VI. Arguments of Parties Regarding Canadian Measures Alleged by Brazil to be Prohibited Export Subsidies in the Sense of SCM Articles 1 and 3¹⁴³

A. Export Development Corporation ("EDC")

1. General arguments of the parties

(a) Alleged concessionary terms of EDC financing; EDC risk level and performance

(i) Arguments of Brazil

6.1 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.

6.2 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

6.3 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.

6.4 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.

6.5 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 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

6.6 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.

6.7 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

6.8 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, conversely, that recipients of EDC funding are paying less for that funding than they would for funding from commercial sources.

6.9 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.

6.10 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.

6.11 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.

6.12 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:

6.13 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

6.14 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

6.15 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.

6.16 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.

(ii) Arguments of Canada

(a) Evidence submitted by Brazil on alleged concessionary terms of EDC financing

6.17 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.

6.18 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:

6.20 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

6.21 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."

6.22 Canada asserts that Brazil has omitted the qualifying subordinate clause of the quoted sentence, and that the full sentence, placed in context, is:¹⁷⁰

6.23 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 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.

6.24 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.

6.25 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:

6.26 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.

To continue with EDC's risk level and performance

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.

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).

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).

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

¹⁶⁹ Exh. CDN-52

¹⁷⁰ Exh. BRA-7, p. 2

¹⁷¹ Exh. BRA-8, p.2

¹⁷² Exh. BRA-9