Brazil-Export Financing Programme for Aircraft
Report of the Panel
IV. Main Arguments of the Parties, Section 4.68 (continued)
4.68 Addressing the argument of
Canada that its interpretation of item (k) would turn the illustrative list of export
subsidies into an exhaustive list and produce absurdities such as the one claimed in
paragraph 4.47 above, Brazil argues that Canada is mistaken and has misconstrued its
interpretation of item (k). Brazil submits that the limiting language in item (a), which
is the counterpart to the "material advantage" language of item (k) is not
"direct." It is "contingent upon export performance." The
"provision by governments of direct subsidies to a firm or an industry" is a
subsidy, but it is not prohibited subsidy unless it is "contingent upon export
performance" (italics in original) 54.
4.69 Brazil also referred to the duty drawback
provisions of item (i) of Annex I. Brazil argues that item (i), as Canada has noted,
contains a "provided, however" clause which defines activity that is not an
export subsidy. However, item (i) has more. In its very first sentence it defines the
prohibited drawback payments as those "in excess of those levied on imported
products that are consumed in the production of the exported product." Drawback
payments not in excess of those levied on imported products consumed in the
production of an exported product are not prohibited. 55
4.70 Brazil also argues that item (h), as Canada
notes, also contains a "provided, however" clause, but it too contains more in
its first sentence. This sentence specifies that the exemption, remission or deferral of
prior-stage cumulative indirect taxes on exported products that are "in excess of the
exemption, remission or deferral of like prior-stage cumulative indirect taxes" sold
for domestic consumption constitutes a prohibited subsidy. Brazil submits that the only
logical reading of this language is that exemption, remission or deferral for exports that
is not in excess of those imposed for domestic consumption is not prohibited 56. (emphasis in original)
4.71 Brazil submits that the "material
advantage" clause of the first paragraph of item (k) serves the same function as
these provisions and that it defines and limits the prior portion of the sentence 57.
2. Whether PROEX payments are payments within the meaning of item
(k) of the Illustrative List
4.72 Canada submits that although PROEX
subsidies are related to export credits, they do not constitute a "payment" by
[governments] of all or part of the costs incurred by exporters or financial institutions
in obtaining credits" within the meaning of the second clause of the first paragraph
of item (k), given their nature as periodic cash grants to the purchasers of civil
aircraft.
4.73 Canada submits that its interpretation of the
first paragraph of item (k) is as follows. The first part of item (k) refers to a
situation where a government lends funds at an interest rate that is below the rate it
would pay for raising such funds. The phrase "the costs incurred … in obtaining
credits" in the second part of item (k) refers to a similar situation, but with
private financing. That is, where an exporter or a financial institution obtains credits
at rates higher than the rates at which it would lend to a purchaser and incurs a cost as
a result, and a government pays for all or part of this difference.
4.74 Canada submits that its interpretation is
consistent with the response given by the Counsel for Brazil to a question from the Panel.
He stated that PROEX payments were not made to cover the development costs of EMBRAER
aircraft. He also stated that these export subsidies were not paid to cover EMBRAER’s
higher cost of funds, but rather they covered the costs incurred when EMBRAER raised the
funds necessary to finance the sales of its aircraft. Canada argues that, implicit in
this, is Brazil’s interpretation of the phrase "payment by [governments] of all
or part of the costs incurred by exporters or financial institutions in obtaining
credits" in item (k): the costs incurred by – in this case – the exporter,
EMBRAER, in raising funds to finance the purchase of its product.
4.75 Canada submits that PROEX subsidies have little
to do with EMBRAER’s cost of raising funds to provide financing. Rather, PROEX export
subsidies are in practice paid when non-Brazilian purchasers finance their purchases
through non-Brazilian lenders. Canada gives the example of Mesa Airlines, an American
regional airline and a purchaser of EMBRAER aircraft 58. Mesa financed its purchase through CoreStates Bank, a US
bank. EMBRAER did not provide the financing in that transaction, but the transaction
received PROEX support. The PROEX subsidy in that case was, therefore, not a
"payment" of the costs incurred by the exporter, EMBRAER within the meaning of
item (k) of the Illustrative List. PROEX export subsidies are not payments to cover the
costs incurred by exporters or Brazilian financial institutions in raising funds used for
financing purchases. They are simply cash grants made for the benefit of purchasers of
Brazilian exported products. They do not, therefore, fall within the first paragraph of
item (k).
4.76 Brazil submits that PROEX export subsidies
constitute a payment by the Government of Brazil of all or part of the costs incurred by
EMBRAER or financial institutions in obtaining credits within the meaning of item (k) of
the Illustrative List. With respect to Canada's argument that PROEX payments do not relate
to costs incurred by Brazilian financial institutions in obtaining credits, Brazil
argues that the relevant language of item (k) refers to "financial
institutions," not to "Brazilian financial institutions" or financial
institutions of any nationality. At any rate, contrary to the assertions of Canada, in
most of the transactions concluded by EMBRAER, "the lender was a Brazilian lender
inside Brazil, not a financial institution outside Brazil."
4.77 With respect to Canada's argument that PROEX
subsidies has little to do with EMBRAER's cost of raising funds, and in
response to a question from the Panel 59,
Brazil stated that it is undisputed that in international transactions involving the sale
of aircraft, it is the exporter who has to provide or arrange for financing. The financial
package offered by the exporter is second only to the characteristics of the aircraft in
making the sale. Brazil asserts that the Canadian manufacturer of regional civil aircraft,
Bombardier, has, or is in the process of establishing its own subsidiary that will provide
or arrange for financing to its customers. EMBRAER does not have such a subsidiary, but is
not relieved from the necessity of putting together a financing package for its customers
that is competitive with what Bombardier offers. In doing so, it incurs costs, such as
those documented by Dr. Finan with regard to the Continental Express transaction. In this
situation, EMBRAER may be viewed as "obtaining" funds with which to provide a
financial package, or "obtaining" an acceptable package from other sources for
the benefit of its customers. 60
4.78 In its response to a question from the Panel
whether there was a difference between the "cost of credit" and "the cost
of obtaining credits 61,"
Brazil stated that in the context of item (k), there is no meaningful distinction between
the cost of credit and the cost of obtaining credit. To "obtain," according to
Webster’s Third International Dictionary, is "to gain or attain
possession." A synonym is "get." When one "gets" a loan one
"obtains" it. That the cost of this action may be spread out over the life of
the loan does not change the fact that it is a cost of obtaining the loan in the first
instance. The "risk spread" added to the basic interest rate is certainly a cost
to "obtain" the credit. This cost varies from borrower to borrower depending on
their respective creditworthiness. The "cost of credit" is the basic cost
charged for borrowers with the best credit ratings. A Brazilian company or financial
institution carries the "risk spread" of Brazil on top of its own credit rating.
This is a cost to "obtain" the credit, which would not be granted unless the
borrower agreed to pay for it, on top of the basic "cost of credit," over the
life of the loan. Other borrowers with better credit ratings - or that do not carry a high
country "risk spread" - would not have to pay for these costs. Therefore the
"risk spread" could not be interpreted as part of the "cost of
credit," since it is applied in a differentiated manner.
3. Whether PROEX payments are used to secure a material advantage in
the field of export credit terms
4.79 Brazil contends that although PROEX
payments confer a benefit within the meaning of Article 1 of the SCM Agreement, "they
are not used to secure a material advantage in the field of export credit terms."
Brazil notes that the term "material advantage" is not defined in the Agreement,
and argues that the words should be given their ordinary meaning. Brazil relies on the
Webster’s Third New International Dictionary of the English Language, which defines
"material" as "being of real importance or great consequence."
"Advantage" is defined as "a more favorable or improved condition."
Thus, any more favourable or improved condition secured by the Brazilian aircraft exporter
or financial institutions as a result of PROEX must be of real importance or of great
consequence, before they would be prohibited under the SCM Agreement.
4.80 Brazil argues that the concept of material
advantage includes comparison – advantage vis-à-vis someone or something. Two
points of comparison are relevant to the determination in this dispute of whether PROEX is
used by Brazil to secure a material advantage in the field of export credit terms: (i)
Brazil risk and (ii) Canada’s subsidies to Bombardier. On either measure alone, PROEX
provides no material advantage. To demonstrate that PROEX subsidies do not secure a
material advantage "in the field of export credit terms", Brazil introduced as
evidence a report prepared by Dr. Finan ("Finan Report").
(a)Summary of the Evidence Presented by Dr. Finan
4.81 The Finan Report states that this dispute
involves a unique set of facts and circumstances that must be acknowledged in order to
correctly evaluate material advantage in the field of export credits, particularly as it
relates to regional aircraft exports 62.
First, the dispute is between a developed country and a developing country both of which
maintain export subsidy programmes that benefit regional aircraft exports. Second, the
dispute concerns the export of capital goods, which have a different set of financing
requirements relative to non-capital goods or agricultural products. Financing is of a
longer term (15 years or longer) and is based on "buyer credits." Third, the
dispute concerns the only two manufacturers of 50-seat regional jet aircraft in the world.
Bombardier was the first to manufacturer this type of aircraft. Canada defined the terms
for export credits that Brazil was required to match. It concludes that Brazil cannot
match Canada’s terms in kind 63.
4.82 The Finan Report identifies the major elements of
export credit terms as (1) the export price of the aircraft, (2) the term of
financing, and (3) interest rates, including factors that determine the interest rate.
4.83 In evaluating material advantage, the Finan
Report evaluated two types of Canadian subsidies, direct and indirect, that benefited
Canada’s export credit terms 64.
The Finan Report assesses whether there is a material advantage given to Brazil through
the export credits of the PROEX programme. To do so, the Finan Report compares PROEX
benefits to the aircraft buyer with the benefits provided to the aircraft buyer by all of
Canada’s subsidy programmes that directly or indirectly affect export credit terms 65.
4.84 The Finan Report submits that evaluating material
advantage requires evaluating one country’s set of export credit terms relative to
another country’s set of export credit terms by applying a consistent methodology
appropriate under the facts and circumstances. The Finan Report further states that
determining the present value of total subsidy benefits is the most reliable way to
determine material advantage in this dispute, with the present value of PROEX benefits
compared to the reduction in the present value of financing costs resulting from all of
Canada’s export supports 66.
The Finan Report argues that Bombardier, its suppliers, and its customers have been and
are subsidized by the Government of Canada and the Provinces of Ontario and Québec
through a variety of programmes; Bombardier was the first to enter the regional jet
market, and therefore Canada was able to define any and all export credit terms. The Finan
Report claims that Brazil was forced to match Canada’s terms in order to be
competitive; and many regional aircraft buyers have weak credit ratings, and thus rely
heavily on the export credit packages offered by official export credit agencies 67. The Finan Report submits that, because
of Brazil’s weak credit rating and sovereign risk element, Brazil is unable to match
Canada’s export credit terms in kind 68.
4.85 The second part of the Finan Report discusses the
PROEX programme, evaluates why it is necessary in Brazil’s situation, and details
Brazil risk. The Finan Report concludes that a Brazilian commercial entity cannot avoid
bearing the additional cost of Brazil sovereign risk when it raises financial capital or
finances a purchase or a sale 69.
4.86 The third part of the Finan Report evaluates the
effect of PROEX on the Brazilian regional aircraft market and concludes that PROEX does
not provide Brazil with a material advantage in the field of export credit terms for
exports of regional aircraft. Based on data provided by EMBRAER, the Brazilian aircraft
manufacturer, the Finan Report details the firm orders for all manufacturers of the
50-seat regional jet, both before and after PROEX came into effect (there are only two
manufacturers of the 50-seat regional jet: Bombardier of Canada and EMBRAER of Brazil).
The Finan Report concludes that even after PROEX began to be applied to exports of
regional aircraft, Bombardier’s CRJ still captured 57 per cent of the firm
orders 70.
4.87 The Finan Report also attempted to quantify the
Canadian export subsidies received by Bombardier or its suppliers, although it notes at
page 3.23 that benefits presented in the report "constitute only a sample, or subset,
of the total benefits conveyed to Bombardier by the Government of Canada and the
provincial governments of Ontario and Québec" because of a lack of transparency in
Canada’s programmes 71.
According to the Finan Report, Canada’s total direct and indirect benefits provided
to Bombardier ranged between US$1,883 thousand to US$2,446 thousand.
4.88 The Finan Report states that based on the
Canadian supports, the total present value for reduced financing costs per CRJ
(Bombardier’s 50 seat jet aircraft) to be delivered in 1998 ranges from
US$4.4 million to US$6.0 million. Again, Brazil argues these figures are only a
subset of the actual figures, which may be much higher 72. Hence, the Finan Report concluded that EMBRAER through the
PROEX programme did not achieve a material advantage in export credit terms relative to
Bombardier 73.
4.89 The Finan Report provided six appendices to
support its presentation, consisting of market return benchmarks, Bombardier regional
aircraft deliveries, EMBRAER’s cost of capital analysis, EMBRAER-Brazil specific
risk, calculation of present value of Canadian subsidies, and a description of the method
applied to determine material advantage.
4.90 Brazil argues that given the language of
the SCM Agreement, together with the unique set of facts and circumstances surrounding
this dispute, there are three methods of determining that PROEX provides no material
advantage: (1) if the lender is situated in Brazil, because there is Brazil risk in the
cost of funds for Brazilian lending institutions; (2) if the lender is outside Brazil,
because EMBRAER bears increased costs associated with obtaining financing for its exports
due to Brazil risk; and (3) independent of the location of the lender, i.e., even if the
lender is outside Brazil, Canada’s subsidies to Bombardier are greater than the
assistance provided by PROEX 74.
4.91 Relying on Dr. Finan’s presentation at the
Second Panel Meeting, Brazil states that there are four categories of transactions
involving EMBRAER aircraft purchases: (1) purchaser is located inside Brazil and financed
by a Brazilian bank; (2) purchaser is located inside Brazil and financed by a
non-Brazilian bank; (3) purchaser is located outside Brazil and financed by a Brazilian
bank; and (4) purchaser is located outside Brazil and financed by a non-Brazil based
financing source 75. Brazil argues
that all deals financed by a bank in Brazil contain Brazil risk (i.e., situations 1 and 3) 76. Brazil also argued that there are no
purchasers inside Brazil financed by a non-Brazilian bank that benefit from PROEX (i.e.,
situation 2) 77. Brazil also argues
that deals outside of Brazil contain Brazil risk (i.e., situation 4) 78. Brazil contends the first two
categories of transactions are not at issue in this dispute (i.e., situations 1 and 2).
According to Brazil, with respect to the latter two categories of transactions (i.e.
situations 3 and 4), deals financed by a bank in Brazil and deals financed outside Brazil
both contain Brazil risk-related costs. In such deals, the Brazil risk-related costs are
an amount greater than PROEX’s assistance. According to Brazil, when compared to the
subsidies given to Bombardier, clearly PROEX also provides no material advantage 79.Therefore, when the two are considered
together, it is clear that it is Canada’s programmes, not Brazil’s, that secure
a material advantage in the field of export credit terms.
4.92 Canada submits that PROEX payments are
used to secure a material advantage in the field of export credit terms. In Canada's view,
a material advantage is "secured when a government provides export credit at rates
that are lower than those prevailing in the international financing market. A material
advantage is secured, in addition, when payments by governments to exporters or financial
institutions result in interest rates that are lower than those prevailing in the
international market." Canada submits in that connection that the relevant rates to
be considered in the international market are the London Inter Bank Offer Rate (LIBOR ) or
the United States (US) Treasuries rate plus a spread that reflects the credit risk of the
transaction. 80
4.93 Canada confirms its view that the appropriate
benchmark is either LIBOR or US Treasury rates in its response to the following question
from the Panel: "Should material advantage be assessed by reference to (a) whether
the export credit terms with respect to a given transaction are improved by virtue of the
government payment or (b) whether the export credit terms available to a purchaser for a
given transaction benefiting from a payment are better than the terms that would have been
available to that purchaser with respect to transactions involving competing
products?":
"[N]either (a) nor (b) is appropriate for assessing material
advantage. With respect to (a), "the government payment" in question could be
applied in combination with an officially supported export credit. In this sense, the
question would not be whether any export credit terms for a given transaction have been
improved, but whether the final export credit terms are in line with terms and conditions
available in the international financing market. Under (b), the point of reference would
continue to be the export credit terms available to a competing product in line with the
international financing market and clearly not subsidised credit terms for other products.
To measure export credit terms as against subsidised credit terms would be unworkable: an
export credit could be found to be an export subsidy if measured against one competing
product, but not so against another. It is Canada’s position that material advantage
should be assessed with respect to the international financing market. The international
financing market is defined by the benchmarks that have been discussed by Canada (LIBOR or
US Treasuries plus a spread that reflects the credit risk of transaction )."
(b) "Brazil risk"
4.94 Brazil submits that in financial markets,
the debt securities of all governments bear a "sovereign risk" premium. This
risk premium reflects the views of participants in those markets as to the likelihood of
repayment on schedule. Generally the debt securities of the United States bear no risk
premium, and, as a result, the sovereign risk of the securities of other governments may
be measured as the amount of additional interest markets demand for those securities
compared to the securities of the United States. For developed countries, such as Canada,
this premium is extremely low or non-existent. For developing countries, such as Brazil,
this premium can be quite high. The risk premium is normally referred to in terms of
"basis points," with 100 basis points equalling one per cent. As described in
detail in the Finan Report 81 ,
which has been tendered in evidence by Brazil, the risk premium for securities issued by
the Government of Brazil in recent years has varied between a high of 1,300 basis points
in the first quarter of 1995 and a low of 400 basis points in the third quarter of 1997.
However, by the third quarter of 1998, Brazilian sovereign risk rose again to over 1,000
basis points, an increase in the spread for Brazil risk of 6 percentage points or 600
basis points due to the Asian debt crisis and the Russian debt default and generally the
volatility in the financial markets.
4.95 Brazil further notes that during the same period,
Canada’s spread remained virtually unchanged. The reality with which Brazil and other
developing countries must live is that with globalised financial markets, when others
catch a cold, they may catch pneumonia. The risk markets assign to both Brazilian and
Canadian debt is reflected in the ratings they receive from financial services. According
to Standard & Poor’s, the rating on Brazil’s foreign currency denominated
debt is BB-, while the rating on Canada’s is AA+. What this means in real terms is
that in the third quarter of 1998 the yield on the Brazilian 10-year C-Bond was 15.75 per
cent, while the yield on the Canadian 10-year bond was 5.25 per cent, or
10.5 percentage points lower. Those 10.5 percentage points represent the difference
between Brazil risk and Canada risk.
4.96 Brazil submits that the "Brazil risk"
has a dramatic and severe impact on all financial terms and transactions involving Brazil,
including export credit terms. The adverse impact in the area of export credit terms is
particularly severe in the regional aircraft industry, where market custom, established by
developed countries long before Brazil became a factor, is that the manufacturer supports
the financing of its aircraft. Brazil submits that one consequence of Brazil risk is that
EMBRAER cannot compete in providing manufacturer’s supports that match in-kind
Bombardier’s 82. Brazil
submits that aircraft trade is conducted in US dollars, which means that direct financing
must be made available in U.S dollars. The OECD rate for US dollars is in the vicinity of
6 per cent. The OECD Agreement permits members to provide financing at the OECD rate for
US dollars regardless of a member's cost of acquiring dollars. While item (k) of the
Illustrative List essentially extends this permission to all WTO Members regardless of
whether they are members of the OECD, for Brazil and other developing countries, the cost
of acquiring dollars is very high because of the compensation demanded by lenders for
"Brazil risk [or developing country risk]."
4.97 First, Brazil argued at the first meeting of the
Panel 83 that paragraph (k) makes
clear that the developed country Members of the WTO that are also Members of the
Organization for Economic Cooperation and Development – the OECD – have taken
care of themselves. It does so by providing that an export credit practice which is in
conformity with the interest rates provisions of "an international undertaking on
official export credits to which at least twelve original Members to this Agreement are
parties as of 1 January 1979" shall not be considered to be a prohibited export
subsidy. This is a reference to the so-called OECD Agreement or Guidelines.
4.98 Brazil argued that developing countries did not
bargain for the OECD "alternative" in the Uruguay Round. They are not members of
the OECD. They have no voice in the OECD. Developing countries bargained for the temporary
exemption of Article 27, and for the "material advantage" language of item (k)
precisely because, when Article 27’s protection expires, they needed something they
could afford in view of the fact that the developed countries had their own self-designed
exemption, which is presently in the second paragraph of item (k).
4.99 Brazil asserts that while it could take advantage
of the OECD provisions in paragraph (k) of Annex I by borrowing US dollars for
approximately 16 per cent and lending them for approximately 6 per cent, absorbing a
10 per cent loss in the process, this is hardly a viable alternative from a fiscal
viewpoint. PROEX is more affordable to Brazil 84
v. Brazil further submits that apart from the high cost of borrowing dollars, the
OECD exemption would not have helped it to counteract the incentives provided by Canada to
Bombardier, especially considering that under the OECD Guidelines, the maximum term for
aircraft financing permitted to OECD members is 10 years, yet Canada established the
practice in the regional aircraft industry of granting 15 year financing to Bombardier 85 . Brazil further notes that
"developing countries did not bargain for the OECD "alternative" during the
Uruguay Round…[T]hey bargained for the temporary exemption of Article 27, and for the
"material advantage" language of item (k) precisely because, when Article 27's
protection expires, they need something they could afford in view of the fact that the
developed countries had their own self-designed exemption which is presently in the second
paragraph of item (k)." In response to a direct question from the panel 86 , to support its position that
developed country Members bargained for the "material advantage" language of
item (k), Brazil submitted a July 7, 1990 proposal made during the Uruguay Round by a
developed country to delete the ‘material advantage’ language of item (k).
Brazil and other developing countries opposed this initiative. The proposal was rejected
and the language remained in item (k) 87
.
4.100 Brazil submits that no private institution in
any country can expect to borrow at rates below those made available to its government.
The risk premium markets assign to private borrowers includes the sovereign risk of their
governments in addition to their own enterprise-specific risk. Thus, any Brazilian
exporter or financial institution attempting to obtain funds for export credits is
handicapped from the start with a 10.5 per cent or 1,050 basis point risk premium compared
to a Canadian counterpart, or a counterpart in other developed countries. Relying on the
Finan report, Brazil asserts that to support the financing of EMBRAER aircraft, EMBRAER
and Brazilian financial institutions must pay a premium of 1,000 basis points or more
above what Bombardier and Canadian financial institutions must pay. PROEX offsets only 380
basis points of this premium. Brazil deduces that "[a]t current rates, approximately
670 basis points (1,050-380) [is] left for the private sector to absorb. Thus, in these
circumstances, when the lender is inside Brazil, PROEX provides no material advantage and
thus is not a prohibited export subsidy. Moreover, in the transactions when the lender was
inside Brazil, the actual interest rate was always above LIBOR or the OECD rate in
practice."
4.101 Further, according to Brazil, when the lender is
a non-Brazilian financial institution outside of Brazil, it may be true that Brazil risk
does not apply to that lender’s cost of funds. But Brazil risk continues to apply to
EMBRAER, and paragraph (k) of Annex I permits equalization payments that reflect the
costs incurred by exporters such as EMBRAER as well as by financial institutions 88 . To establish how "Brazil
risk" is present even in a transaction when the lender is outside of Brazil, Brazil
submitted detailed, confidential business information to the Panel and its expert, Dr.
Finan, gave a detailed presentation to the Panel. 89
4.102 Canada challenges the accuracy of Dr.
Finan's presentations 90 and
submits that the argument by Brazil that PROEX subsidies are necessary to counter alleged
developed country market advantages is baseless and unwarranted on the facts. Canada
further submits that "even assuming that the "Brazil risk" argument is
relevant, the argument must fail… If the [true] purpose of PROEX export subsidies
were to reduce the "Brazil risk," they would be paid to the Brazilian lender
that incurs the higher costs of borrowing, as indeed item (k) provides. Such payments
would be made to a purchaser only in the unlikely circumstance that it has borrowed at a
higher interest rate than that prevailing in the international market. Any such subsidy
would directly reflect and offset such risk premium."
4.103 Canada "challenges vigorously"
Brazil's assertion that it does not "seriously challenge" its argument that
"when the lender is inside of Brazil, PROEX provides no material advantage and thus
is not a prohibited export subsidy." Canada also denies that it accepts that
"when the lender was inside Brazil, the actual interest rate was always above LIBOR
or the OECD rate in practice." Canada submits that when assessing material advantage,
"the question is not whether the lender faces a "Brazil risk" (or any other
special cost) compared to other potential lenders. Rather, it is whether as a result of
the PROEX export subsidy -- whatever the motivation -- the borrower obtains terms that are
better than those available in the international financing market." Canada further
submits that "in most, if not all, instances where financing is done in Brazil, it is
done through…[Brazil's National Bank for Economic and Social Development] (BNDES).
And when BNDES lends money to airlines, it does so at international commercial rates.
Therefore, PROEX export subsidies, when applied to financing offered by BNDES, bring down
the already commercial interest rate to one that is significantly below the international
market." Canada provided evidence which, it submitted, offered proof that PROEX
payments reduced the interest rate levels to be paid by a number of airlines significantly
below "both LIBOR plus a credit risk premium and the OECD rates. 91"
"Continue on to Main Arguments of the Parties, Section
4.104"
54Ibid at paras.
42-44.
55 Ibid at paras.
47-49 and 54-56
56Ibid.
57 Ibid.
58 Canada also
gives the examples of three other airlines, namely Comair, Skywest and Trans State
Airlines. In the case of the first two airlines, Canada argues that PROEX payments
"simply reduced the rate of interest paid, to one significantly below that prevaling
in the international market. Canada claims that PROEX export subsidies were not, in those
cases, "payments"made to defray "costs incurred"by the exporter,
EMBRAER, within the meaning of item (k)."In the case of the last transaction, a term
issued by the financing bank, Bank of America, Canada claims that "offered to arrange
the payment of PROEX subsidies was not conditional on EMBRAER providing the financing. Nor
was it conditional on a Brazilian financial institution providing the financing."
59 The question
from the panel was as follows: "Please explain fully your view that PROEX payments
are payments of "costs incurred by exporters or financial institutions in obtaining
credits": In what sense does EMBRAER or a financial institution, whether Brazilian or
other, "obtain" credits? Could it not rather be said in this case that financial
institutions "provide"credits which are "obtained"by the
purchaser?"
60"Brazil
Risk"is explained in detail in paras. 4.94 - 4.101.
61 The question
from the panel was phrased as follows: "In what sense can interest rate payments be
seen as costs of "obtaining"credits as opposed to costs of carrying them? Is
there a distinction between the "cost of credit"and the "cost of obtaining
credits?"
62 Finan Report,
pp. 1.1 - 1.3.
63 Ibid at p. 3.20.
64 Ibid at p. 18.
65 Ibid at pp. 1.8
- 1.9
66 Ibid at pp. 1.9.
67 Ibid at p. 1.11.
68 Ibid at p. 1.13.
69 Ibid at pp.2.1 -
2.4.
70Ibid at pp. 3.1 -
3.2: the table is found at p.32.
71 Ibid at p.3.9.
72 Ibid at p. 3.23.
73 Ibid at pp.
3.22, 3.25; the table is found at p. 3.25 of the Finan Report.
74 Oral
Presentation of Dr. William F. Finan at Second Panel Meeting, p.3.
75 Oral
presentation of Dr. William F. Fianan At Second Panel Meeting (Graphic Presented to the
Panel).
76 Brazil's
Response to Questions From the PAnel, No. 12.
77 Ibid.
78 Ibid; Dr.
Finan's Oral Statement at the Second Meeting of the Panel (statement contains confidential
business information).
79Supra note 77.
80 In a floating
rate transaction, the lender sets an interest rate thet will be moved up or down in
relation to general movements in interest rates in the wider economy. In the floating-rate
aircraft financing transactions, the benchmark used (to reflect the "general
movements in interest rates") is the three-month or six-month LIBOR. LIBOR is the
rate of interest at which banks offer to lend money to one another in the
"wholesale" money markets in the City of London. Although LIBOR figures are
available for the major currencies, the US dollar tends to be the currency of choice in
international financing activities related to aircraft. They tend to quote interest rates
they would charge as "basis points (or bps) above LIBO"."These "basis
points above LIBOR"are what is known as the "spread"charged by a lender --
tha additional charge that reflects the credit risk of the transaction. This credit risk
is based on the credit quality of the borrower and incorporates other criteria determined
by the lender to be relevant to the transaction, such as the value of any asset being
financed, any security interests in the assets of the borrower, or any third-party
guarantees. If three-month LIBOR were (for example) six per-cent, a bank may choose to
lend to a purchaser at (for example) seven and a quarter per cent, or 125 basis points
(bps) above three-month LIBOR. 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.
81 The credibility
of the Finan Report has been contested by Canada; see paras. 4.122 and 4.139.
82 Finan Report, p.
3.20.
83 Brazil's First
Oral Statement, para. 21.
84 In this respect,
the Panel posed the following question to the parties: "Brazil argues in its first
oral statement (paras. 22-25) that developing country Members may not in practice act in
accordance with the OECD Arrangement because it is too expensive for developing countries
to raise dollars at a high rate of interest reflecting sovereign risk and then lend those
dollars at Arrangement interest rates. The Arrangement, however, applies to öfficial
support"whether given by means of "direct credits/financing, refinancing,
interest rate support, gurantee or insurance"(emphasis added). In the case of
interest rate support, the supporting government presumably need not either provide the
principal amount itself or guarantee its repayment. Accordingly, how is Brazil's argument
relevant in the context of "interest rate support"? Brazil's response was as
follows: "[The] argument was made in the context of the first sentence of the
first paragraph of item (k). With regard to interest rate support, Brazil would note that
David J. Blair, in Trade Negotiations in the OECD... states at page 42: "There are a
number of ways in which governments can give support to export financing. Official support
may include insurance against commercial and political risk, guarantees of repayment, and
insurance against cost-inflation and foreign exchange risk. Governments may also give
financial support in the form of direct credits, refinancing and the subsidisation of
interest rates... Brazil...[reiterates] that, in practice all financing with lenders
inside Brazil has been done at LIBOR-plus or the CIRR rate"(emphasis in original).
85See para. 4.137.
86 Brazil's
Response to Questions From the Panel, No. 42.
87 See para. 4.98
of this report.
88 Brazil's First
Oral Statement, para. 38; Brazil's Second Written Submission, para. 52. Brazil argues that
EMBRAER itself bears Brazil risk, and therefore that the costs it incurs in obtaining
credits for its customers outside of Brazil, from lenders outside of Brazil
89 See para. 4.139
90 See paras. 4.122
and 4.139.
91In particular,
Canada noted that where Brazil has provided financing - for example, in the cases of
Skywest and British Regional - The "post-equalisation" rate of interest offered
by Brazil has been significantly below the OECD Consensus benchmark. Indeed, the effective
interest rates offered by Brazil have been lower than the applicable yield on US
Treasuries, themselves benchmarks for the OECD rate. Canada argued that the evidence it
had adduced had not been contradicted or even questioned by Brazil. Further, the PROEX
subsidy for aircraft is fixed at 3.8 per cent, regardless of the interest rate in the
financing. To be operable in a manner consistent with the Consensus, the PROEX programme
would have to provide for the adjustment of the subsidy in each transaction, so that it
would compensate only for any excess in the financing rate over the OECD Consensus rate.
|
