the ordinary meaning is to “send out (goods …) esp. for sale in another country”.470 Canada submits that there is nothing in the context, object and purpose, or negotiating history of the SCM Agreement that would indicate that the interpretation of “export” should be any different than the ordinary meaning. In this sense, Canada states, there is no indication of these aircraft being exported.
Canada takes issue with Brazil's characterization of its argument regarding Transport Canada's requirements (para. 6.384). Canada states that its argument is that each aircraft was required by contract to be manufactured in compliance with Transport Canada requirements for domestic aircraft.471 This, Canada submits, is evidence that these aircraft were not intended for export, as if they had been intended for export to the United States, they would have been manufactured in compliance with the requirements of that jurisdiction – that is, requirements set by the US Federal Aviation Authority.
In this regard, the Standard Certificate of Airworthiness should be contrasted with an Export Airworthiness Certificate, according to Canada. Canada states that CRJs acquired by Air Canada receive a Standard Certificate of Airworthiness,472 whereas export customers such as those in the United States would receive an Export Airworthiness Certificate473, the significance of which is set out in the Canadian Aviation Regulations.474
Canada submits that according to these Regulations, where the Canadian Minister of Transport agrees to issue an Export Airworthiness Certificate in respect of an aircraft being exported
468 Exh. CDN-93
469 Memorandum to Blair Hankey from William Macan, 14 November 1998, “US Cross-Border Leasing – No ‘Export’ Requirement” (Exh. CDN-79)
470 Hawkins and Allen, eds., The Oxford Encyclopedic English Dictionary (Oxford University Press: Oxford, 1991) definition of “export” at 500 (Exh. CDN-80).
471 Declaration of Michel Bourgeois, 24 November 1998. (Exh. CDN-69).
472 Standard Certificate of Airworthiness, 27 May 1996, CRJ serial number 7120, Canadian Registration Mark C-FWSC (a CRJ delivered to Air Canada) (Exh. CDN-69).
473 Export Certificate of Airworthiness, 17 August 1998, for CRJ serial number 7253 (Exh. CDN-69).
474 See, Transport Canada, online: Transport Canada homepage, www.tc.gc.ca (date accessed: 3 December 1998) Canadian Aviation Regulations – General Information on the CARS (Exh. CDN-81)
WT/DS70/R
Page 142
as conforming to a foreign airworthiness standard, the Minister must verify compliance with any special requirements set out in that foreign standard475, which for aircraft exported to the United States are set out in an Appendix to the Regulations.476 According to Canada, The US Federal Aviation Authority (or FAA) will accept Canadian export airworthiness certification, as long as the aircraft conforms to certain FAA requirements, pursuant to a bilateral agreement between Canada and the United States concerning, among other things, the airworthiness of imported civil aircraft.477
Canada argues that the 24 CRJs in question were not manufactured to comply with the requirements for Export Airworthiness Certificates for the United States – precisely because there was no intention to export them, and they were not exported. Canada states that the results of a Transport Canada database search on all of the CRJs in question show that none of the 24 CRJs acquired by Air Canada under US leveraged leases have been exported478, and that none of them ever was registered by Transport Canada as being imported into Canada.479 Canada also argues that if an aircraft is imported, that fact is registered in the Canadian Civil Aircraft Registry, so in the Registry for a used Dash 8 imported by Air Nova, the entry lists the "Year imported", as 1988. Similarly, the entries for two SAAB 340Bs show they were imported in 1994 and 1995 respectively. Canada submits, however, that the Canadian Civil Aircraft Registry entries for all Air Canada CRJs, shows that the aircraft financed through US leveraged leases have never been imported.
Canada also disagrees with Brazil's argument that it would be illegal under section 202.42 of the Canadian Aviation Regulations for Air Canada to operate foreign-registered aircraft without obtaining a Canadian Certificate of Registration. Canada states that this section does allow for the operation of foreign-registered aircraft on lease operations in Canada, without requiring that those aircraft receive a Canadian Certificate of Registration.480 Canada argues that a Canadian operator operating a foreign-registered aircraft can only do so only for a period of twenty-four months in a thirty month period481, but also notes that under the Convention on International Civil Aviation, Article 18, an aircraft cannot be validly registered in more than one State482, meaning that either the CRJs are registered in Canada, or they are registered elsewhere (such as the United States) – they cannot be registered in two countries at the same time.
Canada asserts that a Canadian Certificate of Registration is necessary for Canadian aircraft, as is clear in the Canadian Aviation Regulations483. The fact that the Air Canada CRJs have Canadian
475 Airworthiness Manual, Chapter 509, CAR 509.03(1) “Authority for Export”; and also see Canadian Aviation Regulations, subpart 9 – Export Airworthiness Certificates (Exh. CDN-82)
476 Canadian Aviation Regulations, Airworthiness Manual, Chapter 509, Appendix B “Foreign Special Requirements” and Annex I “United States of America”. (Exh. CDN-83)
477 Exchange of Notes regarding Agreement Concerning the Airworthiness and Environmental Certification, Approval, or Acceptance of Imported Civil Aeronautical Products: Note No. ETT-1583, dated
31 August 1994, from Lloyd Axworthy, Minister of Transport, to Paul Robinson, Ambassador for the United States of America; Note No. 330, dated 31 August 1994, from John Rouse, Charge d’Affaires to Lloyd Axworthy, Minister of Transport. (Exh. CDN-78)
478 Affidavit of Sheila Dowd, Chief, Aircraft Registry and Leasing, General Aviation Branch of the Department of Transport. (Exh. CDN-84) One aircraft, mark C-FZSI, purchased outright for cash by Air Canada and delivered in January 1997, (see Exh. CDN-70) was sent to Argentina in November 1997, and was subsequently returned to Air Canada in April 1998.
479 Affidavit of Sheila Dowd, (Exh. CDN-84).
480 Canadian Aviation Regulations Section 202.42 is expressly subject to Section 203.03. (Exh. CDN-85) Section 203.03, and Subpart 3 in general, set out the conditions for, among other things, a Canadian operator to operate a foreign-registered aircraft that the operator has leased for use in Canada. (Exh. CDN-86) See also Standard 223, Respecting the Operation of a Leased Aircraft by a Non-registered Owner. (Exh. CDN-87)
481 Canadian Aviation Regulations Section 203.08. (Exh. CDN-86).
482 Convention on International Civil Aviation, International Civil Aviation Organization Doc. 7300/5, 1980 (Exh. CDN-88)
483 Canadian Aviation Regulations Sections 202.13 to 202.17 and 202.69.
WT/DS70/R
Page 143
Certificates of Registration thus does not make it “'all the more evident that the CRJs at issue… are therefore ‘foreign-registered’”, in Canada's view. Canada reiterates that the CRJs cannot be both Canadian registered and foreign-registered. And since these aircraft have never been exported and subsequently imported, they never were foreign-registered.
Rebuttal of Brazil
Brazil takes issue with Canada’s argument that the Clark Report used an “‘incorrect’” definition of the term “sale.” Brazil recalls Canada’s argument that a sale only occurs when a plane is delivered, and that had the Clark Report used this date, it would have accounted for the purchase of two Dash 8 aircraft in 1992 by Air BC, a Canadian airline. Brazil notes that it appears that these aircraft were ordered in 1989, and were therefore not included in the Clark Report’s results.
For Brazil, Canada’s criticism is of no consequence, since it relates to sales occurring before the effective date of the Subsidies Agreement, no matter what the proper definition of “sale.” In any event, the Clark Report adopted the “date of sale” used by Canada in trade remedy cases, where it identifies the “date of sale” as:
generally considered to be the date that the parties establish the terms of sale. The date of order confirmation is usually considered as the date of sale, although the date of sale could be the contract, purchase order or invoice date, whichever establishes the terms of sale.484
This definition would not, according to Brazil, dictate that the correct date of sale is determined by the delivery date.
ARGUMENTS OF THIRD PARTIES485
EUROPEAN COMMUNITIES
Export contingency
The European Communities disagree with Brazil's interpretation of export contingency under Article 3.1(a) SCM Agreement which would include “export propensity” or “intent to increase exports”. The European Communities state that under this approach whenever a benefit is conferred to an export-oriented undertaking with the explicit or implicit aim to increase its exports, the requirement of export contingency would be fulfilled, that is that benefits:
provided by an Organization that “exists solely to support export transactions” (para. 6.49), or given to a specific industry because it is an “export-oriented” sector (para. 6.221), or
a covenant making a transaction contingent upon maintenance of, among other things, an export-oriented sales strategy (para. 6.287), or
loans and guarantees provided to businesses with “an established export trade or significant export potential” (para. 6.311), or
loans and guarantees for the express purpose of assisting businesses to the “conquête de marchés à l'exportation” and support transactions “à l'étranger”. (para. 6.332)
484 Exh. BRA-99.
485 The footnotes and citations and the emphasis in the text are the third parties’.
WT/DS70/R
Page 144
would constitute subsidies contingent upon export performance.
The European Communities consider that Brazil's approach is unjustifiably broad and puts all industrial promotion activities into the category of potential prohibited subsidies, and does not take into account that the SCM Agreement distinguishes three categories of subsidies: prohibited subsidies, actionable subsidies and non-actionable subsidies. The European Communities believe that Brazil's approach is in contradiction to this basic distinction by declaring even certain non-actionable subsidies to be prohibited.
According to the EC Article 3.1(a) of the SCM Agreement and its footnote provide for a much more limited definition of prohibited subsidies. For the European Communities, the plain meaning of the words “contingent” and “tied” require conditionality between the subsidies and exportation or export performance. The European Communities support Canada's view that the effect of a subsidy or the objective of a subsidy cannot on their own be sufficient to establish de facto export contingency under Article 3.1(a). The European Communities submit that exportation, or export earnings must be a condition for the grant of the subsidy, and that such a condition would be fulfilled in cases where exports arise out of the desire of companies to take advantage of a subsidy that can only be obtained if exports take place; in other words, it would cover cases where, although the subsidy is not specifically linked to export performance, in practice companies are obliged to export if they are to fulfil the requirements for obtaining the subsidy.
The European Communities submit that the language of Article 3.1(a) was the result of a long debate between the opposing views represented in the Uruguay Round negotiations. For the European Communities, Canada's explanation of the negotiating history demonstrates that the broad view being put forward by Brazil was excluded from the text through a succession of improvements in the wording. Thus, the European Communities argue what was called the “quantitative approach”, as well as the “export propensity” test to determine whether a given measure was a prohibited subsidy were rejected in the context of the Negotiating Group on Subsidies and Countervailing Measures486, and similarly, the evolution of footnote 4 demonstrates the negotiator's rejection of an objective or intent-based test in favour of a conditions-based test.
The European Communities also agree with Canada's view that as a practical matter the following factors are useful in determining whether a subsidy is in fact contingent upon export performance:
evidence that the subsidy would not have been paid but for the exports flowing from it;
whether there are penalties – in the sense of reduction or withdrawal of payments – if exports do not take place; or
whether there are bonuses or additional payments if exports do take place (para. 5.56).
Regarding a hypothetical example posited by the Panel, in which a WTO Member administered a subsidy programme eligibility for which was restricted de jure to enterprises which qualified, based on past performance, as “export-oriented”, the European Communities remark that the problem becomes clearer when the case is even more hypothetical, i.e., where eligibility is restricted de jure to enterprises which exported a certain product in the year 1900. It would not in the view of the European Communities be within the intent of Article 3.1(a) of the SCM Agreement that this be considered a de jure export contingent subsidy in such circumstances. In the EC’s view, therefore, the term “’export performance‘” in Article 3.1(a) can only mean present or future export performance. The European Communities argue that this interpretation is supported by footnote 4 to
486 Note by the Secretariat, Negotiating Group on Subsidies and Countervailing Measures, Meeting of 27-28 March 1990, MTN.GNG/NG10/17 at para. 3.
WT/DS70/R
Page 145
Article 3.1(a) which refers to subsidies which are “’in fact tied to actual or anticipated exportation or export earnings‘”. The European Communities also consider that export performance within the meaning of Article 3.1 of the SCM Agreement must be measurable in some way.
The European Communities consider that the subsidy described in the example should probably be considered a prohibited subsidy as being de facto contingent, as in the real world, it is hard to see how a subsidy based de jure on past export performance would be granted for any other reason than to enhance anticipated exportation or export earnings in the future. Therefore, in the absence of other evidence to the contrary, for the European Communities it would be reasonable to make this assumption and consider the scheme in question to be a de facto export subsidy. This would in particular be the case if the scheme were kept in force for a period of time or was regularly repeated, as it would then constitute an incentive to all producers to start exporting to benefit from the subsidy.
Regarding the example posited by the United States (paras. 7.41-7.42), the European Communities in response to a question from the Panel indicate that the example is more instructive if slightly varied or clarified such that an eligibility criterion for a grant is that a firm must submit a plan explaining how the construction of the facilities will affect the firm’s exports or export earnings. As there is no de jure link between the granting of the subsidy and future exports, in order to determine whether such a requirement constitutes a de facto export subsidy, the European Communities maintain that it would be necessary to examine how the scheme is applied in practice. For the European Communities, if the grant is granted whether the plan shows that export earnings will increase or decrease, it is clearly not contingent on export performance and not a prohibited subsidy; if the grant is discretionary and only in fact given to firms which will increase their export performance, it may be considered de facto export contingent. The European Communities note that future exports may not be according to the plan for some unforeseen reason without the subsidy being refundable, but that if it is because the plan was fraudulent, the subsidy will no doubt be recoverable and the scheme is therefore contingent upon export performance in the form of a good faith execution of the plan.
For the European Communities, the US example is misleading since it speaks of “’a plan explaining how the construction of such facilities will increase the firm’s exports or export earnings‘”. Such an ‘”increase‘” will in practice inevitably be measurable and the firm will be committed to a good faith execution of the plan. Therefore, in the view of the European Communities the US example sets a condition for the granting of the subsidy which is tied to anticipated exportation or export earnings and would fall squarely within the conditionality approach advanced by the European Communities; the subsidy will be de facto export contingent and prohibited.
The European Communities contend that the US example demonstrates the need to perform a more subtle analysis than proposed by the United States and examine how the eligibility criterion is applied, for example by examining whether applications for aid would be refused where the submitted plans showed that the subsidy (i) would not affect exports at all (i.e., they would remain at the current level), or (ii) would only affect exports in proportion to the overall increase in production (thus maintaining the current domestic/export sales ratio). If the application for aid was accepted in both cases, then in the view of the European Communities it is not necessarily an export subsidy as it is granted also to producers selling on the domestic market.
Annex I (k), first paragraph SCM Agreement
The European Communities comment on the “safe haven” provided by the second paragraph of Item (k) to Annex I of the SCM Agreement for export credit practices which conform to the conditions referred to therein.
WT/DS70/R
Page 146
The European Communities note that Canada does not defend the financing and loan guarantees of EDC on this basis but rather argues that EDC financed loan guarantees “no material advantage” in the field of export credit terms and that its finance is not on “concessionary terms”.
For the European Communities, the OECD Guidelines on Officially Supported Export Credits to clearly satisfy the requirements of paragraph 2 of Item (k) of Annex I SCM Agreement. The European Communities submit that the importance of the OECD Guidelines derives from footnote 5 to Article 3.1 of the SCM Agreement, which states that:
“Measures referred to in Annex I as not constituting export subsidies shall not be prohibited under this or any other provision of this Agreement.”
The European Communities argue that reading this footnote together with paragraph 2 of Item (k) of Annex I SCM Agreement leads to the conclusion that export credit activities which conform with an international undertaking on official export credits, as described in paragraph 2 of Item (k) of Annex I SCM Agreement, and, therefore, all export credit activities which conform with the OECD Guidelines, are not prohibited under Article 3.1 or any other provision of the SCM Agreement. That is, for the purposes of the SCM Agreement, paragraph 2 of Item (k) of Annex I SCM Agreement creates a “safe haven” for Member export credit practices which conform to the OECD Guidelines.
The European Communities defend the principle of application of the OECD Guidelines on officially supported export credits to all the organizations which fall under its scope. It considers that the interest rates allowed by the OECD Guidelines should be used in the case of public financing support. The European Communities argue that publicly supported rates below those of the Arrangement fall outside the “safe haven” and are subject to the full rigours of the SCM Agreement. If they constitute a subsidy and are export contingent, they are prohibited.
In response to a Panel question regarding the EC’s understanding of the term “’interest rate provisions‘” of the relevant undertaking in item (k) of the Illustrative List of Export Subsidies, the European Communities reiterate that the first paragraph of Annex I(k) is to be considered an illustrative prohibition (i.e., it does not exhaustively define the scope of Article 3.1(a) in this sector), and that the second paragraph by contrast contains an exception, not only from the first paragraph but from the whole of the SCM Agreement (the OECD “’safe haven‘”).
For the European Communities, these are important principles to recall since the parties and the other third party do not appear to have correctly construed the provisions of item (k). Particularly noteworthy to the European Communities in this regard is the position of the United States (in para. 7.45) where the US asserts, without any reasoning and in contradiction to the clear text of the SCM Agreement that:
item (k) establishes the exclusive standard for determining whether the practices described therein constitute export subsidies.
For the European Communities it is clear from the text of Article 3.1(a) (“’including those illustrated in Annex I [emphasis added by the European Communities]) and the title to Annex I itself (“’Illustrative List of Export Subsidies‘” [emphasis added by the European Communities]), that in general the paragraphs of Annex I contain illustrative prohibitions. It is only otherwise where the Annex provides that a certain measure does not constitute an export subsidy. In such a case the measure is, by virtue of footnote 5 to Article 3, not prohibited under any provision of the SCM Agreement, according to the European Communities.
The European Communities maintain that this is the case of the second paragraph of item (k) to Annex I of the SCM Agreement which creates a “’safe haven‘” for the export credit practices
WT/DS70/R
Page 147
which conform to the conditions therein. The European Communities submit that the second paragraph of item (k) and in particular the term “’interest rate provisions‘” must be interpreted in the light of the fact that it creates an exception to the whole SCM Agreement.
According to the European Communities, if these words were taken to refer, in the case of the OECD Arrangement, only to the provisions on minimum interest rates, any “’export credit practice‘” which respected those provisions would be deemed not prohibited by the SCM Agreement. Since the term “’export credit practice‘” can cover much else besides the provision of interest rates below those provided for in the Arrangement, the European Communities argue that these other practices would also be deemed not to be prohibited by the Agreement even if they did not comply with the Arrangement, defeating the object and purpose of the provision. It would for example be possible to increase the term of the loan, or alter the repayment profile (which can change the average term of the credit although the total length of two credits are the same) or provide for reductions of premia and fees (Arrangement interest rates are supposed to be for risk free lending – premia for insurance/guarantees is to be paid in addition) so as to reduce the effective cost of the loan in the same way as a reduction in the interest rate.
The European Communities consider that such a wide interpretation of the exception is not warranted and that the exception should be interpreted more narrowly, meaning that the term “’interest rate provisions‘” must be taken to refer to all the provisions of the OECD Arrangement which are related to interest rates, and in particular the whole of Chapter II.
The European Communities consider that this would correspond to the object and purpose of the OECD Arrangement and thus of the second paragraph of item (k) in Annex I of the SCM Agreement, namely to prevent export credits from being used to distort competition by harmonising their conditions. Therefore, the European Communities maintain, the priority is to ensure that export credit conditions are equal and the question of whether they are commercial rates is of less importance (although the European Communities consider the interest rates in the Arrangement to be broadly in line with the total costs of commercial loans with the same terms (credit length, etc.)). In the view of the European Communities, this objective of equalising conditions of competition in export credits would be defeated if export credit practices which can be translated into interest rates were not covered.
SUBMISSION OF THE UNITED STATES
Applicability of the SCM Agreement to subsidies provided prior to the entry into force of the WTO Agreement
The United States argues that, contrary to Canada’s argument (para. 4.74), the SCM Agreement does apply to subsidies provided prior to the entry into force of the WTO Agreement. In the view of the United States, Canada’s arguments are wrong for two reasons. First, putting aside the question of a different intention in the SCM Agreement or the WTO Agreement, Article 28 of the Vienna Convention states that a treaty does not apply to “’any situation which ceased to exist‘” before the entry into force of a treaty. In this case, the United States notes, the “’situation‘” complained about by Brazil is subsidization. The United States argues that depending on the nature of the alleged subsidies provided under DIPP and SDI, it may be the case that subsidization did not cease as of 1 January 1995, the date on which the WTO Agreement entered into force with respect to Canada.
More specifically, the United States argues that it is well-accepted that the benefits of certain types of subsidies can last for a period of years, noting that this concept is recognized in Annex IV of the SCM Agreement, paragraph 7 of which provides as follows: “’Subsidies granted prior to the date of entry into force of the WTO Agreement, the benefits of which are allocated to future production,
WT/DS70/R
Page 148
shall be included in the overall rate of subsidization‘”. (emphasis added by the United States).487 The Tokyo Round Subsidies Code Committee also recognized the concept when it adopted Guidelines on Amortization and Depreciation that called for the allocation of certain subsidies over time.488 Moreover, to the best of the US’ knowledge, all major users of the countervailing duty remedy, including Canada, allocate certain subsidies over time, although the mechanics of allocation methodology may vary from country to country. Finally, for the United States, common sense dictates that, for example, the provision of a $100 million grant to a firm does not constitute a subsidy only with respect to merchandise produced by the firm on the day the grant is received. Instead, the grant benefits merchandise produced over a period of time.
The United States maintains that because certain subsidies bestowed in the past can be deemed to benefit current or future production, it follows that a situation of subsidization can exist long after a subsidy is initially bestowed. Where this is the case, in the United States’s view, the fact that the action creating the situation of subsidization may have been taken prior to 1 January 1995 does not mean that the situation of subsidization is exempt from the obligations and remedies of the SCM Agreement if that situation persists after 1 January 1995.
For the United States, the second reason why Canada is wrong is that the SCM Agreement does, in fact, demonstrate that it applies to subsidies provided prior to 1 January 1995, recalling paragraph 7 of Annex IV, which expressly provides that pre-WTO subsidies must be included for purposes of Article 6.1(a). Thus, the United States maintains, Canada is clearly wrong when it states that the SCM Agreement “’does not . . . apply to events that took place prior to 1 January 1995.’”
The United States submits that under Article 28 of the Vienna Convention, the DIPP and SDI subsidies could be subject to the SCM Agreement.489 Therefore, in the view of the United States, this preliminary objection of Canada’s is unfounded.
In response to a Panel question whether the distinction between the SCM Agreement’s parts II and III is relevant to the question of temporal application of the part II prohibition (in light of the US argument based on the serious prejudice provisions including paragraph 7 of Annex IV), the United States replies that this distinction is not relevant. Here, the United States recalls the reference in paragraph 7 of Annex IV to subsidies “’the benefits of which are allocated to future production‘” (emphasis added by the United States), and argues that the wording of paragraph 7 makes clear that insofar as the allocation of subsidies is concerned, the drafters did not intend paragraph 7 to be a special rule applicable exclusively to serious prejudice cases involving Article 6.1(a) of the SCM Agreement. Instead, the drafters took as a given the notion that the benefits of certain types of subsidies should be allocated to future production (i.e., should be allocated over time), and added paragraph 7 in order to clarify how these types of subsidies figured into the calculation of the ad valorem subsidy rate for purposes of Article 6.1(a). For the United States, the phrase “’the benefits of which are allocated‘” connotes a proposition that is universally applicable, regardless of the context, and had the drafters intended otherwise, they would have used different language, such as the following:
“Subsidies granted prior to the date of entry into force of the WTO Agreement [the benefits of which are] shall be allocated to future production and shall be included in the overall rate of subsidization.”
487 Annex IV deals with the calculation of the total ad valorem subsidization for purposes of Article 6.1(a) of the SCM Agreement.
488 BISD 32S/154. These guidelines were applicable to countervailing duty proceedings.
489 Whether those subsidies are, in fact, subject to the SCM Agreement raises a factual question on which the United States does not opine. To answer this question, the Panel would have to determine (1) whether these subsidies are of the type that should be allocated over time; and (2) if they are, the number of years over which they should be allocated.
WT/DS70/R
Page 149
Moreover, the United States maintains, even if the drafters had intended that paragraph 7 function as an acknowledgement of the propriety of allocating subsidies over time for purposes of Article 6.1(a), the fact that paragraph 7 expressly recognizes, for purposes of Article 6.1(a) of the SCM Agreement, that the benefits of subsidies can last for a period of years does not mean that in other contexts the benefits of subsidies cannot last for a period of years. Put differently, the fact that the drafters expressly recognized in one portion of the SCM Agreement the concept of allocating subsidy benefits over time does not mean that the drafters intended to preclude the application of this concept in other portions of the Agreement.
The United States also notes that it did not cite only paragraph 7 of Annex IV in its third party submission, but also referred to experience under the Tokyo Round Subsidies Code, the countervailing duty practice of Members (including Canada), and common sense, the latter probably being the most important factor of all.
Finally, the United States notes that it also cited paragraph 7 of Annex IV for purposes of responding to Canada’s across-the-board assertion that the SCM Agreement does not apply to subsidies provided prior to 1 January 1995. If nothing else, the United States argues, the existence of paragraph 7 effectively rebuts such a sweeping assertion.
Canada’s interpretation of “export subsidy”
The United States does not take a position on whether the measures at issue are, in fact, export subsidies. The United States agrees with Canada that the first Brazilian submission is somewhat cryptic on the question of what constitutes an export subsidy, and notes that not understanding Brazil’s legal theory, it is difficult to comment on Brazil’s arguments.
The United States strongly disagrees with Canada’s interpretation of the term “export subsidy”. In the view of the United States, Canada’s interpretation is not supported by the relevant rules of treaty interpretation,490 and, if accepted, would eviscerate the stronger disciplines against the use of export subsidies that were one of the principal achievements of the Uruguay Round subsidy negotiations.
The United States notes Canada’s interpretation that: “’Article 3.1 prohibits subsidies that are, in law or in fact, contingent upon or tied to export performance. A subsidy is so contingent or tied when it is available only on condition that exports take place.‘” The United States recalls the following factors identified by Canada as “’useful in determining whether subsidies are in fact contingent upon export performance:
evidence that the subsidy would not have been paid but for the exports flowing from it;
whether there are penalties -- in the sense of reduction or withdrawal of payments -- if exports do not take place; or
whether there are bonuses or additional payments if exports do take place.‘” [emphasis in original] (para. 5.56)
For the United States, these conclusions are premised on a flawed interpretative analysis. First, Canada places great reliance on the presence of the word “performance” in Article 3.1(a), concluding that “’on its plain meaning, Article 3.1(a) applies only to subsidies that are conditional on exports being executed or accomplished.‘”491 However, the United States argues, Article 3.1(a) does
490 Articles 31 and 32 of the Vienna Convention.
491 Id., para. 82.
WT/DS70/R
Page 150
not, as Canada suggests, state that exports must actually be executed or accomplished. Instead, Canada is impermissibly trying to read words into Article 3.1(a) that are not there.492
Second, the United States maintains that throughout its discussion of export subsidy, Canada ignores one key word: “anticipated”. Footnote 4 to Article 3.1(a) provides that a de facto export subsidy exists “’when the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in fact tied to actual or anticipated exportation or export earnings.‘” (emphasis added by the United States). For the United States, the ordinary meaning of “’anticipate‘” is to “’look forward to; colloq. expect.‘”493 Thus, under Article 3.1(a), a de facto export subsidy can exist where the granting of a subsidy is tied to expected exportation or export earnings, according to the United States, and there is no requirement that exportation or export earnings actually have occurred or that penalties be imposed if expectations are not fulfilled. Therefore, the United States submits, the text of Article 3.1(a) contradicts Canada’s assertion that the intent or objective of a subsidy is irrelevant.
Finally, the United States argues, Canada’s discussion of the negotiating history behind Article 3.1(a) is inaccurate and incomplete. In the view of the United States, it is inaccurate because Canada concludes that the evolution of the text somehow demonstrates that intent became irrelevant (paras. 5.75- 5.76), but the word “’anticipated‘” connotes a test based on intent. For the United States, all that the various draft versions of Article 3.1(a) show is that the manner in which the drafters express the intent factor changed over time.
The United States asserts that Canada’s discussion of the negotiating history is incomplete, because it ignores the reason why footnote 4 was added in the first place. According to the United States, footnote 4 originated with a proposal by the European Communities, which the European Communities explained as follows:
The prohibition of export subsidies in Article 9 of the Subsidies Code should be reformulated in order to define clearly its scope.
This prohibition must apply to all export subsidies, that is, all government interventions which confer, through a charge on the public account (in the form of direct financial outlays or revenue foregone, such as tax relief and debt forgiveness), a benefit on a firm or an industry contingent upon export performance.
In addition, since experience has shown that government practices may be easily manipulated or modified in order to avoid this prohibition, it is apparent that a prohibition only of those subsidies which are de jure (that is, expressly) made contingent upon export performance is open to circumvention.
The prohibition, in the present discipline also applies to subsidies de facto contingent upon export. This, however, makes it necessary to provide for clearer guidance in identifying de facto export subsidies . . . de facto export subsidies are those where facts which were known -- or should clearly have been known -- to the government when granting the subsidy demonstrate that the
492 As the Appellate Body stated in European Communities - Measures Affecting Meat and Meat Products (Hormones), WT/DS26/AB/R, Report of the Appellate Body adopted 13 February 1998, para. 181, “The fundamental rule of treaty interpretation requires a treaty interpreter to read and interpret the words actually used by the agreement under examination, and not words which the interpreter may feel should have been used.”
493 The New Shorter Oxford English Dictionary (1993).
WT/DS70/R
Page 151
subsidy, without having been made expressly contingent upon export performance, was indeed intended to increase exports.494
The United States submits that while the final text of footnote 4 does not contain a “’knowledge‘” test as originally articulated by the European Communities,495 the word “’anticipated‘” is consistent with a test that takes into account the intent of the subsidizing government.
For the United States, Canada’s discussion of the negotiating history also is incomplete because it fails to identify another change that was made in order to expand the scope of the export subsidy category. Specifically, instead of requiring that export performance be the “’only‘” or the “’most important‘” element, Article 3.1(a) provides that export performance may be either the sole contingency for the subsidy or merely “’one of several other conditions.‘” Thus, the United States argues, Canada’s constricted interpretation of export subsidy is at odds with this textual evidence of an intent to craft a more liberal test for identifying an export subsidy.
The United States submits that the flaws in Canada’s reasoning can best be demonstrated with an example where a government establishes a subsidy programme that provides large grants for the construction of production facilities. The only eligibility criterion for a grant is that a firm must submit a plan explaining how the construction of such facilities will increase the firm’s exports or export earnings. If a firm does not submit such a plan, it cannot receive a grant. Once the grant is bestowed, however, the government does not impose penalties if the export plan is not fulfilled.
In the view of the United States, according to Canada such a programme would not constitute an export subsidy, because grants would not be contingent on the condition that exports actually take place. In the view of the United States, such a programme clearly would be an export subsidy under Article 3.1(a) because the grants would be tied to anticipated exportation or export earnings.
The United States submits that in this dispute the Panel should not apply Canada’s erroneous standard for identifying an export subsidy. If adopted, this standard would enable governments to engage in the very sorts of manipulation and modifications of subsidy programmes that the drafters of the SCM Agreement sought to curtail.
The relevance of item (k) of the Illustrative List to export financing activities of the EDC
The United States notes that in its discussion of the export financing activities of the Export Development Corporation (“EDC”), Canada appears to distinguish between the activity of the EDC under its corporate account and activities under the Canada Account, appearing to argue that activities under the Canada Account are governed by item (k) of the Illustrative List of Export Subsidies contained in Annex I to the SCM Agreement, but that activities under the corporate account are governed by some other standard.
In the view of the United States, both activities (the corporate account and the Canada Account) are governed by item (k). Thus, the United States submits, if financing provided under either account is inconsistent with the standard set forth in the first paragraph of item (k), the financing is potentially an export subsidy. Conversely, if financing provided under either account is consistent with the standard set forth in the first paragraph, it is not an export subsidy in the US view, because item (k) establishes the exclusive standard for determining whether the practices described therein constitute export subsidies. In addition, the United States argues, even if financing provided under either account should run afoul of the first paragraph of item (k), it still would not be deemed an
494 “Elements of the Negotiating Framework,” Submission by the European Communities, MTN.GNG/NG10/W/31 (27 November 1989).
495 And correctly so, because it would be nonsensical to have a standard based on a government’s knowledge of it own intent.
WT/DS70/R
Page 152
export subsidy if the financing conformed to the provisions of the OECD Consensus referred to in the second paragraph of item (k).
In response to a Panel question regarding the United States’ understanding of the term “interest rate provisions” of the relevant undertaking in item (k) of the Illustrative List of Export Subsidies, the United States responds that the “’interest rate provisions‘” of the undertaking (i.e., the OECD Arrangement) provide rules for the provision by governments of fixed interest rate financing (either through direct lending or interest rate support mechanisms) or pure cover (provision of guarantees or insurance to banks providing either fixed or floating rate financing). The “’interest rate provisions‘” do not provide rules for the provision by governments of floating interest rate financing (e.g., through direct lending). Thus, the United States states, if the EDC were to provide export subsidies in the form of a direct loan at a floating rate of interest, the second paragraph of item (k) would not apply to such financings.
In its comments on the draft descriptive part of this report, the United States requested permission from the Panel to withdraw its submissions in this dispute, and asked the Panel to modify the descriptive part of the reports so as not to reflect those submissions. The United States further asked that, if the Panel declined the United States' request to withdraw its submissions, the descriptive part of the report reflect that this request had been made."496
INTERIM REVIEW
On 25 February 1999, both parties requested the Panel to review, in accordance with Article 15.2 of the DSU, precise aspects of the interim report issued on 17 February 1999. While neither party requested an additional meeting with the Panel, both parties provided written replies to certain of one another's comments.
COMMENTS BY BRAZIL
Brazil identified a number of typographical errors in the interim report, which have been corrected by the Panel.
496 Brazil submitted a letter requesting the Panel to decline the request of the United States, in view of practical and substantive concerns over the request. In Brazil's view, complete deletion of the US submissions would pose logistical problems in these time-constrained proceedings, as both the section containing the US submissions, and all references in the parties' arguments to those submissions, would need to be removed. Substantively, Brazil states that such deletion would prejudice the rights of the parties, because parties (and all Members) have an interest in preserving an accurate and complete record of the proceedings; and because parties may have either developed or foregone particular lines of argument in response to arguments made by the United States. In Brazil's view, granting the request would prejudice these interests, and would be contrary to the twin goals of "security and predictability" identified by Article 3.2 of the DSU as central to the WTO dispute settlement system. Brazil believes that the Panel does not have the authority to fulfill the US request, in view of the provision of Article 10.2 of the DSU that third party submissions "'shall be reflected in the panel report'" (emphasis added by Brazil), and the provision of Rule XXV(2) of the Working Procedures for Appellate Review that third party written submissions, recorded oral statements and written answers to questions are part of the record of the panel proceedings. The Panel, in view of the practical difficulties of removing the US submissions and all references thereto from this report, in view of the strong opposition of one of the parties, and in view of the fact that the Panel had asked the parties to submit comments on specific aspects of the US submissions, declined the United States' request.
WT/DS70/R
Page 153
Regarding para. 9.85, Brazil commented that the Panel initially overlooked Brazil's reaction to Canada's argument set forth in para. 9.84. We amended para. 9.85 accordingly.
Brazil commented that, at para. 9.199, the Panel has mischaracterized its statements regarding EDC equity infusions. With reference to para. 59 of its oral statement at the second substantive meeting (see para. 6.136 above), Brazil denies that it considers the fact that CRJ Capital purchases and/or leases aircraft to be an "essential element" of its claim.
We have amended the final sentence of para. 9.199, in order to clarify that it is the Panel's understanding that the fact that CRJ Capital allegedly purchases and/or leases aircraft is an "essential element" of Brazil's claim regarding EDC equity financing. The Panel's understanding is based on Brazil's submissions throughout the Panel process, in which Brazil repeatedly argued that EDC equity infusions confer a "benefit" because they enable CRJ Capital (which, as indicated at note 233 above, Brazil treats as a Special Purpose Company) to offer airlines lower lease payments than would be available on the market. Brazil's repetition of the factual basis for this argument is reflected inter alia at paras. 6.103, 6.105, 6.106, 6.108, 6.109, 6.110, 6.111, 6.114, 6.116 and 6.135 above. The only reference by Brazil to EDC equity infusions conferring a "benefit" through lower loan payments was included at para. 59 of its oral statement at the second substantive meeting (see para. 6.136 above). We note, however, that at para. 57 of the very same oral statement (see para. 6.135 above), Brazil asserted:
Brazil's claim is that EDC, directly or indirectly, has made equity infusions into CRJ Capital which have facilitated CRJ Capital's ability to lease or sell Canadian regional aircraft at a reduced price." (italics in original, bold emphasis supplied)
For these reasons, and especially in light of Brazil's own description of its claim at para. 6.135 above, we believe that we have correctly identified the fact that CRJ Capital allegedly purchases and/or leases aircraft as an "essential element" of Brazil's claim regarding EDC equity infusions.
With regard to paras. 9.220-9.224, Brazil commented that the Panel could include an additional source as further support for its conclusions regarding Canada Account debt financing. Brazil refers in particular to the Canada Account Profile, which states that a portion of Canada Account financing is extended on "concessional" terms. Brazil asserts that the Canada Account Profile defines "concessional" to include "interest-free or low-interest loans repayable over extended periods." Canada replied that Canada Account "concessional" financing falls outside the Panel's jurisdiction, because it has not been used in respect of civil aircraft since 1995. Canada referred to its argument at para. 6.164 above in this regard.
In the present instance, the Panel was required to determine whether or not the Canada Account debt financing in issue constitutes a "subsidy" within the meaning of Article 1.1 of the SCM Agreement. Whether or not such debt financing is "concessional" within the meaning of the Canada Account Profile is of no relevance to the Panel's findings. For this reason, we have not made the change requested by Brazil.
COMMENTS BY CANADA
Canada suggested a number of changes to the Panel's description of its arguments. To the extent that Canada indicated the source for these changes, they have been made by the Panel. Canada also made a comment on the relevance of certain text in para. 4.126, without requesting any specific changes in this regard. In addition, Canada identified a number of typographical errors in the interim report, which have been corrected by the Panel.
At the request of Canada, we included the phrase ",on 21 December 1998," in the second sentence of para. 9.64.
WT/DS70/R
Page 154
At the request of Canada, we included the phrase "as well as one from Exinvest" in the last sentence of para. 9.193, and pluralized "Officer's Certificates" in para. 9.194.
With regard to para. 9.246, we clarified our finding that there is no prima facie case that the sale by the OAC of its 49 percent share in de Havilland to Bombardier in 1997 constitutes a prohibited export subsidy.
In respect of para. 9.253, Canada requested changes to the description of Canada's reaction to the Panel's request for information. Brazil objected to Canada's request, stating that the Panel's text "accurately describe[d] the purported rationale supporting Canada's refusal to comply with the Panel's request for information. … Moreover, to adopt the modifications … advocated by Canada would result in an inaccurate description of Brazil's position." In order to avoid any uncertainty, the Panel included at para. 9.253 the full text of Canada's response to the Panel's request, and at para. 9.254 the full text of Brazil's comment on Canada's response.
At the request of Canada, we included in para. 9.289 Canada's argument that royalties can be tied "to total sales of the recipient enterprise."
With regard to para. 9.290, Canada asked the Panel to include an additional element in its description of Canada's argument against reliance on the latest WTO Trade Policy Review of Canada. Brazil objected to Canada's request. The Panel's views on the relevance of the latest WTO Trade Policy Review of Canada are set forth at para. 9.274. None of the Panel's findings are based on the latest WTO Trade Policy Review. Since the Panel attaches no importance to this Trade Policy Review, and since Canada's specific argument on this issue is in any event set forth in full at para.
6.199 above, we do not consider it necessary to make the change requested by Canada.
FINDINGS
INTRODUCTION
This dispute concerns prohibited export subsidies allegedly provided by the Government of Canada or its provinces to the Canadian civil aircraft industry.
On 10 March 1997, Brazil requested consultations with Canada concerning "certain subsidies granted by the Government of Canada or its provinces that support the export of civilian aircraft from Canada."497 Consultations were requested under Article 4 of the Agreement on Subsidies and Countervailing Measures (hereinafter "SCM Agreement"). Consultations were held in Geneva on 30 April 1997, but the parties failed to reach a mutually satisfactory solution.
On 10 July 1998, Brazil requested the establishment of a panel498 pursuant to Article 4.4 of the SCM Agreement and Article 6 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (hereinafter the "DSU"). This Panel was established on 23 July 1998, with the following standard terms of reference:
"To examine, in the light of the relevant provisions of the covered agreements cited by Brazil in document WT/DS70/2, the matter referred to the DSB by Brazil in that document and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements".
497 WTO document WT/DS70/1.
498 WTO document WT/DS70/2.
WT/DS70/R
Page 155
PRELIMINARY ISSUES
The parties requested a number of preliminary rulings, and raised a number of preliminary issues, concerning the jurisdiction of the Panel, the temporal application of the SCM Agreement, fact- finding by the Panel, procedures for the protection of business confidential information, deadlines for the submission of new evidence and allegations, deadlines for the submission of affirmative defences, the Panel's right to seek information in respect of defences not raised, the Panel's right to seek information without finding the establishment of a prima facie case, and the Panel's right to seek information in respect of allegations not made.
Jurisdiction of the Panel
Canada has raised two preliminary issues concerning the jurisdiction of the Panel. The first issue is whether certain measures identified in Brazil’s request for a panel had been the subject of a request for consultations and were in fact consulted upon, and thus could be the subject of a panel request under Article 4.4 of the SCM Agreement. The second issue is whether certain of the measures identified in Brazil’s request for a panel were sufficiently specific as required by Article 6.2 of the DSU. Canada asked the Panel to issue the requested rulings prior to the deadline for the parties’ first submissions.
Article 4.4 of the SCM Agreement
Arguments of the parties
Canada recalls that, with regard to the Export Development Corporation ("EDC"), Brazil's request for establishment of a panel referred to "financing and loan guarantees provided by the [EDC], including equity infusions into corporations established to facilitate the export of civil aircraft", whereas Brazil's request for consultations referred to "[EDC] equity infusions into corporations specially established to facilitate the export of aircraft", and "EDC loan guarantees for exported aircraft".
Canada submits that Brazil's claim concerning EDC "financing" is outside the Panel's jurisdiction because that claim was not included in Brazil's request for consultations. Canada notes that the only EDC activities referred to in Brazil's request for consultations were "equity infusions" and "loan guarantees". Canada asserts that "financing" is a form of activity distinct from either "equity infusions" or "loan guarantees".
Canada recalls that Article 4.1 - 4.4 of the SCM Agreement provides:
Whenever a Member has reason to believe that a prohibited subsidy is being granted or maintained by another Member, such Member may request consultations with such other Member.
A request for consultations under paragraph 1 shall include a statement of available evidence with regard to the existence and nature of the subsidy in question.
Upon request for consultations under paragraph 1, the Member believed to be granting or maintaining the subsidy in question shall enter into such consultations as quickly as possible. The purpose of the consultations shall be to clarify the facts of the situation and to arrive at a mutually agreed solution.
If no mutually agreed solution has been reached within 30 days of the request for consultations, any Member party to such consultations may refer the matter to the
WT/DS70/R
Page 156
Dispute Settlement Body ("DSB") for the immediate establishment of a panel, unless the DSB decides by consensus not to establish a panel. (footnote omitted)
Canada submits that there must be a rational connection between "the subsidy in question" identified in the complaining party's request for consultations under Article 4.2 of the SCM Agreement, and the "matter" referred to the DSB in the request for establishment of a panel under Article 4.4 of the SCM Agreement. Canada submits that the "matter" must necessarily follow from the "subsidy in question". Canada asserts that the SCM Agreement is not observed, and due process is not served, if Canada is forced to respond to a "matter" on which no consultations were requested.
Brazil makes a number of arguments in support of its request that the Panel reject this preliminary issue raised by Canada. First, Brazil asserts that the identification of the EDC in its request for consultations, combined with the broad language of that request of a panel (" … certain subsidies granted by the Government of Canada or its provinces that support the export of civilian aircraft from Canada"), covered all subsidy aspects of the EDC. Second, Brazil asserts that EDC "financing" was discussed during consultations with Canada. Third, Brazil denies that there is a marked distinction between "financing", and "equity infusions" and "loan guarantees". Brazil argues that "financing" is a broader, more general term encompassing direct lending, debt and equity support. Fourth, Brazil argues that a request for establishment will often not be identical to a request for consultations, because of the refinement of claims likely to result from consultations.
Evaluation by the Panel
We note that Canada's argument is based exclusively on Article 4.1 - 4.4 of the SCM Agreement.499 In focusing on Article 4.1 - 4.4 of the SCM Agreement to establish the jurisdiction of the panel, we consider that Canada has overlooked Article 7.1 of the DSU. Article 7.1 defines the terms of reference of a panel, providing that a panel's terms of reference are "to examine … the matter referred to the DSB by (name of party) in [the] document [containing the party's request for establishment of a panel…" Thus, except where the parties agree on special terms of reference, the terms of reference are typically determined on the basis of the complaining party's request for establishment. The terms of reference are critical to the preliminary issue before us, since the Appellate Body stated generally in India - Patents that "[t]he jurisdiction of a panel is established by that panel's terms of reference …"500 Likewise, the Appellate Body stated in Brazil - Desiccated Coconut that "a panel's terms of reference … establish the jurisdiction of the panel by defining the precise claims at issue in the dispute." 501
In our view, a panel's terms of reference would only fail to be determinative of a panel's jurisdiction if, in light of Article 4.1 - 4.4 of the SCM Agreement applied together with502 Article 4.2 -
4.7 of the DSU, the complaining party's request for establishment were found to cover a "dispute" that had not been the subject of a request for consultations. Article 4.4 of the SCM Agreement permits a Member to refer a "matter" to the DSB if "no mutually agreed solution" is reached during consultations. In our view, this provision complements Article 4.7 of the DSU, which allows a
499 Brazil argues that EDC "financing" was consulted on. If this were the case, we consider that this fact would normally dispose of the preliminary issue raised by Canada. However, Canada does not accept Brazil's argument that EDC "financing" was consulted on, and Brazil has provided no support for its position. We are therefore unable to establish whether or not EDC "financing" was in fact consulted on.
500 India - Patent Protection for Pharmaceuticals and Agricultural Chemical Products, WT/DS50/AB/R, adopted 16 January 1998, para. 93.
501 Brazil - Measures Affecting Desiccated Coconut, WT/DS22/AB/R, adopted 20 March 1997, page
22.
502 According to the Appellate Body in Guatemala - Anti-Dumping Investigation Regarding Portland
Cement from Mexico, wherever possible, special or additional rules and procedures for dispute settlement in Annex 1 of the DSU (such as Article 4.1 - 4.4 of the SCM Agreement) should be read so as to complement the provisions of the DSU (WT/DS60/AB/R, adopted 25 November 1998, paras. 64-66).
WT/DS70/R
Page 157
Member to refer a "matter" to the DSB if "consultations fail to settle a dispute". Read together, these provisions prevent a Member from requesting the establishment of a panel with regard to a "dispute" on which no consultations were requested. In our view, this approach seeks to preserve due process while also recognising that the "matter" on which consultations are requested will not necessarily be identical to the "matter" identified in the request for establishment of a panel. The two "matters" may not be identical because, as noted by the Appellate Body in India - Patents, "the claims that are made and the facts that are established during consultations do much to shape the substance and the scope of subsequent panel proceedings"503
The terms of reference of this Panel are to:
examine, in the light of the relevant provisions of the covered agreements cited by Brazil in document WT/DS70/2, the matter referred to the DSB by Brazil in that document and to make such findings as will assist the DSB in making the recommendations or in giving the rulings provided for in those agreements.504
Accordingly, our terms of reference are determined by document WT/DS70/2, i.e., Brazil's request for establishment of this Panel. This document refers expressly to "financing … provided by the Export Development Corporation …" In principle, therefore, EDC "financing" falls within our jurisdiction. As noted above, EDC "financing" would only fall outside our jurisdiction if EDC "financing" were not part of the "dispute" on which Brazil had requested consultations. In our view, Brazil requested consultations in respect of a "dispute" concerning prohibited export subsidies allegedly provided to the Canadian civil aircraft industry by inter alia EDC. This "dispute" is also the subject of Brazil's request for establishment of this Panel. Since the EDC "financing" identified in Brazil's request for establishment of a panel was part of the same "dispute" with respect to which consultations were requested, we find that EDC "financing" falls within the Panel's jurisdiction.
Canada asked the Panel to issue the requested ruling on the Panel's jurisdiction prior to the deadline for the parties’ first written submissions. In our view, there is no requirement in the DSU for panels to rule on preliminary issues prior to the parties' first written submissions. Nor is there any established practice to this effect, for there are numerous panel reports where rulings on preliminary issues have been reserved until the final report.505 Furthermore, there may be cases where the panel wishes to seek further clarification from the parties before providing a preliminary ruling. Indeed, we considered it necessary to request such clarification in the present case. In our view, the possibility for obtaining such clarification would be lost - or at least significantly undermined - if a panel were required to rule on preliminary issues before the deadline for the parties' first written submissions. For these reasons, we rejected Canada's request for a preliminary ruling on this issue prior to the deadline for the parties' first submissions.
503 India - Patent Protection for Pharmaceuticals and Agricultural Chemical Products, WT/DS50/AB/R, adopted 16 January 1998, para. 94.
504 WTO document WT/DS70/3.
505 See, for example, European Communities - Bananas (WT/DS27//R, adopted 25 September 1997), and European Communities - Computer Equipment (WT/DS62, 67/R, adopted 22 June 1998).
WT/DS70/R
Page 158
Specificity of Brazil's request for establishment of a panel
Arguments of the parties
Canada requests a preliminary ruling that Brazil's request for establishment of a panel does not meet the requirements of Article 6.2 of the DSU. Canada recalls that Article 6.2 of the DSU provides, in relevant part, that the request for the establishment of a panel shall:
... identify the specific measures at issue and provide a brief summary of the legal basis of the complaint sufficient to present the problem clearly. ...
According to Canada, certain items in Brazil's request for establishment are inconsistent with Article 6.2 because they are too vague to provide Canada with sufficient information concerning the claims at issue, and their lack of precision prejudices Canada’s due process right to know the case against it. In support, Canada cites the Appellate Body in European Communities - Bananas, where it stated that:
Article 6.2 of the DSU requires that the claims, but not the arguments, must all be specified sufficiently in the request for establishment of a panel in order to allow the defending party and any third parties to know the legal basis of the complaint. If a claim is not specified in the request for the establishment of a panel, then a faulty request cannot be subsequently 'cured' by a complaining party's argumentation in its first written submission to the panel or in any other submission or statement made later in the panel proceeding.506
Canada submits that the reasoning of the Appellate Body is applicable to the case where a defending party is not given specific information as to the factual basis of the claims against it. Canada argues that in cases with an accelerated time-table, the time allotted to a defending party is simply not adequate from a due process perspective for the preparation of a credible defence against a vague accusation. Canada also asserts that Brazil's request for establishment is more vague than its request for consultations. According to Canada, consultations afford parties the opportunity to give more precision to a dispute, not less.
Brazil denies that its request for establishment is inconsistent with Article 6.2 of the DSU. Brazil asserts that due process must flow two ways. Brazil asserts that it would violate all notions of basic fairness to blame Brazil for any alleged lack of specificity in stating the factual bases for its claim stemming from Canada's failure to disclose those facts crucial to the process envisioned in the DSU. Brazil also argues that the aforementioned extract from the Appellate Body in European Communities - Bananas required only that the "legal basis" of the complaining party's claim be "sufficiently precise." Brazil argues that Canada's claims concerning any alleged lack of specificity of the factual basis of Brazil's claims is therefore irrelevant. Brazil submits that it identified the two components of a claim essential to maintain Canada's due process rights pursuant to Article 6.2 of the DSU: (1) the particular provisions of the SCM Agreement with which Canada is charged with failure to comply; and (2) the particular Canadian measures demonstrating that failure. Brazil states that it therefore identified its claims with "sufficient precision", as required by Article 6.2 and the Appellate Body in European Communities - Bananas.
The preliminary issue raised by Canada concerns the following items identified in Brazil's request for establishment:
"financing" provided by the EDC ;
506 European Communities - Regime for the Importation, Sale and Distribution of Bananas, WT/DS27/AB/R, adopted 25 September 1997, para. 143.
WT/DS70/R
Page 159
funds provided to the "civil aircraft industry" by Technology Partnerships Canada and "predecessor programs"; and
"benefits" provided under the Canada-Québec Subsidiary Agreement on Industrial Development and the Société de Développement Industriel du Québec.
EDC "financing"
Canada recalls its claim that this matter was not included in Brazil's request for consultations, and states that nothing in the course of consultations could, therefore, guide Canada as to the nature of Brazil's claim. Canada argues that, given the accelerated procedure of Article 4 of the SCM Agreement, it is manifestly contrary to Canada's due process rights to have to respond to a claim that could potentially cover hundreds of clients, many thousands of financing transactions over several years, and a portfolio of C$10 billion.
Brazil submits that there is a veil of secrecy around the EDC, and that this veil of secrecy should not stand in the way of Brazil's right to pursue its claim. Brazil asserts that EDC "financing" was discussed in consultations, and that Canada was therefore on notice of Brazil's claim. Brazil also notes that in a preliminary submission to the Panel, Canada defined the term EDC "financing" to mean "EDC's financing activities (direct lending)", suggesting that Canada appears to understand fully the meaning of this term.
Funds provided to the "civil aircraft industry" by Technology Partnerships Canada and "predecessor programs"
Canada asserts that the term "civil aircraft industry" is over-broad for the purpose of Article
6.2 of DSU, because "[i]t includes firms ranging from machine shops and metal treatment facilities to those involved in advanced instrumentation and communications equipment. In Canada, this comprises over 200 enterprises employing over 38,000 workers."507 Canada asserts that Brazil is not unaware of the broad nature of the "civil aircraft industry", since it provided an all-encompassing definition of the "aircraft definition" during consultations. With reference to Korea - Alcoholic Beverages, Canada notes that there is no identified tariff heading and no previous decision to guide Canada as to what Brazil might mean by the "civil aircraft industry".
Canada also argues that Brazil failed to identify with adequate specificity which "predecessor programs" to Technology Partnerships Canada ("TPC"), and which activities or transactions, are being challenged by Brazil.
Brazil notes that Canada refers to the definition of "aircraft industry" provided by Brazil during consultations. Brazil asserts that Canada therefore had sufficient knowledge of the meaning of the term "aircraft industry". Brazil denies Canada's argument that the term "civil aircraft industry" is over-broad for the purpose of Article 6.2 of DSU, quoting the Appellate Body's statement in European Communities - Computer Equipment that Article 6.2 is not violated through a "lack of precision of the terms … in the request for the establishment of a panel".508 Brazil asserts that Article
6.2 does not require the complaining party to circumscribe a narrow category of products to which the challenged measures apply. Brazil also challenges Canada's argument that Brazil should have specified the products subject to the challenged measures by use of tariff headings. Brazil cites from the panel in Korea - Alcoholic Beverages which, in interpreting the Appellate Body in European Communities - Computer Equipment, stated that "a panel request based on a broader grouping of
507 See para. 4.47 above.
508 European Communities - Customs Classification of Certain Computer Equipment, WT/DS62, 67, 68/AB/R, adopted 22 June 1998, para. 70.
WT/DS70/R
Page 160
products [than those included in an identified tariff heading] was sufficiently specific for the purposes of Article 6.2".
With regard to TPC "predecessor programs", Brazil argues that Canada must be well aware that there is only one "predecessor program" to the TPC, namely the Defence Industry Productivity Program ("DIPP").
Benefits provided under the Canada-Québec Subsidiary Agreement on Industrial Development and the Société de Développement Industriel du Québec
Canada asserts that the phrase "benefits provided" is not sufficiently specific for the purpose of Article 6.2, since it is not clear which aspect of these programmes, or which activities or transactions under these programmes, are considered by Brazil to confer a "benefit".
Brazil argues that Canada was on notice of these matters, since its claims concerning benefits under the abovementioned programmes were included in Brazil's request for consultations and were on the table for discussion. Brazil asserts that Canada failed to provide details of the Canada - Québec Subsidiary Agreements on Industrial Development ("Subsidiary Agreements") or the Société de Développement Industriel du Québec ("SDI") during consultations, and that it would be contrary to the letter and spirit of both the consultation provision and Article 6.2 of the DSU to reward Canada's failure to disclose freely the facts of these programmes by limiting the scope of the Panel's jurisdiction.
Evaluation by the Panel
We will consider the preliminary issue raised by Canada in light of the following general observations. First, we note that Canada refers on a number of occasions509 to the accelerated timetable of a "fast-track" case, suggesting that any impact on Canada's due process rights caused by the alleged absence of specificity in Brazil's request for establishment is compounded in an accelerated timetable. However, although Article 4.2 of the SCM Agreement requires the Member requesting consultations to provide a "statement of available evidence", there is nothing in either the DSU or the SCM Agreement to suggest that requests for establishment of panels for "fast-track" cases should be any more precise than requests for establishment of panels in "standard" WTO dispute settlement cases.
Second, in European Communities - Computer Equipment the Appellate Body was required to consider the specificity of the US panel request, which referred inter alia to "all types of LAN [Local Area Network] equipment". In doing so, the Appellate Body stated that:
Whether these terms are sufficiently precise to "identify the specific measure at issue" under Article 6.2 of the DSU depends, in our view, upon whether they satisfy the purposes of the requirements of that provision.
In EC - Bananas, we stated that:
It is important that a panel request be sufficiently precise for two reasons: first, it often forms the basis for the terms of reference of the panel pursuant to Article 7 of the DSU; and, second, it informs the defending party and the third parties of the legal basis of the complaint.
509 See, for example, paragraphs 40, 43 and 45 of Canada's preliminary submission regarding the jurisdiction of the Panel, dated 23 October 1998.
WT/DS70/R
Page 161
The European Communities argues that the lack of precision of the term, LAN equipment, resulted in a violation of its right to due process which is implicit in the DSU. We note, however, that the European Communities does not contest that the term, LAN equipment, is a commercial term which is readily understandable in the trade. The disagreement between the European Communities and the United States concerns its exact definition and its precise product coverage. We also note that the term, LAN equipment, was used in the consultations between the European Communities and the United States prior to the submission of the request for the establishment of a panel and, in particular, in an "Information Fiche" provided by the European Communities to the United States during informal consultations in Geneva in March 1997. We do not see how the alleged lack of precision of the terms, LAN equipment and PCs with multimedia capability, in the request for the establishment of a panel affected the rights of defence of the European Communities in the course of the panel proceedings. As the ability of the European Communities to defend itself was not prejudiced by a lack of knowing the measures at issue, we do not believe that the fundamental rule of due process was violated by the Panel.510 (bold emphasis supplied) (footnotes deleted)
We consider it appropriate to apply a similar standard in determining whether Brazil's request for establishment meets the requirements of Article 6.2 of the DSU in the present case. In particular, we shall consider whether any alleged imprecision in Brazil's request for establishment affected Canada's due process rights of defence in the course of the Panel proceedings. Indeed, we understand Canada to advocate a similar interpretation of Article 6.2, since Canada asserts that Brazil's "lack of precision prejudices Canada's due process right to know the case against it. These claims are therefore inconsistent with Article 6.2 of the DSU."511 (emphasis supplied). Thus, we understand Canada to argue that Brazil's request for establishment would not be inconsistent with Article 6.2 of the DSU if the alleged lack of precision did not prejudice Canada's due process right to know the case against it.
We note Canada's argument that a party's request for establishment of a panel should be more specific than its request for consultations. As a general rule, it may be true that a request for establishment will be more specific than a request for consultations. However, we consider that Article 6.2 of the DSU is concerned exclusively with a party's request for establishment. Thus, the consistency of a party's request for establishment with Article 6.2 of the DSU should be judged exclusively in light of the specificity of the request for establishment, and not in light of the specificity of the party's earlier request for consultations.
Finally, we recall that Canada asked the Panel to rule on the consistency of Brazil's request for establishment with Article 6.2 of the DSU prior to the deadline for the parties’ first written submissions. We recall our finding that there is no requirement in the DSU for panels to rule on preliminary issues prior to the parties' first written submissions. Nor is there any established practice to this effect, for there are numerous panel reports where rulings on preliminary issues have been reserved until the final report.512 Furthermore, we have stated above that we will decide this preliminary issue by determining whether any alleged imprecision in Brazil's request for establishment prejudiced Canada's due process right of defence during the panel process. We can necessarily only undertake such an analysis at the end of the panel process. For these reasons, we
510 European Communities - Customs Classification of Certain Computer Equipment, WT/DS62, 67, 68/AB/R, adopted 22 June 1998, paras. 68, 69 and 70.
511 Canada's preliminary submission regarding the jurisdiction of the Panel, dated 23 October 1998,
para. 37.
512 See, for example, European Communities - Bananas (WT/DS27//R, adopted 25 September 1997),
and European Communities - Computer Equipment (WT/DS62, 67/R, adopted 22 June 1998).
WT/DS70/R
Page 162
rejected Canada's request for a preliminary ruling on this issue prior to the deadline for the parties' first submissions.
EDC "financing"
According to Brazil, the term "financing" does not imply a distinct form of 'direct lending', but is a broader, more general term encompassing direct lending, debt and equity support".513 By its own admission, therefore, Brazil could have chosen a more specific term in its request for establishment. However, this does not necessarily mean that Brazil's use of the term "financing" prejudiced Canada's due process right of defence during the Panel process.
In addition to EDC "financing", Brazil's request for establishment refers to EDC "loan guarantees" and "equity infusions". Given the express references to "loan guarantees" and "equity infusions", and given that EDC is primarily a lending institution,514 we consider that there is very little else left under the term "financing" besides direct lending. For this reason alone, we believe that Canada would have been on notice that Brazil intended to advance claims concerning direct lending by the EDC under the concept of "financing". We attach further importance to the fact that, in its preliminary submission concerning the jurisdiction of the Panel, Canada itself interpreted Brazil's reference to "financing" as a reference to "direct lending".515 For these reasons we consider that, insofar as Brazil advances claims concerning EDC direct lending in the course of Panel proceedings, Canada's due process right of defence is not impaired by any lack of precision in the term "financing". We therefore find that the term "financing" is sufficiently clear and specific for the purpose of Article
6.2 of the DSU, insofar as claims concerning EDC direct lending are concerned.
Funds provided to the "civil aircraft industry" by Technology Partnerships Canada and "predecessor programs"
Civil aircraft industry
We do not consider that the mere fact that the scope of a measure is identified in the request for establishment by reference to a broad product or industry grouping necessarily renders that request for establishment inconsistent with Article 6.2 of the DSU. We believe that the Appellate Body was of a similar opinion in LAN Equipment, where it shared the US concern that:
if the EC arguments on specificity of product definition are accepted, there will inevitably be long, drawn-out procedural battles at the early stage of the panel process in every proceeding. The parties will contest every product definition, and the defending party in each case will seek to exclude all products that the complaining parties may have identified by grouping, but not spelled out in 'sufficient' detail.
Although the Appellate Body's remarks were made in the context of a reference to a broad product grouping in the complaining party's request for establishment, we can see no basis for not adopting a similar approach when the request for establishment refers to a broad industry sector, such as the "civil aircraft industry". If a complaining party believes that a measure affects a broad industry sector, in our view that complaining party should be entitled to challenge that measure insofar as it affects the totality of the industry concerned, without having to spell out the individual components of that industry, and without running afoul of Article 6.2 of the DSU.
513 Brazil's reply to Canada's preliminary submission, dated 30 October 1998, page 4.
514 According to EDC's 1997 Annual Report (page 27), loan interest was by far EDC's greatest source of income during 1997. EDC's corporate plan provides for this trend to continue through 1998.
515 Canada's preliminary submission regarding the jurisdiction of the Panel, dated 23 October 1998,
para. 33.
WT/DS70/R
Page 163
Furthermore, despite the breadth of the industry sector selected by Brazil, Canada was nevertheless able to define and establish the outer-limits of that industry. Canada stated specifically that "[i]t includes firms ranging from machine shops and metal treatment facilities to those involved in advanced instrumentation and communications equipment. In Canada, this comprises over 200 enterprises employing over 38,000 workers."516 Given Canada's ability to define the outer-limits of the "civil aircraft industry", we fail to see how Canada's due process right of defence could be prejudiced by an alleged lack of precision in the terms employed by Brazil.
For these reasons, we find that the phrase "civil aircraft industry" is sufficiently specific for the purpose of Article 6.2 of the DSU.
Predecessor programmes
Despite the reference to "predecessor programs" in the plural, Brazil has argued that, in fact, there is only one predecessor program, i.e., the DIPP. Given the existence of only one predecessor programme, the reference to "predecessor programs" in the plural could have caused some uncertainty to Canada. However, Brazil has demonstrated517 that the TPC programme was perceived by Industry Canada to constitute the "new" programme replacing the DIPP. Thus, despite any potential for uncertainty concerning the exact parameters of Brazil's claim against "predecessor programs", Canada should have known that Brazil's "predecessor programs" claim would at least include DIPP. For this reason, we do not consider that the term "predecessor programs" would have prejudiced Canada's due process rights during the Panel process insofar as claims regarding the DIPP are concerned. In these circumstances, we find that the reference to "predecessor programs" is "readily understandable"518 and sufficiently specific for the purpose of Article 6.2 of the DSU insofar as claims concerning DIPP are concerned.
"Benefits" provided under the Canada-Québec Subsidiary Agreement on Industrial Development and the Société de Développement Industriel du Québec
Article 1.1(b) of the SCM Agreement explicitly provides that, in order to constitute a "subsidy", a "financial contribution" by a government or public body must confer a "benefit". In the context of a dispute under the SCM Agreement, therefore, the term "benefits" should be "readily understandable"519 to an informed reader. Canada asserts that the reference to "benefits" does not make it clear which aspect of the relevant programmes, or which activities or transactions under these programmes, are considered by Brazil to confer a "benefit". However, and especially in the context of claims against the application of alleged subsidy programmes that could involve tens or even hundreds of transactions, we consider that such detailed information would normally be included in the arguments adduced by the complaining party in its various submissions to a panel. The mere fact that such detailed information is not included in the request for establishment does not in and of itself prejudice the respondent's due process right of defence. For these reasons, we find that the term "benefits" is sufficiently specific for the purpose of Article 6.2 of the DSU.
Temporal application of the SCM Agreement
Canada has raised a preliminary issue concerning Brazil's claim against an alleged export subsidy granted in April 1989. Canada requested a preliminary ruling that the SCM Agreement does not apply to contributions and transactions that took place prior to 1 January 1995, the date of entry
516 See para. 4.47 above.
517 Brazil's reply to Canada's preliminary submission, dated 30 October 1998, para. 4.16.
518 European Communities - Customs Classification of Certain Computer Equipment, WT/DS62, 67, 68/AB/R, adopted 22 June 1998, para. 70.
519 European Communities - Customs Classification of Certain Computer Equipment, WT/DS62, 67, 68/AB/R, adopted 22 June 1998, para. 70.
WT/DS70/R
Page 164
into force of the WTO Agreement. Canada's request is based on Article 28 of the Vienna Convention on the Law of Treaties, which provides:
Unless a different intention appears from the treaty or is otherwise established, its provisions do not bind a party in relation to any act or fact which took place or any situation which ceased to exist before the date of the entry into force of the treaty with respect to that party.
Brazil agreed with Canada that, with regard to prohibited subsidies, the SCM Agreement does not apply to contributions made prior to 1 January 1995. For this reason, Brazil withdrew its claim concerning the 1989 measure in issue.
In light of Brazil's decision to withdraw its claim concerning the 1989 measure in issue, we do not consider it necessary to rule on the preliminary issue raised by Canada.
Fact-finding by the Panel
Arguments of the parties
In a letter to the Panel dated 23 October 1998, Brazil stated that it would be "handicapped" in pursuing its case "because Canada declined to provide transaction-specific details concerning [certain] measures during consultations." Brazil referred to the following statement by the Appellate Body in India Pharmaceuticals to argue that Canada was not justified in declining to disclose relevant details during consultations:
All parties engaged in dispute settlement under the DSU must be fully forthcoming from the very beginning both as to the claims involved in a dispute and as to the facts relating to those claims. Claims must be stated clearly. Facts must be disclosed freely. This must be so in consultations as well as in the more formal setting of panel proceedings. In fact, the demands of due process that are implicit in the DSU make this especially necessary during consultations.
Brazil argued that Canada's refusal to disclose the details of the operations of certain measures with regard to regional civil aircraft meant that "not all of the pertinent facts relating to its claim [were] before the Panel." Brazil then referred to the following statement by the Appellate Body in India Pharmaceuticals:
If, in the aftermath of consultations, any party believes that all the pertinent facts relating to a claim are, for any reason, not before the panel, then that party should ask the panel in that case to engage in additional fact-finding.
For these reasons, Brazil requested the Panel "to engage in additional fact-finding by requesting Canada to present to the panel and the parties, at the first meeting of the Panel, the complete details of all operations of the Export Development Corporation, the Canada Account, the Technology Partnerships Canada and its predecessor programs, the Canada-Québec Subsidiary Agreement on Industrial development, and the Société de Développement Industriel du Québec with regard to the civil aircraft industry, including all grants, loans, equity infusions, and loan guarantees, or any other direct or indirect financial contribution of any kind."
Canada submits that it is a well-established practice for a WTO panel, having received the first submissions and evidence of the parties, and having heard their first substantive oral arguments, to ask the parties for information additional to that submitted by the parties. According to Canada, there is no support from the Appellate Body, however, or from the DSU, WTO practice, or international law and practice for turning the panel process into something akin to a commission of
WT/DS70/R
Page 165
inquiry. Canada submits that there is also no provision in the DSU and no precedent in GATT or WTO jurisprudence for subjecting a responding party to a discovery process.
Evaluation by the Panel
We note that, by virtue of Article 13.1 of the DSU, we "have the right to seek information … from any … body which [we] deem[] appropriate." We note further that, according to the Appellate Body in Argentina - Textiles and Apparel, Article 13 of the DSU constitutes "a grant of discretionary authority: a panel is not duty-bound to seek information in each and every case…" 520 We also recall the Appellate Body's statement in European Communities - Hormones that "Article 13 of the DSU enables a panel to seek information … as it deems appropriate in a particular case…"521
We did not consider it appropriate to seek any information before receiving at least the first written submissions of both parties. We considered that it was only on the basis of these first written submissions that we could properly determine what, if any, additional information might need to be sought. In this regard, we recall that the Appellate Body in India - Patents referred to "additional fact-finding" by a panel in a context where pertinent facts are "not before the panel".522 In our view, the Appellate Body could not have been referring in that case to a situation where information is not before the panel because the panel has not yet received any submissions from the parties. Any contrary view would be absurd, since it would at once defeat the very purpose of the parties making written submissions.
Brazil and Canada filed their first written submissions on 3 and 16 November 1998 respectively. Brazil renewed its 23 October 1998 request at the first meeting with the Panel on 26 November 1998. The Panel sent the parties written questions on 27 November 1998, in order to seek clarification of a number of issues raised in the first written submissions and at the first substantive meeting with the Panel. These questions were not intended to elicit the detailed information referred to in Brazil's request of 23 October 1998.
Following receipt of the parties' second written and oral submissions, and the parties' written replies to the Panel's questions of 27 November 1998, a number of relevant transactions had been identified in the record. Accordingly, on 10 and 13 December 1998 we exercised our discretionary authority under Article 13.1 and asked Canada to provide detailed information (including the terms and conditions of various loans and contributions etc., and internal assessment documents concerning such loans and contributions etc.) concerning some of those transactions.523 However, we did not consider it appropriate to seek detailed information in respect of transactions (if any) not identified in the record.
In the circumstances of the present case, we did not consider it appropriate to exercise our discretionary authority under Article 13.1 to make generalized requests for information. Instead, we only sought detailed information of relevant loans, funds, contributions, assistance etc. identified in the record. Whereas more generalized requests for information (of the sort envisaged in Brazil's submission of 23 October 1998) may be appropriate for bodies such as commissions of enquiry, we do not consider them appropriate for a panel acting under Article 13.1 of the DSU. In cases involving alleged prohibited export subsidies, we appreciate that a complaining party may have difficulty in obtaining information necessary to support its case, especially where details of the alleged subsidy has
520 Argentina - Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items, WT/DS56/AB/R, adopted 22 April 1998, para. 84.
521 EC Measures Concerning Meat and Meat Products (Hormones), WT/DS26,48, AB/R, adopted 13 February 1998, para. 147.
522 India - Patent Protection for Pharmnaceutical And Agricultural Chemical Products, WT/DS50/AB/R, adopted 16 January 1998 para. 94.
523 Our questions also addressed more general issues relevant to the case.
WT/DS70/R
Page 166
not been notified under Article 25 of the SCM Agreement, and where the respondent Member chooses not to divulge relevant information during dispute settlement proceedings. This raises obvious systemic issues regarding the effectiveness of the SCM Agreement notification procedures, and of the WTO dispute settlement procedures generally. However, we consider that such systemic issues are a matter for serious consideration by the WTO Members in the appropriate fora, and not a matter to be resolved by this Panel acting under Article 13.1 of the DSU.
Procedures governing Business Confidential Information
Arguments of the parties
Early in this proceeding, Canada asked the Panel to adopt special working procedures for the protection of Business Confidential Information (hereinafter "BCI"). To this end, Canada submitted a set of draft procedures to the Panel. Canada stated that its draft procedures struck a balance between
the need for "reasonable access" to BCI by the Panel and the other disputing parties, and (2) the need to provide private business interests with adequate protection for their proprietary business information.
Brazil agreed in principle with Canada's request, but recommended a number of changes to the draft procedures submitted by Canada.
Evaluation by the Panel
We note that procedures concerning the protection of confidential information are provided for in Article 18.2 of the DSU. We note, however, that Article 12.1 of the DSU effectively permits panels to adopt working procedures in addition to those set forth in the DSU, after consulting the parties to the dispute.524 Given the sensitive nature of the BCI that could be submitted to the Panel in the present case, and given the agreement between the parties on the need for additional protection of such Business Confidential Information, the Panel decided to adopt special Procedures Governing Business Confidential Information that go beyond the protection afforded by Article 18.2 of the DSU. The Panel agreed with Canada that such Procedures should strike a balance between "reasonable access" to BCI, and the need to protect the integrity of such BCI.
The Panel sent proposed Procedures Governing BCI to the parties on 4 November 1998. Brazil provided comments on the proposed Procedures later the same day. Brazil noted that the proposed Procedures only provided for BCI to be submitted to, and stored at, the premises of the WTO or the premises of the submitting party's embassy or mission. Brazil expressed its concern that the Procedures proposed by the Panel did not allow a party to keep a set of BCI submitted by the other at its Geneva Mission. Brazil considered that this would "unduly restrict the ability of the parties to deal effectively with [BCI] during the course of this very rapid proceeding." Brazil proposed changes that would allow either party to keep a set of BCI submitted by the other party in a secure location, namely a locked safe, at its Geneva mission, and that would grant the other party to visit the other party's Geneva mission to review the secure location and propose changes in the arrangements. Brazil asserted that such changes would "achieve a reasonable balance between the competing interests of providing reasonable access to information and providing adequate safeguards for [the BCI] of private business interests."
The Panel wrote to Canada to solicit its views on the changes proposed by Brazil. Canada replied that, in its view, the change proposed by Brazil did "not afford private interests the protection provided for in the special procedure devised by the Panel [on 4 November 1998]." Canada also expressed the concern that disagreement between parties over the adequacy of protection afforded
524 Argentina - Measures Affecting Imports of Footwear, Textiles, Apparel and Other Items, WT/DS56/AB/R, adopted 22 April 1998, para 79.
WT/DS70/R
Page 167
BCI at one another's Geneva missions "could in itself cause unwanted delays in providing access to the information and thus undermine the very reason for Brazil's request."
Having taken into account the comments made by both parties, the final Procedures Governing BCI adopted by the Panel (see Annex I) provided for the submission of BCI at inter alia the Geneva mission of the other party. The Procedures stated that such BCI shall be stored in a safe in a locked room at the premises of the relevant Geneva mission, with restrictions imposed on access to the locked room and safe. The Procedures also provided for either party to visit the other party's Geneva mission and review the proposed location of the safe, and to propose any changes.
In a subsequent submission, Canada stated that because "the modified procedures do not provide the requisite level of protection for [BCI]. … Canada would not be in a position to submit [BCI] under the modified procedures." Canada submitted, however, that the Procedures proposed by the Panel on 4 November 1998 "struck a reasonable balance between the interests of the parties to have access to evidence submitted to the Panel and to provide protection for [BCI]."
In light of Canada's comments, it appeared to the Panel that Canada was essentially concerned that the level of protection afforded by the final Procedures was less than that afforded by the Procedures proposed on 4 November 1998. Accordingly, the Panel addressed the following question to Canada:
Canada stated in its oral presentation [at the first substantive meeting] (paras. 20-21) that “the original procedures adopted by the Panel governing confidential business information provided…a sufficient level of protection for that information”, but that “the procedures governing confidential business information, as modified at Brazil’s request, … do not provide the requisite level of protection…” (emphasis in original). Could Canada please explain how the level of protection under the procedures ultimately adopted by the Panel for the protection of confidential business information differs, in substance, from what would have been afforded by the procedures to which Canada indicates it could have agreed. For example, under the first set of procedures, what other than good faith would prevent a representative of a party that had signed a non-disclosure form from taking verbatim notes on the confidential business information on the premises of either the WTO or the opposing party, removing those notes from those premises and then disclosing them to non- authorized persons? Would not exactly the same degree of good faith be involved under either set of procedures?
In response, Canada stated:
Canada continues to believe that the Panel’s amended confidentiality procedures do not adequately protect private sector interests and the interests of the Government of Canada in maintaining effective control over the dissemination of business confidential information. In particular they may not be adequate to protect the Government of Canada from potential liability under domestic law, in the event the confidentiality provisions are not strictly observed.
In Canada’s view, the essential difference between the confidentiality procedures originally proposed by the panel and those ultimately adopted are that the original procedures entrusted the commercial confidential materials to the care of a neutral third party, i.e. the WTO Secretariat. The modified procedures oblige Canada to entrust the materials to the care of the opposing Party, which Party, or at least nationals of the Party on whose behalf the present case is being pursued, may have an immediate and commercial interest in these materials.
WT/DS70/R
Page 168
Under the original procedures, any breach of these procedures is likely to take the form of detailed or verbatim notes on a document, as is implied in the Panel’s question. Under the modified procedures, breach could take the form of a photocopy of the actual document. The difference between the commercial and legal effects of the two potential forms of breach are substantial.
As to commercial effects, the uncertain authenticity of notes on a document will diminish its utility and correspondingly decrease the potential commercial injury that might be caused by its disclosure to an interested party. For example, in negotiations for the sale of aircraft, a purchaser might pass a copy of the term sheet proposed by one manufacturer/seller to the agents of a competing manufacturer/seller, as a means of inducing the first manufacturer/seller to lower its price or otherwise improve the terms of its offer. A copy of a term sheet is a far more effective inducement for the manufacturer/seller to vary his proposal than the assertion by the purchaser that the competing manufacturer/seller has offered better terms, even if the specific content of the alleged better terms is detailed by the purchaser.
As to legal liability, any claim, based on the production of detailed or verbatim notes, that business confidential information was improperly disclosed would constitute hearsay under Canadian law, and its evidentiary value would be limited. A photocopy of an original document, however, would constitute best evidence. (emphasis supplied)
The Panel considers that any special procedures (i.e., in addition to Article 18.2 of the DSU) for the protection of BCI in the context of WTO dispute settlement will necessarily depend on the good faith of the parties and their representatives. This is true of the final Procedures adopted by the Panel, and is equally true of the 4 November 1998 Procedures.
We recall that both the 4 November 1998 Procedures, and the final Procedures, provided that BCI "shall not be copied, … except as specifically provided in these Procedures."525 In response to a request from Canada, the Panel subsequently, on 21 December 1998, amended the final Procedures to clarify that the term "copied" included verbatim transcription. Despite the prohibition of copies, therefore, both sets of Procedures effectively allowed approved persons to make detailed (but not verbatim) handwritten notes of a party's BCI.
Canada suggests that it is the potential for approved persons representing one party to make photocopies of BCI submitted by the other party (because the BCI is entrusted to the care of that other party) that renders the final Procedures unacceptable. Canada effectively argues that there was no potential to make photocopies under the 4 November 1998 Procedures, because the BCI was entrusted to the WTO Secretariat, and that this potential exists under the final Procedures because they envisage BCI being provided to the other party to the dispute. Canada submits that "the uncertain authenticity of notes on a document will diminish its utility and correspondingly decrease the potential commercial injury that might be caused by its disclosure to an interested party". Canada asserts that photocopies of documents have greater commercial effects than detailed handwritten notes of the contents of such documents.526 We note that Canada has advanced no additional arguments in support of its assertion that the protection afforded to BCI by the final Procedures is less than that afforded by the 4 November 1998 Procedures.
525 Section VIII.4 of the 4 November 1998 Procedures, and Section VIII.5 of the final Procedures.
526 Canada also points to the superiority of photocopies over handwritten notes for the purposes of
Canadian rules of evidence. As Canadian rules of evidence are not relevant in this proceeding, we shall not address this argument.
WT/DS70/R
Page 169
We are not convinced by Canada's arguments in this regard. In particular, although detailed handwritten notes of BCI documents may be less authentic than photocopies of documents containing BCI, we consider that handwritten notes of BCI would nevertheless have some commercial value to certain unscrupulous individuals. The Procedures seek to protect the substance of the BCI, not the form of the BCI. This is because it is the substance of the BCI that has the potential to cause significant commercial prejudice in any form, be it handwritten note or photocopy. Accordingly, we attach no importance to Canada's argument that the final Procedures provide unscrupulous approved persons with the opportunity to make photocopies of BCI documents, whereas the 4 December 1998 Procedures would only have provided unscrupulous approved persons an opportunity to make handwritten notes of BCI. Under both sets of Procedures, the substance of the BCI could have been disclosed by unscrupulous approved persons bent on this course.
Furthermore, assuming arguendo that a photocopy of BCI does have greater commercial effect than detailed handwritten notes of BCI, we are not convinced that the possibility of approved persons making photocopies of documents containing BCI was totally precluded under the 4 November 1998 Procedures. In order to exclude any potential for approved persons to make photocopies under the 4 November 1998 Procedures, the Secretariat would have had to continually monitor all approved persons' access to the BCI. However, the Secretariat is not responsible for the enforcement of the Procedures, and would not therefore have engaged in such monitoring. Thus, even under the 4 November Procedures, there was a potential for unscrupulous and determined approved persons, having been granted access to the relevant BCI by the Secretariat, to temporarily remove BCI documents for photocopying.
In our opinion, the important distinction between the 4 November 1998 Procedures, and the final Procedures, is that the latter would facilitate the work of the parties in preparing themselves for these "fast-track" proceedings, without impairing the protection afforded to the substance of the BCI. The timetable of the proceedings is such that party representatives would be likely to spend large periods of time in Geneva. As noted above, Canada itself has recognised the need for a party to have "reasonable access" to BCI submitted by the other party. In the context of a fast-track case in particular, we do not consider that there is "reasonable access" to the BCI if a party is required to adjust its work in respect of that BCI to the official working hours of the WTO Secretariat, excluding evenings and weekends. Under the final Procedures, authorised representatives of the parties would have had the convenience of access to the BCI of the other party at any time of day or night, rather than during the working hours of the WTO Secretariat. In our view, the final Procedures therefore strike a reasonable balance between (1) the need for "reasonable access" to BCI by the Panel and the other disputing parties, and (2) the need to provide private business interests with adequate protection for their proprietary business information.
For these reasons, we reject Canada's arguments concerning the inadequacies of the final Procedures Governing Business Confidential Information.
Deadline for the submission of new evidence or allegations
In its submission dated 16 November 1998, Canada requested a preliminary ruling that the complaining party may not adduce new evidence or allegations after the end of the first substantive meeting of the Panel with the parties. Canada submitted that, given the accelerated procedure under Article 4 of the SCM Agreement, the late submission of allegations or evidence would be prejudicial, as Canada would be effectively denied an adequate opportunity to respond.
Brazil asked the Panel to reject Canada's request. Brazil argued that the request is without support in WTO law. In addition, Brazil argued that in Argentine Footwear the Appellate Body affirmed the Panel's authority to permit the submission of new facts at any point in the proceedings, provided that the other party is given an adequate opportunity to respond.
WT/DS70/R
Page 170
Submission of new evidence
We recall that the Appellate Body found in Argentine Footwear that neither Article 11 of the DSU, nor the Working Procedures in Appendix 3 of the DSU, establish precise deadlines for the presentation of evidence by parties to a dispute. There is nothing in the DSU, or in the Appendix 3 Working Procedures, to suggest that a different approach should be taken in "fast-track" cases under Article 4 of the SCM Agreement.
In our opinion, an absolute rule excluding the submission of evidence by a complaining party after the first substantive meeting would be inappropriate, since there may be circumstances in which a complaining party is required to adduce new evidence in order to address rebuttal arguments made by the respondent. Furthermore, there may be instances, as in the present case,527 where a party is required to submit new evidence at the request of the panel. For these reasons, we rejected Canada's request for a preliminary ruling that the Panel should not accept new evidence submitted by Brazil after the first substantive meeting.
Submission of new allegations
We also consider that we are not bound to exclude the submission of new allegations after the first substantive meeting. We can see nothing in the DSU, or in the Appendix 3 Working Procedures, that would require the submission of new allegations to be treated any differently than the submission of new evidence. Indeed, one could envisage situations in which the respondent might present information to a panel during the first substantive meeting that could reasonably be used as a basis for a new allegation by the complaining party. Provided the new allegation falls within the panel's terms of reference, and provided the respondent party's due process rights of defence are respected, we can see no reason why any such new allegation should necessarily be rejected by the panel as a matter of course, simply because it is submitted after the first substantive meeting with the parties. We consider that this approach is consistent with the Appellate Body's ruling in European Communities - Bananas that "[t]here is no requirement in the DSU or in GATT practice for arguments on all claims relating to the matter referred to the DSB to be set out in a complaining party's first written submission to the panel. It is the panel's terms of reference, governed by Article 7 of the DSU, which set out the claims of the complaining parties relating to the matter referred to the DSB."528
Deadline for the submission of affirmative defences
By letter dated 26 November 1998, and during its oral statement at the first substantive meeting on 26 November 1998, Brazil expressed its view, on the basis of certain statements in Canada's first written submission to the Panel, that "Canada might intend to assert that the provisions of paragraph (k) of Annex I to the [SCM] Agreement excuse its otherwise prohibited subsidies." Brazil asserted that reliance on paragraph (k) would amount to a positive defence for which Canada bears the burden of proof. According to Brazil, a good faith interpretation of the DSU requires a party making an affirmative defence to set out the grounds for that affirmative defence in its first written submission to the panel. Accordingly, Brazil asked the Panel "not to accept any evidence or claims of defense in the nature of affirmative defences with regard to the Canada Account, the EDC
527 At the time of the second substantive meeting, we asked the parties a series of questions that could have led to the submission of new evidence or arguments. In order to ensure due process, we allowed each party 18 days (i.e., equivalent to the time between the deadline for the respondent's first submission and the deadline for rebuttal submissions) in which to comment on any new evidence or arguments adduced by the other party in response to our questions.
528 European Communities - Regime for the Importation, Sale and Distribution of Bananas, WT/DS27/AB/R, adopted 25 September 1997¸ para. 145
WT/DS70/R
Page 171
or any other measure at issue which is not submitted to the Panel and to Brazil prior to the scheduled close" of the first substantive meeting. Brazil asserted that "this is particularly important in this fast- track proceeding, where the time constraints on the Panel could limit, or even preclude, the panel's giving Brazil a reasonable time to respond to new information after" the second written submission, or particularly after the second substantive meeting.
Canada did not comment on Brazil's request.
As noted above, there is nothing in the DSU, or in the Appendix 3 Working Procedures, to prevent a party submitting new evidence or allegations after the first substantive meeting. We can see no basis in the DSU to treat the submission of affirmative defences after the first substantive meeting any differently. Thus, although it is desirable that affirmative defences, as with any claim, should be submitted as early as possible, there is no requirement that affirmative defences should be submitted before the end of the first substantive meeting with the parties. Provided that due process is respected, we see nothing to prohibit the submission of affirmative defences after the first substantive meeting with the parties.
In the present case, we considered that Brazil's due process rights would not be respected if Canada were able to submit an affirmative defence after 12 December 1998, i.e., after the second substantive meeting with the Panel. We therefore required Canada to invoke any affirmative defence it intended to raise in this case by 12 December 1998. In the event, Canada did not adduce any affirmative defences.
Panel's right to seek information in respect of defences Canada has not made
In introductory comments to its 21 December 1998 response to questions from the Panel, Canada expressed the "concern" that "the Panel has requested the production of evidence in respect of defences Canada has not made." Canada asserts that "this method of proceeding ignores the principle of judicial economy." Consistent with the principle of judicial economy, Canada submits that the Panel should address only those issues which must be addressed in order to resolve the matter in issue in the dispute. By way of example, Canada argues that it is not necessary for this Panel to determine whether impugned programmes, activities or transactions are “subsidies”, if it finds that they are not “contingent … on export performance”, and vice versa.
Brazil has expressly endorsed Canada's views on the principle of judicial economy generally. However, Brazil has not expressly endorsed Canada's view that the principle of judicial economy should necessarily prevent the Panel from asking a party questions on issues in respect of which it has not adduced defences.
We note that, according to Canada, Brazil must demonstrate two distinct and necessary elements in order to pursue its claims in the present case. First, Brazil must demonstrate the existence of a subsidy. Second, Brazil must demonstrate that the subsidy is contingent on export performance. Canada effectively submits that it has the option of choosing whether to defend itself on both these elements, or only one of them. In principle, we agree with Canada in this regard.
Canada asserts, however, that if Canada chooses to defend on only one element, then the Panel is precluded from seeking information on the other element in respect of which Canada has not raised a defence. We do not accept this argument.
With regard to certain measures before the Panel, Canada chose to defend itself on the issue of export contingency. Thus, Canada does not advance detailed arguments on the question of subsidization. Despite the absence of any defence on the issue of subsidization, Canada states expressly that it does not admit that the relevant measures constituted subsidies. Canada defends itself on the issue of export contingency, because it believes that the Panel will reject Brazil's claim on
WT/DS70/R
Page 172
export contingency. However, Canada has ignored the possibility that the Panel could find in favour of Brazil on the question of export contingency. If the Panel were to find against Canada on the question of export contingency, the Panel would then be required to make findings on the subsidy issue, particularly given Canada's express statement that it does not admit that the relevant measures constitute subsidies.529 If the Panel were prevented from seeking information on the subsidy issue because of Canada's decision not to defend itself on that issue, the basis for the Panel's findings on subsidization would be weak at best. It is for these reasons that we reject Canada's argument that a party's decision not to put in a defence on a particular issue, when that party denies or refuses to admit elements of the claim, should prevent the Panel from seeking information on that issue.
Panel's right to seek information in the absence of a preliminary ruling on the establishment of a prima facie case
In introductory comments to its 21 December 1998 response to our questions, Canada expressed the "concern" that "the Panel has requested the production of evidence in respect of matters where in Canada's view Brazil has clearly not made out a prima facie case." Canada also notes that "the Panel has not ruled on whether Brazil has made out a prima facie case on any of the impugned programmes, activities or transactions." With regard to the latter consideration, Canada noted at the second substantive meeting with the Panel that "the fact that the Panel has not ruled on what Canada considers to be an essential preliminary issue, that is, whether Brazil has made out a prima facie case, makes it difficult for Canada to defend itself."
Brazil characterized Canada's concern as "illogical", noting that "[n]either of the Parties will know what the Panel thinks of the case until a point in this proceeding at which neither will be able to present new evidence."
We understand that there are two elements to the "concern" raised by Canada. First, Canada appears to argue that a panel should rule on whether the complaining party has established a prima facie case as a preliminary matter. Second, Canada appears to argue that a panel should only seek information on matters with respect to which it has ruled as a preliminary matter that a prima facie case has been established. We do not accept Canada's argument.
We note that there is nothing in the DSU that requires a panel to rule on the establishment of a prima facie case as a preliminary matter. Indeed, the DSU makes no provision for any form of preliminary or interim ruling. Furthermore, we are not aware of any WTO panels that have made preliminary rulings on whether the complaining party has established a prima facie case. Whereas a critical decision for panels in every case is whether or not a prima facie case has been established, and whether or not the respondent has rebutted any such prima facie case, in practice these decisions are made in light of all the evidence adduced by the parties, including both written and oral submissions. Such decisions are, therefore, necessarily made in the latter stages of the panel process.
We recall Canada's assertion that it had experienced difficulty in defending itself as a result of the Panel's failure to issue a preliminary ruling on whether Brazil had established a prima facie case. Without expressing any opinion on Canada's assertion that it experienced difficulty in defending itself in the present case, we consider that Canada could defend itself fully during the Panel proceedings simply by assuming that Brazil had, or would, establish a prima facie case in respect of all its claims.
Furthermore, a panel's right to seek information is governed by Article 13.1 of the DSU. There is nothing in Article 13.1 to suggest that a panel's right to seek information is restricted to
529 Canada states that it is not necessary for this Panel to determine whether impugned programmes, activities or transactions are “subsidies”, if it finds that they are not “contingent … on export performance”, and vice versa. A contrario, we understand Canada to argue that it is necessary for the Panel to determine whether impugned programmes etc. are subsidies if it finds that they are contingent on export.
WT/DS70/R
Page 173
matters in respect of which the complaining party has been deemed (as a preliminary matter) to have established a prima facie case. The only express restriction on a panel's right to seek information is the Article 13.1 obligation for a panel to "inform the authorities" of a Member before seeking information or advice from any individual or body within the jurisdiction of that Member. In our opinion, any requirement that panels should provide preliminary rulings on whether the complaining party has established a prima facie case before seeking information or advice under Article 13.1 could render that provision ineffective. This is because in certain circumstances a panel may consider it appropriate to seek information or advice precisely in order to determine whether the complaining party has established a prima facie case. Canada's approach to a panel's right to seek information would preclude this possibility, and therefore render Article 13.1 ineffective in such circumstances.
For these reasons, we reject Canada's criticism of the Panel for having requested information on matters in respect of which there had not been a preliminary ruling on whether or not Brazil had established a prima facie case.
Panel's right to seek information in the absence of an allegation by the complaining party
Canada objects that the Panel has requested the production of evidence in respect of transactions where Brazil has not made an allegation.
Brazil does not comment on Canada's objection.
We note that Canada has raised this objection in respect of questions asked by the Panel concerning: contributions made under the SDI programme generally; specific SDI contributions under the Aerospace Industry Development Fund; the EDC debt financing to ASA; and the Subsidiary Agreement assistance to Rolls Royce and Lamines CTEK.
We recall that Brazil presented very broad allegations effectively covering all assistance granted to the Canadian civil aircraft industry under the various programmes identified in its request for establishment. Accordingly, any assistance provided to the civil aircraft industry under the relevant programmes was effectively covered by the scope of Brazil's allegations. Thus, to the extent that the Panel sought information under Article 13.1 with regard to specific instances of assistance (provided under the relevant programmes) identified in the record, the Panel necessarily only sought information in respect of assistance covered by Brazil's allegations.
We note, however, that in its submission dated 4 December 1998, Brazil stated that it was not making any specific allegation against the Subsidiary Agreement assistance to Rolls Royce or Lamines CTEK identified in the record, since such assistance did not concern the regional aircraft industry. On the basis of this statement, the Panel ultimately decided not to take these two transactions into account when reviewing Brazil's claim concerning Subsidiary Agreement assistance to the regional aircraft industry.530 In normal circumstances, the Panel would not have sought additional information regarding these transactions under Article 13.1 of the DSU. However, the request for information on these transactions was sent to Canada on 10 December 1998. At the time the request was sent, the Panel had not had sufficient time to reach a final decision on the full implications of Brazil's 4 December 1998 statement concerning these transactions. Accordingly, it was considered appropriate to request detailed information concerning these transactions, to cover the possibility that this information might be relevant to the Panel's deliberations at a future date. Consistent with the Panel's finding that these transactions were not covered by Brazil's allegations concerning Subsidiary Agreement assistance to the Canadian regional aircraft industry, the requested information ultimately did not become relevant to the Panel's deliberations.
530 See para. 9.259 above.
WT/DS70/R
Page 174
DEFINITION OF "SUBSIDY" WITHIN THE MEANING OF ARTICLE 1 OF THE SCM AGREEMENT
A critical issue in this case concerns the definition of "subsidy" within the meaning of Article 1 of the SCM Agreement. Leaving aside the issue of "income or price support[s]" (Article 1.1(a)(2)), Article 1 of the SCM Agreement provides that a "subsidy" exists when there is a "financial contribution" by a "government or any public body within the territory of a Member" (Article 1.1(a)(1)) that confers a "benefit" (Article 1.1(b)).
The parties do not disagree on the notion of a "financial contribution" by a "government" or "public body". However, there is significant disagreement between the parties on the meaning of the term "benefit" in Article 1.1(b) of the SCM Agreement. This disagreement influences much of the parties' argumentation before the Panel. Therefore, before proceeding with our analysis of Brazil's claims, we shall first provide our interpretation of "benefit".
Arguments of the parties
Canada asserts that a "benefit" is conferred when a public financial contribution by a public body (i) imposes a cost on the government, and (ii) results in an advantage above and beyond what the market could provide. Canada claims that this interpretation of the term "benefit" is based on the ordinary meaning of "benefit", the context in which it is found, and the object and purpose of the SCM Agreement read as a whole.531
According to Canada, the ordinary meaning of "benefit" is "an advantage". However, in the context of the SCM Agreement, Canada argues that this ordinary meaning is overly broad, since it could include "normal commercial activity", such as a commercial contract (entered into by a government) that accords an advantage to a firm relative to its competitors. For this reason, Canada interprets "benefit" in light of Annex IV amd Article 14 of the SCM Agreement, which it relies on as relevant context for the interpretation of Article 1.
Canada notes that Article 14 of the SCM Agreement applies commercial benchmarks as guidelines for the 'benefit to recipient method' of calculating the amount of a subsidy.532 For this reason, Canada accepts that commercial benchmarks are relevant in determining the existence of "benefit" within the meaning of Article 1. However, Canada also refers to Annex IV of the SCM Agreement as relevant context. Annex IV concerns the calculation of the total ad valorem subsidization with regard to the presumption of "serious prejudice" under Article 6.1(a). Paragraph 1 of Annex IV provides that [a]ny calculation of the amount of a subsidy for the purpose of paragraph 1(a) of Article 6 shall be done in terms of the cost to the granting government". Canada asserts that if "benefit" is defined exclusively in terms of commercial benchmarks (i.e., without any reference to net cost to government), a subsidy might be deemed to exist under Article 1 because it was on below- market terms, but it might nonetheless have no value within the meaning of Article 6.1(a) if it does not involve any net cost to the government. By applying Annex IV as context for the interpretation of Article 1 of the SCM Agreement, Canada asserts that net cost to the government should be a condition for establishing "benefit" within the meaning of Article 1 of the SCM Agreement.
With regard to government credit, Canada asserts that item (k) of the Illustrative List of Export Subsidies of Annex I of the SCM Agreement provides specific contextual guidance as to what constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement. On the basis of this
531 The Panel notes that Canada therefore purports to interpret the term "benefit" consistent with Article
31.1 of the Vienna Convention on the Law of Treaties.
532 The Panel notes that Article 14(a) provides, for example, that "government provision of equity capital shall not be considered as conferring a benefit, unless the investment decision can be regarded as inconsistent with the usual investment practice … of private investors …".
WT/DS70/R
Page 175
specific contextual guidance, Canada submits that there are two elements in determining whether particular credit terms are subsidies: (1) does a government provide credit at rates below those which it has to pay for the funds so employed, and (2) does such credit secure a material advantage in the field of export credit terms? Canada emphasised, however, that neither item (k), nor other items in the Illustrative List, identify a contrario what would not constitute a "subsidy" within the meaning of Article 1. For Canada, such an a contrario reading would turn the Illustrative List into an exclusive list.
With regard to the object and purpose of the SCM Agreement, Canada states that "the 'mischief' that the Agreement seeks to discipline are measures that distort the market by (i) imposing a cost on the treasury of the providing Member, and (ii) conferring an advantage to the recipient above and beyond the market".
For Brazil, neither the ordinary meaning of Article 1.1, nor its context, nor the object and purpose of the Subsidies Agreement, nor Canada itself, outside the confines of these proceedings, suggest, much less require, a “net cost to government” test. Brazil argues that the proper test is evident from the ordinary meaning of Article 1.1: a subsidy exists where a government contributes something, and in so doing grants an aid, which gives help or support, or an advantage, thereby improving the recipient’s condition above and beyond the market.
Brazil notes that Canada agrees that demonstrating a “benefit to the recipient” is one part of the “benefit” test. Brazil argues, however, that Canada invents a second, additional requirement – that the government, in making its contribution, realize a “net cost.” Brazil rejects each of Canada’s reasons for this additional asserted requirement. First, Brazil disputes Canada’s argument that applying the ordinary meaning of the term “benefit” – which in Brazil’s view means “advantage” or “aid” – would not “adequately narrow the term”. Brazil argues that no provision of the Vienna Convention requires that the ordinary meaning of a term be narrowed in order for it to be valid. Brazil notes Canada’s statement that a broad definition of “benefit” could mean that a “commercial contract” could possibly be considered a “subsidy,” but argues that such a contract is not, without something more, a subsidy prohibited by the terms of the Subsidies Agreement.
Second, Brazil disagrees with Canada’s view that paragraph (a) of Article 6.1 of the SCM Agreement provides contextual support for its net cost argument. Brazil disagrees with Canada that if “benefit” under Article 1.1 means “benefit to the recipient,” a government contribution identified under Article 1.1 as a “subsidy” could have a value of zero under Article 6.1 and Annex IV. Brazil asserts that valuation is only one of several ways to establish “serious prejudice” under Article 6.1 of the Subsidies Agreement. Moreover, for Brazil, Article 6.1 is not relevant to an export subsidy, which need not be quantified or subject to specific valuation in order to trigger the prohibition of Article 3.
Furthermore, Brazil also disagrees that item (k) of Annex I to the Subsidies Agreement offers “context” supporting Canada’s proposed test. According to Brazil, Annex I does not speak to whether government activity constitutes a subsidy, but rather to whether government activity constitutes a prohibited export subsidy. Brazil submits that a measure may constitute a subsidy, but not be on the Illustrative List of Export Subsidies included in Annex I.
Third, concerning the object and purpose of the SCM Agreement, Brazil finds incomprehensible Canada's implication that the type of benefit estimated by Brazil for TPC's $87 million contribution to Bombardier is not, consistent with the object and purpose of the Subsidies Agreement, “trade distortive.”
Furthermore, with regard to Canada's own administrative practice, Brazil notes that Revenue Canada’s Special Import Measures Act Handbook states that while the determination of “benefit” on a government loan is generally related to whether the government recovers its costs, such a test will not always capture the true effect of the subsidy. According to Brazil, the Handbook states that:
WT/DS70/R
Page 176
“[I]t is also possible that a benefit would accrue to an exporter or an importer as a result of a government guarantee which would not necessarily result in a cost to the government. The benefit could be a lower interest rate or a loan at a commercial rate which the company would otherwise not get without government involvement.”
Furthermore, according to Brazil, in defining a “subsidy,” the Canadian Handbook states that a “benefit” can be direct or indirect:
“A direct financial or other commercial benefit is one which accrues directly to the person, firm or industry which is the intended recipient, such as an outright grant of funds to a producer of goods. An indirect benefit is one which does not accrue directly, but which alters the economic environment within which firms operate, and hence the level of their costs.”
For Brazil, therefore, Canada’s position concerning the definition of “subsidy” and “benefit” before this Panel is irreconcilable with that adopted in its own law.
Interpretation by the Panel
In interpreting the term "benefit" in Article 1.1(b) of the SCM Agreement, we recall that Article 31.1 of the Vienna Convention on the Law of Treaties provides that a treaty shall be interpreted "in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose." We note that Canada purports to respect Article 31.1 of the Vienna Convention in its interpretation of the term "benefit". However, for the following reasons, we are not persuaded that the application of Article 31.1 results in the interpretation of "benefit" advocated by Canada.
First, in our opinion the ordinary meaning of "benefit" clearly encompasses some form of advantage. We do not consider that the ordinary meaning of "benefit" per se includes any notion of net cost to the government. As Canada itself has noted, the dictionary definition of "benefit" refers to "advantage", and not to net cost.533 In order to determine whether a financial contribution (in the sense of Article 1.1(a)(i)) confers a "benefit", i.e., an advantage, it is necessary to determine whether the financial contribution places the recipient in a more advantageous position than would have been the case but for the financial contribution. In our view, the only logical basis for determining the position the recipient would have been in absent the financial contribution is the market. Accordingly, a financial contribution will only confer a "benefit", i.e., an advantage, if it is provided on terms that are more advantageous than those that would have been available to the recipient on the market.
We note that the relevant context supports our interpretation of the ordinary meaning of "benefit". In particular, we note that Article 14 provides guidelines for calculating "the benefit to the recipient conferred pursuant to paragraph 1 of Article 1." These guidelines employ a commercial benchmark, whereby a financial contribution "shall not be considered as conferring a benefit" unless that financial contribution is made on terms that are more advantageous than would have been available to the recipient on the commercial market. Although Article 14 applies expressly for the purposes of Part V of the SCM Agreement (countervailing measures), and although the commercial benchmark approach of Article 14 is not the only test for calculating the amount of a subsidy, we recall that Article 14 refers expressly to commercial benchmarks for identifying explicit situations in which an Article 1.1 "benefit" shall not arise. We see no reason why the commercial benchmarks applied in Article 14 for the purpose of determining when an Article 1.1 "benefit" does not arise should not serve as relevant context for determining when an Article 1.1 "benefit" does arise.
533 See para. 5.30 above.
WT/DS70/R
Page 177
Second, we do not accept Canada's argument that "benefit" must be interpreted more narrowly than its ordinary meaning of "advantage". Canada asserts that a narrow interpretation is necessary in order to exclude "normal commercial activity" such as commercial contracts (by a government) that accord advantages to firms relative to their competitors. In light of the preceding paragraph, however, we consider that using a commercial benchmark to identify the existence of a "benefit" will not create the problem alluded to by Canada. That is, under such a benchmark, "normal [governmental] commercial activity" will not be deemed to confer a "benefit", i.e., an advantage, provided it really is "normal commercial activity". In other words, such activity will not be deemed to confer a "benefit" provided the terms of the contract are not more advantageous than those that would have been negotiated on the market for an equivalent transaction.
Third, if "benefit" were to include the notion of net cost to government, it could exclude from the definition of "subsidy" situations explicitly identified in Article 1.1(a)(1) itself as constituting government financial contributions even though no cost to the government might be involved. Specifically, Article 1.1(a)(1)(iv) identifies as a "financial contribution" the situation where a government directs a private body to make "financial contributions" within the meaning of Article 1.1(a)(1)(i)-(iii). In such a situation, the net cost could be incurred entirely by the private body rather than the government. Canada's interpretation of "benefit" (i.e., to include net cost to government) would render Article 1.1(a)(1)(iv) meaningless, since a form of "financial contribution" explicitly included in Article 1.1(a) would automatically (i.e., because it would never meet the net cost to government test) be excluded by Article 1.1(b).
Fourth, we do not accept Canada's reliance on Annex IV.1 of the SCM Agreement as contextual support for considering net cost to government in determining "benefit". Annex IV.1 concerns the calculation of the amount of a subsidy for the purpose of Article 6.1(a) of the SCM Agreement. Canada effectively argues that the "cost to government" referred to in paragraph 1 of Annex IV constitutes the definition, at least for the purpose of determining the existence of serious prejudice under Article 6.1(a), of "benefit" in the sense of Article 1.1(b). However, Canada's is neither the only interpretation of this provision, nor the most convincing. In our opinion, the need to calculate the value of a subsidy only arises once the existence of the subsidy, and therefore the "financial contribution" and "benefit", have been established. Because "benefit" must be established before the value of the alleged subsidy may be considered, provisions concerning the valuation of subsidies are not necessarily relevant for the purpose of establishing the existence of a subsidy (and therefore "benefit"). The contextual relevance of Annex IV.1 differs significantly from that of Article 14, referred to in para. 9.113 above, since the latter provision deals expressly with the calculation of "benefit" "pursuant to paragraph 1 of Article 1". Annex IV.1, by contrast, refers only to the calculation of the amount of a subsidy, and does not apply expressly for the purpose of calculating "benefit" within the meaning of Article 1.1 of the SCM Agreement.
Fifth, we are unable to accept Canada's argument that item (k) of the Illustrative List of Annex I of the SCM Agreement constitutes contextual guidance for determining the existence of "benefit" in the specific context of government credit under Article 1. In our view, item (k) of the Illustrative List applies in determining whether or not a prohibited export subsidy exists. We do not consider, and the parties have not argued, that item (k) determines whether or not a "subsidy" exists within the meaning of Article 1 of the SCM Agreement. Whereas Canada might have argued that item (k) could be relied on for the purpose of determining when an Article 1 subsidy does not exist, Canada expressly declined to rely on this a contrario reading of item (k). We therefore take no view on the a contrario approach as such. 534
534 In this regard, we note an apparent inconsistency in Canada's arguments as we see no difference between Canada's reliance on item (k) as specific contextual guidance for the interpretation of "subsidy" under Article 1, and the very a contrario reading of item (k) that Canada has expressly rejected for the purpose of defining when government credit constitutes a "subsidy" within the meaning of Article 1 of the SCM
WT/DS70/R
Page 178
Furthermore, whereas Article 31.1 of the Vienna Convention requires "the ordinary meaning to be given to the terms of the treaty in their context", we do not consider that this requires us to ignore the ordinary meaning of the term "benefit" simply to accommodate a conflicting contextual interpretation. As noted above, we consider that the ordinary meaning of "benefit" refers to advantage, to the exclusion of any notion of net cost to the government.
Sixth, we note that the SCM Agreement does not contain any express statement of its object and purpose. We therefore consider it unwise to attach undue importance to arguments concerning the object and purpose of the SCM Agreement. In our view, however, the avoidance of net cost to government is not the object and purpose of the multilateral disciplines contained in the SCM Agreement. Rather, as suggested by Canada itself at para. 96 of its first submission, we consider that the object and purpose of the SCM Agreement could more appropriately be summarised as the establishment of multilateral disciplines "on the premise that some forms of government intervention distort international trade, [or] have the potential to distort [international trade]".
For all the above reasons, we reject Canada's 'net cost to government' interpretation of the term "benefit" in Article 1.1(b) of the SCM Agreement. Thus, leaving aside situations of alleged "income or price supports" within the meaning of Article 1.1(a)(2), we consider that a "financial contribution" by a government or public body confers a "benefit", and therefore constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement, when it confers an advantage on the recipient relative to applicable commercial benchmarks, i.e., when it is provided on terms that are more advantageous than those that would be available to the recipient on the market.
EXPORT DEVELOPMENT CORPORATION
Brazil challenges the EDC programme as a per se export subsidy prohibited by Article 3 of the SCM Agreement. Brazil also challenges the EDC as applied, including funding, support, funds and benefits granted under the auspices of the EDC programme to the regional aircraft industry.535
Brazil submits that the EDC is an agency of the Government of Canada which was established by the Export Development Act “for the purposes of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.”536 Brazil refers to EDC materials to the effect that "EDC's mandate is to help Canadian business compete and succeed in the global marketplace", and to attempt to satisfy "the seemingly endless appetite of Canadian exporters for financial support"537 through a variety of financial and risk absorption services, including export trade insurance, sales, financing, loan guarantees, and equity investments. Brazil asserts that these benefits are not available from
Agreement. That is, we consider that Canada is effectively seeking to rely on an a contrario reading of item (k), while at the same time asserting that it rejects this approach.
535 We note that Brazil's request for establishment of a panel (WT/DS70/2) refers to various alleged prohibited export subsidies provided to the "Canadian industry producing civil aircraft". Our terms of reference, therefore, relate to assistance provided to the Canadian civil aircraft industry. With the exception of its first written submission, however, all findings requested by Brazil concerning assistance provided under the EDC
programme refer to the "regional aircraft industry". Similarly, all arguments adduced by Brazil concern the regional aircraft industry. We consider that the Canadian regional aircraft industry is a sub-set of the Canadian civil aircraft industry. For these reasons, we only examine Brazil's claim against assistance provided under the EDC programme insofar as such assistance relates to the Canadian regional aircraft industry. Brazil stated that "[r]egional aircraft generally range from 30 to 70 seats, and serve markets up to about 1,600 kilometres apart". Brazil noted in addition that customized regional aircraft may occasionally include up to 78 seats, and that Bombardier and Fairchild Dornier USA are currently developing regional aircraft that may range up to 90 seats (see para. 1.2 of Brazil's first written submission).
536 Export Development Act, Section 10(1).
537 Export Development Corporation, 1995 Chairman and President’s Message (“Message”), pages 2
and 4.
WT/DS70/R
Page 179
private financial institutions. Brazil asserts that all assistance granted to the regional aircraft industry under the EDC programme is contingent on export.
Brazil asserts that EDC provides financing assistance to Canadian regional aircraft exporters in a variety of ways, four of which are specifically challenged by Brazil in these proceedings: debt financing, loan guarantees, residual value guarantees, and equity financing. Before addressing Brazil's specific arguments concerning these EDC activities, it is first necessary to examine Brazil's claim regarding the EDC programme per se.
Is the EDC programme per se a prohibited export subsidy?
To assess Brazil's claim against the EDC programme, we must first determine whether the EDC programme per se mandates the grant of prohibited export subsidies in a manner inconsistent with Articles 3.1(a) and 3.2 of the SCM Agreement. In this regard, we recall the distinction that GATT/WTO panels have consistently drawn between discretionary legislation and mandatory legislation. For example, in United States – Tobacco the panel "recalled that panels had consistently ruled that legislation which mandated action inconsistent with the General Agreement could be challenged as such, whereas legislation which merely gave the discretion to the executive authority … to act inconsistently with the General Agreement could not be challenged as such; only the actual application of such legislation inconsistent with the General Agreement could be subject to challenge".538
During the Panel proceedings, we addressed the following question to Brazil:
In light of GATT practice regarding the distinction between mandatory and permissive legislation, could Brazil please state whether it considers that the various “programs” referred to in its oral statement to the Panel [at the first substantive meeting] (26 November 1998) require Canada to act in a manner inconsistent with Article 3.1(a) of the SCM Agreement. Why?
In response, Brazil stated that:
In Brazil’s view, EDC has interpreted its own and the Canada Account’s mandate to require it to fund projects that give “Canadian exporters an edge when they bid on overseas projects.”539 Brazil has described in detail above how this mandate requires EDC and the Canada Account to develop and structure funding schemes that give Canadian exporters and their customers better terms than would be available on the market, and that therefore confer “benefits” within the meaning of Article 1.1(b) of the Subsidies Agreement. Furthermore, Brazil has established that EDC and the Canada Account are required to fund exports, as opposed to domestic sales. As a result, the programs themselves are de jure contingent upon export, within the meaning of Article 3 of the Subsidies Agreement.
Leaving aside the issue of export contingency, we thus understand Brazil to argue that EDC is effectively required to grant subsidies. However, we find nothing in Brazil's various submissions in support of this argument. The only factual evidence proffered by Brazil in support of its argument is the quote from EDC's mandate that EDC was established "for the purposes of supporting and developing, directly or indirectly, Canada’s export trade and Canadian capacity to engage in that trade
538 United States - Measures Affecting the Importation, Internal Sale and Use of Tobacco, BISD 41S/131, para. 118, adopted on 4 October 1994 ("United States – Tobacco").
539 CanadExport On-Line, Focus on Export Development Corporation, pg. 2 (statement of Mr. Paul Labbé, former President of the EDC).
WT/DS70/R
Page 180
and to respond to international business opportunities.”540 This statement by itself clearly cannot be viewed as a requirement to provide prohibited export subsidies. Nor has Brazil demonstrated otherwise that such support and development necessarily involves subsidization. Although such support and development might conceivably take the form of subsidization, there is nothing to suggest that this will necessarily be the case. In our view, a mandate to support and develop Canada's export trade does not amount to a mandate to grant subsidies, since such support and development could be provided in a broad variety of ways.
We consider that Brazil effectively concedes that the EDC mandate does not require the grant of export subsidies when it states that the EDC mandate has been interpreted to require the EDC to fund projects that give “Canadian exporters an edge when they bid on overseas projects.”541 For Brazil, this "edge" necessarily refers to subsidization. Even if the grant of an "edge" did imply the grant of subsidies,542 and even if in practice the EDC programme were applied so as to grant subsidies, this would not mean that, in law, the EDC mandate requires the grant of subsidies. Rather, in such circumstances the grant of subsidies would be the result of the exercise of the administering authority's discretion in interpreting its mandate. We again recall that the panel in US – Tobacco recollected "that panels had consistently ruled that legislation which mandated action inconsistent with the General Agreement could be challenged as such, whereas legislation which merely gave the discretion to the executive authority … to act inconsistently with the General Agreement could not be challenged as such…"
For these reasons, we find that Brazil has failed to demonstrate that the EDC programme as such mandates the grant of subsidies. Rather, the EDC programme constitutes discretionary legislation. In light of the distinction that GATT/WTO panels have consistently drawn between discretionary legislation and mandatory legislation, we find that we may not make any findings on the EDC programme per se. We therefore confine our analysis to Brazil's claims concerning the actual application of the EDC programme in the regional aircraft sector.
Does the EDC programme as applied provide prohibited export subsidies?
We recall that Brazil challenges four types of EDC financing assistance allegedly provided to the Canadian regional aircraft industry: debt financing, loan guarantees, residual value guarantees and equity financing. We shall address each of these alleged forms of EDC financing assistance in turn, for the purpose of determining whether they constitute "subsid[ies]" within the meaning of Article 1 of the SCM Agreement. Only if we make affirmative determinations in this regard will we consider whether such subsidies are "contingent … upon export performance" within the meaning of Article 3.1(a).
EDC debt financing
Arguments of the parties
Brazil asserts that Canada grants subsidies in the form of direct financing at concessionary rates for up to 90 percent of the cost of an aircraft. According to Brazil, EDC financing of up to 90 percent (or more) of an aircraft's cost constitutes a direct transfer of funds by grant or loan, within the meaning of Article 1.1(b) of the Agreement. Furthermore, "EDC's provision of financing of up to 90 percent (or more) of an aircraft's cost over a 15-year or 15-year-plus period at concessionary rates confers the obvious benefit, within the meaning of Article 1.1, of lowering the price of an exported
540 Export Development Act, Section 10(1).
541 CanadExport On-Line, Focus on Export Development Corporation, page 2 (statement of Mr. Paul Labbé, former President of the EDC).
542 Canada expressly denies that giving an "edge" implies the grant of subsidies. Canada asserts that the "edge" given to Canadian exporters derives from EDC's experience and expertise.
WT/DS70/R
Page 181
aircraft for the purchaser. No private financial institution or investor would provide this degree of financing on concessionary terms …"543 Brazil's claim against EDC debt financing is therefore based on its view that a "benefit" is conferred when a financial contribution is provided at terms that would not be available to the recipient on the market.
Brazil refers to a statement by Mr. Labbé, a former EDC President, to demonstrate the "benefit" afforded to Canadian exporters by EDC debt financing:
EDC’s financing support gives Canadian exporters an edge when they bid on overseas projects. . . . Trade deals increasingly depend on complex and tightly negotiated financing arrangements where a few basis points in interest rates can make or break the deal. Exporters are having to bid not just on the basis of quality and price, but also on the basis of the financing package supporting the sale.544
According to Brazil, "EDC’s help – to the tune of a 'few basis points' – must be better than that which would otherwise be commercially available, or an EDC financing package would not, in EDC’s former President’s words, 'give Canadian exporters an edge.' This 'edge' is a 'benefit' conferred upon exporters, within the ordinary meaning of Article 1.1 of the Subsidies Agreement."
Brazil also asserts that "one of the benefits of EDC financing is that a Canadian exporter can 'advis[e] potential foreign buyers that Canadian financing may be available for their purchase,' thereby 'enhanc[ing] the competitiveness of [the exporter’s] sales proposal.'545 The Canadian regional aircraft exporter’s product is more attractive to a purchaser, in turn, for the simple reason that it costs less than it would without the Canadian Government’s help."546
Brazil argues that EDC debt financing confers a "benefit" because it does not provide for the collection of any risk premium. Brazil notes statements by EDC personnel to the effect that EDC "absorb[s] the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries."547 Brazil cites EDC's net interest margin (3.03 percent in 1996, and 2.82 percent in 1997) to establish that it is not being compensated for this extra risk and, conversely, that recipients of EDC funding are paying less for that funding than they would from commercial sources. Brazil asserts that, to accept a loan portfolio of EDC's risk class, private investors would demand a spread of 1,242 basis points over riskless US Treasuries of 15 plus years maturity, or 17.73 percent. Brazil submits that the difference between this figure and EDC's net interest margin demonstrates that EDC's lending activities do not effectively cover the real and quantifiable costs associated with the riskiness of EDC's loan portfolio. Brazil considers that EDC's net interest margin should be higher than that of commercial banks, since EDC's loan portfolio is higher risk than commercial banks' loan portfolio. According to Brazil, it is a well-known and accepted financial concept that investments with higher risks demand higher returns.
Brazil refers to EDC's loan portfolio to establish that EDC's debt financing is riskier than that undertaken by commercial banks. Brazil notes that impaired or non-performing loans comprise 14.4 percent of gross loans receivable. Brazil asserts, on the basis of Standard & Poor's analysis, that an equivalent ratio of 3 percent "can cause concern", and even a ratio of 7 percent would indicate that the future of the relevant institution "may be doubtful".548 Brazil asserts that EDC's 14.4 percent ratio of gross impaired loans to gross loans receivable compares to ratios of approximately one percent for The Royal Bank of Canada, the Bank of Montreal, Canadian Imperial Bank of Commerce, and the
543 See para. 6.60 above.
544 CanadExport On-Line, Focus on EDC, page 2.
545 Infoentrepreneurs website, page 13.
546 See para. 6.59 above.
547 Export Development Corporation, 1995 Chairman and President's Message, page 2.
548See para. 6.13 above.
WT/DS70/R
Page 182
average of all US FDIC-insured commercial banks. In addition, Brazil notes that approximately 57.6 percent of EDC's performing loan portfolio has been classified by EDC itself as "below investment grade" or "speculative grade". Brazil also refers to EDC's ratio of allowance (or reserve) for losses on loans to gross loans receivable. Brazil asserts that banks are required to maintain a reserve for loan losses in order to cover possible future loan losses. Brazil argues that the level of reserve maintained by a bank therefore reflects management's judgment regarding the quality of its loan portfolio. In this regard, Brazil notes that EDC's allowance for loan losses as a percent of total loans is 13.2 percent, compared with between one and two percent for each of the three major Canadian banks and the US industry average. Brazil also refers to EDC's ratio of net write-offs to gross loans receivable. Brazil asserts that EDC's ratio of 0.15 percent is "absurdly low", and questions both the criteria applied by the EDC in determining when loans are written off, and whether the criteria are similar to those applied by commercial banks.
Furthermore, Brazil relies on the following statement by an EDC official before the Canadian Parliament to demonstrate that a risk premium is not collected by EDC:
If we are not to lose money, we should be making at least the rate of inflation on our capital base which is our aim. That goal is a long cry from the 15 percent or 20 percent return on equity that would be required to survive in the private sector.
Brazil submits that EDC cannot be collecting the requisite risk premium if it is merely making the rate of inflation.
Brazil argues that, even if all EDC loans are secured, private lenders still demand a spread of at least 150 basis points above the riskless US Treasuries of identical tenor. Brazil bases this calculation on the fact that EDC provides debt financing for CRJ aircraft for 15 years or longer. Brazil asserts that the difference between the yield risk for riskless US Government 15-year securities (5.31 percent) and the yield for investment grade-rated unsecured non-rail transportation bonds (6.88 percent) is approximately 150 basis points.
Brazil also submits even if the net cost to government approach to "benefit" advocated by Canada were accepted generally, it would not be appropriate in the specific context of EDC debt financing since, in order to give the appearance of meeting its costs, EDC receives relief from the Canadian government. Brazil refers to the April 1996 Report of Canada's Standing Senate Committee on Banking, Trade and Commerce, which notes that the Canadian Government provided debt relief of
$151 million to two of EDC's "problem" accounts. Brazil recalls that the Senate Report concludes that "looking at the bottom line can give a misleading impression of how EDC is faring on its loan portfolio."549
Brazil notes that, in response to a question from the Panel, Canada submitted as Business Confidential Information an EDC Standing Board Resolution of 17 June 1992, which applies to all business conducted by EDC under its corporate account, including the regional aircraft sector. Brazil notes Canada's assertion that, in accordance with this Resolution, EDC has lent above its cost of funds under its corporate account with respect to the regional aircraft sector since 1 January 1995. Brazil notes, however, that a clause in the Resolution expressly allows the EDC to derogate from the Resolution in certain circumstances. Brazil submits that a question remains whether the derogation has been applied for EDC lending in the regional aircraft sector.
Brazil also submits that, even if EDC were considered to be profitable overall, this fact tells the Panel nothing about whether EDC meets its costs on its lending activities to the regional aircraft industry. Brazil has adduced evidence with respect to one alleged instance of EDC of debt financing in the regional aircraft sector. Brazil asserts that EDC debt financing was provided for 30 Bombardier
549 See para. 6.7 above.
WT/DS70/R
Page 183
CRJs purchased by ASA Holdings, Inc. and its subsidiary, Atlantic Southeast Airlines (hereinafter referred to collectively as "ASA"), in April 1997.
Canada denies that EDC debt financing in the regional aircraft sector constitutes "subsid[ies]" within the meaning of Article 1.1 of the SCM Agreement. Canada asserts that Brazil's allegations concerning EDC are not supported by the evidence adduced.
Canada asserts that the EDC is a corporation incorporated under the laws of Canada that is wholly-owned by the Government of Canada. Canada states that EDC operates on commercial principles with the objectives of (a) supporting and developing, directly or indirectly Canada’s export trade; and (b) supporting and developing, directly or indirectly Canada’s capacity to engage in exports, and respond to international business opportunities.550 Canada submits that EDC is self- sustaining, and that it earns a significant net interest margin that is equal to or better than most commercial financial institutions of similar rating. Canada asserts that the commercial viability of EDC’s activities should be viewed in the context of the fact that the core lending business of many major banks is "increasingly unrewarding”.551 According to Canada, at 3.03 percent, EDC’s net interest margin -- a better measure of performance than “return on equity” -- is better than most commercial banks of similar or better credit rating.
Canada denies Brazil's allegation that EDC provides debt financing at "concessionary" rates. Canada submits that EDC’s financing activities are based on commercial pricing. According to Canada, rates for EDC financing reflect commercial benchmarks and spreads that are in accordance with commercial credit ratings -- and, where this is not available, internal EDC credit ratings in accordance with prudent commercial practices. Canada also submits that EDC’s financing terms and structures are consistent with market trends and practices.
Canada states that the EDC always lends above its cost of funds, and therefore does not incur a net cost on its financing activities. Canada also states that the EDC operates on the basis of commercial principles, and therefore does not provide an advantage above and beyond the market. For these reasons, Canada argues that EDC financing does not constitute a subsidy.
Canada notes Brazil's general argument that EDC attempts to satisfy "the seemingly endless appetite of Canadian exporters for financial support", Canada asserts that Brazil has selectively quoted from the relevant source materials. Canada states that Brazil omits a connecting sentence that substantially qualifies the passage. Canada notes that the full paragraph from which Brazil quoted states:
“It will be hard to maintain the pace of 1995 and earlier, but EDC has a lot of growing to do before it begins to satisfy the seemingly endless appetite of Canadian exporters for financial support and advice. However, EDC cannot nor should not strive to be the solution for all the challenges faced by Canadian exporters. EDC complements the banks and other financial intermediaries, but cannot substitute for them.” [emphasis added by Canada]
Canada asserts that, by saying that EDC cannot substitute for banks and other financial institutions, the Chairman and the President of the EDC were acknowledging that EDC should not try to satisfy the “endless appetite of Canadian exporters.” Canada concludes that, contrary to Brazil’s conclusion from the misquoted passages, EDC manifestly does not attempt to satisfy the "endless appetite of Canadian exporters".
550 Export Development Act, Section 10.
551 “The trials of megabanks”, The Economist, October 31, 1998, 23 at 23.
WT/DS70/R
Page 184
Canada also notes Brazil's reliance on the statement in the Chairman and President's Message that EDC's "goal is to help absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries". Canada notes that Brazil relies on this statement to support its argument that "no private financial institution or investor would provide this degree of financing on concessionary terms." Canada states that the full text from which this statement is extracted reads:
In addition to the shift from sovereign to commercial loans, the complexity, scale and duration of financing are changing, and thereby changing the risks associated with insuring and financing Canadian exports.
To reinforce its capacity to manage these changing risks, EDC has established a new Financial Services Office and procedures for evaluating loan portfolios on an industry, geographic, and individual transaction basis. Our goal is to help absorb risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries, by diversifying the Corporation’s business both on a country and sectoral basis. We are determined to achieve this goal through growth in both emerging and established markets. [emphasis added by Canada]
Canada submits that, contrary to Brazil’s assertion, the passage, when quoted in full, does not support the proposition that EDC enters into financing on concessionary terms. Canada asserts that the sentence quoted by Brazil is concerned with portfolio diversification by the EDC, which Canada considers to be an elementary and prudent market activity, which has nothing to do with whether EDC financing confers a benefit.
Canada asserts that the statement by Mr. Labbé, a former EDC President to the effect that "EDC's financing support gives Canadian exporters an edge when they bid on overseas projects" is not evidence that EDC finances below market. Canada asserts that earlier in the same article, Mr. Labbé is quoted as stating that:
What we bring to the table is a wide variety of financial solutions and insurance support, as well as extensive market and sectoral expertise… We have teams dedicated to different market sectors such as information technology and industrial equipment so that we understand your business as well as you do.
Canada argues that, taken in context, the "edge" that EDC’s financing support gives Canadian exporters derives from their knowledge of the various export markets, from their expertise in “complex and tightly negotiated financing arrangements” and from their awareness that a “few basis points in interest rates can make or break the deal”. Canada adduces evidence to demonstrate that this sort of knowledge and expertise is something that almost all financial institutions trumpet in their promotional literature.
With regard to Mr. Labbé's statement before the Canadian Parliament that EDC's goal is to avoid losing money by "making at least the rate of inflation", and his acknowledgement that the rate of inflation would be below the return "that would be required to survive in the private sector", Canada asserts that net interest margin is a better measure of performance than return on equity in the context of EDC debt financing. Canada defines net interest margin as the difference between gross interest income and gross interest expense on all interest bearing assets, divided by the value of interest bearing assets. Thus, Canada asserts that net interest margin is a useful measure of how well a financial institution's assets are performing. Canada asserts that EDC's net interest margin is better than most commercial banks of similar or better credit rating. Although Canada concedes that EDC does not pay corporate income tax, and does not normally pay a dividend, Canada asserts that these facts are not relevant for the net interest margin comparison because the net interest margin is calculated before tax and before dividends are paid. Canada asserts that whether or not a financial institution pays corporate income taxes or dividends does not affect its cost of funds, or the risk
WT/DS70/R
Page 185
margin charged. Canada notes Brazil's argument that, even if all EDC loans are secured, private lenders still demand a spread of at least 150 basis points above the riskless US Treasuries of identical tenor. Canada notes, however, that Brazil's argument is based on the difference between the yield for riskless US Government securities (5.31 percent) and the yield for investment grade-rated unsecured non-rail transportation bonds (6.88 percent). Canada submits that the yield on unsecured bonds does not support an argument regarding secured lending.
Furthermore, Canada rejects Brazil's argument that EDC's net interest margin is insufficient to cover the ratio of non-performing loans in its portfolio. Canada argues that the net interest margin already takes non-performing loans into account.552 Thus, any further deduction from the net interest margin to cover the cost of funding non-performing loans would constitute double-counting.
With regard to EDC's non-performing loan ratio, Canada acknowledges that, overall, EDC's non-performing loans comprise 14.4 percent of gross loans receivable. Canada also acknowledges that, overall, 57.6 percent of EDC's performing loans are classified as below investment grade. Canada asserts, however, that EDC loans in the aircraft sector are secured against the asset, thereby improving the rating of the credit. Canada asserts whether or not a loan is secured or unsecured loan has an impact on the credit risk. Canada submits that, after considering the value of such security, 91 percent of EDC's aircraft portfolio is investment grade or higher. Thus, Canada denies that EDC is engaged in high risk lending in the regional aircraft sector, as alleged by Brazil.
Canada notes Brazil's argument that EDC's allowances for loan losses as a percentage of total loans is 13.2 percent, compared with one or two percent fore major Canadian banks and the US industry average. Canada asserts that EDC allowances for loan losses are set to ensure that EDC's total portfolio maintains an equivalent rating of at least AA. Canada asserts that it is more important to consider how much is lost on the portfolio, and whether adequate provision has been made for such loss. In this regard, Canada submits that EDC's write-offs against provision have never exceeded 0.15 percent of gross loans receivable over the past eight years. Over the same period, Canada asserts that none of the Canadian commercial banks had write-offs of less than 0.21 percent, and which in one case was as high as 1.52 percent. Canada also refers to material adduced by Brazil indicating that, according to the US Federal Deposit Insurance Corporation, net write-offs of the US banking industry amounted to 0.63 percent of average loans in 1997, down from a peak of 1.59 percent in 1991. Canada concludes therefore that EDC's write-offs against provisions are more conservative than any of these other institutions. In addition, Canada refers to a letter form Canada's Auditor General contained in EDC's 1997 Annual Report , which confirms that the EDC financial statements are presented in accordance with Generally Accepted Accounting Principles ("GAAP"). Canada asserts that essential elements of GAAP are the criteria applied, and that it consistently applies those criteria, when valuing and writing-off loans.
Canada notes Brazil's reference to "Cdn $151 million in direct government relief to EDC for two of its 'problem' accounts."553 Canada asserts that this amount was paid to EDC as indemnification for the forgiveness of the sovereign debts of Poland and Egypt by the EDC in accordance with instructions from the Government of Canada resulting from Canada's international commitments through the Paris Club. Canada asserts that the relevant sovereign debts were unrelated to the civil aircraft sector.
552 Canada asserted that "when a loan becomes non-performing, all financial institutions, including the EDC, cease to recognize interest income on that non-performing loan. However, all financial institutions continue to carry the interest expense of that non-performing loan. In other words, the gross interest income is the interest income on performing loans, but interest expense is the cost of funding all loans in the portfolio, performing and non-performing" (see para. 6.34 above).
553 See para. 6.7 above.
WT/DS70/R
Page 186
In response to a question from the Panel, Canada submitted as Business Confidential Information an EDC Standing Board Resolution of 17 June 1992, which applies to all business conducted by EDC under its corporate account, including the regional aircraft sector. Canada asserts that, in accordance with this Resolution, EDC has lent above its cost of funds under its corporate account with respect to the regional aircraft sector since 1 January 1995.
Evaluation by the Panel
In examining Brazil's claim against EDC debt financing, we recall that the Appellate Body stated in EC - Hormones that:
"[t]he initial burden lies on the complaining party, which must establish a prima facie case of inconsistency with a particular provision of the SPS Agreement on the part of the defending party, or more precisely, of its SPS measure or measures complained about. When that prima facie case is made, the burden of proof moves to the defending party, which must in turn counter or refute the claimed inconsistency."
Thus, in order for Brazil's claim against EDC debt financing in the Canadian regional aircraft sector to succeed, a prima facie case must be made that EDC debt financing in the Canadian regional aircraft sector constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement. In particular, there must be a prima facie case that a "financial contribution" by a government or public body confers a "benefit".
We are in no doubt that EDC debt financing in the Canadian regional aircraft sector constitutes a "financial contribution" within the meaning of Article 1.1(a) of the SCM Agreement, since it constitutes a "direct transfer of funds". Similarly, we are in no doubt that the EDC constitutes a relevant "public body". Canada has not disputed that EDC debt financing constitutes "financial contributions" by a "public body".
Brazil's claim that EDC debt financing confers the requisite "benefit" is essentially based on statements by EDC officials, an analysis of EDC's financial performance, and the EDC debt financing granted to ASA. In analysing Brazil's claim, we recall that a "benefit" is conferred within the meaning of Article 1.1(b) of the SCM Agreement when a financial contribution is provided on terms that are more advantageous than those that would have been available to the recipient on the market.554
Statements by EDC officials
Brazil refers to a statement concerning EDC's alleged attempt to satisfy "the seemingly endless appetite of Canadian exporters for financial support" through a variety of financial and risk absorption services. Canada has demonstrated, however, that the statement in full expressly provides that "EDC cannot nor should not strive to be the solution for all the challenges faced by Canadian exporters." In context, therefore, we consider that the statement adduced by Brazil does not support a conclusion that EDC attempts to satisfy "the seemingly endless appetite of Canadian exporters for financial support" through the provision of subsidized debt financing.
Brazil also refers to a statement by Mr. Labbé, a former EDC President, that "EDC's financing support gives Canadian exporters an edge when they bid on overseas projects." Brazil notes that this statement was made in the context of references to financing packages supporting export sales, and suggests that the "edge" in question is an edge in financing terms. However, Canada has demonstrated that the statement was also made in the context of references to "market and sectoral expertise", suggesting that the relevant "edge" is the ability of EDC officials to assemble better
554 See para. 9.120 above
WT/DS70/R
Page 187
structured financial packages on the basis of their knowledge and expertise. Given the possibility for divergent contextual interpretations of Mr. Labbé's reference to the "edge" provided by EDC debt financing, this statement provides no firm guidance as to whether EDC provides exporters with an "edge" through subsidization.
Brazil also refers to a statement in the EDC Chairman and President's Message that EDC's "goal is to help absorb the risk on behalf of Canadian exporters, beyond what is possible by other financial intermediaries." Brazil relies on this statement to argue that "no private financial institution or investor would provide this degree of financing on concessionary terms." Canada has demonstrated, however, that when taken in full, this statement does not establish that EDC engages in subsidized debt financing. Rather, it simply demonstrates that EDC engages in portfolio diversification.
In addition, Brazil has relied on Mr. Labbé's statement before the Canadian Parliament that EDC's goal is to avoid losing money by "making at least the rate of inflation", and Mr. Labbé's acknowledgement that the rate of inflation would be below the return on equity "that would be required to survive in the private sector." In response, Canada asserts that return on equity is not an appropriate measure of EDC's performance, and that a more appropriate measure of performance is EDC's net interest margin. We note that Brazil does not expressly disagree with Canada's assertion that net interest margin is a better measure of EDC's debt financing performance than return on equity. Indeed, Brazil itself referred to EDC's net interest margin as a relevant economic indicator in its first written submission to the Panel. In our view, net interest margin is a more appropriate measure of performance than return on equity in the context of debt financing. We consider that return on equity is a more appropriate measure of performance in the investment sector, where equity shares are held by the investor. For the above reasons, we do not accept Brazil's argument that Mr. Labbé's statement concerning EDC's return on equity necessarily demonstrates that EDC engages in subsidized debt financing.
EDC's financial performance
In the context of EDC's financial performance, we understand Brazil to argue that EDC provides concessionary, and therefore subsidized, debt financing because it provides debt financing on terms that do not cover the risk margin of its loan portfolio. Although EDC's net interest margin appears to compare favourably with that of commercial banks, we understand Brazil to argue that a true comparison cannot be made because of the greater riskiness of EDC's loan portfolio relative to that of commercial banks. Brazil does not dispute Canada's argument that its net interest margin compares favourably with that of certain commercial banks.
By way of preliminary remark, we recall that Brazil itself initially referred to EDC's net interest margin ("a mere 2.82 percent in 1997, and 3.03 percent in 1996")555 to indicate the poor financial performance of EDC. We find unconvincing, therefore, that Brazil subsequently seeks to indicate that in fact EDC's net interest margin does not constitute a sufficient basis for comparison.
Furthermore, we consider that Canada has demonstrated that the net interest margin is a sufficient basis for comparing EDC's debt financing performance with that of commercial banks. In our view, Canada has demonstrated that EDC's net interest margin already takes into account any additional riskiness in EDC's loan portfolio compared with commercial banks. We recall Canada's argument that, when a loan becomes non-performing, in calculating net interest margin the EDC ceases to recognize interest income on that non-performing loan, but continues to carry the interest expense of that non-performing loan. Thus, the gross interest income is the interest income on performing loans, but interest expense is that of all loans in the portfolio, performing and non- performing. This argument has not been challenged by Brazil .
555 See para. 6.7 above.
WT/DS70/R
Page 188
We do not accept Brazil's argument that EDC's net interest margin reflects a poor financial performance indicative of concessionary, and therefore subsidized, financing as even if all EDC's loans for aircraft were secured, private lenders would still demand a spread of at least 150 points above riskless US Treasuries of identical tenor. As noted by Canada, that assertion is based on the yield for unsecured bonds. We consider that assertions based on unsecured bonds provide no guidance in reviewing returns on secured lending.
Brazil has also sought to establish the added risk of EDC's loan portfolio by emphasising EDC's allowance for losses. We note in this regard that Canada has expressly acknowledged that "the average portfolio of [EDC] business is of a poorer risk quality" than that of commercial banks. As also noted by Canada, however, "EDC's provision charge is higher than that for a commercial bank", and therefore the risk in EDC's loan portfolio is offset by these allowances, or provision. Furthermore, EDC's higher allowance for losses does not undermine any appraisal of EDC's debt financing made on the basis of net interest margin, since the allowance, and therefore the higher risk of the EDC portfolio, is reflected in EDC's net interest margin. In this regard, we recall Canada's express statement that "the provision charge is funded by the Net Interest Margin".556
We recall Brazil's argument that the fact that 57.6 percent of EDC's performing loans generally are classified as below investment grade demonstrates that EDC's loan portfolio is riskier than that of commercial banks. However, Canada explains that EDC's aircraft portfolio is not as risky as its general portfolio. In particular, Canada asserts that all loans in the civil aircraft sector, and therefore the regional aircraft sector, are secured, and that as a result in fact 91 percent of EDC's aircraft portfolio is investment grade or higher. Brazil has provided no basis for us to dispute Canada's argument, or otherwise suggested that a loan portfolio comprising 91 percent investment grade loans is inconsistent with commercial banks' aircraft portfolios.
We are not persuaded that our analysis of Brazil's arguments concerning EDC's debt financing performance should be influenced by indemnification payments from the Government of Canada to EDC following the writing-off of sovereign debt pursuant to Canada's Paris Club commitments. We do not consider that EDC action in response Canada's Paris Club commitments is indicative of whether or not EDC debt financing in the Canadian regional aircraft sector confers a "benefit". In any event, we recall that the relevant sovereign debt was unrelated to the civil aircraft sector.
Before concluding our examination of Brazil's arguments concerning EDC's financial performance, we recall the EDC Standing Board Resolution of 17 June 1992, which applies to all business conducted by EDC under its corporate account, including the regional aircraft sector. We note that, according to this Resolution, EDC's lending yield must cover cost plus a minimum risk margin (which varies according to the credit rating of the recipient). Brazil makes no attempt to suggest that this policy is inconsistent with that of commercial banks. Although we acknowledge that under the Resolution EDC may derogate from this policy, Brazil has made no attempt to establish557 that such derogation has been exercised with respect to EDC debt financing in the regional aircraft sector.
In light of the above, we are not convinced by Brazil's argument that EDC's net interest margin does not provide sufficient basis for comparing EDC's debt financing performance with that of commercial banks. Brazil's position is based on its argument that EDC's net interest margin does not reflect the risk of EDC's loan portfolio. However, in light of the above we consider that the riskiness
556 Section II(a), Analysis of EDC's Financial Performance, EDC Corporate Finance and Control, 12 December 1998.
557 We recall that the burden is on Brazil, as the complaining party, to adduce evidence sufficient to support its case.
WT/DS70/R
Page 189
of EDC's loan portfolio is reflected in its net interest margin. Once again, we recall that EDC's net interest margin compares favourably with that of certain commercial banks.
EDC debt financing for ASA
Brazil has referred to the EDC debt financing provided to ASA in April 1997, in support of its claim that EDC debt financing is provided on concessionary, and therefore subsidized, terms.
On the basis of information in the record, on 13 December 1998 we asked Canada to provide details of the terms and conditions of the alleged EDC debt financing to ASA, together with a copy any relevant finance agreement. Canada refused to provide the information requested by the Panel, on the grounds that "Brazil has made no allegation concerning ASA", and that the information requested is Business Confidential Information.
In commenting on Canada's response to our request, Brazil asked the Panel to "adopt adverse inferences, presuming that the information withheld is prejudicial to Canada's position."
We recall our earlier rejection of Canada's criticism of the Panel's Procedures Governing Business Confidential Information, and therefore regret deeply Canada's refusal to provide the requested information. With regard to Canada's assertion that Brazil failed to make any specific allegation concerning the ASA transaction, we understand Brazil's claim against EDC debt financing to cover all instances of EDC debt financing in the Canadian regional aircraft sector. We therefore reject Canada's assertion that Brazil has made no allegation concerning the ASA transaction.
In adducing evidence regarding the ASA transaction, Brazil does not assert, much less provide evidence to show, that EDC provided ASA with debt financing at below-market rates. The only information adduced by Brazil concerning the financing terms for this transaction is contained in ASA's 1997 annual report. At pages 15/16 of the annual report, reference is made to loans or leases "with interest payable at various interest rate options determined by reference to either U.S. treasury rates or LIBOR". Brazil makes no attempt to specify what these rates are, or how they are calculated, or that the rates referred to are below-market. Accordingly, we find that Brazil's arguments concerning ASA provide no basis for finding that either this specific instance of EDC debt financing, or EDC debt financing in the regional aircraft sector generally, confers a "benefit" within the meaning of Article 1.1(b) of the SCM Agreement.
For the above reasons, we do not believe that the evidence and arguments adduced by Brazil in respect of statements by EDC officials or EDC's overall financial performance demonstrates subsidized debt financing. In particular, we find no basis for accepting Brazil's argument that EDC debt financing confers a "benefit" because it does not provide for a margin to cover the additional risk of EDC's loan portfolio. Indeed, we note that as a matter of policy, EDC debt financing generally (i.e., in all sectors) is designed to cover cost and provide a minimum risk margin. Brazil has not demonstrated that this general policy is derogated from in respect of debt financing in the regional aircraft sector. Certainly, the evidence adduced by Brazil concerning the ASA debt financing transaction in no way indicates that the general EDC debt financing policy of covering costs and a minimum risk margin has not been applied in the regional aircraft sector.
We note that Brazil asked us to make "adverse inferences" in light of Canada's refusal to provide details of the ASA transaction. In certain circumstances when direct evidence is not available, we consider that a panel may be required to make such inferences when there is sufficient basis to do so. This is especially true when direct evidence is not available because it is withheld by a party with sole possession of that evidence. In the present instance, however, we do not consider that there is sufficient basis for an inference that EDC debt financing in the Canadian regional aircraft sector confers a "benefit". In particular, Brazil has made no attempt to demonstrate that EDC debt financing was provided to ASA on below-market terms. Furthermore, Brazil has not demonstrated,
WT/DS70/R
Page 190
on the basis of its arguments concerning statements by EDC officials and EDC's financial performance, that EDC debt financing generally confers a "benefit". Had Brazil done so, we may have been required to make the inferences requested by Brazil.
Thus, we find that there is no prima facie case that EDC debt financing confers a "benefit", and therefore constitutes a "subsidy", within the meaning of Article 1 of the SCM Agreement. In the absence of such prima facie case, we cannot find in favour of Brazil's claim that EDC debt financing in the regional aircraft sector takes the form of prohibited export subsidies, contrary to Article 3.1(a) and 3.2 of the SCM Agreement, and we therefore reject that claim
EDC loan guarantees
Brazil claims that EDC offers long-term loan guarantees to purchasers or lessors of Canadian regional aircraft, contrary to Article 3.1(a) and 3.2 of the SCM Agreement. Brazil asserts that loan guarantees constitute the "potential direct transfer of funds or liabilities" within the meaning of Article
1.1 of the SCM Agreement. Brazil asserts that EDC loan guarantees confer a "benefit" within the meaning of Article 1.1 because such guarantees allow Bombardier to deduct as much as 20 per cent from the selling price of a CRJ.
In support of its claim against EDC loan guarantees to purchasers or lessors of Canadian regional aircraft, Brazil has adduced evidence of two alleged EDC loan guarantees. First, Brazil claims that in April 1995 EDC provided loan guarantees to a Bombardier customer for CRJs. Brazil refers to a press report of statements by an Industry Canada official in support of this claim. In this press report, one Industry Canada official is quoted as saying that "[i]n this case, the federal loan guarantee can shave as much as 20% off the selling price of a Canadair jet." Second, Brazil claims that EDC loan guarantees were provided to Comair Holdings Inc. ("Comair") in 1997. Brazil refers to 10-K forms filed by Comair with the US Securities and Exchange Commission in support of this claim. The relevant 10-K forms state that "Comair expects to finance the aircraft … through a combination of working capital and lease, equity and debt financing, utilizing manufacturers' assistance and government guarantees to the extent possible."
On the basis of information in the record concerning alleged EDC loan guarantees, we asked Canada to provide details, including terms and conditions, of any loan guarantees issued by EDC for transactions concerning the civil aircraft sector since 1 January 1995. We also asked Canada to provide details of any loan guarantees provided by EDC to Comair in 1997.
In response to our question concerning loan guarantees to Comair, Canada denied that EDC had provided loan guarantees to Comair in 1997.558 In light of Canada's denial, and considering that Brazil has merely adduced evidence of Comair's expectation of government guarantees, rather than Comair's actual receipt of government guarantees, we reject Brazil's allegation that EDC granted an export subsidy in the form of a loan guarantee to Comair in 1997.
Furthermore, Canada informed us that the April 1995 loan guarantee referred to in the press report cited by Brazil was provided for a domestic sale of CRJs by a government agency other than EDC. In light of this information, and in the absence of information to the contrary from Brazil, we find that Brazil's allegation concerning the alleged April 1995 EDC loan guarantee falls outside our terms of reference because the April 1995 guarantee does not constitute a "loan guarantee provided by the Export Development Corporation."559
In its reply to our question requesting details of any EDC loan guarantees granted since 1 January 1995, Canada informed us that the EDC had granted two loan guarantees since 1 January
558 See para. 6.100 above.
559 WTO document WT/DS70/2.
WT/DS70/R
Page 191
1995. The first transaction was in support of the sale of two used de Havilland Twin Otters and two used de Havilland Dash 8-102s to an airline operating in the South Pacific. The second transaction was in support of pre-shipment financing of a flight inspection system sold to a sovereign Latin American buyer. Canada asserts that both guarantees were provided at "commercial rates". Canada refused to provide us with business confidential details of these transactions, because it claimed that Brazil had failed to make a prima facie case against EDC loan guarantees, and because it claimed a lack of adequate procedures to protect business confidential information.
In its comments on Canada's reply, Brazil asserted that the Panel should adopt "adverse inferences" where Canada has expressly refused to provide documentary information specifically requested, presuming that the information withheld constitutes inculpatory evidence of Canada's infringement of the SCM Agreement.
As noted above at para. 9.181, in certain circumstances we consider that a panel may be required to make inferences on the basis of relevant facts when direct evidence is not available. This is especially true when direct evidence is not available because it is withheld by a party with sole possession of that evidence. In the present case, however, we do not consider that there is sufficient basis for any such inference. The only evidence adduced by Brazil in support of its claim that EDC grants subsidies in the form of loan guarantees to purchasers or lessors of Canadian regional aircraft has been fully rebutted by Canada. In particular, Brazil's evidence concerning alleged EDC loan guarantees that "can shave as much as 20% off the selling price of a Canadair jet" has been shown to relate to non-EDC loan guarantees. Thus, this statement is in no way relevant to EDC loan guarantees provided for the export of Canadian regional aircraft. Accordingly, we decline Brazil's request to infer that EDC loan guarantees to purchasers or lessors of Canadian regional aircraft constitute prohibited export subsidies contrary to Article 3.1(a) and 3.2 of the SCM Agreement.
For the above reasons, we find that there is no prima facie case that EDC provides prohibited export subsidies in the form of loan guarantees to purchasers or lessors of Canadian regional aircraft, contrary to Article 3.1(a) and 3.2 of the SCM Agreement. We reject Brazil's claim accordingly.
EDC residual value guarantees
Brazil claims that in certain instances EDC offers prohibited export subsidies in the form of residual value guarantees to lessors of regional aircraft, protecting against the risk that the residual value of the used aircraft will be lower than anticipated. In support, Brazil relies on a press article that refers to a "suggestion that the residual value of [an] aircraft [sold in 1992] may have been guaranteed."560 Brazil asserts that EDC residual value guarantees constitute the potential direct transfer of funds or liabilities within the meaning of Article 1.1 of the SCM Agreement. Brazil asserts that EDC residual value guarantees confer an Article 1.1 "benefit" because, "by protecting [a special purpose company] against the risk that the residual value of a used aircraft at the end of a lease will be lower than anticipated, the [special purpose company] is relieved of the burden of absorbing a loss from a lower-than-expected residual value, and can pass any savings along to the airline customer in the form of lower lease payments."561
Canada asserts that Brazil's claim is based on "an article that notes a 'suggestion' that a deal completed in 1992 'may have' involved a residual value guarantee." Canada denies the factual basis of Brazil's claim, and states that "neither the EDC nor the Government of Canada has provided residual value guarantees through CRJ Capital or any other means in support of civil aircraft." Canada provides an Officer's Certificate from CRJ Capital as well as one from Exinvest to this effect.
560 "A very tactical regional response", Airfinance Journal, November 1994, pages 18-20.
561 See para. 6.139 above.
WT/DS70/R
Page 192
On the basis of Brazil's allegation that Canada "deliberately attempted to mislead the Panel" in its response to another question from the Panel not related to EDC residual value guarantees, and "because of the risk that Canada is also misrepresenting the truth" regarding EDC residual value guarantees (because the Officer's Certificates presented by Canada only expressly concern CRJ Capital and Exinvest, and not EDC), Brazil asserts that "the Panel should not accept Canada's denial as an effective rebuttal of Brazil's prima facie case." 562
We note that the only evidence adduced by Brazil to support its claim that EDC provides residual value guarantees to lessors of regional aircraft is a 1994 press article containing the "suggestion" that residual value guarantees may have been granted in 1992. In light of Canada's express denial that EDC provides residual value guarantees to lessors of regional aircraft, we find that there is no factual basis to Brazil's claim that the EDC has provided prohibited export subsidies in the form of residual value guarantees to lessors of regional aircraft.
For the above reasons, we find that there is no prima facie case that EDC provides residual value guarantees to lessors of regional aircraft, contrary to Article 3.1(a) and 3.2 of the SCM Agreement. We reject Brazil's claim accordingly.
EDC equity financing
Brazil asserts that EDC, directly or indirectly, has made equity infusions into CRJ Capital which have facilitated CRJ Capital's ability to lease or sell Canadian regional aircraft at a reduced price, contrary to Article 3.1(a) and 3.2 of the SCM Agreement. Brazil argues that EDC made an equity infusion into Structured Finance, Inc., later called Exinvest, and that Exinvest subsequently created a special purpose company called CRJ Capital. Brazil relies on information from EDC sources to support its claim that CRJ Capital acts as an aircraft leasing company. For example, Brazil argues that, according to press reports, Mr. Henri de Sonquières, Vice-President of Financial Services and Transportation at EDC, characterized CRJ Capital as a leasing company, and stated that the plan is for it to be used in the lease or sale of as many as 75 CRJs. Brazil also argues that, in interviews, EDC officials stated that CRJ Capital purchases part of a plane, syndicates the rest to private-sector lenders, and leases the aircraft to an airline.
Brazil argues that such equity infusions constitute direct transfers of funds by equity infusion within the meaning of Article 1.1 of the SCM Agreement. Brazil also argues that such equity infusions provide a "benefit" within the meaning of Article 1.1 because "EDC’s direct or indirect equity investment in CRJ Capital frees CRJ Capital up to accept a lower lease or loan payment from a lessor or purchaser of a Canadian regional aircraft than it would be able to accept in the absence of that equity investment, or to facilitate the ability of another SPC to do so. CRJ Capital is designed not to earn a profit during the term of the lease; therefore, only the debt portion of the capital used to finance the lease needs to be serviced during this period. No payment is made to equity investors. Thus, the greater the percentage of equity capital in CRJ Capital, the lower the percentage of debt capital that must be serviced. The benefit, in short, is lower lease or debt payments for the airlines."563
Canada acknowledges that Exinvest is a wholly owned subsidiary of EDC, and that EDC has made an equity infusion into CRJ Capital through Exinvest. However, Canada provides a CRJ Capital officer's certificate denying that CRJ Capital has "purchased and/or leased any aircraft", or "taken any ownership interest in any aircraft." According to Canada, CRJ Capital simply provides conventional financing for aircraft sales. Canada therefore denies what the Panel considers to be an essential element of the factual basis to Brazil's claim concerning EDC's equity financing, namely that CRJ Capital purchases and/or leases aircraft.
562 See para. 6.144 above.
563 See para. 6.138 above.
WT/DS70/R
Page 193
We note that Brazil considers that Canada's denial that CRJ Capital purchases and/or leases any aircraft should not be considered credible, since Brazil's allegation regarding CRJ Capital is based on EDC materials and statements by EDC officials. However, in the face of Canada's express denial that CRJ Capital operates as an aircraft leasing company, and an Officer's Certificate to that effect, we are unable to attach any weight to the press reports relied on by Brazil in support of its claim concerning EDC equity infusions into CRJ Capital. Accordingly, we find that there is no factual basis on which to establish a prima facie case that EDC has made equity infusions into CRJ Capital that have facilitated CRJ Capital's ability to lease or sell Canadian regional aircraft at a reduced price. We therefore reject Brazil's claim that EDC has granted prohibited export subsidies to the Canadian regional aircraft industry in the form of equity infusions into CRJ Capital.
Brazil also alleges that EDC has made an equity infusion into another special purpose company called Canadian Regional Aircraft Finance Transaction ("CRAFT"), which Brazil considers may be a successor to or replacement of sorts for Exinvest. Brazil asserts that CRAFT was launched in April 1998 as an aircraft securitization structure to provide lease and loan financing for customers buying Bombardier’s CRJ and Dash 8 aircraft. Brazil argues that EDC and the Government of Québec provided preferred capital to the CRAFT structure. Brazil states that Standard & Poor's estimate that EDC and the Government of Québec’s combined participation could exceed US $300 million.564 Brazil presumes that investment by the Canadian federal and provincial governments was required because the return to the other shareholders would be below the return demanded by private investors given the level of risk they would have to assume in the financing structure. Brazil asserts that, by accepting below market returns on their investment, EDC and the Québec Government allow the CRAFT SPC to reduce the monthly lease cost to the aircraft lessee. Brazil asserts that CRAFT falls within the Panel's terms of reference because it constitutes an EDC "equity infusion[] into [a] corporation[] established to facilitate the export of civil aircraft."565
Canada asserts that Brazil's allegations regarding CRAFT are "completely false". Canada notes that Brazil's understanding of the alleged role of EDC and the Government of Québec in CRAFT is based on a Standard & Poor's Presale Report on CRAFT that makes no reference to participation by the EDC or the Government of Québec. Canada asserts that, according to additional materials submitted by Brazil, the source of the information concerning participation by the EDC and the Government of Québec is Embraer. Canada has provided an attestation from the Director of CRAFT that "at no time has the Canadian Government or the Québec Government or any agency of the Canadian Government or Québec Government, in particular the Export Development Corporation, participated in the financing of CRAFT."
In light of Canada's express denial of any EDC equity participation in CRAFT, supported by a written attestation by a senior CRAFT official, we find that there is no factual basis to establish a prima facie case that the EDC has provided a prohibited export subsidy in the form of an equity infusion into CRAFT. We reject Brazil's claim accordingly.
CANADA ACCOUNT
Brazil claims that the Canada Account programme per se provides prohibited export subsidies to the Canadian regional aircraft industry,566 contrary to Article 3.1(a) and 3.2 of the SCM Agreement.
564 Standard & Poor’s Presale Report, “CRAFT No. 1 Trust 1998 – A,” page 3.
565 WTO document WT/DS70/2.
566 We note that Brazil's request for establishment of a panel (WT/DS70/2) refers to various alleged prohibited export subsidies provided to the "Canadian industry producing civil aircraft". Our terms of reference,
therefore, relate to assistance provided to the Canadian civil aircraft industry. With the exception of its first written submission, however, all findings requested by Brazil concerning assistance provided under the Canada Account refer to the "regional aircraft industry". Similarly, all arguments adduced by Brazil concern the
WT/DS70/R
Page 194
Brazil also challenges the Canada Account as applied, including funding, support, funds and benefits granted in the regional aircraft sector under the auspices of the Canada Account programme.
Brazil notes that EDC’s 1995 Annual Report describes the Canada Account in the following terms:
Canada Account funds are used to support export transactions which the federal government deems to be in the national interest but which, for reasons of size or risk, the Export Development Corporation (EDC) cannot support through regular export credits. Transactions are negotiated, executed and administered by EDC on behalf of the government, and are accounted for separately on the books of the Department of Foreign Affairs and International Trade (DFAIT). These activities are known collectively as the Canada Account.567
Brazil asserts that the Canada Account exists for the sole purpose of supporting exports, and that Canada Account support is therefore contingent upon export performance within the meaning of Article 3.1(a) of the SCM Agreement.
Without conceding that the Canada Account programme, or assistance granted thereunder, is contingent on export, Canada argues primarily that Canada Account assistance is not a "subsidy" within the meaning of Article 1 of the SCM Agreement. Canada asserts that Canada Account operations are financing and loan guarantee activities undertaken by the EDC on the account of the Government of Canada. Thus, any obligations under the Canada Account are funded by the Government of Canada. Canada asserts that the Canada Account is not a pool of money used to subsidise export sales.
Is the Canada Account programme per se a prohibited export subsidy?
In order to assess Brazil's claim against the Canada Account programme, we must first determine whether the Canada Account programme mandates the grant of prohibited export subsidies in a manner inconsistent with Articles 3.1(a) and 3.2 of the SCM Agreement. In this regard, we recall our earlier discussion568 on the distinction that GATT/WTO panels have consistently drawn between discretionary legislation and mandatory legislation.
During the Panel proceedings, we addressed the following question to Brazil:
In light of GATT practice regarding the distinction between mandatory and permissive legislation, could Brazil please state whether it considers that the various “programs” referred to in its oral statement to the Panel [at the first substantive meeting] (26 November 1998) require Canada to act in a manner inconsistent with Article 3.1(a) of the SCM Agreement. Why?
In response, Brazil stated that:
In Brazil’s view, EDC has interpreted its own and the Canada Account’s mandate to require it to fund projects that give “Canadian exporters an edge when they bid on overseas projects.” Brazil has described in detail above how this mandate requires EDC and the Canada Account to develop and structure funding schemes that give Canadian exporters and their customers better terms than would be available on the market, and that therefore confer “benefits” within the meaning of Article 1.1(b) of
regional aircraft industry. For this reason, we only examine Brazil's claim against assistance provided under the Canada Account insofar as such assistance relates to the Canadian regional aircraft industry.
567 EDC 1995 Annual Report, “Canada Account Profile”.
568 See para. 9.124 above.
WT/DS70/R
Page 195
the Subsidies Agreement. Furthermore, Brazil has established that EDC and the Canada Account are required to fund exports, as opposed to domestic sales. As a result, the programs themselves are de jure contingent upon export, within the meaning of Article 3 of the Subsidies Agreement. (footnote deleted)
Leaving aside the issue of export contingency, we understand Brazil to argue that Canada Account assistance is, as a matter of law, granted in the form of subsidies. However, we find nothing in Brazil's various submissions in support of this argument. As Brazil itself notes, "Canada Account funds are used to support export transactions which the federal government deems to be in the national interest but which, for reasons of size or risk, the Export Development Corporation (EDC) cannot support through regular export credits."569 Brazil has failed to demonstrate that such "support" necessarily involves subsidization. Although such "support" might conceivably take the form of subsidization, there is nothing to suggest that this must, in law, be the case. Indeed, we consider that such "support" could be provided in a broad variety of ways other than subsidization.
In our view, Brazil effectively concedes that the Canada Account mandate does not require the grant of subsidies when it states that EDC has interpreted the Canada Account mandate to require it to fund projects that give “Canadian exporters an edge when they bid on overseas projects.”570 For Brazil, this "edge" necessarily refers to subsidization. Even if the "edge" did imply the grant of subsidies,571 and even if in practice the Canada Account were applied so as to grant subsidies, this would not mean that, in law, the Canada Account mandate requires the grant of subsidies. Rather, in such circumstances, the grant of subsidies would be the result of the exercise of the administering authority's discretion in interpreting the Canada Account mandate. We again recollect "that legislation which mandated action inconsistent with the General Agreement could be challenged as such, whereas legislation which merely gave the discretion to the executive authority … to act inconsistently with the General Agreement could not be challenged as such…"572
For these reasons, we find that Brazil has failed to demonstrate that the Canada Account programme as such mandates subsidies that are contingent upon export performance. Rather, the Canada Account programme constitutes discretionary legislation. In light of the distinction that GATT/WTO panels have consistently drawn between discretionary legislation and mandatory legislation, we find that we may not make any findings on the Canada Account programme per se. We therefore confine our analysis to Brazil's claims concerning the actual application of the Canada Account programme.
Does the Canada Account programme as applied provide prohibited export subsidies?
Brazil asserts that the Canada Account provides grants, interest-free loans, low-interest loans, guarantees and other give-aways and support which confer an Article 1.1(b) "benefit" on Canadian exporters of regional aircraft by artificially facilitating their sale. In particular, Brazil asserts that in the absence of Canada Account subsidies, high-risk buyers would pay higher interest rates and would be required to make higher down-payments.
Brazil notes that, according to the "Canada Account Profile" contained in the EDC 1995 Annual Report, there are two types of Canada Account expenditure: "budgetary expenditures" covering concessionary support, and "non-budgetary expenditures" covering non-concessionary support. Brazil asserts that concessionary assistance in the form of "budgetary expenditures" would
569 EDC 1995 Annual Report.
570 CanadExport On-Line, Focus on Export Development Corporation, page 2 (statement of Mr. Paul Labbé, former President of the EDC).
571 Canada expressly denies that giving an "edge" implies the grant of subsidies. Canada asserts that the "edge" given to Canadian exporters derives from EDC's experience and expertise.
572 See para. 9.124 above.
WT/DS70/R
Page 196
certainly confer "benefits". However, because of the limited Canada Account definition of concessionary financing, Brazil asserts that non-concessionary support in the form of "non-budgetary expenditures" could also confer "benefits" when it allows the recipient to borrow at rates that the recipient would not be able to obtain on the market. Brazil states that the Canada Account Profile limits its definition of concessionary financing to "interest-free or low-interest loans repayable over extended periods." Brazil asserts that this definition would seem, for example, not to characterize as concessional those loans that, while extended at "market" interest rates, do not account for the particularly high risk-status of the borrower. In other words, a nominally non-concessional rate would amount to a concessional rate if extended to a borrower with a poor credit rating. Thus, Brazil argues that while ostensibly non-concessional loans may be at market rates, the loans would be concessional if they were extended at lower rates than the particular recipient would have been able to obtain on the market.
In support of its claim, Brazil submitted a November 1998 press report573 in which Dr. Allaire, a senior Bombardier official, acknowledged that Bombardier had used the Canada Account for "a very small number of transactions" under terms of financing described as "close to commercial."
On the basis of information in the record, we asked Canada to identify all projects / transactions in the civil aircraft sector between 1 January 1995 and 30 June 1998 which involved Canada Account financing. In response , Canada informed us that during the relevant period Canada Account debt financing had been provided for two "export transactions" in the civil aircraft sector: one transaction involved delivery of 3 Dash 8-300s to South African Express in 1995, and the other transaction involved delivery of 3 Dash 8-300s to LIAT in 1996. This information was presented in a chart. As the source of this chart is cited as EDC and Bombardier, and as Canada informed us of all Canada Account assistance in the civil aircraft sector since 1 January 1995, we understand that the two Canada Account transactions described by Canada are the same Bombardier transactions referred to by Dr. Allaire in the aforementioned November 1998 press report.
In light of all the above information, we asked Canada to provide details of the terms of the two instances of Canada Account debt financing acknowledged by Canada. Canada refused to provide this information to the Panel. Canada asserted that:
The information requested by the Panel is sensitive business confidential information. Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute.
In commenting on Canada's refusal to provide the information requested by the Panel, Brazil asked the Panel to "adopt adverse inferences, presuming that the information withheld is prejudicial to Canada's position."
Is there a prima facie case against Canada Account financing in the regional aircraft sector?
We recall that, in order for Brazil's claim against Canada Account financing in the regional aircraft sector to succeed, there must be a prima facie case that Canada Account assistance in the regional aircraft sector constitutes a prohibited export subsidy. For a prima facie case of "subsidy" within the meaning of Article 1 of the SCM Agreement, there must be a prima facie case that a "financial contribution" by a government or public body confers a "benefit".
573 "Ottawa slams Brazil's Bombardier claims", Globe & Mail, 23 November 1998 (Exh. BRA-87).
WT/DS70/R
Page 197
We are in no doubt that the Canada Account debt financing in issue constitutes a "financial contribution" by a public body, since Brazil has demonstrated that such assistance debt financing constitutes a "direct transfer of funds" by a public body, within the meaning of Article 1.1(a)(i) of the SCM Agreement.
We recall that Canada Account debt financing would confer a "benefit" if it was provided on terms more advantageous than those that would be available to the recipient on the market. With regard to Dr. Allaire's press statement, we understand Brazil to argue that "close to commercial" terms means that the terms are more advantageous for the recipient than those available on the market.
Canada asserts that Dr. Allaire’s comment was made in reference to the Commercial Interest Reference Rates established by the OECD Consensus. Canada argues that, according to the OECD Consensus, the Commercial Interest Reference Rates ("CIRRs") “represent commercial lending interest rates in the domestic market of the currency concerned.”574 According to Canada, these commercial reference rates are, therefore, by definition, “close to commercial rates”. Canada states that "these commercial reference rates may be above or below the market for a particular credit".575
By stating that OECD Consensus CIRRs "may be above or below the market for a particular credit", we consider that Canada is implicitly acknowledging that the "close to commercial rates" referred to by Dr. Allaire could themselves have been "below the market". For this reason, we find a prima facie case that the Canada Account debt financing in issue confers a "benefit", and therefore constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement.
Has Canada rebutted the prima facie case?
We recall that once a prima facie case is established, "the burden then shifts to the other party, who will fail unless it adduces sufficient evidence to rebut the presumption." Canada has asserted that "it has not put in a defense regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement." Nor has Canada sought to rely on the safe haven provided for in item (k) of the Illustrative List of Export Subsidies in Annex I of the SCM Agreement.576 In the absence of additional submissions by Canada, we find that Canada has not rebutted the prima facie case against the Canada Account debt financing in issue.
In light of the above, we find that the Canada Account debt financing in issue constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement.577
Is the Canada Account debt financing contingent on export?
We must now consider whether Canada Account debt financing in the regional aircraft sector is "contingent … upon export performance". Brazil asserts that as an alternative to EDC, and as the financier of last resort for export transactions, the Canada Account grants funds contingent in law upon export performance. Brazil also argues that Canada does not challenge Brazil's assertion that Canada Account assistance is contingent on export.
During the Panel proceedings, we asked Canada whether it "concede[s] that EDC and Canada Account activities are 'contingent … upon export performance' within the meaning of Article 3.1(a) of
574 OECD Consensus, at 12.
575 See para. 6.167 above.
576 Although Canada initially stated that "Canada Account financing and loan guarantees for exports
committed since the entry into force of the SCM Agreement have been consistent with the interest rate provisions of the OECD Consensus", Canada subsequently asserted that it is "not invoking the second paragraph of item (k) [of the Illustrative List in Annex I of the SCM Agreement] as a positive defense to any of the claims made by Brazil."
577 As a result of this finding, it is not necessary to consider Brazil's request for "adverse inferences".
WT/DS70/R
Page 198
the SCM Agreement". Canada replied "[n]o, Canada does not concede this point." We also asked Canada "[g]iven the mandate of the EDC, is it reasonable to assume that any transaction financed with EDC assistance is necessarily an export transaction. If not, why not?" Canada replied inter alia that:
EDC's mandate allows it to offer a full range of risk management services and financing products 'for the purpose of supporting and developing, directly or indirectly, Canada's export trade and Canadian capacity to engage in that trade and to respond to international business opportunities.' EDC offers, therefore, a variety of services and products, some of which are contingent on export, and others - such as foreign investment insurance, domestic credit insurance, funding investments overseas, and various equity investments - that are not contingent on export.
We also asked Canada whether "all EDC debt financing takes the form of export credits". Canada replied that "it would appear that all of the debt financing provided by the EDC since 1 January 1995 in the civil aircraft sector has taken the form of export credits, with the exception of debt financing of parts sold to Canadian civil aircraft manufacturers and debt financing for 5 Canadair regional Jets for Air Canada [which] are domestic transactions …" In response to another question from the Panel, Canada asserted that "[t]he term 'export credits' refers to direct financing for export of goods."
We conclude from the preceding paragraph, and from Canada's description of the relevant transactions as "export transactions",578 that the Canada Account debt financing in issue takes the form of export credits and, in Canada's own words, was granted "for export of goods". In our opinion, export credits granted "for the purpose of supporting and developing, directly or indirectly, Canada's export trade" are expressly contingent in law on export performance.579 We therefore find that the Canada Account debt financing in issue is "contingent in law… upon export performance" within the meaning of Article 3.1(a) of the SCM Agreement.
In light of our above findings, we conclude that the Canada Account debt financing at issue constitutes "subsid[ies] contingent in law … upon export performance" prohibited by Article 3.1(a) of the SCM Agreement. We also conclude that, in granting this prohibited export subsidy, Canada necessarily acted in violation of Article 3.2 of the SCM Agreement.
SALE TO BOMBARDIER BY THE ONTARIO AEROSPACE CORPORATION OF A 49 PERCENT INTEREST IN DE HAVILLAND INC.
Brazil claims that the sale by the Ontario Aerospace Corporation of its 49 percent share to Bombardier constitutes a prohibited export subsidy contrary to Articles 3.1(a) and 3.2 of the SCM Agreement. We will first examine Brazil's claim that the sale constitutes a "subsidy" within the meaning of Article 1 of the SCM Agreement. Only in the event of an affirmative finding on this claim will we examine the question of export contingency.
Brazil asserts that in January 1992 Bombardier and the Government of Ontario, through the Ontario Aerospace Corporation ("OAC"), purchased 51 and 49 percent, respectively, of the shares in Boeing's de Havilland Division, for a total purchase price of $100 million. At the same time, Bombardier reserved the option to purchase from the Government of Ontario the latter’s remaining 49 percent share in de Havilland for a total of Cdn $49 million. In January 1997, Bombardier exercised this option, 580 issuing to the Government of Ontario and OAC a 15-year promissory note bearing
578 see para. 9.217 above.
579 We note that Canada does not include export credits or debt financing for export transactions in its illustrative list of EDC services that in its view are not contingent on export (see para. 6.52 above).
580 During proceedings, Canada stated that the 1992 deal contained a "put/call" option, and that the 1997 sale was triggered by OAC's "put", not by Bombardier's "call".
WT/DS70/R
Page 199
interest at seven percent with annual principal repayments of Cdn $4.9 million in years six through
15.581 Brazil argues that, during the period 1992-1997, OAC provided to de Havilland four distinct types of subsidies which made the purchase of OAC’s 49-percent interest even more attractive to Bombardier: interest-free shareholder loans in the amount of Cdn $200 million; cash grants of up to Cdn $100 million; sales financing support for de Havilland’s Dash 8 aircraft, including coverage of losses; and, reimbursement of restructuring costs of up to Cdn $370 million. Brazil asserts that "increases in de Havilland's equity … did not accrue to [OAC], whose return was strictly limited to its initial investment in de Havilland."582
Canada did not submit any arguments on the question of whether the de Havilland sale constitutes a "subsidy" within the meaning of Article 1.1 of the SCM Agreement. Canada explicitly stated that it did not put in a defence on whether the transaction is a "subsidy".
In response to questions from the Panel, Brazil confirmed that it is not challenging the four distinct types of subsidies allegedly provided by OAC to de Havilland during the period 1992-1997. Rather, Brazil asserts that "the $49 million purchase price represented a major windfall for Bombardier and a subsidy within the meaning of Article 1.1 of the [SCM] Agreement, given that contributions the present value of which constitute $874.7 million were made by the Canadian and Ontario Governments to de Havilland between 1992, when the $49 million purchase price was set, and 1997, when Bombardier completed the purchase of de Havilland."583 According to Brazil, the "benefit" accrued to Bombardier when it purchased the remaining share of de Havilland in 1997 at a price fixed in 1992 that "did not reflect the value added by [the] contributions"584 made by the Canadian and Ontario Governments between 1992 and 1997. According to Brazil, "paying this $49 million debt will not bring Bombardier anywhere close to offsetting the tremendous benefit that accrued to it when [Bombardier] purchased the remaining shares of de Havilland in 1997",585 because Bombardier "has pledged to pay to Ontario the same amount Ontario paid for its share in de Havilland five years earlier, regardless of any increase in de Havilland's value in the interim."586
In our view, Brazil's claim of "benefit", and therefore subsidization, in the context of the de Havilland sale is based on Brazil's understanding that the value of de Havilland's equity increased between 1992 and 1997. According to Brazil, Bombardier derived a "benefit" from such increase because, in acquiring OAC's 49 percent share in 1997, it was merely required to pay the 1992 value of that share. The "benefit", in other words, resides in the difference between the 1992 value of OAC's 49 percent equity share (i.e., $49 million), and the increased 1997 value of that 49 percent equity share (which the purchase price, fixed in 1992, did not account for).
On the basis of information in the record, on 27 November 1998 we asked Canada to specify the total value of de Havilland's equity on the date of the sale of OAC's 49 percent share to Bombardier. Canada replied that OAC sold its share in de Havilland as a result of political imperative. In view of this political imperative, and because the purchase price had been fixed in 1992, no analysis was done to determine the value of de Havilland in 1997. Canada therefore stated that it did not know the total value of de Havilland's equity at the time of sale. At the second substantive meeting on 12 December 1998, however, Canada asserted that the value of OAC's 49 percent equity share in de Havilland decreased between 1992 and 1997. According to Canada, de
581 Initially, Brazil had argued that the Government of Ontario forgave the 15-year promissory note effective March 31, 1996. Canada adduced evidence, however, to demonstrate that the promissory note has not been forgiven, and that Bombardier has commenced repayments in line with the promissory note. Brazil has not disputed the evidence adduced by Canada.
582 See para. 6.268 above.
583 See para. 6.273 above.
584 See para. 6.273 above..
585 See para. 6.284 above.
586 See para. 6.283 above.
WT/DS70/R
Page 200
Havilland had a negative equity value in January 1997. In support, Canada provided the Panel with a letter from the Vice President, Legal Services, of Bombardier, who stated that "[t]he book value of de Havilland in 1997 was irrelevant to the price paid at the time. The 1992 agreement set the price ($49 million) for the Government of Ontario's share of de Havilland. According to the audited statements of de Havilland, the net equity value of de Havilland when Ontario exercised its put in January, 1997 was negative …"
Brazil made two comments on the negative value of de Havilland's equity in January 1997. First, Brazil stated that "even if de Havilland's net equity value was negative at the time of its sale in 1997, it does not change the fact that it would have been even more negative if government contributions, the present value of which constitute $875 million, had not been made to de Havilland between 1992 and 1997. These contributions increased the value of Ontario's equity beyond the price set in 1992, but Ontario was given no share in this increase."587
Second, Brazil states that "a negative equity value does not give the complete picture of the value of a shareholder's equity. It does not mean that de Havilland was worth nothing - or less than nothing - to Ontario and Bombardier when purchased in 1997. De Havilland's book of orders at the time of sale, for example, although not factored into net equity value, represented value which an equity investor selling its shares could reasonably expect to be captured in the selling price of its shares." 588
With regard to Brazil's first comment, Canada responds that "[w]hether de Havilland's net equity value would have been more negative without the contributions is immaterial: Ontario's sale price was still well in excess of what the equity was worth. Hence, there is no benefit conferred by the sale." We agree with Canada's response.
Canada makes three observations with regard to Brazil's comment on the value of de Havilland's order book. First, Canada notes that "Brazil has adduced no evidence as to what the value of the de Havilland order book was on the date of sale, let alone demonstrated that the value would have been sufficient to compensate the purchaser for the negative net equity value of de Havilland at the time. Thus, Brazil has not demonstrated that a benefit was conferred by the sale. Second, any such valuation would have to reflect the possibility that orders may not result in actual sales … Third, an order book does not indicate the viability of an aircraft manufacturer: Fokker’s order book just prior to its bankruptcy in early 1996 contained firm orders for 81 aircraft (… source: Lundkvist)." We concur with these observations.
In light of the information concerning the value of de Havilland's equity at the time of sale in January 1997, we asked Canada to "provide the audited statements of de Havilland that form the basis of this statement, along with all accompanying notes and any other relevant documentation." Canada replied:
Canada notes that the information requested by the Panel – detailed audited statements of de Havilland along with all accompanying notes and any other relevant documentation - is highly sensitive business confidential information that is not in the possession of the Government of Canada or the Government of Ontario. Canada’s desire to present to the Panel such information as may help it to arrive at a decision must be balanced against the concerns of private parties not Party to this dispute.
Canada further notes that Brazil has adduced no evidence whatever in support of its assertions about the increase in de Havilland’s value as a result of alleged subsidies between 1992 and 1997. As well, Canada has shown, in respect of Brazil’s
587 See para. 6.284 above.
588 See para. 6.285 above.
WT/DS70/R
Page 201
unfounded and incorrect allegations about the alleged forgiveness of the $49 million promissory note, that Brazil’s assertions about de Havilland suffer from a fundamental lack of credibility. Brazil’s allegations about the rise in value of de Havilland should be viewed, in Canada’s respectful submission, in that light.
Finally, Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Canada notes that Brazil is in agreement with Canada regarding the relevance of the principle of judicial economy to the issues to be determined in this case.
Canada does not consider it appropriate to adduce evidence in response to baseless and false allegations and a case that has not been established, and in support of a defence Canada has not made.
In comments on Canada's response, Brazil stated:
Although Canada claims that the de Havilland audited financial statements used to evaluate the company in January 1997 are “not in the possession of the Government of Canada or the Government of Ontario,” Brazil finds it inconceivable that Ontario, which at the time owned 49 percent of de Havilland, would not have been furnished a copy of de Havilland’s financial statements.
It is Canada, and not Brazil, that “suffer[s] from [the] fundamental lack of credibility” noted in Canada’s reply. First, Canada has failed to supply the background documents supporting the statements in the letter from Mr. Desjardins submitted as evidence in support of its defense. Second, Canada has failed effectively to rebut evidence submitted by Brazil from the Public Accounts of Ontario detailing contributions, the present value of which constitute $875 million, made by the Canadian federal and Ontario governments to de Havilland between 1992, when the
$49 million purchase price for Ontario’s share in de Havilland was set, and 1997, when Bombardier completed the purchase of Ontario’s share. Canada’s argument that this massive injection of capital did not effect the value of Ontario’s equity is the only claim lacking credibility.
We are surprised that Ontario was not furnished with a copy of de Havilland’s financial statements at the time of the 1997 sale. We also regret that Canada chose not to provide us with copies of the audited statements on which Canada's statement concerning the January 1997 value of de Havilland was based. However, in the absence of any evidence to the contrary, we have no basis for rejecting Canada's assertion, based on a signed statement from a de Havilland executive, that de Havilland had a negative equity value at the time of the January 1997 sale.
We note that, according to Brazil, "Canada has failed effectively to rebut evidence submitted by Brazil from the Public Accounts of Ontario detailing contributions, the present value of which constitute $875 million, made by the Canadian federal and Ontario governments to de Havilland between 1992, when the $49 million purchase price for Ontario’s share in de Havilland was set, and 1997, when Bombardier completed the purchase of Ontario’s share." We recall, however, that "Brazil does not consider that [such contributions] were themselves subsidies contingent upon export performance."589 Accordingly, Canada's alleged failure to effectively rebut evidence adduced by Brazil with regard to such contributions is not relevant to Brazil's claim.
We recall that Brazil's claim was premised on an increase in the value of de Havilland's equity between 1992 and 1997. In light of the unrebutted evidence of a decrease in the value of OAC's 49
589 See para. 6.273 above.
WT/DS70/R
Page 202
percent equity share in de Havilland between 1992 and 1997, we consider that there is no factual basis to Brazil's claim of subsidization in the context of the sale of OAC's 49 percent share in de Havilland to Bombardier in 1997. Indeed, as a matter of logic, we cannot accept that an Article 1.1 "benefit" is conferred when a purchaser is required to pay $49 million for an equity share valued in the negative. It makes no difference to this analysis that the amount of the deficit is less than it might have been absent alleged government contributions. For these reasons, we find that there is no prima facie case that the sale by the OAC of its 49 percent share in de Havilland to Bombardier in 1997 constitutes an export subsidy contrary to Article 3.1(a) and 3.2 of the SCM Agreement, and we reject Brazil's claim accordingly.
BENEFITS PROVIDED UNDER THE CANADA-QUEBEC SUBSIDIARY AGREEMENTS ON INDUSTRIAL DEVELOPMENT
Arguments of the parties
Brazil claims that benefits provided to the regional aircraft industry590 under the Canada- Québec Subsidiary Agreements on Industrial Development constitute prohibited export subsidies contrary to Article 3.1(a) and 3.2 of the SCM Agreement.
Brazil asserts that under the Subsidiary Agreements, the Governments of Canada and Québec jointly fund major industrial projects aimed at improving the competitiveness of Québec’s economy. Brazil argues that Subsidiary Agreement assistance has "as an historical matter" been granted to the Canadian regional aircraft industry on terms described as "conditionally repayable". Brazil asserts that assistance under the Subsidiary Agreements may be in the form of repayable or non-repayable contributions. Brazil states that both types of contribution confer a "benefit" within the meaning of Article 1.1(b) of the SCM Agreement. Brazil states that conditional repayment confers a benefit "because the recipient has no down-side risk; if the project is unsuccessful, the contribution need not be repaid."591 Even if the loan is repaid, Brazil asserts that a "benefit" is conferred if "the rate of return is such that the lender is not compensated for either the risk that it would have received no payment or the extended repayment period during which neither principal nor interest is due."592 Brazil notes that Canada notified the 1988 Memorandum of Understanding that provided the authority for the 1992 Subsidiary Agreement to the WTO SCM Committee. In its notification, Canada informed the Committee that "[a]ssistance was provided in the form of repayable or non-repayable contributions."593
According to Brazil, the purpose of the Subsidiary Agreements is explicitly to foster development of export markets for Québec products by “[p]romot[ing] access to new foreign as well as domestic markets.”594 Brazil argues that financing provided under the auspices of the Subsidiary Agreements is often explicitly directed at export-oriented projects and industries. Brazil cites a study prepared by the Canadian Reform Party which reports that the regional aircraft sector has been a significant recipient of benefits under the Subsidiary Agreements. According to Brazil, Subsidiary Agreement funds disbursed to the Canadian regional aircraft industry were granted to an industry that is totally export oriented, precisely because it is an export industry and was anticipated to remain so.
590 We note that Brazil's request for establishment of a panel (WT/DS70/2) refers to various alleged prohibited export subsidies provided to the "Canadian industry producing civil aircraft". Our terms of reference, therefore, relate to various forms of assistance provided to the Canadian civil aircraft industry. With regard to Subsidiary Agreement assistance, however, arguments adduced by Brazil concerning the issue of export contingency are restricted to the "regional aircraft industry". For this reason, we only examine Brazil's claim against Subsidiary Agreement assistance insofar as it is provided to the Canadian regional aircraft industry.
591 See para. 6.180 above.
592 See para. 6.180 above.
593 WTO document G/SCM/N/16/CAN, 25 June 1997, para. XXIX:5.
594 1985 Canada-Québec Subsidiary Agreement on Industrial Development, Art. 2.2(d). Id. at Schedule A, para. (d).
WT/DS70/R
Page 203
For these reasons, Brazil asserts that the Subsidiary Agreement as applied, but not per se, grants prohibited export subsidies to the regional aircraft sector, contrary to Article 3 of the SCM Agreement. In support of its claim, Brazil has adduced evidence with regard to a number of alleged instances of Subsidiary Agreement assistance. Brazil has referred to press releases concerning the grant of a $2.5 million repayable loan to Rolls Royce Canada Ltd., and a $4.6 million repayable loan to Lamines CTEK. However, in response to a question from the Panel, Brazil stated that it has not claimed that the assistance described in these press releases constitutes a subsidy within the meaning of Article 1 of the SCM Agreement, "among other reasons because this assistance is not [granted] to the regional aircraft industry."
Brazil also submitted a study prepared by the Canadian Reform Party which identifies five instances of assistance allegedly provided to Bombardier under the Subsidiary Agreement programmes. Brazil refers to this study to demonstrate that, "as an historical matter",595 Subsidiary Agreement assistance has been granted to the Canadian regional aircraft industry "on terms described as 'conditionally repayable'".
The final Subsidiary Agreement transaction identified by Brazil concerns a payment of
$6,586,547 to Bombardier Canadair Group Montreal Que, identified in the 1996-1997 Public Accounts of Canada. This payment is presented by Brazil as proof that "Subsidiary Agreement assistance has gone to the Canadian regional aircraft industry".
While not conceding this issue, Canada notes that it has not put in a defence on whether Subsidiary Agreement assistance to the Canadian regional aircraft industry takes the form of "subsidies" within the meaning of Article 1 of the SCM Agreement. Canada argues primarily that Subsidiary Agreement assistance is not contingent on export. In particular, Canada notes that contributions under the Subsidiary Agreement promote access “to new foreign as well as domestic markets."
On the basis of information in the record, we asked Canada to submit details of the Subsidiary Agreement contributions to Rolls Royce and Lamines CTEK identified by Brazil. Canada replied:
Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Canada notes that Brazil agrees with Canada regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Accordingly, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to further support Canada’s submission that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
Canada notes two additional points. First, Brazil has made no specific allegation about contributions to Rolls Royce or Lamines CTEK. Neither has it has adduced any evidence whatever in support of its vague and unspecific allegations concerning “benefits” under the Subsidiary Agreement. Canada does not consider it appropriate to adduce evidence in response to allegations not made and a case that has not been established. Second, As Canada noted in paragraph 99 of its Second Oral Submission, the contribution to Lamines CTEK was in respect of certain electronic materials not related to the civil aircraft sector. As such, it is not within the Panel’s jurisdiction.
Finally, most of the information requested by the Panel is sensitive business confidential information. Canada’s desire to present to the Panel such information as
595 See para. 6.300 above.
WT/DS70/R
Page 204
may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute.
Accordingly, Canada, with the consent of the Government of Quebec, has provided the Rolls Royce contribution agreement, as Business Confidential Information, with clauses relating to repayment terms blocked (see BCI Tab 4).
In commenting on Canada's refusal to provide this information in full, Brazil stated:
It appears that Canada has not provided in BCI Tab 4 all of the information requested by the Panel concerning Subsidiary Agreement assistance. First, the repayment terms are redacted from the agreement setting out the terms of the $2.5 million contribution to Rolls Royce. The Panel should therefore adopt adverse inferences, presuming that these terms tie repayment to export performance. Furthermore, Brazil notes that while Canada has provided the Rolls Royce contribution agreement, it appears to have provided as “Annex A” thereto the project description for the Lamines CTEK contribution, rather than the project description associated with the Rolls Royce contribution. The Panel should adopt adverse inferences, presuming from Canada’s failure to produce the annex containing the project description for the Rolls Royce contribution that this description contains inculpatory information suggesting that the contribution is tied in fact to actual or anticipated export.596
Evaluation by the Panel
We shall commence our examination of Brazil's claim by determining whether there is a prima facie case that Subsidiary Agreement assistance to the Canadian regional aircraft industry has taken the form of "subsidies" within the meaning of Article 1 of the SCM Agreement. Only if we reach an affirmative conclusion in this regard will we consider whether such assistance is "contingent
… upon … export performance" within the meaning of Article 3.1(a).
Brazil initially argued that Subsidiary Agreement assistance could be provided in the form of non-repayable contributions. In support, Brazil relied on Canada's notification to the WTO SCM Committee, made pursuant to Article 25.2 of the SCM Agreement. We note that, however, that Article 25.7 provides:
Members recognize that notification of a measure does not prejudge either its legal status under GATT 1994 and this Agreement, the effects under this Agreement, or the nature of the measure itself.
In our view, Article 25.7 of the SCM Agreement effectively precludes us from finding a prima facie case that Subsidiary Agreement assistance is provided in the form of non-repayable contributions simply on the basis of Canada's notification of that programme.
Brazil also submitted a study prepared by the Canadian Reform Party. This study refers to five instances of alleged Subsidiary Agreement assistance. The only instance of Subsidiary Agreement defined as "non-repayable" relates to a 1989 project for "Snowmobile Chassis made out of composite materials". This study therefore lends no support to Brazil's claim concerning non- repayable Subsidiary Agreement contributions to the Canadian regional aircraft industry.
596 The Panel notes that Canada subsequently corrected its administrative error and provided Annex A to the Rolls Royce contribution agreement on 22 January 1999.
WT/DS70/R
Page 205
Brazil has adduced no other evidence to demonstrate that Subsidiary Agreement assistance generally is provided in the form of non-repayable contributions, or that non-repayable contributions are provided to the Canadian regional aircraft industry in particular.
We note that Brazil has conceded that the two Subsidiary Agreement contributions to Rolls Royce and Lamines CTEK did not concern the Canadian regional aircraft industry. Accordingly, we shall not take these contributions into account when reviewing Brazil's claim against Subsidiary Agreement assistance to the Canadian regional aircraft industry.
Brazil also relies on the aforementioned Canadian Reform Party study to demonstrate that, "as an historical matter",597 Subsidiary Agreement assistance has been granted to the Canadian regional aircraft industry "on terms described as 'conditionally repayable'".598 Brazil asserts that "conditionally repayable" contributions confer a "benefit" if the rate of return is such that the lender is not compensated for either the risk that it would have received no payment or the extended repayment period during which neither principal nor interest is due. We note that the study identifies five instances of alleged Subsidiary Agreement assistance. However, Brazil fails to demonstrate that any of the five instances of alleged Subsidiary Agreement assistance concerns the regional aircraft industry. Since four of the alleged instances of Subsidiary Agreement assistance identified in the study are described as relating to projects concerning "Snowmobile Chassis made out of composite materials", "Continued development of VENUS I, "equipment for centre of excellence", and "Equipment for centre of excellence", and since none of the alleged Subsidiary Agreement assistance is expressly labelled as relating to regional aircraft projects, we are not convinced that any of the Subsidiary Agreement assistance referred to in the study concerns Canadian regional aircraft. Furthermore, none of the five instances of Subsidiary Agreement assistance identified in the study is described as being "conditionally repayable", as alleged by Brazil. For these reasons, we consider that this study provides no support to Brazil's claim that Subsidiary Agreement assistance is provided to the regional aircraft industry on "conditionally repayable" terms that confer a "benefit" within the meaning of Article 1.1(b) of the SCM Agreement.
We note that Brazil also refers to a Subsidiary Agreement contribution to Bombardier Canadair Group Montreal Que, identified in the 1996-1997 Public Accounts of Canada, as proof that Subsidiary Agreement assistance has gone to the Canadian regional aircraft industry. We agree with Brazil that this contribution does appear to have been provided to a Canadian manufacturer of regional aircraft. However, Brazil fails to provide any evidence, or even to argue, that the relevant contribution confers a "benefit" within the meaning of Article 1.1(b) of the SCM Agreement. As noted above, Brazil claims that Subsidiary Agreement assistance confers a "benefit" if it is either non- repayable, or if it is conditionally repayable and the rate of return is such that the lender is not compensated for either the risk that it would have received no payment or the extended repayment period during which neither principal nor interest is due. However (assuming arguendo that Brazil's analysis of "benefit" in such circumstances is accurate), Brazil makes no attempt to establish that this particular contribution is either non-repayable, or that it is conditionally repayable and the rate of return is such that the lender is not compensated for either the risk that it would have received no payment or the extended repayment period during which neither principal nor interest is due. In such circumstances, the evidence adduced by Brazil concerning the Subsidiary Agreement contribution to Bombardier provides no support for Brazil's claim that Subsidiary Agreement assistance to the Canadian regional aircraft industry takes the form of "subsidies".
For the above reasons, we find that there is no prima facie case in support of Brazil's claim that Subsidiary Agreement assistance to the Canadian regional aircraft industry has taken the form of subsidies within the meaning of Article 1 of the SCM Agreement.
597 See para. 6.300 above.
598 See para. 6.298 above.
WT/DS70/R
Page 206
In light of the above finding, it is not necessary for us to consider whether Subsidiary Agreement assistance to the regional aircraft industry is contingent on export. We recall that Brazil asked the Panel to adopt "adverse inferences" that the Subsidiary Agreement contributions to Rolls Royce and Lamines CTEK were de facto export contingent. Since it is not necessary for us to examine the issue of export contingency, we decline to consider Brazil's request for "adverse inferences" on this issue.
Accordingly, we reject Brazil's claim that Subsidiary Agreement assistance to the Canadian regional aircraft industry takes the form of prohibited export subsidies, contrary to Article 3.1(a) and
3.2 of the SCM Agreement.
SOCIETE DE DEVELOPPEMENT INDUSTRIEL DU QUEBEC
Arguments by the parties
Brazil submits that SDI assistance has been provided to the Canadian regional aircraft industry599 in the form of prohibited export subsidies, contrary to Articles 3.1(a) and 3.2 of the SCM Agreement. Brazil asserts that, under the Société de Développement Industriel du Québec (“SDI”), the Government of Quebec sponsors export programs "designed to promote the export of Quebec- produced goods and services.”600 Referring to the SDI Act, Brazil states that SDI was formed with
the objective “to promote economic development in Québec, particularly by encouraging the development of businesses, the growth of exports, research and the development of new techniques.”601 SDI provides Quebec businesses with “guarantee[s] of payment or repayment of a financial obligation” and “loan[s] at the current market rate”602 for the purpose of assisting those businesses with “[l]a conquète de marchés à l’exportation,”603 and to assist with “tout projet d’exportation” or “de démarrage à l’exportation.”604 Brazil asserts that SDI not only funds exports to other of the Canadian Provinces, but also specifically supports transactions “à l’étranger".
Brazil asserts that SDI funding is extended on conditionally repayable terms that confer a "benefit" within the meaning of Article 1.1 of the SCM Agreement "because the recipient has no down-side risk -- that is to say, if the project is unsuccessful, SDI funds need not be repaid …"605 Brazil claims that SDI funding is "contingent …in fact … upon export performance" within the meaning of Article 3.1(a) because (1) one of the objectives of the SDI is the "growth of exports" and
when SDI funds were disbursed to the Canadian regional aircraft industry, they were given to an industry that is overwhelmingly export oriented, precisely because it is an export industry and was anticipated to remain so.
Brazil notes that SDI was recently “regrouped” for administrative purposes into a new corporation, known as Investissement-Québec (“IQ”).606 On the basis of materials from the SDI
599 We note that Brazil's request for establishment of a panel (WT/DS70/2) refers to various alleged prohibited export subsidies provided to the "Canadian industry producing civil aircraft". Our terms of reference, therefore, relate to various forms of assistance provided to the Canadian civil aircraft industry. With regard to SDI assistance, however, arguments adduced by Brazil concerning the issue of export contingency are restricted to the "regional aircraft industry". For this reason, we only examine Brazil's claim against SDI assistance insofar as it is provided to the Canadian regional aircraft industry.
600 Government Assistance Programs - A Practical Handbook (15th ed., 1995), para. 135.230.
601 An Act Respecting the Société de Développement Industriel du Québec, para. 2.
602 Id. at para. 135.240.
603 SDI Website at 2.
604 Gouvernement du Québec, Ministère de L’Industrie, du Commerce, de la Science et de la Technologie, Répertoire des services offerts à l’exportation, page 53.
605 See para. 6.323 above.
606 Materials from SDI/Investissement-Québec website, at 1 (Exhibit 65).
WT/DS70/R
Page 207
website, Brazil states that IQ maintains the resources, including the staff, of SDI.607 Brazil notes that, according to the WTO November 1998 Trade Policy Review of Canada, the IQ programme “provides export guarantees for projects considered too risky by private financial institutions.”608 The Trade Policy Review also states that IQ assistance “is available for export development or expansion, or purchases of foreign companies, or for contract finance.”609 Brazil asserts that IQ confers an Article
1.1 "benefit" by offering guarantees for projects which are “considered too risky by private financial institutions,” because it provides resources which would otherwise not be available.
Brazil understands Canada to argue that SDI activities are "immunized" from the Article 3 prohibition because SDI also provides funding to domestic projects. Brazil does not contest that some SDI or IQ funds may have gone to other industries with sales in domestic markets. However, Brazil claims that "when SDI or IQ funds were disbursed to the Canadian regional aircraft industry, they were given to an industry that is overwhelmingly export oriented, precisely because it is an export industry and was anticipated to remain so."
Canada claims that Brazil's allegations about the SDI are unfounded and incorrect. While expressly denying that SDI assistance in the civil aircraft sector takes the form of "subsidies" within the meaning of Article 1.1 of the SCM Agreement, Canada states that it has not put in a defence regarding whether such SDI assistance constitutes "subsidies". Rather, Canada argued primarily that SDI assistance in the civil aircraft sector is not in law or in fact contingent upon export performance. Canada notes that, according to Article 2 of the SDI Act, the objective of the SDI is to "favoriser le développement économique du Québec, notamment en encouragent le développement des entreprises, la croissance des exportations et les activités de recherche et d’innovation." Canada claims that Brazil itself acknowledges that "exportation" in this context means exports outside Québec, including the other provinces of Canada. Canada submits that SDI has as a general objective the enhancement of Québec’s competitiveness and, as a desirable but not necessary result, an increase in Québec’s exports. Canada asserts that export performance is not a criterion of success for either the programme as a whole or its activities;610 nor is export performance a condition for receiving SDI contributions. Canada asserts that 458 SDI activities were authorised in 1997-1998, of which 53 percent involved loans or loan guarantees of less than C$200,000. Canada states that SDI comprises four sets of eligibility criteria, one of which concerns "export development". According to Canada, "export development" includes credit insurance, financing for foreign direct investment by Québec companies, and inventory financing. Canada asserts that one out of 39 financing activities and 96 out of 419 guarantees authorised were carried out under the exportation criterion.611 Canada claims that none of these related to the civil aircraft sector.
Canada also asks the Panel to ignore Brazil's "inappropriate use" of the WTO's November 1998 Trade Policy Review of Canada. Canada asserts that section A(i) of the WTO Trade Policy Review Mechanism provides that the review mechanism "is not, however, intended to serve as a basis for the enforcement of specific obligations under the Agreements or for dispute settlement procedures
…."
Canada confirms that it does consider that the fact that SDI funds are available to both domestic and export markets does immunise them to the prohibition of Article 3.1(a). According to Canada, Article 3.1(a) does not prohibit subsidies that are aimed at both domestic and export markets.
On the basis of information in the record, we asked Canada to provide details of three Aerospace Industry Development Fund contributions, and five Private Investment and Job Creation
607 Id. at 1, 3.
608 WTO document WT/TPR/S/53 (19 November 1998), at page 59.
609 Id.
610 Société de développement industriel du Québec, 1997-1998 Annual Report, at 36-37, 40.
611Id. at 22, Table 1.
WT/DS70/R
Page 208
Promotion Fund ("FAIRE") transactions, all identified in the SDI Annual Report 1997-1998. Canada refused to provide this information in full. Canada asserted that it had not put in a defence regarding whether the Aerospace Industry Development Fund contributions are "subsidies", and that Brazil had not made a specific allegation against these contributions. Canada submitted some of the information requested by the Panel, but redacted information concerning repayment terms relevant to determining whether or not the relevant contributions constitute "subsidies". With regard to FAIRE, Canada asserted that the five activities were not in the civil aircraft sector. In commenting on Canada's failure to respond to the Panel in full in respect of the Aerospace Industry Development Fund contributions, Brazil asked the Panel to adopt "adverse inferences" that these contributions are de facto contingent on export.
Evaluation by the Panel
Brazil challenges assistance provided to the Canadian regional aircraft industry under the SDI and IQ programmes. We shall examine both types of assistance in turn, considering first whether such assistance takes the form of "subsidies" within the meaning of Article 1 of the SCM Agreement. Only if we find subsidization will we consider whether such assistance is "contingent … upon … export performance", within the meaning of Article 3.1(a) of the SCM Agreement.
Assistance provided under the IQ programme
Brazil refers to the WTO November 1998 Trade Policy Review of Canada to argue that IQ assistance to the regional aircraft industry confers a "benefit" by "provid[ing] export guarantees for projects considered too risky by private financial institutions". We note that section A(i) of the WTO Trade Policy Review Mechanism provides that the review mechanism "is not, however, intended to serve as a basis for the enforcement of specific obligations under the Agreements or for dispute settlement procedures …." Accordingly, we attach no importance to the Trade Policy Review of Canada in considering Brazil's arguments concerning IQ assistance to the regional aircraft industry.
Brazil has failed to adduce any evidence of IQ assistance to the Canadian regional aircraft sector. Accordingly, there is no basis for a prima facie case that IQ assistance has been provided to the regional aircraft industry in the form of export subsidies, contrary to Articles 3.1(a) and 3.2 of the SCM Agreement.
Assistance provided under the SDI programme
In its first submission, Brazil included a claim concerning a joint SDI / DIPP conditionally- repayable contribution of $43 million made in April 1989 for the development of Bombardier's 50- seat Canadair Regional Jet ("CRJ"). However, Brazil subsequently withdrew this claim on the grounds that the contribution is not subject to the SCM Agreement because, in its view, the SCM Agreement does not apply to allegedly prohibited export subsidies granted prior to 1 January 1995.612 In these circumstances, we will not consider this contribution in resolving Brazil's claim concerning SDI assistance to the Canadian regional aircraft industry.
The SDI Annual Report 1997-1998 refers to three contributions under the Aerospace Industry Development Fund. We consider that these contributions are covered by Brazil's claim, since its claim effectively covers all SDI assistance to the regional aircraft industry.613 We note that these contributions are described in the SDI Annual Report as "loans and shares". In its first submission, Brazil expressly cites SDI documentation that refers to "loan[s] at the market rate". Since we consider that a "loan at the market rate" would not confer a "benefit" within the meaning of Article 1.1(b) of
612 See para. 9.43 above.
613 WTO document WT/DS70/2.
WT/DS70/R
Page 209
the SCM Agreement,614 these three Aerospace Industry Development Fund contributions provide no support for Brazil's claim against SDI assistance to the Canadian regional aircraft industry.
The SDI Annual Report 1997-1998 also refers to five FAIRE transactions. However, Canada has asserted that these transactions did not relate to the civil aircraft industry. As Brazil does not contest this assertion, these transactions provide no support for Brazil's claim against SDI assistance to the Canadian regional aircraft industry.
Brazil has adduced no further evidence in support of its claim against SDI assistance to the regional aircraft industry. Accordingly, there is no prima facie case that SDI assistance has been provided to the regional aircraft industry in the form of subsidies within the meaning of Article 1 of the SCM Agreement
For the above reasons, we reject Brazil's claims that SDI and IQ assistance to the Canadian regional aircraft industry takes the form of prohibited export subsidies, contrary to Articles 3.1(a) and
3.2 of the SCM Agreement.
We recall that Brazil asked the Panel to adopt "adverse inferences" that the SDI contributions under the Aerospace Industry Development Fund were de facto export contingent. In light of the above finding, it is not necessary for us to consider whether SDI assistance to the regional aircraft industry is contingent on export. Accordingly, we decline to consider Brazil's request for "adverse inferences" on this issue.
I. TECHNOLOGY PARTNERSHIPS CANADA AND DIPP
Brazil asserts that prohibited export subsidies have been provided to the Canadian regional aircraft industry615 under the TPC programme and its predecessor programme, DIPP,616 contrary to Articles 3.1(a) and 3.2 of the SCM Agreement. Brazil is not challenging the TPC or DIPP programmes per se. Rather, Brazil is challenging the actual application of the TPC and DIPP programmes in the Canadian regional aircraft sector. We will first consider Brazil's claim against TPC assistance in the regional aircraft sector, and then Brazil's claim against the DIPP as applied. Furthermore, in reviewing the TPC programme as applied, we shall first address Brazil's arguments concerning subsidization. Only if we find in favour of Brazil on this issue will we consider whether TPC assistance to the regional aircraft industry is contingent on export.
1. Does the application of the TPC programme result in subsidies for the Canadian regional aircraft industry?
Arguments of the parties
Brazil states that, according to TPC documentation, TPC was created in 1996 "to address the need by established companies in specific industrial segments to ensure that near-market products -- those with a high potential to stimulate economic growth and job creation -- actually reach the marketplace.” Brazil notes that the specific sectors eligible for assistance from TPC are
614 We recall our interpretation of the meaning of "benefit" at para. 9.120 above.
615 We note that Brazil's request for establishment of a panel (WT/DS70/2) refers to various alleged prohibited export subsidies provided to the "Canadian industry producing civil aircraft". Our terms of reference, therefore, relate to assistance provided to the Canadian civil aircraft industry. With the exception of its first written submission, however, all findings requested by Brazil concerning TPC refer to the "regional aircraft
industry". Similarly, all arguments and evidence adduced by Brazil concerns the regional aircraft industry. For this reason, we only examine Brazil's claim against TPC assistance insofar as such assistance relates to the Canadian regional aircraft industry.
616 In response to a question from the Panel, Brazil confirmed that the phrase "predecessor programs" in its request for establishment is restricted to the DIPP.
WT/DS70/R
Page 210
environmental technologies, enabling technologies, and aerospace and defence industries. Brazil submits that TPC’s predecessor was the Defense Industry Productivity Program (“DIPP”). Brazil cites to a press report that suggests that DIPP “had been used primarily by aerospace companies.” Brazil asserts TPC funding “is intended to cover outstanding commitments under the DIPP."
According to Brazil, TPC explicitly targets conditionally repayable investments in projects that result in a high technology product for sale in export markets. Brazil submits that such investments are conditionally repayable on a royalty basis, meaning that repayment will only occur if the underlying project achieves a certain degree of profitability. Brazil asserts that the provision of funds on such terms confers a "benefit" within the meaning of Article 1.1 of the SCM Agreement because the recipient has no down-side risk. If the project is unsuccessful, TPC and DIPP loans need not be repaid. Brazil also submits that, even if the Canadian government recovers its investment, its anticipated rate of return is well below that expected by a reasonable market investor, and does not compensate the lender for either the risk that it would have received no payment or the extended repayment period during which neither principal nor interest are due. According to Brazil, a reasonable market investor would expect a rate of return of 16.91 - 21.92 percent for airframe development expenses. Brazil also refers to the November 1998 WTO Trade Policy Review of Canada, which identifies TPC as a "subsidy" in the form of "grants".
Brazil has adduced evidence with regard to a number of specific alleged TPC contributions. In particular, Brazil has adduced evidence with regard to: a $87 million contribution to Bombardier for the development of the CRJ-700 (1996); a $57 million contribution to de Havilland for the development of the Dash 8-400 (1996); a $100 million contribution to Pratt & Whitney for the development of the PW150 turboprop engine used in the Dash 8(1997); a portion of a $12.7 million contribution to Allied Signal that concerns the development of power management generating system for the Dash 8-400 (1997); and a $9.9 million contribution to Sextant for the development of avionics systems for the Dash 8 and a flight control system for the CRJ (1998).
$87 million to Bombardier
Brazil challenges a 1996 $87 million TPC contribution granted to Bombardier to assist the development of Bombardier's 70-seat CRJ project, known as the CRJ-700. Brazil asserts that this contribution is conditionally repayable on a royalty basis. Brazil asserts that TPC would only receive a return on its investment upon the sale of the 401st aircraft. Assuming a royalty rate of $580,000 per aircraft, with 25 aircraft sold annually, Brazil submits that TPC's expected rate of return on this loan is
1.76 percent. Brazil submits that a commercial investor would expect to make a return of 16.91 -
21.92 percent for such an investment. Brazil asserts that an Article 1.1 "benefit" is conferred because the contribution was made to Bombardier on terms more favourable than those available to Bombardier from a private investor.
$57 million TPC contribution to de Havilland; and $100 million TPC contribution to Pratt & Whitney
Brazil also claims that a 1996 TPC $57 million contribution to de Havilland to develop a 70- seat "stretch" version of the Dash 8 is inconsistent with Articles 3.1(a) and 3.2 of the SCM Agreement. Brazil estimates that this contribution would provide TPC a return of 3.02 percent. A similar claim is advanced by Brazil against a $100 million TPC contribution to Pratt & Whitney Canada, for work on the firm's 6,500 SHP PW150 turboprop engine. Brazil estimates a 3.31 percent return for TPC on this contribution. Brazil asserts that the rates of return for both contributions are "well below that which would be expected by a market investor", i.e., 16.91 - 21.92 percent, and therefore confer a benefit.
WT/DS70/R
Page 211
$12.7 million TPC contribution to Allied Signal; and $9.9 million TPC contribution to Sextant Avionique
Brazil also made specific claims against a 1997 $12.7 million TPC contribution to Allied Signal, a portion of which is for the development of the power management generating system for the Dash 8-400, and a 1998 $9.9 million contribution to Sextant Avionique Canada Inc. (hereinafter "Sextant") to be used for the development of the avionics system for the Dash 8-400 and the flight control system for the CRJ-700. However, Brazil stated that it was unable to analyse the precise rates of TPC return on these contributions because of insufficient information in the public project announcements.
While expressly denying that TPC transactions constitute subsidies within the meaning of Article 1 of the SCM Agreement, Canada did not put in a defence regarding whether TPC contributions are subsidies. Canada's argues primarily that TPC is not contingent upon export performance. However, Canada made a number of comments concerning the terms on which TPC assistance is provided. Canada states that TPC contributions usually constitute a portion - on average less than 30 per cent - of the eligible costs of a project incurred by the recipient. Canada asserts that TPC contributions are repayable, generally on a royalty basis tied to the market success of a project. Canada considers that TPC contributions constitute investments rather than loans, since TPC shares in the rewards of successful projects, and TPC returns are not limited to a fixed amount. Canada asserted that TPC royalty-based repayment is not tied to profits, but to sales. Royalties can either be tied to sales of a specific product to be developed under the project, to sales of a family of products to be developed under a project, or to total sales of the recipient enterprise.
Canada objects to Brazil's reliance on the November 1998 WTO Trade Policy Review of Canada. Canada asserts that the Trade Policy Review was conducted pursuant to the WTO Trade Policy Review Mechanism, section A(i) of which provides that the review mechanism "is not, however, intended to serve as a basis for the enforcement of specific obligations under the Agreements or for dispute settlement procedures …."
Canada confirmed that each of the specific contributions identified by Brazil were made under the TPC programme.
Additional information on the specific TPC contributions identified by Brazil
$87 million to Bombardier
On the basis of the information adduced by Brazil concerning the $87 million TPC contribution to Bombardier, we asked Canada to provide inter alia: complete details of, and to comment on, the terms and conditions of the $87 million Bombardier contribution; copies of all application documents (including any pre-screening submissions, application forms, supporting business plans, sales and/or export projections including “expected clients, units, prices and timing”, repayment plan of the applicant, etc.); and assessments and funding decisions by TPC pertaining to that contribution. We also asked Canada to specify commercial borrowing rates in Canada for debt of comparable size, risk level and maturity prevailing at the time the contribution was made by TPC. Furthermore, we asked Canada to comment on the estimated rate of return calculated by Brazil for the
$87 million Bombardier contribution.
In its response, Canada stated:
Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Brazil has explicitly indicated its agreement with Canada in its letter of 13 December 1998, regarding the relevance of the principle of judicial economy to the issues to be
WT/DS70/R
Page 212
determined in this case. Accordingly, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to support Canada’s submissions that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
Most of the information requested by the Panel is highly sensitive business confidential information. Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. These private parties, and others in the process of submitting applications under the TPC programme, have already expressed reluctance in sharing information, or additional information, concerning their business plans. Such reluctance, if it were to continue, would have a serious deleterious impact on the functioning of the TPC programme.
[…]
Bombardier agreed to release specific documents relating to the CRJ-700 programme that illustrate the operation of TPC. Exceptionally sensitive confidential information in those documents was redacted to protect the commercial interests of Bombardier.
Regarding our request for internal TPC project assessments for the $87 million contribution, Canada stated:
Canada is unable to comply to the Panel’s request for all project assessments and funding decisions documents. Since the level of contribution was in excess of $20 million, Cabinet approval was required for this investment. The recommendation, options, communications strategy, supporting rationale and analysis are contained in a Memoranda to Cabinet (MC) and as such constitutes a cabinet confidence and cannot be divulged. Similarly, the Project Summary Form (PSF) in this case was required to be presented to the Minister of Industry Canada for signature, and therefore constitutes Ministerial advice that cannot be released.
In an effort to comply to the greatest extent possible with the Panel’s request, Canada will briefly summarize the basis for its investment in the CRJ-700 project. The key considerations were as follows:
the timing of the project coincided with planned Canadair workforce reductions and would create or maintain 1,000 jobs during the development phase alone.
the CRJ-700 was planned to advance the state-of-the-art in regional jet operating performance and would provide tangible benefits to operators in Canada and abroad.
the project would offer real opportunities for participation by domestic aerospace sub- contractors to expand their sales base and enhance their technical capabilities.
project risk was considered manageable in light of Bombardier’s proven track record and because the technical risk was mitigated by the fact this aircraft was a stretch version of an existing platform - the CRJ50.
market risk was also manageable given the CRJ50's strong position in the regional jet market, and the preference by airlines to utilize families of aircraft to reduce part inventories and training requirements.
WT/DS70/R
Page 213
independent market forecasts and Industry Canada’s sectoral experts confirmed that the size of the potential market for a 60-90 seat regional jet to be in excess of 1,000 units through 2010.
With regard to our request for commercial borrowing rates, Canada responded that:
the relevant consideration for determining the rate of return for the TPC contributions in question was the Government of Canada’s cost of funds. At the time these transactions were entered into, Canada’s long term bonds were yielding approximately six to seven percent […]. TPC repayment terms are negotiated so that on a net present value basis they are cost-revenue neutral or better […].
In response to our request for comments on the rate of return estimated by Brazil, Canada submitted that a number of assumptions supporting Brazil's estimation "have no basis in fact". Canada also asserts that the methodology of Brazil's estimation of TPC's return is "less than credible", and that the estimation contains a number of "fundamental problems" that have led to "fantastical estimates". On the basis of its own assumptions and methodology, Canada asserts that TPC's rate of return on the Bombardier contribution would be "in the order of [X - BCI] percent." Canada asserts that a return of [X-BCI] percent clearly covers the cost of funds.
Brazil provided a number of comments on Canada's replies to our questions. Brazil states inter alia that Canada "has wholly or selectively redacted any and all information which could be at all informative to the Panel." For this reason, Brazil asks the Panel to "adopt adverse inferences, presuming that all information withheld by Canada is inculpatory", and that the $87 million Bombardier contribution is an export subsidy prohibited by Article 3 of the SCM Agreement.
Brazil also submits that only if CRJ-700 meets the most optimistic market forecasts would TPC reach the maximum rate of return estimated by Canada. In any event, Brazil asserts that even Canada's own estimate of TPC's rate of return on the $87 million Bombardier contribution is below market, i.e., below 16.91 - 21.92 percent.
$57 million TPC contribution to de Havilland; and $100 million TPC contribution to Pratt & Whitney
On the basis of the information adduced by Brazil concerning the TPC contributions to de Havilland and Pratt & Whitney, we asked Canada to provide inter alia: complete details of, and comment on, the terms and conditions of these contributions; copies of all application documents (including any pre-screening submissions, application forms, supporting business plans, sales and/or export projections including “expected clients, units, prices and timing”, repayment plan of the applicant, etc.); and assessments and funding decisions by TPC pertaining to those contributions. We also asked Canada to specify commercial borrowing rates in Canada for debt of comparable size, risk level and maturity prevailing at the time the contribution was made by TPC. Furthermore, we asked Canada to comment on the estimated rate of return calculated by Brazil for these contributions.
In its response, Canada stated:
Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Brazil has explicitly indicated its agreement with Canada in its letter of 13 December 1998, regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Accordingly, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to support Canada’s submissions that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
WT/DS70/R
Page 214
Most of the information requested by the Panel is highly sensitive business confidential information. Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and legal rights of private parties not Party to this dispute. These private parties, and others in the process of submitting applications under the TPC programme, have already expressed reluctance in sharing information, or additional information, concerning their business plans. Such reluctance, if it were to continue, would have a serious deleterious impact on the functioning of the TPC programme.
Canada has requested the interested private parties to release Canada from obligations arising under the business confidentiality clauses of Canada’s arrangements with them. With the exception of Bombardier, these interested parties have indicated that are not prepared to allow Canada to release business confidential information, or have not responded to the request for release […]
In response to our request for comments on the rates of return estimated by Brazil, Canada relied on the same arguments as for the aforementioned $87 million TPC contribution to Bombardier. Thus, Canada asserted that a number of assumptions supporting Brazil's estimation "have no basis in fact". Canada also asserted that the methodology of Brazil's estimation of TPC's return is "less than credible", and that the estimation contained a number of "fundamental problems" that have led to "fantastical estimates". Unlike the Bombardier contribution, however, Canada did not provide specific arguments seeking to rebut the rates of return estimated by Brazil.
As with Brazil's reaction to Canada's refusal to provide detailed information concerning the
$87 million Bombardier contribution, Brazil asks the Panel to "adopt adverse inferences, presuming that all information withheld by Canada is inculpatory", and that the two TPC contributions in issue are subsidies prohibited by Article 3 of the SCM Agreement.
$12.7 million TPC contribution to Allied Signal; and $9.9 million TPC contribution to Sextant Avionique
On the basis of the information adduced by Brazil concerning the TPC contributions to Allied Signal and Sextant, we asked Canada to provide inter alia: complete details of, and comment on, the terms and conditions of these contributions; copies of all application documents (including any pre- screening submissions, application forms, supporting business plans, sales and/or export projections including “expected clients, units, prices and timing”, repayment plan of the applicant, etc.); and assessments and funding decisions by TPC pertaining to those contributions. We also asked Canada to specify commercial borrowing rates in Canada for debt of comparable size, risk level and maturity prevailing at the time the contribution was made by TPC.
In its response, Canada stated:
Canada notes that it has not put in a defence regarding whether these contributions are subsidies within the meaning of Article 1 of the SCM Agreement. Brazil has explicitly indicated its agreement with Canada in its letter of 13 December 1998, regarding the relevance of the principle of judicial economy to the issues to be determined in this case. Accordingly, to the extent that any documents are produced by Canada in response to this question of the Panel, they are provided to support Canada’s submissions that the contributions at issue are not “contingent on export performance” within the meaning of Article 3 of the SCM Agreement.
Most of the information requested by the Panel is highly sensitive business confidential information. Canada’s desire to present to the Panel such information as may help it arrive at a decision must be balanced against the commercial interests and
WT/DS70/R
Page 215
legal rights of private parties not Party to this dispute. These private parties, and others in the process of submitting applications under the TPC programme, have already expressed reluctance in sharing information, or additional information, concerning their business plans. Such reluctance, if it were to continue, would have a serious deleterious impact on the functioning of the TPC programme.
Canada has requested the interested private parties to release Canada from obligations arising under the business confidentiality clauses of Canada’s arrangements with them. With the exception of Bombardier, these interested parties have indicated that are not prepared to allow Canada to release business confidential information, or have not responded to the request for release […]
In light of Canada's refusal to provide the detailed information requested by the Panel, Brazil asks the Panel to "adopt adverse inferences, presuming that all information withheld by Canada is inculpatory", and that the two TPC contributions in issue are subsidies prohibited by Article 3 of the SCM Agreement.
Evaluation by the Panel
We recall that, in order for Brazil's claim against TPC assistance to the Canadian regional aircraft industry to succeed, there must be, first, at least a prima facie case that such TPC assistance takes the form of "subsidies" within the meaning of Article 1 of the SCM Agreement. In particular, there must be a prima facie case that TPC assistance to the regional aircraft industry constitutes "financial contributions" by a government or public body (Article 1.1(a)) that confer "benefits" (Article 1.1(b)).
Is there a prima facie case that TPC assistance to the Canadian regional aircraft industry takes the form of subsidies?
We are in no doubt that TPC contributions constitute "financial contributions" by a public body within the meaning of Article 1.1 of the SCM Agreement, as they are direct transfers of funds by the government of Canada, in the sense of Article 1.1(a)(1)(i). Canada has not disputed this fact.
As to the question of "benefit", Brazil has adduced evidence demonstrating that at least three specific TPC contributions in the regional aircraft sector have been negotiated on terms that do not provide for a commercial rate of return. The total value of these contributions is $244 million. These contributions therefore account for approximately 68 per cent of TPC contributions to the aerospace and defence sector during the period 1996-1997.617 We recall that, in our opinion, a "benefit" is conferred when a financial contribution by a public body is made on terms more favourable than those available to the recipient on the market. Accordingly, we find that Brazil's arguments concerning these three specific contributions establish a prima facie case that TPC assistance to the Canadian regional aircraft industry confers "benefits" within the meaning of Article 1.1(b) of the SCM Agreement.
In light of the above, we find a prima facie case that TPC assistance to the Canadian regional aircraft industry has taken the form of "subsidies" within the meaning of Article 1.1 of the SCM Agreement.
Has Canada rebutted the prima facie case of subsidization?
We must now consider whether Canada has rebutted the prima facie case that TPC assistance to the Canadian regional aircraft industry has taken the form of "subsidies".
617 TPC Annual Report 1996-1997. The TPC programme was launched in 1996.
WT/DS70/R
Page 216
We note that Canada does not challenge Brazil's claim that TPC assistance to the Canadian regional aircraft industry constitutes "financial contributions" by a public body, within the meaning of Article 1.1(a) of the SCM Agreement.
As a general matter, Canada asserts that "TPC repayment terms are negotiated so that on a net present value basis they are cost-revenue neutral or better". Furthermore, Canada asserts that "the relevant consideration for determining the rate of return for the [five] TPC contributions in question was the Government of Canada's cost of funds." With regard to the $87 million contribution to Bombardier specifically, Canada states that its estimation of the return (i.e., [X-BCI]) percent is sufficient to cover TPC's cost of funds. Canada therefore seeks to rebut the prima facie case by arguing that TPC contributions do not result in any net cost to the Canadian Government. However, we recall our earlier finding618 that net cost to government is not a relevant consideration for the purpose of determining the existence of a "benefit" within the meaning of Article 1.1(b) of the SCM Agreement. Canada's arguments concerning net cost do not, therefore, rebut the prima facie case of "benefit".
Canada has criticised the methodology behind Brazil's estimation of TPC's return on the three specific contributions discussed above. In the event that such criticisms were to demonstrate that TPC's rate of return corresponded to a commercial benchmark, these criticisms would be highly relevant in the present case. However, Canada is unable to demonstrate that this is the case.619 Canada only provided detailed arguments concerning Brazil's estimation of TPC return with regard to the $87 million Bombardier contribution. Canada argues that Brazil has underestimated TPC's return on the Bombardier contribution by [X-BCI] percent. Canada does not demonstrate that TPC's return on the Bombardier contribution is actually equivalent to a commercial benchmark. In fact, in response to a Panel request for “commercial borrowing rates” for comparable investments, Canada explicitly confirms that it does not seek a commercial return, noting that “the relevant consideration for determining the rate of return for the [five] TPC contributions in question was the Government of Canada’s cost of funds”. In order to rebut the prima facie case of "benefit", we consider that Canada must do more than simply demonstrate that the amount of specific "benefit" estimated by Brazil may be incorrect, or that TPC’s rate of return covers Canada’s cost of funds. Rather, Canada must demonstrate that no "benefit" is conferred, in the sense that the terms of the contribution provide for a commercial rate of return. To the extent that Canada's criticisms of Brazil's methodology and estimates fail in this regard, and in light of Canada’s admission that it has sought only to cover its cost of funds, we consider that Canada has failed to rebut the prima facie case of "benefit".620
We note that, in conjunction with its reply, Canada submitted a number of documents concerning the $87 million Bombardier contribution, including Bombardier's application for TPC
618 See Section C above.
619 We note that Canada does not dispute Brazil's argument that a commercial investor would expect a rate of return of 16.91 - 21.92 percent on TPC's $87 million contribution to Bombardier. With regard to the $87 million TPC contribution to Bombardier, Canada's own estimate of TPC's return on that contribution is [X-BCI] percent. Thus, even Canada's own estimate of TPC's return is substantially below the commercial investor benchmark proffered by Brazil and not rebutted by Canada. In this context, we recall as well the statement by a Canadian official in the context of EDC (see para. [135] above), that commercial investors demand a return on equity of between 15 and 20 percent. We find this statement to be useful corroboration of Brazil’s estimated commercial rates of return, as in our view, TPC contributions are similar to equity investments in that, as with equity investments, TPC contributions will only be repaid if the funded projects are commercially successful.
620For this reason, we do not consider it necessary to address the merits of Canada's criticisms of the methodology behind Brazil's estimates of TPC's rate of return.
WT/DS70/R
Page 217
assistance, business plans, market assessments and sales forecasts. However, these documents were extensively redacted, which in our view significantly undermined their practical value for the purpose of assessing Brazil's claim of "benefit". Canada provided no specific documentation whatsoever concerning the terms of the remaining four TPC contributions identified by Brazil. As noted above, however, Canada did state in answer to the Panel’s question that on all five of the contributions identified by Brazil, the goal was only to cover Canada’s cost of funds. We note in addition that, more generally, Canada has provided as a business confidential document a memorandum concerning TPC’s repayment policy. Canada states that, according to this policy, “TPC’s repayment terms are negotiated so that on a net present value basis, they are cost-revenue neutral or better”. We consider this to be an admission by Canada that TPC generally, as a matter of policy, does not seek a commercial rate of return on its contributions, including those to the Canadian regional aircraft industry. We note that Canada’s admission to this effect with respect to the five contributions identified by Brazil confirms us in this view.
For the above reasons, we find that not only has Canada failed to rebut the prima facie case that TPC contributions to the Canadian regional aircraft industry constitute "subsidies" within the meaning of Article 1 of the SCM Agreement, in fact Canada has confirmed that case by admitting that TPC neither seeks nor earns a commercial rate of return on these contributions.621
Accordingly, we find that TPC assistance to the Canadian regional aircraft industry constitutes "subsidies" within the meaning of Article 1.1 of the SCM Agreement.
Is TPC assistance to the Canadian regional aircraft industry contingent on export performance?
In order to complete our examination of Brazil's claim against TPC assistance to the Canadian regional aircraft industry, we must now consider whether such assistance is "contingent … upon export performance" within the meaning of Article 3.1(a) of the SCM Agreement.
Brazil submits that TPC assistance to the Canadian regional aircraft industry is contingent in fact upon export since it is provided by TPC precisely because the Canadian regional aircraft industry is export-oriented and is anticipated to remain so. Brazil argues that the export-orientation of the Canadian regional aircraft industry is the condition for the grant of the subsidies, in the sense that the subsidies "would not have been granted were it not for the virtually total export orientation of the Canadian regional aircraft industry."
Brazil notes that the 1996-1997 TPC annual report itself recognizes the aerospace industry as "highly export oriented". Furthermore, Brazil refers to funding statistics in the TPC 1996-1997 annual report to allege that TPC is captive to the Canadian aerospace industry, and more specifically, the regional aircraft industry. Brazil submits that such support is, as an historical matter, par for the course. Brazil asserts that TPC’s predecessor, DIPP, channelled subsidies to the aerospace sector totalling approximately $2 billion.622 Brazil also asserts that statements made by the Canadian Government in announcing its $267 million in grants to the regional aircraft industry demonstrate that those grants are tied to the Canadian Government’s anticipation, based on ample historical evidence, that the industry will maintain its 100 percent export orientation. Brazil refers to Industry Minister
621 In light of this finding, we do not consider it necessary to address Brazil's request that the Panel adopt "adverse inferences" in examining the question of whether the TPC financing at issue constitutes a "subsidy". We regret, however, that Canada chose not to provide the Panel with the full details requested by the Panel concerning this financing, and we believe that had it been necessary to do so, the Panel may have been required to make inferences on the basis of the information available.
622 Canadian Taxpayers Federation study, “Corporate Welfare – A Report on Sixteen Years of Industry Canada Financial Assistance,” April 16, 1998.
WT/DS70/R
Page 218
John Manley's statement that “[a]erospace is a crucial sector for Canada’s economy, with exports growing at 10% per year.” Brazil also refers to a statement by Mr. Herb Gray, Leader of the Government in the House of Commons and then-Solicitor General of Canada, concerning the $57 million TPC contribution to de Havilland that “[t]hese two outputs of the Dash 8-400 project -- the creation of jobs and the building of exports -- are just what the government had in mind when we established Technology Partnerships Canada earlier this year.” Brazil asserts that neither these statements by senior Canadian Government officials nor the statements about the industry in TPC’s promotional materials were crafted in a vacuum. According to Brazil, at the time the Canadian Government was making these statements, it was aware of the fact that every single sale of Dash 8 series aircraft made since 1992, and every single sale of the CRJ since its development and commercialization, had been for export.
Brazil argues that the Canadian regional aircraft industry is devoted to exports. Brazil acknowledges that TPC assistance may be granted to other industries with domestic sales. Brazil asserts, however, that such assistance is not relevant to its claim concerning TPC assistance in the regional aircraft sector. According to Brazil, the Canadian Government has made it very clear that it maintains massive amounts of support to the Canadian regional aircraft industry precisely because it is an export industry and precisely because they anticipate that it will continue to be an export industry. According to Brazil, export orientation and performance are central to TPC’s decision- making processes in the aerospace and defense sector, since maintaining exports is identified as a key factor to consider in awarding TPC grants and evaluating TPC projects, and since information about a project’s export performance is requested and maintained by TPC. Brazil submits that this represents further evidence that TPC grants to the industry are “in fact tied to actual or anticipated exportation or export earnings” within the meaning of Article 3 of the Subsidies Agreement.
Brazil asserts that TPC practices embody precisely the meaning of the term “export subsidies in fact." Brazil submits that to fail to apply the de facto export subsidy provision in these circumstances would be to reduce the prohibition of de facto export subsidies to inutility, and to give WTO Members a license to provide export subsidies with abandon, subject only to the limitation that they avoid using the word “export” in their laws and regulations.
Canada asserts that TPC is not contingent upon export performance. Canada submits that TPC provides support to a broad base of sectors and technologies that touch on virtually all industrial sectors of Canada. Specifically, eligible sectors and technologies include the aerospace and defence sector (including defence conversion), environmental technologies and “enabling” technologies, which include biotechnologies, information and communication technologies, and advanced materials and advanced manufacturing technologies. Canada states that, as of September 30, 1998 TPC has approved 65 projects representing a total of $582 million in multi-year investments; of these, 48 projects ($174.5 million) involved environmental and enabling technologies, with the balance going to the aerospace and defence sector. Canada submits that the basic objectives of the TPC programme are set out in its Charter, according to which TPC aims, among other things, “to maintain and build the industrial technology and skill base essential for internationally competitive products and services”. Canada asserts that the basic objectives of TPC are to:
be an investment approach supportive of the government’s priorities for jobs, growth and sustainable development with all repayments of TPC contributions being recycled to help fund the programme;
be strongly market driven and results-oriented;
focus on activities in environmental technologies, strategic enabling technologies (e.g., advanced manufacturing and processing, advanced material processes and applications, applications of biotechnology and applications of selected information technologies) and aerospace and defence (including defence conversion); and
WT/DS70/R
Page 219
adhere to the twin principles of international competitiveness and national access by putting in place the necessary program machinery, rules and processes to ensure that competitive and capable high technology small and medium-sized enterprises from all regions of the country are encouraged to participate and have fair access to the program.
Canada asserts that TPC does not have “export performance” as a condition, in law or in fact, of project support. According to Canada, nothing in the application documents or the contribution agreements of TPC identifies export performance as a condition for eligibility for or approval of contributions. Canada submits that there is no requirement, in law or in fact, that the products resulting from the research and development investment by the Government of Canada be exported. The recipient does not suffer additional penalties because it did not sell into export markets. If additional contributions by the government are due, they are not terminated because exports do not take place. And royalties payable to the government are not increased on domestic sales if export sales are not made. Canada asserts that recipients' repayment obligations are not in anyway affected by whether the sales are made in Canada or outside Canada.
Canada argues that TPC contributions are simply conditional on success – and on presenting a successful business plan. Canada provides the example of a prospective recipient, such as the developer of a water-treatment facility for aircraft, in support of this argument. Canada hypothesises that the recipient comes to the administrators of TPC with the project, and suggests that it can sell 10,000 units, in export markets. According to Canada, the questions that the administrators of the programme will put to the proponent of the project will not concentrate on the export aspects of the project. Rather, they will ask whether the project is going to be viable. That is, whether the business plan makes sense so that the contribution is not wasted. Canada submits that the same questions will be asked of a project proponent who asks for a contribution for the development of arctic flight navigation equipment that can be used only in Canada. That is to say, is the project viable?
Canada acknowledges the phrase “maintain and build upon the technological capabilities and production, employment and export base [of Canada]" in the TPC eligibility criteria for the aerospace and defence sector. Canada asserts, however, that the export base can be maintained and built-upon in different ways. Subsidies that develop a country’s global competitiveness and thus maintain and build upon its export base are not inconsistent with the SCM Agreement for that reason alone. To so argue would be to suggest that the SCM Agreement permits only subsidies that are, at best, neutral as to competitiveness and productivity. According to Canada, that would render illegal just about every industrial and labour adjustment programme the world over.
Canada understands Brazil to argue that TPC is inconsistent with Article 3 because some of its contributions have been made in a sector that is export-oriented and to companies that export. Canada asserts that this is not the test in Article 3. According to Canada, official acknowledgements of the importance of the aerospace sector to the Canadian economy, and the fact that the creation of jobs and building of exports are objectives of TPC, are irrelevant to the issue of export contingency. Furthermore, Canada submits that the “export propensity” of the aerospace sector is a fact of the market rather than a condition or requirement of the programme. Canada asserts that the world aircraft industry is one of the most globalized industries, with few countries producing all the necessary technology domestically and all relying on economies of scale for profitability. Canada submits that the Canadian aerospace sector is no different. In addition, Canada argues that in view of the small size of the Canadian market, the significant dependence of the Canadian economy in general and the manufacturing sector in particular on exports, it would not be unusual if a large proportion of the sales of Canadian manufacturers are made in markets other than Canada. Canada notes that Brazil acknowledges that TPC funding is also available for projects that result in sales in domestic markets. Canada understands Brazil to argue that the mere fact that companies in the civil aviation sector engage in exports turns a programme that is also available for domestic markets into an export contingent subsidy. In this regard, Canada also denies Brazil's argument that every single sale of
WT/DS70/R
Page 220
Dash 8 series aircraft made since 1992, and every single sale of the CRJ since its development and commercialization, had been for export.623
Canada suggests that Brazil, by emphasising the term "anticipated" in footnote 4, is advocating a pure "intent" test. Canada criticises such an approach because it would render the general objective of a subsidy determinative of whether that subsidy is "contingent … in fact … upon export performance". Canada argues that any prospective analysis inherent in the term "anticipated" cannot undermine the conditionality of the phrase "tied to". Canada argues that conditionality must be read into the phrase "tied to" because (1) this is the ordinary meaning of "tied to", and (2) the phrase "tied to" elaborates on the term "condition" in the main text of Article 3.1(a). Thus, Canada argues that "'tied to … anticipated exportation' cannot and does not translate into 'granted with the general intent that exports somehow increase'. Rather, it means that 'one of the conditions for the grant of the subsidy is the expectation that exports will flow thereby'." Canada also expressed this approach using the phrase "but for", whereby a subsidy is "contingent … in fact … upon export performance" or "tied to … anticipated exportation or export earnings" if the subsidy "would not [have been] paid but for the expectation that exports would ensue" (emphasis in original). In response to a question from the Panel whether a subsidy would be "tied to … anticipated exportation or export earnings … when one of the reasons for the grant of the subsidy is the expectation that exports will flow thereby, or when the subsidy is granted because of the expectation that exports will flow thereby", Canada acknowledged that the reason why a subsidy is granted could be relevant to the extent that it "establishes the condition for the grant of the subsidy".
On the basis of the information adduced by Brazil, we asked Canada to provide inter alia the "project assessments and funding decisions by TPC" for each of the five TPC subsidies in issue. We recall that Canada refused to provide the Panel with this material. Canada asserts that the relevant material constitutes "cabinet confidence" or "Ministerial advice" that cannot be divulged. The Panel also asked Canada to provide the full TPC Interim Reference Binder, which provides guidance to TPC employees on the administration of the TPC programme. Canada complied with this request.
Reacting to Canada's refusal to provide "cabinet confidence" or "Ministerial advice", Brazil asserted that "[t]he decision to withhold these documents, which are exclusively within the possession and control of the Canadian government, must carry adverse consequences for Canada, or the good faith obligation contained in Article 3(10) of the DSU will become meaningless. The Panel should presume that these documents contain information prejudicial to Canada's position."
Concerning the TPC Interim Reference Binder which Canada did provide, Brazil identified the following extracts which, in its view, demonstrate that TPC’s funding decisions for the aerospace and defense industry are tied to export:
Section 3.2.3 of TPC’s “Terms and Conditions” states that “[c]ontributions under the Aerospace and Defence component will be directed to projects that will maintain and build upon the technological capabilities and production, employment and export base extant in the aerospace and defence sector”;
Section 3.3 of the TPC Charter, entitled “Aerospace and Defence (including Defence Conversion)” states that “[i]nvestments will be directed to projects that build on and maintain technological capabilities and the production, employment and export base of the sector”;
623 Canada's principal argument is that Brazil has overlooked a domestic transaction involving one Dash 8, and that Brazil has misclassified one transaction concerning 26 CRJs as for export, when in fact the transaction was domestic. We do not consider it necessary to make findings on the accuracy of Brazil's argument, since in this paragraph Canada effectively acknowledges the "export propensity" of the aerospace sector, and that a "large proportion" of Canadian civil aircraft are exported.
WT/DS70/R
Page 221
Part B of Schedule B of the TPC Aerospace and Defense Generic Model Agreement specifically calls for any representations by an applicant regarding “export markets penetration through marketing partnership agreements with foreign companies”;
Schedule C from the same Model Agreement, representing a form for “Report[s] on Estimated & Actual Sales and Royalties” requires the reporting of export sales revenues;
Page 10 of the 1996-1997 TPC Business Plan notes that TPC’s “approach” in the aerospace and defense sector is to “[d]irectly support the near market R & D projects with high export market potential”;
Page 12 of the 1996-1997 TPC Business Plan records the proportion of the aerospace and defense industry’s revenue allocable to exports.
We note that Brazil does not claim that the TPC programme is de jure export contingent. Rather, Brazil asserts that TPC contributions in the regional aircraft sector are "contingent … in fact
… upon exportation", within the meaning of Article 3.1(a) of the SCM Agreement. We recall that, according to note 4 of the SCM Agreement, the "contingent … in fact … upon exportation" standard is met when:
the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in fact tied to actual or anticipated exportation or export earnings. The mere fact that a subsidy is granted to enterprises which export shall not for that reason alone be considered to be an export subsidy within the meaning of this provision.
We note that Brazil effectively claims that TPC assistance to the Canadian regional aircraft industry is "contingent … in fact … upon export performance" (Article 3.1(a)) because it is "in fact tied to … anticipated exportation or export earnings" (footnote 4 of the SCM Agreement). In order to examine Brazil's claim, we must first examine the legal standard against which Brazil's claim must be measured. We shall do so on the basis of the principles of treaty interpretation contained in Article 31.1 of the Vienna Convention on the Law of Treaties.624
Interpretation of "contingent … in fact … upon export performance" and "in fact tied to … anticipated exportation or export earnings"
In our view, the ordinary meaning of "contingent" in Article 3.1(a) is "dependent for its existence on something else, "conditional; dependent on, upon".625 The ordinary meaning of "tied to" in footnote 4 is "restrain or constrain to or from action; limit or restrict as to behaviour, location, conditions, etc.".626 The phrase "tied to" requires a specific connection between the grant of the subsidy and "actual or anticipated exportation or export earnings" in order for a subsidy to be "contingent … in fact … upon export performance" within the meaning of Article 3.1(a) of the SCM Agreement. When read in the context of the "contingency" referred to in Article 3.1(a), we consider that the connection between the grant of the subsidy and the anticipated exportation or export earnings required by "tied to" is conditionality. We note that the parties agree with this interpretation. Canada has repeatedly emphasised the conditionality inherent in "tied to". For its part, Brazil has "agree[d] with Canada's statement … that 'tied to … anticipated exportation' means that 'one of the conditions
624 Article 31.1 of the Vienna Convention on the Law of Treaties provides that a treaty shall be interpreted "in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose."
625 The New Shorter Oxford English Dictionary, Vol. 1 (Oxford: Clarendon Press, 1993).
626 Id.
WT/DS70/R
Page 222
for the grant of the subsidy is the expectation that exports will flow thereby.'" We concur with the parties that this is an appropriate and useful formulation of the nature of the requisite conditionality.
We consider that we can examine most effectively whether there exists the requisite conditionality between the grant of TPC assistance to the Canadian regional aircraft industry and anticipated exportation or export earnings, by determining whether the facts demonstrate that such TPC assistance would not have been granted to the regional aircraft industry but for anticipated exportation or export earnings. Again, we note that the parties effectively agree that a "but for" test is appropriate for the purpose of determining whether a subsidy is "tied to" anticipated exportation or export earnings. Canada has stated that a subsidy is "tied to" anticipated exportation or export earnings if the subsidy "would not [have been] paid but for the expectation that exports would ensue" (emphasis in original). Brazil, in turn, has argued that export-orientation is the condition for the grant of the TPC subsidies in issue because these subsidies "would not have been granted were it not for the virtually total export orientation of the Canadian regional aircraft industry."
The parties disagree on which factors are relevant in determining whether TPC assistance to the Canadian regional aircraft industry would not have been granted but for anticipated exportation or export earnings. Brazil posits that this test is fulfilled when a subsidy is granted to a recipient because it has been, and is anticipated to remain, export-oriented. Canada has advanced a number of arguments rejecting Brazil's position. First, Canada asserts that Brazil's approach would "mean that there would be one law for larger economies that are not dependent on international trade and another for smaller economies that are." Second, Canada asserts that Brazil's approach would be over-broad, since "any subsidy that would help in the development of the competitive advantage of a state -- that would make its industries more efficient globally -- would then be prohibited as an export subsidy. Third, Canada asserts that "Brazil's interpretation would render government planning for WTO- consistent subsidies impossible", since "greater efficiency could lead to greater exports."
Canada seems to be arguing that export-orientation should not be taken into account at all in considering whether de facto export contingency exists. In our opinion, Canada's arguments are based on a misunderstanding of Brazil's approach to de facto export contingency. Canada's comments are based on its view that "Brazil argues that Article 3.1(a) applies not only to subsidies that are conditional upon export performance, but also those subsidies that have an 'export propensity'."627 However, we do not understand Brazil to argue that any subsidy that could lead to increased exports is de facto export contingent. Instead, Brazil refers to subsides that are granted precisely because they are expected to lead to increased exports.628
Canada also advances an additional argument for rejecting Brazil's export-orientation approach, in the form of the final sentence of footnote 4: "[t]he mere fact that a subsidy is granted to enterprises which export shall not for that reason alone be an export subsidy within the meaning of this provision." Canada argues that Brazil's approach to de facto export contingency is inconsistent with this sentence. Again, however, we consider that Canada has missed the essence of Brazil's argument. Brazil does not in fact argue that the mere fact that a subsidy is granted to an exporter renders that subsidy de facto export contingent. Rather, Brazil argues629 that "[w]hile Article 3.1(a) provides that the mere granting of a subsidy to an exporting entity will not 'for that reason alone' make it a prohibited export subsidy, the Panel is not here faced with such a case." We understand Brazil to argue that subsidies granted to exporters are not "for that reason alone" de facto export contingent;
627 See para. 5.77 above.
628 Brazil asserts in response to Question 19 from the Panel (27 November 1998) that "Brazil's entire
claim does not, as Canada claims, turn on the 'export propensity' of the Canadian regional aircraft industry alone. The Panel is faced with a situation in which several factors converge, together illustrating that the Canadian Government and the provinces have supported the Canadian regional aircraft industry precisely because it is a total export industry, and precisely because they anticipate that this performance will continue".
629 See para. 5.104 above.
WT/DS70/R
Page 223
they are de facto export contingent if, in addition, they are granted precisely because the recipient was export-oriented and anticipated by the granting authority to remain so.
For these reasons, we do not consider that Canada has effectively demonstrated that export- orientation should not be taken into account in the context of determining whether a subsidy would not have been granted but for anticipated exportation or export earnings. This does not mean, however, that export-orientation alone can necessarily be determinative.
In our view, no fact should automatically be rejected when considering whether the facts demonstrate that a subsidy would not have been granted but for anticipated exportation or export earnings. We note that footnote 4 provides that the "facts" must demonstrate de facto export contingency. Footnote 4 therefore refers to "facts" in general, without any suggestion that certain factual considerations should prevail over others. In our opinion, it is clear from the ordinary meaning of footnote 4 that any fact could be relevant, provided it "demonstrates" (either individually or in conjunction with other facts) whether or not a subsidy would have been granted but for anticipated exportation or export earnings. We consider that this is true of the export-orientation of the recipient, or of the reason630 for the grant of the subsidy, just as it is true of a host of other facts potentially surrounding the grant of the subsidy in question. In any given case, the relative importance of each fact can only be determined in the context of that case, and not on the basis of generalities.
We would emphasise, however, that our finding that a broad range of facts could be relevant in this context does not mean that the de facto export contingency standard is easily met. On the contrary, footnote 4 of the SCM Agreement makes it clear that the facts must "demonstrate" de facto export contingency. That is, de facto export contingency must be demonstrable on the basis of the factual evidence adduced.
Putting this into more concrete terms, we consider that the factual evidence adduced must demonstrate that had there been no expectation of export sales (i.e., “exportation” or “export earnings”) “ensuing” from the subsidy, the subsidy would not have been granted. To us, this implies a strong and direct link between the grant of the subsidy and the creation or generation of export sales. That is, we consider that the “but for” test is concerned principally with anticipated export sales. Thus, the closer a subsidy brings a product to sale on the export market, the greater the possibility that the facts may demonstrate that the subsidy would not have been granted but for anticipated exportation or export earnings. Conversely, the further removed a subsidy is from sales on the export market, the less the possibility that the facts may demonstrate that the subsidy would not have been granted but for anticipated exportation or export earnings. In this light, subsidies for pure research, or for general purposes such as improving efficiency or adopting new technology, would be less likely to satisfy the “but for” test than subsidies that directly assist companies in bringing specific products to the (export) market.
(v) Application of the "but for" test
We recall that, in the present instance, there must be a prima facie case that TPC assistance to the Canadian regional aircraft industry is "contingent … in fact … upon export performance", in the sense of being "tied to … anticipated exportation or export earnings. In light of the above analysis, there must be a prima facie case that the facts demonstrate that TPC assistance to the Canadian regional aircraft industry would not have been granted but for "anticipated exportation or export earnings." In our opinion, the following considerations, based on materials and arguments submitted by the parties are especially relevant in this regard:
630 We note that Canada has accepted (see para. 5.98 above) that the reason for the grant of the subsidy could be relevant to the extent that the reason "establishes the condition for the grant of the subsidy".
WT/DS70/R
Page 224
as Canada admits, the Canadian aerospace sector exports a large proportion of its output, due to the small size of the Canadian domestic market;631 (emphasis supplied)
the 1996-1997 TPC Business Plan notes 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"; (emphasis supplied)
section 3.2.3 of the "Terms and Conditions" set forth in the TPC Interim Reference Binder states “TPC will provide contributions to specific industrial development projects in order to enable Canadian Aerospace and Defence Industries to compete fairly and openly on the world competitive stage”; (emphasis supplied)
the Industry Minister's Message included in the 1996-1997 TPC Annual Report states:
"Aerospace and defense also make a significant contribution to our economic well- being. The sector is highly export oriented. Exports accounted for about 70 percent of sales, or $7.4 billion, in 1995. And there is the prospect of real growth in this area. Canada's aerospace sector currently ranks sixth in the world. With investments from TPC, and with industry's concerted efforts, this sector will be better equipped to compete effectively in the world marketplace and could grow to fourth place"; (emphasis supplied)
the 1996-1997 TPC Annual Report states that "[t]he 12 largest firms [in the A&D sector] account for most of the R&D and shipments, of which 80 percent are exported. … TPC is proud to be an investment partner in this export-oriented success story"; (emphasis supplied)
an Industry Canada press release, issued in respect of the $100 million TPC contribution to Pratt & Whitney, quotes Industry Minister Manley as stating "[A]erospace is a critical sector for Canada's economy, with exports growing at a rate of 10% per year. TPC's investment in these projects will help increase the global competitiveness of this industry, while supporting jobs in Montreal, in Halifax and across the country, generating economic growth and export dollars"; (emphasis supplied)
Concerning the $57 million contribution to de Havilland for the development of the Dash 8-400, Mr. Herb Gray, Leader of the Government in the House of Commons and then-Solicitor General of Canada, stated that “[t]hese two outputs of the Dash 8-400 project -- the creation of jobs and the building of exports -- are just what the government had in mind when we established Technology Partnerships Canada earlier this year”; (emphasis supplied)
TPC website material states that "TPC approved projects are forecasted to generate sales of more than $65 billion (mostly exports) and create or maintain 13,166 direct and indirect jobs"; (emphasis supplied)
the TPC Applications Kit requires applicants to describe "potential broad economic and social benefits, such as: job creation and retention, increased exports, new investment ….". The Applications Kit also states that TPC "invests in projects that have the potential to create jobs, generate export, launch new industries, and transform or strengthen the competitiveness of industry"; (emphasis supplied)
631 See paras. 6.241 and 6.242 above.
WT/DS70/R
Page 225
according to the TPC Interim Reference Binder, Schedule C of the TPC Aerospace and Defence Sector Generic Model Agreement requires applicants to distinguish between domestic sales and exports when reporting forecast and actual sales; (emphasis supplied)
the TPC Interim Reference Binder requires TPC employees, when completing Project Summary Forms, to explain the reasons for recommending support or rejection of the project. In particular, employees are given the following instructions:
"Strategic Considerations, Benefits, Indicators
In justifying the recommendation in strategic terms emphasise "business results". The following may be considered:
Link the project with departmental strategies and priorities as relevant:
improvement of international competitiveness; when sales will result directly from the project, report annual sales in terms of domestic and export sales and any import replacement”; (emphasis supplied)
section 3.2.3 of the "Terms and Conditions" set forth in the TPC Interim Reference Binder states that TPC will “fill a financial void…where government action is required to level the competitive playing field, at the near-market end of the spectrum”; and that “[c]ontributions under the Aerospace and Defence component will be directed to projects that will maintain and build upon the technological capabilities and production, employment and export base extant in the aerospace and defence sector”; (emphasis supplied)
Industry Canada website material concerning TPC "Application information" states that one factor considered by TPC in determining the need for federal government involvement is whether "assistance is required to level the playing field against international competitors; (emphasis supplied)
section 3.3 of the TPC Charter, entitled “Aerospace and Defence (including Defence Conversion)” states that “[i]nvestments will be directed to projects that build on and maintain technological capabilities and the production, employment and export base of the sector”; (emphasis supplied)
the 1996-1997 TPC Business Plan records the proportion of the aerospace and defense industry’s revenue allocable to exports; (emphasis supplied) and
the TPC contributions identified by Brazil have been for the development of specific products, and have been provided expressly on the basis of the projected sales of those products, the market for which is known to be almost entirely outside Canada; the statistics maintained by TPC, and the public statements about TPC, separately recount, and emphasize, the amount of export sales “generated” by these contributions. (emphasis supplied)
In our view, these facts demonstrate that TPC funding in the regional aircraft sector is expressly designed and structured to generate sales of particular products, and that the Canadian Government expressly takes into account, and attaches considerable importance to, the proportion of those sales that will be for export, when making TPC contributions in the regional aircraft sector. In this regard, we note again in particular that TPC contributions in the aerospace and defence sector, including the regional aircraft industry, are provided for "near-market projects with high export potential" (emphasis added). To us, therefore, these facts demonstrate that TPC assistance to the Canadian regional aircraft industry would not have been granted but for some expectation of
WT/DS70/R
Page 226
exportation or export earnings.632 Accordingly, we find that there is sufficient basis for a prima facie case that the facts demonstrate that TPC assistance to the Canadian regional aircraft industry is "in fact tied to … anticipated exportation or export earnings", and therefore "contingent … in fact … upon export performance" within the meaning of Article 3.1(a) of the SCM Agreement.
We do not consider that Canada has effectively rebutted the prima facie case.
Throughout the Panel proceedings, and during its final oral submission to the Panel, Canada's basic rebuttal argument has been that the TPC is not conditional on exports taking place. In particular, Canada argues that there are no penalties if export sales are not realised. While this argument may be relevant in determining whether a subsidy would not have been granted but for actual exportation or export earnings, we find this argument insufficient to rebut a prima facie case that a subsidy would not have been granted but for anticipated exportation or export earnings.
Furthermore, Canada has failed to adduce any evidence demonstrating that TPC assistance to the Canadian regional aircraft industry would have been granted irrespective of anticipated exportation or export earnings. We note that Canada has made an assertion concerning the "key considerations" behind the $87 million TPC contribution to Bombardier (see para. 9.294 above), in order to demonstrate the absence of export contingency. However, Canada has provided no evidence to support this assertion, or to show that the export-related considerations, described above, that permeate the general information about the policy and operation of TPC (such as anticipated exportation or export earnings) were not also taken into account by the TPC administrators. In the absence of supporting evidence, such assertions are insufficient to rebut the prima facie case, based on a considerable volume of documentary evidence, that TPC assistance to the Canadian regional aircraft industry, including but not limited to the $87 million contribution to Bombardier, is de facto export contingent.
We also note that Canada has submitted documents designated as Business Confidential Information concerning the $87 million subsidy to Bombardier, in order to demonstrate that this contribution is not export contingent. However, the material submitted by Canada has been redacted to such an extent that it is simply of no value to the Panel, and there is no means for the Panel to be certain that information pointing to de facto export contingency has not been redacted. Moreover, Canada has outright refused (on the basis of “Cabinet privilege”) to provide what in the Panel’s view are the most relevant of the documentation that it requested regarding the five contributions identified by Brazil: the Project Summary Forms, and the memoranda and other documents supporting the recommendations and decisions to provide these contributions. For this reason, the Business Confidential documents submitted by Canada are not sufficient to rebut a prima facie case that TPC assistance to the Canadian regional aircraft industry, including but not limited to the $87 million contribution to Bombardier, would not have been granted but for anticipated exportation or export earnings.
We recall Canada’s own formulation (with which Brazil and we concur) of the nature of the required conditionality as being that “one of the conditions for the grant of the subsidy is the expectation that exports will flow thereby”. We believe that the facts available demonstrate that one of the conditions of the grant of TPC contributions to the Canadian regional aircraft industry is indeed such an expectation, in the form of projected export sales anticipated to “flow” directly from these contributions.
632 We note that other conditions for TPC assistance in addition to anticipated exportation or export earnings, may include the creation of jobs, the launching new industries, or transforming or strengthening the competitiveness of Canadian industry. However, the existence of such additional conditions does not preclude a finding that TPC contributions are "contingent … upon export performance", since Article 3.1(a) of the SCM Agreement applies when export contingency exists "solely or as one of several other conditions…"
WT/DS70/R
Page 227
For the above reasons, we find that TPC assistance to the Canadian regional aircraft industry is "contingent … in fact … upon export performance" within the meaning of Article 3.1(a) of the SCM Agreement.633
(vi) Conclusion
In light of the above findings, we conclude that TPC assistance to the Canadian regional aircraft industry constitutes "subsidies contingent … in fact … upon export performance", contrary to Articles 3.1(a) and 3.2 of the SCM Agreement.
2. Assistance provided under the DIPP programme
Brazil has adduced very little evidence in support of its claim that DIPP assistance to the Canadian regional aircraft industry constitutes subsidies contingent on export, contrary to Articles 3.1(a) and 3.2 of the SCM Agreement. In short, Brazil argues simply that the DIPP programme was TPC's predecessor.
Brazil identified one instance of joint DIPP / SDI assistance provided in April 1989. However, Brazil subsequently withdrew this claim on the grounds that the contribution is not subject to the SCM Agreement because, in its view, the SCM Agreement does not apply to allegedly prohibited export subsidies granted prior to 1 January 1995.
Canada made no specific arguments concerning Brazil's claim against DIPP assistance to the Canadian regional aircraft industry.
In the absence of any relevant evidence adduced by Brazil, we have no basis for determining the terms on which DIPP assistance has been provided to the Canadian regional aircraft industry, or whether such assistance was contingent on export. In the absence of any relevant evidence, we therefore find that there is no basis for a prima facie case that DIPP assistance to the Canadian regional aircraft industry constitutes "subsidies contingent … in fact … upon export performance", contrary to Articles 3.1(a) and 3.2 of the SCM Agreement.
X. CONCLUSIONS AND RECOMMENDATION
In conclusion, we:
reject Brazil's claim that EDC assistance to the Canadian regional aircraft industry constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement;
find that Canada Account debt financing since 1 January 1995 for the export of Canadian regional aircraft constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement;
633 As a result of this finding, there is no need to address Brazil's request for adverse inferences in light of Canada's refusal to submit certain BCI or documents protected by cabinet privilege. We explained above why we reject Canada's arguments based on the inadequate protection of BCI. With regard to cabinet privilege, we note that in certain circumstances, such as national security, a Member may consider itself justified in withholding certain information from a panel. However, in such circumstances, we would expect that Member to explain clearly the basis for the need to protect that information. In the present case, Canada has invoked cabinet privilege for the purpose of protecting documents concerning the approval of contributions under the TPC. Canada has failed to explain why such information needs to be protected. In the absence of any such explanation, we are not at all convinced of the merits of Canada's reliance on cabinet privilege in the present case.
WT/DS70/R
Page 228
reject Brazil's claim that the sale by the Ontario Aerospace Corporation of its 49 percent interest in de Havilland Inc. to Bombardier in January 1997 constitutes an export subsidy inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement;
reject Brazil's claim that Subsidiary Agreement assistance to the Canadian regional aircraft industry constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement;
reject Brazil's claim that SDI and IQ assistance to the Canadian regional aircraft industry constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement;
find that TPC assistance to the Canadian regional aircraft industry constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement; and
reject Brazil's claim that DIPP assistance to the Canadian regional aircraft industry constitutes export subsidies inconsistent with Article 3.1(a) and 3.2 of the SCM Agreement.
Pursuant to Article 3.8 of the DSU, the findings in sub-paragraphs (b) and (f) of the preceding paragraph also constitute a case of prima facie nullification or impairment of benefits accruing to Brazil under the SCM Agreement, which Canada has not rebutted.
In light of the above findings, we are required to make the recommendation provided for in Article 4.7 of the SCM Agreement. Accordingly, we recommend that Canada withdraw the subsidies identified above without delay.
Article 4.7 further provides that "the panel shall specify in its recommendation the time- period within which the measure must be withdrawn." In other words, we are required to specify what period would represent withdrawal "without delay". Taking into account the procedures that may be required to implement our recommendation on the one hand, and the requirement that Canada withdraw its subsidies "without delay" on the other, we conclude that Canada shall withdraw the subsidies identified in sub-paragraphs (b) and (f) of the preceding paragraph within 90 days.
WT/DS70/R
Page 229
ANNEX 1
PROCEDURES GOVERNING BUSINESS CONFIDENTIAL INFORMATION AND DECLARATION OF NON-DISCLOSURE
BASIC PRINCIPLE
The treatment of information as Business Confidential under these procedures imposes a substantial burden on the Panel and the parties. The indiscriminate designation of information as Business Confidential could limit the ability of a party to fully include in its litigation team individuals who have particular knowledge and expertise relevant to presenting the party's case, impede the work of the Panel and complicate the Panel's task in formulating credible public findings and conclusions. Finally, the Panel recalls that all WTO Members are obliged under Article 25.9 of the SCM Agreement to provide information regarding the nature and extent of any subsidy "in a comprehensive manner" and with "sufficient details to enable the other Member to assess their compliance with the terms" of the SCM Agreement. Accordingly, while the Panel recognizes that the parties have a legitimate interest in protecting sensitive Business Confidential information, the Panel expects that parties will exercise the utmost restraint in designating information as Business Confidential.
DEFINITIONS
“approved person” means
a Panel member;
a representative;
a Secretariat employee; or
a PGE member,
who has filed with the Chairman of the Panel a Declaration of Non-disclosure.
“conclusion of the Panel” means when, pursuant to DSU Article 16.4, the Panel report is;
adopted;
not adopted; or
the Panel report is appealed and the report of the Appellate Body is adopted.
“Business Confidential information” means any information that has been designated as Business Confidential by the party submitting the information, and that is not otherwise available in the public domain.
“Declaration of Non-disclosure” means a copy of the declaration set out in Annex II, signed and dated by the person making the declaration.
“designated as Business Confidential” means:
for printed information, clearly marked with the notation ‘BUSINESS CONFIDENTIAL INFORMATION’ and with the name of the party that submitted the document;
for binary-encoded information, clearly marked with the notation ‘BUSINESS CONFIDENTIAL INFORMATION’ on a label on the storage medium, and clearly annotated with the notation ‘BUSINESS CONFIDENTIAL INFORMATION’ in the binary-encoded files; and
WT/DS70/R
Page 230
for uttered information, declared by the speaker to be “Business Confidential information” prior to the disclosure.
“dispute” means Brazil's challenge to certain Canadian measures under Article 4 of the WTO Agreement on Subsidies and Countervailing Measures, WT/DS70, entitled “Canada – Measures Affecting the Export of Civilian Aircraft”.
“DSU” means the WTO Understanding on Rules and Procedures Governing the Settlement of Disputes.
"Geneva mission" means the buildings and grounds of Brazil and Canada at Ancienne Route 17B, 1218 Grand-Saconnex and Rue du Pré-de-la-Bichette 1, 1202 Geneva, respectively.
“information” means:
printed information;
binary-encoded information stored in computer diskettes, computer disc drives,
CD roms, or other electronic media; or
uttered information,
including without limiting the generality of the foregoing, offers, agreements, reports, forecasts, compilations, studies, plans, presentations, charts, graphs, pictures and drawings.
“Panel” means the WTO panel established pursuant to DSU Article 6 by the 23 July 1998 decision of the WTO Dispute Settlement Body to examine the dispute.
“Panel meeting” means a substantive meeting of the Panel with the parties or the interim review meeting of the Panel with the parties, as described in the working procedures of DSU Appendix 3.
“Panel member” means a person selected pursuant to DSU Article 8 to serve on the Panel.
“Panel process” means the process of the Panel as described in DSU Articles 12, 15 and 16, until and including the conclusion of the Panel.
“party” means Brazil or Canada.
“PGE member” means a person appointed to the Permanent Group of Experts established pursuant to SCM Agreement Article 24, and who has been requested to assist the Panel pursuant to Article 4.5 of the SCM Agreement.
“premises of the WTO” means buildings and grounds of the WTO at Centre William Rappard, Rue de Lausanne 154, Geneva, Switzerland.
“representative” means:
an employee of a party;
an agent for all purposes of a party; or
a legal counsel or other advisor of a party,
who has been authorized by a party to act on behalf of such party in the course of the dispute and whose authorization has been notified to the Chairman of the Panel and to the other party, but in no circumstances shall this definition include an employee, officer or agent of a private company engaged in aircraft manufacturing.
WT/DS70/R
Page 231
“SCM Agreement” means the WTO Agreement on Subsidies and Countervailing Measures. “Secretariat” means the Secretariat of the World Trade Organization.
“Secretariat employee” means a person employed or appointed by the Secretariat who has been authorized by the Secretariat to work on this dispute and whose authorization has been notified to the Chairman of the Panel, including without limiting the generality of the foregoing, translators and transcribers present at the Panel hearings.
“secure location” means a locked storage receptacle on the premises of the WTO chosen by the Secretariat to provide secure storage for Business Confidential information.
“submit” means:
the filing by a party of printed or binary-encoded information at the Secretariat during the dispute;
the filing by a party of printed or binary-encoded information with the Panel during a Panel hearing; or
the uttering of information during a Panel hearing.
"third party" means a Member having notified its interest in the dispute to the DSB pursuant to DSU Article 10.
SCOPE
These procedures apply to all Business Confidential information submitted during the Panel process.
OBLIGATION ON PARTIES
Each party shall ensure that its representatives comply with these procedures.
SUBMISSION BY A PARTY
When submitting information, a party may designate all or any part or parts of that information as Business Confidential information. Business Confidential information shall be submitted in two copies: one copy of the Business Confidential information shall be submitted to the Secretariat; the other copy of the Business Confidential information shall be submitted to the other party at its Geneva mission.
If, taking into the account the Basic Principle stated in Article I, the Panel considers that a party has designated as Business Confidential information which is not reasonably entitled to such treatment, the Panel may decline to consider such information. In such a case, the party submitting the information may, at its discretion:
withdraw the information, in which case the Panel and the other party shall promptly return the information to the party submitting it; or
withdraw the designation of the information as Business Confidential.
When submitting printed or binary-encoded Business Confidential information, the party shall also provide:
WT/DS70/R
Page 232
a non-Business Confidential edited version, redacted in such a manner as to convey a reasonable understanding of the substance of the information;
a non-Business Confidential summary in sufficient detail to convey a reasonable understanding of the substance of the information; or
in exceptional circumstances, a written statement:
that such a non-Business Confidential edited version or non-Business Confidential summary cannot be made, or
that such a non-Business Confidential edited version or non-Business Confidential summary would disclose facts that the party has a proper reason for wishing to keep business confidential.
If the Panel considers that a non-Business Confidential edited version or summary does not fulfill the requirements of paragraph 3(i) or (ii), or that such exceptional circumstances as justify a statement pursuant to paragraph 3(iii) do not exist, the Panel may decline to consider the Business Confidential information in question. In such a case, the party submitting the information may, at its discretion,
withdraw the information, in which case the Secretariat and the other party shall promptly return the information to the party submitting it; or
comply with the provisions of paragraph 3 to the satisfaction of the Panel.
When uttering Business Confidential information at a Panel meeting, the speaker shall also provide a brief non-Business Confidential oral statement in sufficient detail to convey a reasonable understanding of the substance of the information that will be uttered.
STORAGE
The Secretariat shall store all Business Confidential information submitted in the secure location when not in use by an approved person.
Each party shall store all Business Confidential information submitted to it by the other party in a safe in a locked room at the premises of its Geneva mission, when not in use by a representative who is an approved person. Only a representative who is an approved person shall be given authority to unlock the locked room containing the safe, and the locked safe. If requested, either party may visit the other party's Geneva mission to review the proposed location of the safe, and to propose any changes. Any disagreements between the parties regarding the location of the safe, or any other aspect related to safeguarding of the Business Confidential information will be decided by the Panel.
An approved person shall take all necessary precautions to safeguard Business Confidential information when in use.
OBLIGATION NOT TO DISCLOSE
Where Business Confidential information has been submitted pursuant to these procedures, no approved person who views or hears such information shall disclose that information, or allow it to be disclosed, to any person other than another approved person, except in accordance with these procedures.
The Panel shall not disclose Business Confidential information in its interim and final reports, but may make statements of conclusion drawn from such information.
WT/DS70/R
Page 233
DISCLOSURE
The Secretariat shall make available for viewing or hearing only on the premises of the WTO any Business Confidential information requested by an approved person.
Each party shall promptly, and in a convenient manner, make available for viewing on the premises of its embassy or other diplomatic mission in the capital of the other party or, at the request of a approved person, on the premises of its embassy or other diplomatic mission at some other location, any Business Confidential information requested by an approved person.
Business Confidential information stored at the Geneva mission of a party may only be viewed by a representative of that party who is an approved person.
An approved person viewing or hearing Business Confidential information may take written summary notes of that information for the sole purpose of the Panel process.
Business Confidential information shall not be copied, distributed, or removed from the premises of the WTO, or from the premises of a party's Geneva mission, or from the premises of the embassy or other diplomatic mission referred to in paragraph 2, except as specifically provided in these Procedures.
Notwithstanding paragraph 5. above, a Panel member may remove a copy of Business Confidential information from the premises of the WTO. Any copies of Business Confidential information removed from the premises of the WTO by a Panel member shall be used exclusively by that Panel member for the purpose of working on the dispute, and shall be returned to the Secretariat upon conclusion of the Panel. Copies of Business Confidential information removed from the premises of the WTO by a Panel member shall be stored in a locked receptacle.
DISCLOSURE AT A PANEL MEETING
A party that wishes to submit Business Confidential information during a Panel meeting may request the Panel to exclude persons who are not approved persons from the meeting. The Panel shall exclude such persons from the meeting for the duration of the submission of such information.
DISCLOSURE TO THIRD PARTIES
Article 10.3 of the DSU provides that "[t]hird parties shall receive the submissions of the parties to the dispute to the first meeting of the panel." Accordingly, disclosure shall be granted to representatives of third parties of Business Confidential information contained in the first submissions of the parties on the premises of the WTO, or on the premises of an embassy or other diplomatic mission of the party submitting the Business Confidential information consistent with Section VIII, paragraph 2. The provisions of these procedures shall apply mutatis mutandis to any such disclosure.
TAPES AND TRANSCRIPTS
Any tapes and transcripts of Panel meetings at which Business Confidential information is uttered shall be treated as Business Confidential information under these procedures.
RETURN AND DESTRUCTION
At the conclusion of the Panel the Secretariat and the parties shall:
WT/DS70/R
Page 234
return any printed or binary-encoded Business Confidential information in their possession to the party that submitted such Business Confidential, unless that party agrees otherwise; and
destroy all tapes and transcripts of the Panel hearings that contain Business Confidential information, unless the parties mutually agree otherwise.
If the Panel Report is appealed, the Secretariat shall transmit any printed or binary encoded Business Confidential information, plus all tapes and transcripts of the Panel hearings that contain Business Confidential information, to the Appellate Body as part of the record of the Panel proceedings. The Secretariat shall transmit such information to the Appellate Body separately from the rest of the record and shall inform the Appellate Body of the special procedures that the Panel has applied with respect to such Business Confidential information. The parties shall comply with any directive of the Appellate Body regarding disclosure of Business Confidential information to parties or third parties as the Appellate Body may deem appropriate.
WT/DS70/R
Page 235
ANNEX II DECLARATION OF NON-DISCLOSURE
In accordance with the Procedures Governing Business Confidential information contained Annex I to the Working Procedures of the Panel on Canada – Measures Affecting the Export of Civilian Aircraft (the Procedures), I agree to the following:
Words defined in the procedures have the same meaning in this Declaration of Non-Disclosure as in the Procedures.
I acknowledge having received a copy of the Procedures, a copy of which is attached.
I acknowledge having read and understood the Procedures.
I agree to be bound by, and to adhere to, the provisions of the Procedures and, accordingly, without limitation, to treat confidentially all Business Confidential information that I may view or hear from time to time in accordance with the Procedures.
Executed on this day of , 1998. BY:
Name: Title:
(Advisors only) Affiliation or employment: