UNITED STATES � FINAL COUNTERVAILING DUTY
 DETERMINATION WITH RESPECT TO CERTAIN
 SOFTWOOD LUMBER FROM CANADA
 

Report of the Panel

(Continued)  

D. FIRST ORAL STATEMENT OF THE UNITED STATES

4.147 The following summarizes the United States' arguments in its first oral statement.

1. Opening Statement of the United States of America at the First Meeting of the Panel

(a) Financial Contribution

4.148 The first legal issue in this dispute is whether Canadian provincial timber sales systems constitute the provision of a �good� within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement. The text and context of Article 1.1(a)(1)(iii), and the object and purpose of the SCM Agreement, all lead inexorably to the conclusion that standing timber is a �good.�

4.149 Canada argues that �goods� is limited to items that are tradeable across borders and subject to tariff classification. To the contrary, the ordinary meaning of �goods� is broad, encompassing all property and possessions, including things to be severed from the land, such as standing timber. The sole exclusion in Article 1.1(a)(1)(iii) for �general infrastructure� underscores the intent that the provision sweep broadly. �Infrastructure� is not tradeable across borders. Nevertheless, infrastructure that is not �general� must fall within Article 1.1(a)(1)(iii). To conclude otherwise is to render the explicit exclusion for infrastructure that is �general� entirely meaningless.

4.150 The uncontested facts leave no doubt that the provinces sell timber. There is one reason and one reason only that companies enter into provincial timber contracts, which we generally refer to as �tenures.� They do so to obtain the government-owned timber for their mills. Through tenures, the provinces are providing a good � timber � to lumber producers. Accordingly, the provinces provide a financial contribution within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement.

(b) Benefit

4.151 We turn to the methodology used to determine whether and to what extent the provinces confer a benefit on lumber producers through the sale of timber for less than adequate remuneration. This methodology poses two distinct issues. First, there is a question of legal interpretation of �benefit� within the meaning of the SCM Agreement. Second, there is the issue of the application of that legal concept to the particular facts of this case. It is imperative to examine the legal question before turning to the facts.

4.152 A financial contribution confers a �benefit� if it �makes the recipient �better off� than it would otherwise have been, absent that contribution.�¹⁵ In determining whether a benefit has been conferred, the �marketplace� is the appropriate basis for comparison, i.e., the issue is whether �the recipient has received a �financial contribution� on terms more favourable than those available to the recipient in the market.�¹⁶

4.153 The guidelines in Article 14(d) of the SCM state that a benefit is conferred if the government provides the good for �less than adequate remuneration.� Article 14(d) also states that the adequacy of remuneration shall be determined in relation to �prevailing market conditions� for the good in the country of provision. The concept of a comparison �market� therefore is central to the concept of �benefit� generally, and to adequate remuneration specifically. The Brazil�Aircraft panel concluded that the concept of a comparison market necessarily means a �commercial market, i.e., a market undistorted by government intervention.�¹⁷ The United States agrees.

4.154 As the EC states in its third-party submission, �market� conditions exist where prices are �determined by independent operators following the principles of supply and demand.� Thus, as the EC implicitly acknowledges, not all observed prices are necessarily �market� prices. We agree. �Market� prices are prices between independent buyers and sellers in a competitive market where prices are determined by the forces of supply and demand. Such prices represent what is commonly referred to as �fair market value.� It therefore follows logically that adequate remuneration is fair market value. That is also how the term adequate remuneration is defined in Canadian law.

4.155 Article 14(d) sets forth the principle underlying the adequate remuneration inquiry � it must be made �in relation to prevailing market conditions� in the country of provision. However, Article 14(d) does not set out rules governing the specific types of data that may be used in conducting that analysis.

4.156 Where reliable commercial market prices are available in the country of provision, ignoring such prices would be inconsistent with Article 14(d). Where, however, no such prices exist or are unreliable, an investigating authority may use prices commercially available on world markets as the basis for an assessment of the adequacy of remuneration, provided that those prices are informative as to the fair market value of the goods in the country of provision.

4.157 There is no real dispute that there are circumstances under which an investigating authority may look to sources outside the country of provision for data to assess the fair market value of goods in the country of provision. The US � Softwood Lumber III panel,¹⁸ Canada, the EC and Japan have all implicitly or explicitly acknowledged this. The real issue in this dispute is what factual circumstances warrant the use of price data from sources outside the country of provision to determine the fair market value of goods in the country of provision.

4.158 To determine the adequacy of remuneration in the underlying investigation, the United States calculated province-specific, species-specific market benchmark prices for stumpage. The United States requested data on private market transactions for stumpage in each province for the purpose of calculating those market benchmark prices. Three of the six provinces � Alberta, Manitoba and Saskatchewan � did not provide any data on private market prices for stumpage. Thus, there should be no dispute that the United States acted consistently with Article 14(d) in using commercially available prices from sources outside Canada to determine the fair market value of timber in those provinces.

4.159 British Columbia (�B.C.�) submitted a survey containing a few average prices. The volume of the private timber on which those averages were based, however, represented less than one-half of one per cent of the total timber harvested by the survey respondents. Moreover, the survey did not contain the detail or underlying support that would be necessary to calculate market benchmark prices. Ontario provided a limited survey and analysis of private stumpage sales in the province. Similarly, Quebec submitted an average price for private stumpage in the province, which was based on a survey.

4.160 The United States determined that there were no commercial market conditions � that is, a market undistorted by the government�s financial contribution � in any of the provinces. The evidence, including the governments� dominant market share, the lack of incentive for sawmills to pay more for private stumpage than they pay the government, and statements by provincial officials and forestry economists concerning the impact of the government prices on the sale of private timber, is more than adequate to support that determination. Thus, although there is data on observed prices in certain provinces, the record demonstrates that those observed prices are not commercial market prices. Therefore, the United States acted consistently with Article 14(d) in using data from sources outside of Canada as the starting point for determining the fair market value of timber in Canada. Moreover, numerous adjustments were made to reflect conditions of sale in Canada.

(c) Calculation Issues

4.161 Canada claims that the United States was required under Article 19 of the SCM and Article VI:3 of the GATT 1994 to conduct an �upstream� subsidy analysis. Canada makes this claim with respect to two distinct situations.

4.162 With respect to remanufacturers: The United States determined the total amount of the benefit, but did not determine what portion of the benefit individual sawmills or remanufacturers received. The United States allocated the total subsidy benefit over all sales of the products resulting from the lumber production process. The total amount of the subsidy benefit does not change, however, regardless of how the benefit is allocated. Thus, allocating a portion of the benefit to remanufacturers cannot overstate the total subsidy benefit. Moreover, nothing in Article 19 precludes this method of calculation. Article 19.3 specifically contemplates that a producer�s exports may be subject to countervailing duties without knowing whether or to what extent that particular producer received a benefit. Article 19.3 simply obligates Members to provide expedited reviews for such exporters to calculate individual subsidy rates.

4.163 With respect to the second �upstream� subsidy situation, i.e., the alleged independent loggers: This is the only situation that could have any impact on the calculation � rather than the allocation � of the total amount of the benefit to producers of the subject merchandise. The record evidence indicates, however, that sales by independent loggers could only account for a very small portion of the volume of Crown timber entering sawmills. In addition, the evidence suggests that all or most of the sales by independent loggers may not be at arm�s-length. Moreover, an upstream subsidy analysis requires company-specific data and analysis. Canada�s claim that the United States was required to conduct this type of company-specific analysis in the investigation is without foundation in Article 19 of the SCM or Article VI:3 of GATT 1994.

(d) Specificity

4.164 Pursuant to Article 2.1(c) of the SCM Agreement, a subsidy is specific when the users of the subsidy are limited to certain enterprises or industries or to a limited group of enterprises or industries. The users of provincial stumpage are limited to timber processing facilities, which constitute a very limited group of industries. In accordance with Article 2.1, therefore, the subsidy from provincial stumpage is specific. Canada�s claims to the contrary are based on its own definition of specificity, not the definition in Article 2.1. Article 2.1 does not require an investigation into the motives of Members that provide subsidies, does not require an analysis of the number of products made by the users of the subsidy, and does not require that a subsidy be limited to the producers of the subject merchandise, or that a �group of industries� must share common characteristics.

2. Closing Statement of the United States of America at the First Meeting of the Panel

(a) Financial Contribution

4.165 Canada criticizes the United States� reliance on the definition of �goods� in Black�s Law Dictionary, which cross-references the US Uniform Commercial Code (�UCC�). Canada, however, relies on that same definition. Nevertheless, Canada states that the UCC provision cross-referenced in Black�s �expressly excludes� standing timber, �except in certain limited circumstances that do not apply here.� The UCC is, of course, not controlling in this forum. However, the relevant provision reveals that sales of standing timber are expressly �included� � not �excluded� � from the term �goods� as used in the UCC. We also refer the Panel to our first written submission, which quotes a similar definition of �goods� in the British Columbia Sale of Goods Act.

(b) Benefit

4.166 With respect to record evidence concerning private stumpage prices, Canada makes a number of statements at the first Panel meeting which we would like to comment on:

4.167 Canada asserts that Timber Damage Assessments are based on private transactions representing approximately 6 per cent of Alberta�s timber harvest. According to Alberta�s questionnaire response, however, only 1 per cent of the harvest in Alberta comes from private land.

4.168 Canada states that B.C. submitted evidence that demonstrates that B.C. operates its stumpage system consistent with market principles. That evidence merely established that B.C. made a profit on its timber sales, which does not mean that it is receiving adequate remuneration.

4.169 Regarding paragraph 58 of Canada�s oral statement, we have several comments:

4.170 The Final Determination analyzes the reliability of Canadian private timber prices in detail.

4.171 The Economists, Inc. study that the United States relied on concludes that the �existence of an administered market that is willing to supply the preponderance of market demand at an artificially low price drives the price that can be attained in the non-administrative sector below the level that would obtain if the administered market were not subsidized.�

4.172 The student thesis Canada refers to is actually a 1995 doctoral dissertation that analyzes data as recently as 1993.

(c) Market Distortion

4.173 Canada claims that the United States ignored Canada�s evidence �and simply assumed trade distortion.� Rather, the United States determined that US law does not require an analysis of whether a subsidy has market distorting effects. Likewise, there is no obligation in the SCM to find the existence of trade distortion to impose countervailing duties.

(d) Calculation Issues

4.174 Canada implies that Article 19.4 effectively imposes obligations with respect to the calculation of the subsidy. Canada, however, fails to cite to any language in Articles 10, 19.1, 19.4 or 32.1 of the SCM or Article VI:3 of the GATT 1994 establishing any such obligations.

4.175 Canada asserts that �there is no single log conversion factor�, yet the Canadian Government itself publishes a single conversion factor. And if the United States had used Canada�s published conversion factor, the calculated subsidy rate would have been greater.

(e) Administrative Reviews

4.176 With respect to administrative reviews, Canada improperly attempts to bring hypothetical future measures by the United States before this Panel. As the US � Softwood Lumber III panel stated, �the WTO dispute settlement system allows a Member to challenge a law as such or its actual application in a particular case, but not its possible future application.�¹⁹

(f) Conclusion

4.177 Canada went to great lengths to criticize the United States for interpreting the words in Article 14(d). For example, Canada criticized the United States for interpreting �adequate remuneration� to mean �fair market value� even though Canada�s own regulations define adequate remuneration as �fair market value.� Having criticized the United States for interpreting the language in Article 14(d), Canada then proceeded to criticize the United States for failing to interpret the specificity provisions in Article 2.1(c). In both instances, the United States interpreted the provisions in accordance with the ordinary meaning of their terms, in context, and applied the provisions accordingly.

E. SECOND WRITTEN SUBMISSION OF CANADA

4.178 The following summarizes Canada's arguments in its second written submission.

1. Financial Contribution

4.179 What is the scope of the word �goods� in Article 1.1(a)(1)(iii) of the SCM Agreement? The word �goods� in Article 1, read in context and in the light of the object and purpose of the WTO Agreement, has the same meaning and scope as the word �goods� elsewhere in the WTO Agreement. And elsewhere in the WTO Agreement, the word �goods� has the same meaning and scope as the word �products� in Article II of GATT 1994. Therefore, interpreted in accordance with the principles of treaty interpretation, the word �goods� in Article 1.1(a)(1)(iii) refers to tradable items that are capable of bearing a tariff classification.

4.180 May the transfer of tree harvesting rights by provinces properly be characterized as �provision of goods� by a government within the meaning of Article 1.1(a)(1)(iii)? Provinces enter into tenure agreements or grant licences to harvest timber. Under these tenures or licences, the harvester has certain proprietary interests in standing trees and must undertake a series of obligations, some of which are related to the land and others to the volume of harvest. Under these tenure agreements and licences, the trees to be harvested are not identified and many of the obligations must be performed regardless of any harvest. This transfer of a right to harvest standing trees does not amount to the provision of goods.

4.181 After having argued that �goods� includes all property, the US now states that this case is not about �intellectual property� or other property rights; that its earlier assertions about �goods� encompassing all property rights were arguments in the alternative. Again misrepresenting the nature of the transactions at issue, it continues to argue that living trees with roots firmly in the ground are �goods� for the purposes of the SCM Agreement.

4.182 In its closing arguments at the First Substantive Meeting, the US suggested that sales of �standing timber� are included in the definition of goods in the UCC. This definition is in turn reproduced, albeit imperfectly, in the Black�s Law Dictionary as one definition for the term �goods�. The US has not credibly contested Canada�s Vienna Convention analysis of the word �goods�. The only issue appears to be whether a particular definition of �goods�, found in the UCC, supports the US position. It does not.

4.183 The UCC definition of �goods� provides that, ��Goods� means all things, including specially manufactured goods, that are movable at the time of identification to a contract for sale and future goods. � The term does not include � general intangibles.� [emphasis added]

4.184 The UCC further provides that, �A contract for the sale apart from the land �[of] other things attached to realty and capable of severance without material harm � or of timber to be cut is a contract for the sale of goods � whether the subject matter is to be severed by the buyer or by the seller even though it forms part of the realty at the time of contracting, and the parties can by identification effect a present sale before severance.� [emphasis added]

4.185 First, this definition flatly contradicts the US assertion that �goods� somehow includes all property rights, including the right to harvest timber. The basic definition is that of movable things; and the basic definition expressly excludes �general intangibles�. Second, the reference to �timber to be cut� is not part of the ordinary meaning. Rather, it is a special provision covering certain �things� that are not ordinarily �movable� and that would not, absent the clarification, be covered by the ordinary meaning of the word �goods�. Third, what is included in the UCC definition is not �standing timber�, but �timber to be cut�. �Timber to be cut� refers to individual trees identified to be cut in a sales contract, as indeed expressly set out in this definition. This is to be contrasted with standing trees that are subject to tenure agreements and licences. Those trees may or may not be cut during the term of the tenure, but in any event are not specifically identified to be cut; after all, they may not even have been planted.

4.186 The inclusion of �timber to be cut� in the UCC definition of a �good� implies the exclusion of all other timber � that is, timber subject to tenures and licenses � from the scope of the UCC definition. In US law, therefore, standing trees are not �goods�. And because the definition of the word �goods� in the UCC does not cover �intangibles�, in US law the right to harvest does not amount to �goods�.

4.187 The US relies heavily on its incorrect assertion that tenure agreements and licences are �timber sales contracts�. The US supports this assertion by arguing that the tenure agreements and licences result in the timber harvesters owning felled timber and paying for that timber. Canada has demonstrated that there is a distinction between tenure agreements and licences on the one hand, and �sales of goods contracts� on the other.

4.188 Under the SCM Agreement, this distinction makes a crucial legal difference. A timber sales contract concerns a contract that identifies individual standing trees to be cut and hauled away. In contrast, a tenure or licence grants a right of harvest in an area of land in return for certain rights and obligations. No trees are identified to be cut � and in fact, due to disease, fire and environmental reasons, it may well be that no trees are cut. And yet, certain forest maintenance and fire protection obligations continue to run for the length of the tenure or licence in the area covered by the tenure or licence.

2. Benefit

4.189 Canada demonstrated that USDOC�s rejection of valid in-country benchmark evidence and its choice of cross-border benchmarks to determine adequacy of remuneration is inconsistent with Articles 1.1(b) and 14(d) of the SCM Agreement. Canada�s submissions are based on three complementary points:

• first, Article 14(d) requires the use of benchmarks that are based on prevailing market conditions in the country of provision;
   

• second, substantial evidence of prevailing market conditions in Canada that could have been used as benchmarks to determine �adequacy of remuneration� was improperly rejected; and
   

• third, the US has offered no valid defence of its actions.

4.190 Canada�s response to the United States� most recent assertions follows the same structure.

4.191 The US claims that this is a dispute over the factual circumstances that warrant the use of price data from sources outside the country of provision. It is not. This is a legal dispute over whether the plain meaning of Article 14(d) requires that benefit be determined and measured on the basis of prevailing market conditions in the country of provision � in this case Canada.

4.192 Article 14(d) provides that where a government has provided goods, a benefit is conferred if remuneration is not �adequate�. It also provides that, �[A]dequacy of remuneration shall be determined in relation to prevailing market conditions for the good � in the country of provision�. [emphasis added]

4.193 First, the ordinary meaning of �in relation to� is �on the basis of� or �in comparison with�. This phrase prescribes a comparison to �prevailing market conditions�.

4.194 Second, the ordinary meaning of the term �prevailing� is �as they exist�. Accordingly, Article 14(d) requires a comparison to existing market conditions.

4.195 Third, �in the country of provision� means �in the country of provision.� It cannot refer to �prevailing market conditions� in some other country.

4.196 This interpretation is consistent with the Appellate Body�s analysis of Article 1.1(b) in Canada � Aircraft. There, the Appellate Body found that in determining whether the financial contribution conferred a benefit on the recipient, the �marketplace is the appropriate basis for comparison.� The Appellate Body also found that Article 14 provides context for Article 1.1(b). The marketplace, in a provision of goods context under Article 14(d), is the existing �market� in the country of provision.

4.197 Further contextual evidence that Article 14(d) does not permit cross-border comparisons is found in the Accession Protocol of China. The Protocol provides for the application of the SCM Agreement generally, but in language clearly indicating exceptional circumstances, it specifically permits the use of �methodologies for identifying and measuring the subsidy benefit which take into account the possibility that prevailing terms and conditions in China may not always be available as appropriate benchmarks.� If Article 14(d) already permitted the consideration of conditions outside �the country of provision�, Members would not have considered it necessary to provide for this exceptional treatment. This understanding is confirmed by the USTR on its official website.

4.198 In the CVD FD, USDOC concluded there were no usable benchmarks in Canada. This conclusion is flatly contradicted by the record. Canada provided extensive evidence to USDOC concerning prevailing market conditions in Canada. More specifically, this evidence included private timber sales, cost-revenue comparisons, an economic analysis of provincial stumpage charges, and private sector assessments of timber value. USDOC was obligated by Article 14(d) to use this evidence.

4.199 The reasons given by USDOC for rejection of this evidence are baseless. First, USDOC relied on the Preamble to its regulations to irrebutably presume price suppression. No analysis of actual price suppression was ever conducted. Second, the �economic� report which USDOC chose to use as support for its presumption of price suppression is purely theoretical and based on a flawed economic model. It did not include consideration of any actual transaction prices.

4.200 Canada submitted economic studies during the course of the investigation that demonstrated that it is the nature of the competition in the market that determines whether the market produces market prices, not the size of the market. If the market has the indicia of a competitive market � which it does in this case � the size of the government presence does not affect private prices.

4.201 Finally, the US efforts to dismiss evidence provided on in-country benchmarks as either not useable or not representative, are not credible. This evidence includes evidence that satisfies the third benchmark � consistency with market principles � in the US regulations. The US itself confirmed that it uses this benchmark where �no world market price for [the goods in question are] commercially available in the countries under investigation�. Here, there is no �world market price� for stumpage and US stumpage is not available in Canada. Accordingly, the evidence submitted by the Canadian provinces that demonstrated that their stumpage systems are operated in a manner consistent with market principles should have been considered by USDOC.

4.202 The US claim that it made adjustments that took into account the prevailing market conditions is not supported by the record. USDOC has admitted in previous lumber investigations that it would be �arbitrary and capricious� to use cross-border comparisons as it is not possible to identify a country or countries where some market conditions are the same as those in the country of provision and where all differences in market conditions can be identified and adjusted. These differences include: differences in economic conditions such wages, capital costs, taxes and government regulatory policies, comparative advantages in resources, timber characteristics, operating conditions, measurement systems and tenure rights and obligations.

4.203 Moreover, the US assertions respecting the adjustments that it did perform are incorrect. For example, the US claims that it �averaged the [US] price data by species to match the species in the relevant province.� For several provinces there were little or no price data by species. In almost all (ninety-nine per cent) of the US sales used by USDOC as a benchmark for the B.C. coast, the purchaser bid a single lump-sum price for all timber of all species. The result of the use of these benchmarks was to overstate the value in the US sales of the lower-value species common in BC and other provinces.

4.204 In the case of Qu�bec, adjustment categories considered by USDOC were not derived from an analysis of differences in market conditions between Qu�bec�s public forest and Maine�s private forest, but from carefully surveyed cost differences between the public and private forests within Qu�bec. If USDOC had attempted to analyze and compare market conditions in Qu�bec and Maine, it would have discovered that much of the data needed to make necessary adjustments were unavailable.

4.205 In order to argue that it is �well established� that the Article 1.1(b) test for benefit is a �but for� test, the US focuses on the word �absent� in an excerpt from Canada - Aircraft. In doing so, the US both misinterprets the Appellate Body�s statement and ignores the context of the sentence.

4.206 If the balance of the paragraph is considered, it is clear that the Appellate Body was saying that a �benefit�, as used in Article 1.1(b), �implies some kind of comparison� and that the �marketplace� must be the basis for this comparison. The Appellate Body has therefore, directed a comparison that is to be informed by what the recipient could have actually obtained in the market and not a comparison to some artificial �undistorted� market, as the US asserts.

4.207 The Appellate Body then notes that, �Article 14, which we have said is the relevant context in interpreting Article 1.1(b), supports our view that the marketplace is an appropriate basis for comparison.� As Article 14 provides relevant context in interpreting Article 1.1(b), the �marketplace� that is to be used for comparison purposes must be the markets referred to in Article 14. In a situation involving the government provision of goods, Article 14(d) refers to �prevailing market conditions � in the country of provision�. Accordingly, in this situation this refers to the in-country market as it exists.

4.208 Even though the US now contends that it has never advocated a �hypothetical undistorted market�, its arguments indicate the contrary. In the NAFTA proceeding, for example, it specifically describes adequate remuneration as, �the price that the purchaser would pay in an open and competitive market but for the government�s financial contribution.� [emphasis added].

4.209 Canada addressed what had been, to that point, the most recent attempt by the US to construct an argument that supported USDOC�s use of US prices as the appropriate benchmark. Canada showed how the US was attempting, through an elaborate exercise in word substitution, to turn Article 14(d) into a provision that measures adequacy of remuneration by comparing the government price to a constructed �fair market value�, rather than to prevailing market conditions in the country of provision.

4.210 The US has changed tack once again. It now asserts that where no �commercial market prices� exist in the country of provision or where they are �unreliable�, an investigating authority may use, �prices commercially available on world markets as the basis for an assessment of the adequacy of remuneration provided that those prices are informative as to the fair market value of the goods in the country of provision�. Accordingly, in the US view, the prices it uses �outside� the country of provision �to assess the fair market value of goods in the country of provision� need only to be available on the �world market� (and not in the country of provision). In support of this proposition, the US points to alleged acknowledgements by the US � Softwood Lumber III panel, Canada, Japan and the EC.

4.211 The US asserts that Canada has acknowledged that �the use of world market prices is appropriate in the case of a government monopoly for the good in question,� and describes the EC regulation as providing that the use of world market prices is appropriate when �market benchmark prices in the country of provision do not exist or are unreliable.� It goes so far as to claim that the �US regulations which Canada has not challenged establish essentially the same rule as that found in the EC regulation.�

4.212 None of these sources supports the US position. First, the EC regulation refers to �world market� prices and �terms and conditions prevailing in the market of another country�, and the US regulations to �world market price[s]�, that are available to purchasers in the country of provision of the good. The United States� own regulation makes clear that before a �world market price� is used it must be �reasonable to conclude that such price would be available to purchasers in the country in question.� Similarly, the EC regulation states that �when appropriate, the terms and conditions prevailing in the market of another country or on the world market which are available to the recipient shall be used.�

4.213 Second, Canada has consistently taken the position that a price that is available to purchasers in the country of provision makes that price part of the prevailing market conditions in the country of provision. This is why Canada stated in US � Softwood Lumber III  that in the context of a government monopoly over domestic production import prices for the same good if available to purchasers in Canada, could be used as a benchmark to measure adequacy of remuneration.

4.214 Third, the US mischaracterizes what was said by the Panel in US � Softwood Lumber III . In the paragraph cited by the US, the panel stated that �prices of imported goods in the market of provision can indeed form part of the prevailing market conditions in the sense of Article 14(d)�. The Panel further stated that �the text of Article 14(d) SCM Agreement � does not provide for � a world market test�.

4.215 Central to all of these positions is that regardless of whether domestic prices in the country of provision, �world market prices�, import prices or other indicia of prevailing market conditions are used as benchmarks, such prices or conditions must be �available� in the country. If this requirement is satisfied the benchmark will constitute part of the prevailing market conditions in the country of provision.

4.216 That is not the case here. US stumpage is not available in Canada. The good that USDOC contended was available in Canada at a world market price is logs, a product that is produced from standing timber. Canadian producers may purchase US timber �on the stump�, but only where the stumps are � in the US. Stumpage is inherently local. The US is therefore reduced to arguing that prices in another country, not available in the country of provision, may be used because they are commercially available on the world market. This position is clearly inconsistent with the words of Article 14(d) and is not supported by Canada�s earlier statements, the findings of the Panel in US � Softwood Lumber III  r, or either of the EC or US regulations.

4.217 In its Closing Statement, the US asserts that it �did not simply assume trade distortion.� Rather, it claims, it �determined that US law does not require an analysis of whether a subsidy has market distorting effects.� At �the heart� of USDOC�s final determination, however, was a conclusion that these charges result in countervailable market distortion in the US lumber market. According to USDOC, �the whole point of this investigation is to quantify and remedy the impact� of the Canadian �administered pricing� system on the US market, as its �mere existence� has �a resulting impact � on Canadian lumber production� that �distort[s] the US market.� It may not now contend that �the heart� of the case � its �whole point� � was of no consequence to its determination.

3. Pass-Through

4.218 The pass-through issue in this case may be summarized as whether the US, in presuming rather than demonstrating the pass-through of the alleged stumpage subsidy to certain producers of subject lumber, has violated its obligations under the SCM Agreement and GATT 1994. The producers in question are those who buy log or lumber inputs at arm�s-length. Canada has demonstrated in its submissions that a subsidy may never be presumed in a countervailing duty investigation. The US argument to the contrary is that the SCM Agreement allows it to presume the pass-through of a subsidy in aggregate investigations, regardless of any evidence establishing arm�s-length transactions. However, there is no irrebuttable presumption in the SCM Agreement that allows a Member to disregard evidence establishing no subsidy. Because the US has failed to conduct any pass-through analysis regarding independent harvesters, the volume of Crown timber harvested by these entities and the amount of subsidy derived from that volume, for example, must therefore be excluded from the total amount of the subsidy. Further, no countervailing measure can lawfully be imposed on the products of lumber remanufacturers purchasing at arm�s length.

4. Specificity

4.219 Even if the US had correctly determined that provincial stumpage programmes were a subsidy, it failed in the CVD FD to correctly determine that the programmes are �specific�. The US finding under the �limited users� factor in Article 2.1(c) is incorrect. Even if there was a �limited number of users�, the US nevertheless failed to determine that provincial stumpage programmes were �specific to certain enterprises� based on the total configuration of facts and evidence of the case.

4.220 Stumpage programmes are not used by a �limited number� of certain enterprises under Article 2.1(c). The record evidence in this case establishes that thousands of enterprises, in at least 23 standard industry categories, used provincial stumpage programmes during the period of investigation. The survey submitted by the Canadian parties showed that companies using provincial stumpage programmes during the period of investigation manufactured a minimum of 201 distinct products, many of which are not produced by USDOC�s purported group of a �limited number of industries�.

4.221 The US failed to analyze the record evidence regarding which enterprises and industries actually used and allegedly benefited from stumpage programmes. In the CVD FD, it simply asserted that �pulp and paper mills and the saw mills and remanufacturers which are producing the subject merchandise� were the sole users of stumpage programmes. While the US argues that �no other industries� use or allegedly benefit from stumpage, it can point to no evidence in support of its assertion. Indeed, this finding is directly contradicted by the US acknowledgement of significant amounts of log harvesting done by enterprises other than the supposed �sole users� of stumpage. The EC and Japan agree that the �limited users� finding is flawed.

4.222 Moreover, even if the US had properly identified the users of stumpage programmes, its impermissible use of the entire economy as a benchmark would mean that virtually all programmes would be specific, since virtually no programme is used by �everyone�. In particular, every government programme involving the provision of a good would be specific. Unlike money, goods are not fungible and have a limited base of users who will expend the effort to procure the good. On the facts, therefore, the US determination is inconsistent with Article 2.1(c) and fails to satisfy the requirements of Article 2.4.

4.223 The existence of an alleged limited number of users does not ipso facto establish specificity in fact. The factors in Article 2.1(c) simply indicate that a subsidy may be specific despite being non-specific in law. A determination does not satisfy Article 2 unless it is clearly substantiated that the subsidy is in fact specific; this requires cogent reasoning and an assessment of all the evidence to translate a �limited users� finding into a determination that a government is restricting access to a subsidy in fact. Provincial stumpage programmes are not specific. The US failed to clearly substantiate its specificity determination, and has therefore violated its WTO obligations.

4.224 To determine that a subsidy is �specific � under Article 1.2 is to determine, under Article 2, that a government limits access to the subsidy to certain enterprises over other eligible enterprises. The ordinary meaning of the term �is specific to� is that the subsidy is made available to certain enterprises but not available to others. The specificity requirement is meant to capture instances where a government targets a subsidy to certain enterprises.

4.225 This is confirmed by the structure of Article 2.1; paragraph (c) must be read in the context of paragraphs (a) and (b). Under Article 2.1(a), a subsidy is specific in law where a government explicitly limits access to it. Article 2.1(b) provides an exception that where access is limited by objective and neutral eligibility requirements, the limitations do not establish the favouritism necessary to find specificity. Under Article 2.1(c), a subsidy is specific in fact where, instead of limiting access to a subsidy in law, a government does so through its implementation of the programme in fact. The �appearance of non-specificity�, of which Article 2.1(c) speaks, is an appearance that government is not limiting access to a programme, when in fact it might be.

4.226 According to the US, specificity is deemed to exist where any one of the four factors in Article 2.1(c) is shown to exist without regard to why they might exist. This interpretation impermissibly reduces indicative factors to irrebuttable presumptions.

4.227 Article 2.1(c) provides that where there is an appearance of non-specificity in law, �other factors may be considered� to determine whether the government is limiting access to certain enterprises in fact. Unlike the US interpretation of paragraph (c), Canada�s interpretation is based on the ordinary meaning of the word �specific�, read in the context of paragraphs (a) and (b) and the chapeau of Article 2.1. For example, it would be nonsensical under Article 2.1(c) to find that a subsidy is specific in fact solely on the basis of �the manner in which discretion has been exercised by the granting authority in the decision to grant a subsidy�, without any consideration of whether that manner directs the subsidy to certain enterprises. Likewise, the third factor is �the granting of disproportionately large amounts of subsidy to certain enterprises� by a government. Where there is no other explanation for the existence of such amounts, the inference under Article 2.1(c) is that the government is directing the subsidy to certain enterprises. On a plain reading of the third and fourth factors, the analysis required under Article 2.1(c) goes to inferring government action in the application of the programme.

4.228 The specificity requirement deals with government action, and thereby requires direct evidence establishing such action. Where the government action limiting access to a programme is not explicit, evidence will necessarily be circumstantial � allowing an investigating authority to infer the targeting of a subsidy to certain enterprises based on the existence of the factors listed in Article 2.1(c). However, to �clearly substantiate� a determination under Article 2.4, an investigating authority must provide cogent analysis that establishes a high level of confidence. As the EC correctly observes, the distinction between de jure and de facto determinations is one of evidence; Article 2.1(c) does not create a different test for specificity. In assessing the evidence pertaining to the factors, the many reasons why there might be �limited users� or �predominant use�, for example, are highly relevant. If these factors do not indicate government targeting action, the subsidy is not specific in fact.

4.229 The negotiating history surrounding Articles 1.2 and 2 of the SCM Agreement confirms Canada�s interpretation that specificity relates to a determination that government action restricts access to the alleged subsidy. The specificity concept concerns government limitations on access, not simply patterns of �use� on the part of alleged recipients � the latter being merely indicative of the former.

4.230 The concept of �specificity� has been with the WTO and GATT for nearly two decades. It was first considered by the Committee on Subsidies and Countervailing Measures ("SCM Committee") in 1985, when the SCM Group of Experts provided it with Draft Guidelines for the Application of the Concept of Specificity in the Calculation of the Amount of a Subsidy Other than an Export Subsidy ("Draft Guidelines"). The Draft Guidelines set out a series of rules based on what government did to restrict the availability of a programme. Where access was restricted to certain enterprises based on non-neutral criteria, the programme was considered de jure specific. On de facto specificity, the lone paragraph (f) addressed government action affecting availability in terms of a �de facto deliberate[] granting [of] an advantage to certain industries.� The draft language on de facto specificity evolved over the course of the Uruguay Round negotiations, but at all times remained concerned with the granting by government of selective access to a subsidy programme.

4.231 The US failed to convince other members to abandon the specificity requirement during the Uruguay Round. Accordingly, US countervailing duty law, the US application of the requirement in the CVD FD, and the US specificity arguments before this Panel, all attempt to re-write Article 2 of the Agreement for a �next-best� result: that a finding of �limited users� is dispositive of specificity. This interpretation of Article 2 is contrary to the ordinary meaning of the text of Article 2, read in context and in the light of the object and purpose of the Agreement and the negotiating record; it also renders the provision meaningless. The US has failed in the CVD FD to address the other explanations for any alleged pattern in the �use� of stumpage, and failed to explain why its incorrect finding of a �limited number of industries� means that stumpage is specific in fact to certain enterprises.

4.232 The US has also violated its obligations under the SCM Agreement because it failed to take into account, as explicitly required under Article 2.1(c), record evidence concerning the extent of diversification of provincial economies. In British Columbia, for example, forestry-related activity is responsible for a substantial share of economic activity. The value of forestry-related shipments totalled $C20.2 billion in 2000, accounting for more than half the value of all manufactured shipments in the province. Forestry product sales were valued at $C15.6 billion in 1998 and $C16.8 billion in 1997. Indeed, since 1995, forestry product shipments have represented approximately half the value of manufactured shipments in British Columbia. The significant share of manufacturing shipments translates to a substantial contribution to provincial gross domestic product. In 1999, the various forest products industries accounted directly for 24 per cent of the goods-producing industries� provincial GDP, and 6 per cent of the total provincial GDP. The Final Determination does not even address this significant factual information.

5. Other Claims

(a) Calculations

4.233 The US did not correctly determine the amount of the subsidy, did not correctly calculate the subsidy per unit rate, and therefore applied countervailing duties at a rate in excess of the alleged subsidization per unit of the �subsidized and exported product.� By imposing these countervailing duties the US violated Article 19.4 of the SCM Agreement. The errors of the US include:

• use of a manifestly incorrect log scale conversion factor;
   

• massive inflation of the subsidy amount used in the numerator in its subsidy per unit calculation; and
   

• understatement of the denominator in the flawed calculation methodology it chose.

4.234 In response to Canada�s claims, the US simply argues that, �[t]he sole calculation requirement in Article 19.4 is a requirement to calculate the subsidy on a per-unit basis; Article 19.4 does not establish any other requirements concerning how the subsidy is to be calculated.� The US position is untenable.

4.235 The US concedes that Article 19.4, �establishes an upper limit on the amount of the countervailing duty that may be levied� and that it requires calculation of the subsidy �on a per-unit basis.� Thus, the US agrees with Canada that Article 19.4 establishes requirements in calculating a subsidy rate. However, the US claims that it has no obligation to correctly calculate that rate. The US is therefore arguing that a Member may impose countervailing duties at any rate it wishes, as long as it is not in excess of a subsidy per unit rate, however arrived at.

4.236 Article 19.4 is the sole provision establishing an upper limit to the countervailing duty rate to be imposed. Other provisions set out what constitutes a subsidy and what the total amount of that subsidy is; only Article 19.4 sets out any discipline on what the maximum countervailing duty rate may be. To give proper effect to Article 19.4, that discipline must require a per unit subsidy rate correctly calculated, based on a subsidy amount correctly derived, as against a practice correctly determined to be a subsidy.

4.237 The US alleges that the conversion factor mentioned in the Minnesota Public Stumpage Price Review and Price Index (�Price Report�) �only applied to the data contained in Table 2�, while the US used only �sawtimber data in Table 1.� But the statement on the cover of the Price Report says that, �All reported volumes and values were converted to a sawlog and pulpwood basis for inclusion in Table 1.�

4.238 Where �converted� data are based on a specific conversion factor, to ignore the specified factor is to render the data hopelessly distorted and therefore useless. In this case, as a result of the choice of conversion factor, subsidies were found where none existed; and, in any event, subsidies were found using an improper cross-border analysis. The use of a single, and incorrect, conversion factor highly inflated the alleged subsidy per unit rate.

4.239 A conversion factor was necessary in the first place because the US elected to compare tree volumes and values in certain US jurisdictions, to widely disparate trees across Canada. The effective use of a single factor for most of these conversions created significant distortions in the assessment of whether any stumpage benefit even exists, as well as in the subsequent subsidy calculations. The US had a draft report prepared by an economist of the US Forest Service on the record that demonstrates the irrationality of its approach to conversion factors. The report notes that �[t]o properly compare prices, conversion factors must be tailored to the measurement system used in order to adjust for any bias that may have developed.� The bias referred to is the tendency for contemporary prices expressed in board feet to underestimate lumber yield and boost stumpage prices to a degree that cubic measurements do not. In addition, the very publication that the US used to choose its conversion factors explicitly stated that smaller logs, like those in Canada, require use of much larger factors than the US used.

4.240 The alleged subsidy at issue was the provision of �timber� at less than adequate remuneration. The allegedly subsidized merchandise subject to the countervailing duty was softwood lumber and certain products manufactured from that lumber. The alleged subsidy per unit may not properly be calculated using more than the volume of wood actually used in the subject merchandise. The portion of a log that becomes sawdust is not used in the production of lumber; equally, logs destined for the production of posts are manifestly not used in the production of softwood lumber.

4.241 The alleged subsidy attributable to the subject merchandise was easy to trace and just as easy to calculate: out of the total volume of logs entering sawmills, the portion used in the manufacture of lumber created the only possible �subsidy� to the production of lumber and lumber products. Logs used in the production of posts or poles, and the portion of logs that ended up as chips, did not involve a subsidy attributable to softwood lumber.

4.242 An accurate calculation of the subsidy per unit rate of the exported product required that only the volume of logs used in the softwood lumber products be reflected in the numerator; the denominator would then be limited to the output lumber products from those logs. Because Article 19.4 refers to the subsidy per unit rate of the allegedly �subsidized and exported product�, the proper starting point would have been for USDOC to have determined the correct amount of the subsidy attributable to the subject merchandise � softwood lumber products.

4.243 USDOC claims that it corrected for its overbroad numerator by expanding the denominator to include the value of the non-lumber products whose volumes had dramatically increased the numerator. USDOC failed to do this; however, even if it had, this approach would not cure the bias or the subsidy inflation. Non-lumber products account for a significant volume of products produced from softwood logs entering sawmills. Yet, typically, the value of non-lumber products such as chips or sawdust is low when shipped from the sawmill. Further, chips and sawdust are input products to other more expensive products such as pulp and paper. By including the volume of these non-lumber products while not accounting in the denominator for the full value of the products included in the numerator, the per unit subsidy calculation was significantly inflated.

4.244 The United States, having adopted a methodology that grossly inflated the subsidy amount, then did not include the value of all the non-lumber products in its denominator. It seriously understated the denominator because 1) it did not include softwood �residual� products produced in sawmills, and 2) it used a wholly inaccurate and understated sales value for remanufactured products.

(b) Conduct of the Investigation

4.245 With respect to its choice of Minnesota as benchmark state for Saskatchewan and Alberta, the US claims that the responding parties should have had notice of the possibility of that choice. For this reason, it argues, its failure to notify the responding parties of the change from Montana to Minnesota did not violate Article 12.8.

4.246 Canada recalls footnote 225 of the First Written Submission of the US. An oblique reference, in a countervailing duty petition, to the entirety of the US as a potential source of benchmark does not amount to the disclosure of the �essential fact� of Minnesota as a benchmark state by the US. Canada notes the new US argument that only a few of the fifty states were reasonably to be considered as candidates for serving as benchmark. In this respect, Canada recalls the findings of the US in the Preliminary Determination in this case, where USDOC stated, �[I]nformation on the record indicates that stumpage in the United States along the border with Canada is comparable to Canadian stumpage �. we only compared stumpage prices in each Canadian province with stumpage prices in states bordering that province.� [emphasis added]

4.247 Canada also recalls the statement of the US to the panel in US � Softwood Lumber III, �Even within the United States, not all timber prices are appropriate. � Commerce did not use prices from the large timber-growing areas in the southeast, or from any non-contiguous state, because as a matter of commercial reality, Canadian lumber producers do not consider it commercially viable to transport logs over that long a distance.� [emphasis added]

4.248 The only notice given to the responding parties was that data from a state �bordering� the province in question would be used. A state that is at least 700 km from the closer of the two provinces is not a �contiguous� state. Minnesota is not a state that borders either Saskatchewan or Alberta. There was no notice of the choice of Minnesota as a benchmark state. The US violated Article 12.8.

F. SECOND WRITTEN SUBMISSION OF THE UNITED STATES 

4.249 The following summarizes the United States' arguments in its second written submission.

4.250 The standard of review set forth in Article 11 of the DSU applies to disputes arising under the SCM Agreement. Article 11 requires a panel to �make an objective assessment of the matter before it, including an objective assessment of the facts of the case and the applicability of and conformity with the relevant covered agreements.� In conducting this inquiry, the Panel may only address those provisions of the covered agreements that Canada cited in relation to specific claims in its request for the formation of a panel. With few citations to the record, Canada recites a laundry list of facts and figures in support of its arguments, failing to note the substantial record evidence that contradicts its arguments. Canada is asking the Panel to step into the shoes of USDOC and engage in a de novo review and evaluate the facts. It is well-established, however, that panels may not engage in such an exercise.

4.251 The parties� answers to the Panel�s questions have confirmed that companies enter into tenure agreements with the provinces for the sole purpose of obtaining timber. In return for fulfilling the tenure obligations and paying the stumpage fee, the tenure holder acquires ownership of the timber, not a �right� to harvest timber.

4.252 The ordinary meaning of �goods� is broad and encompasses, at the very least, all tangible property, including �growing crops, and other identified things to be severed from real property.� Simply labeling standing timber a �natural resource� does not remove it from the ordinary meaning of the term �goods.� Canada has acknowledged that provincial tenures identify specific, defined areas of forest from which the tenure holder may harvest trees. The identified trees to be severed from provincial land fall squarely within the ordinary meaning of the term �goods.�

4.253 Canada�s argument suggests that tenures are simply about forest management obligations that, almost incidentally, also confer an intangible �right� to harvest. Canada�s analytical approach is based on the flawed premise that the existence of a financial contribution is a matter to be determined from the government�s perspective. However, a financial contribution, as defined in Article 1.1(a)(1)(iii) of the SCM Agreement, exists whenever the government provides a good. As the US � Softwood Lumber III panel recognized, the existence of a subsidy is determined from the perspective of the benefit to the recipient, not the perspective of the government. Specifically, the US � Softwood Lumber III panel found that in spite of the fact that the provincial governments have certain policy objectives, �the fact of the matter remains that from the harvesting company�s point of view, the only reason to enter into such tenure or licensing agreements is to cut trees for processing or sale.�

4.254 As detailed in the United States� response to the Panel�s questions, the record demonstrates that tenure holders do not acquire a freely transferable �right� to harvest. All provinces prohibit the transfer of tenures without government approval. Without any citation to record evidence, Canada asserts that the right to harvest is freely transferable, without approval. There is an obvious conflict, however, between the record evidence and Canada�s suggestion that tenure holders may freely sell or subcontract the �right to harvest.� Despite Canada�s efforts to sever the right to harvest from the sale of the trees, the facts demonstrate that tenure holders are buying trees. In doing so, the provinces make a financial contribution within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement.

4.255 As the United States pointed out in its first written submission, the Canada � Aircraft panel stated that a benefit exists where �the financial contribution places the recipient in a more advantageous position than would have been the case but for the financial contribution.� In reviewing that report, the Appellate Body affirmed that a benefit exists where �the �financial contribution� makes the recipient �better off� than it would otherwise have been, absent that contribution.� In determining the existence of a benefit, therefore, the issue is the position of the recipient �but for� or �absent� the government�s financial contribution.

4.256 Moreover, the Appellate Body has stated that the point of comparison is �the marketplace,� i.e., a benefit exists where the financial contribution is received on terms more favorable than those available in the market. Finally, as the United States has pointed out, following the reasoning of the Appellate Body in Canada � Aircraft, the Brazil � Aircraft panel concluded that the concept of a comparison market necessarily means a �commercial market, i.e., a market undistorted by the government�s financial contribution.�

4.257 Canada erroneously argues that the Brazil � Aircraft report is inapposite because that panel was considering Article 14(b) of the SCM Agreement, which establishes commercial lending rates as the benchmark for a government loan. The panel found that the financial contribution at issue was in the form of a non-refundable payment, however, rather than in the form of a loan. Thus, as Canada argued in that case, the payments at issue were �essentially grants.� In the passage cited by the United States, the Brazil � Aircraft panel was not discussing Article 14(b) of the SCM Agreement. In fact, there is no reference at all to Article 14 of the SCM Agreement in the panel�s report. Rather, the panel was discussing the concept of benefit generally. The panel�s reasoning, which follows logically from the findings of the Appellate Body, is compelling. Only by comparison to a market undistorted by the government�s financial contribution is it possible to determine whether the recipient is better off than it otherwise would have been absent the financial contribution. That is true regardless of the form of the financial contribution.

4.258 Article 14 does not redefine the concept of benefit in Article 1.1(b), as interpreted by the Appellate Body and prior panels. Article 14 merely provides guidelines that must be followed in establishing �methods� for applying that concept to particular types of financial contributions. Each guideline in Article 14, including Article 14(d), must therefore be interpreted in a manner that is consistent with the interpretation of the term �benefit� as used in Article 1.1(b) of the SCM Agreement.

4.259 When the government provides a good, Article 14(d) states that a benefit is conferred if the government receives �less than adequate remuneration� for that good. Applying the reasoning of the Appellate Body, �less than adequate remuneration� must mean a price less than would otherwise be available in the marketplace absent the government�s financial contribution. The proper benchmark therefore is an independent market-driven price for the good, i.e., fair market value � which is also the standard applied under Canadian law.

4.260 Article 14(d) does not specify the type of evidence that must be used to establish the fair market value of goods in the country of provision. Rather, Article 14(d) establishes the general principle that adequate remuneration (fair market value) must be determined �in relation to prevailing market conditions� in the country of provision. There is no basis in the SCM Agreement to conclude that �benefit� means anything less when the government provides a good than when it makes any other type of financial contribution. Thus, �market� conditions must be interpreted in a manner consistent with the concept of benefit in Article 1.1(b) of the SCM Agreement. Following the reasoning of the Appellate Body and the Brazil � Aircraft panel, therefore, the point of comparison under Article 14(d) must be prevailing commercial market conditions, i.e., a market undistorted by the government�s financial contribution, in the country of provision. It is the view of the United States that where such benchmark prices exist in the country of provision, they must be used. However, where such benchmark prices do not exist in the country of provision or are unreliable, a Member may, consistent with Article 14(d), rely on data from outside the country of provision to assess the fair market value of the goods in the country of provision. This is the case with respect to Canadian timber.

4.261 As the United States demonstrated in its response to the Panel�s questions, the actual data on private stumpage prices is virtually non-existent for four of the six provinces. Canada does not dispute the fact that Manitoba and Saskatchewan did not provide any data on private stumpage prices. With respect to Alberta, as the United States explained in response to the Panel�s questions, the record demonstrates that the Timber Damage Assessments (�TDAs�) that Alberta provided are simply voluntary guidelines for settling disputes for damages. Moreover, TDAs are based on log prices and do not differentiate between private and Crown logs. With respect to B.C., Canada acknowledges that the evidence provided establishes that the private market for stumpage in B.C. is �limited� and that �nearly all private wood fibre sales are of logs rather than standing timber.� The data on sales of standing timber accounted for only 0.1 per cent of the B.C. harvest.

4.262 Recognizing the lack of any private stumpage benchmark data for these four provinces (Alberta, B.C., Manitoba, and Saskatchewan), Canada continues to argue, without any basis in the SCM Agreement, that evidence demonstrating that the provinces made a profit on their timber sales suffices. A benefit, however, is not determined based on the cost to the government in making the financial contribution, e.g., whether the government incurred a loss. The issue is whether there is a benefit to the recipient.

4.263 With respect to Ontario and Quebec, as discussed in response to the Panel�s questions, some data on private stumpage prices exists. However, prices that are distorted by the government�s financial contribution do not reflect �market� conditions. It is undisputed that Canadian timber sales are overwhelmingly dominated by the provincial governments. The fact that the government is a price leader does not in and of itself establish the absence of independent commercial market conditions. As the Appellate Body has cautioned, however, governments have the ability to obtain certain results from the market by shaping the circumstances and conditions in which the market operates. The dominance of the government in the marketplace can, therefore, warrant further examination in determining the adequacy of remuneration for government-owned goods. Ultimately, as noted by the European Communities, the issue must be determined on a case-by-case basis.

4.264 The facts of this case demonstrate that the provincial governments not only dominate the Canadian timber market, but also that the governments do not participate in the market as commercial actors. Rather, the provincial governments administer the sale of the overwhelming majority of Canadian timber in a manner designed to further social policy goals. The evidence further demonstrates that, as a result, the administration of provincial timber sales systems distorts the small private timber market.

4.265 All of the provinces generally restrict the sale of Crown timber to purchasers that own a processing facility in the province. These local processing requirements artificially reduce the demand for Crown timber. In addition, the vast majority of Crown timber is under some form of renewable (�evergreen�), long-term tenure. These tenures are not freely transferable. The existence of evergreen, non-transferrable tenures creates significant barriers to entry into the timber market. For example, Canada offers no support for its assertion that a newcomer in Quebec could obtain a provincial tenure, even though 100 per cent of the Crown timber is currently under tenure or reserved and not subject to harvest. Nor does Canada reconcile this claim with its subsequent statement that the vast majority of mills in Quebec are �shut out of the public forest.� Moreover, �transfers� of tenure are, in fact, normally cases in which the entity that holds the tenure is acquired. No new tenure is created in such cases.

4.266 Other aspects of tenures artificially increase supply. For example, B.C. imposes minimum cut requirements and restricts mill closures, thus forcing tenure holders to harvest timber even in down markets and thereby artificially increasing the supply of Crown timber. Nevertheless, the record demonstrates that, during the period of investigation, tenure holders in all of the provinces except Alberta did not harvest their full AAC, thus demonstrating an ample supply of additional Crown timber available at administered prices.

4.267 Central to each of the provincial systems is, of course, administered rather than competitive pricing of that timber supply. The evidence demonstrates that the price leaders, i.e., the provincial governments, do not price to the market. Rather, they administer prices under systems designed to promote employment and keep mills operating even in down markets.

4.268 A study by Economists Inc. (�EI Study�), which the United States found compelling, addressed the impact of the administered provincial systems on the private market. The EI Study applied generally accepted economic analysis to demonstrate that when a single supplier controls the overwhelming portion of market share, that supplier will necessarily influence non‑administered prices. It found that �[t]he lower the market share of the firms in the non‑administered sector relative to the administered sector, the less the ability of the non‑administered sector to raise price above the administered (subsidized) level.� The essence of this analysis is that private (non‑administered) sellers cannot increase prices significantly above administered price levels of the competing supply because, if they did, demand would shift to the administered sector. The study referred to record evidence establishing that provincial governments not only supply the vast majority of timber, but also are willing to provide yet more timber to the major licensees that comprise most of the demand. The study showed that if private landowners attempted to raise their prices significantly above government‑set levels, the major licensees would rely relatively more on administered timber sales and less on private or auctioned timber. Therefore, while local variations will exist, overall the government price effectively sets the average province‑wide price as well.

4.269 Canada criticizes this study and cites to other evidence it finds compelling. The issue, however, is not which evidence Canada finds persuasive. The issue is whether the record evidence supports the United States� determination. The EI Study and other record evidence, some of which is summarized in the United States� first written submission and oral statements at the first substantive meeting of the Panel, support that determination. There also is significant documentary evidence confirming this economic analysis and demonstrating that each provinces� system of public timber sales distorted private timber prices.

4.270 The evidence demonstrates that the provincial governments shape the timber market to achieve public policy goals, and that the government-controlled, public-policy driven sector of the timber market distorts the private timber market. There was therefore ample support for the United States� conclusion that there are no independent, market-driven timber prices available in Canada. The United States� use of data from other sources to assess the fair market value of timber in Canada was therefore warranted and was consistent with Article 14(d) of the SCM Agreement.

4.271 The evidence also supports the United States� decision to use timber prices in the northern United States, properly adjusted to reflect conditions of sale in Canada, as an alternative source for assessing the fair market value of timber in Canada. The rationale for the United States� choice of data for establishing market benchmarks is sound. As the United States explained in its first written submission, the value of timber depends on the demand for the downstream products. US and Canadian timber satisfies the same demand in the integrated North American lumber market, and Canada has no comparative advantage in serving that demand. In addition, the record demonstrates that US timber is commercially available to lumber producers in Canada. The US timber is also comparable to Canadian timber, and the United States established species-specific, province-specific benchmarks to conduct the comparison.

4.272 In response to the Panel�s questions, Canada stated that out-of-country benchmarks that are available to purchasers in the country of provision would be suitable benchmarks because they �would then form part of prevailing market conditions �in� the country of provision.� As discussed in the Final Determination, the evidence demonstrates that US timber is, in fact, commercially available to lumber producers in Canada. Further, Canada asserts that the United States could have used an alternate methodology, i.e., evaluation of the government�s prices based on evidence of the government�s costs and profitability. The standard is not the cost to the government, but rather the benefit to the recipient. We have demonstrated that the use of prices for US timber in the northern United States, adjusted for differences in conditions of sale in Canada, is consistent with Article 14(d) of the SCM Agreement.

4.273 Canada cites to Articles 10, 19.1, 19.4, and 32.1 of the SCM Agreement and Article VI:3 of GATT 1994 as the basis for its calculation-related claims. As the United States previously indicated, Canada�s claims under Articles 10, 19.1, and 32.1 of the SCM Agreement are necessarily derivative claims that cannot succeed because Canada has failed to establish that the United States has breached its obligations under another provision of the SCM Agreement.

4.274 Canada admits that neither Article 19.4 of the SCM Agreement nor Article VI:3 of GATT 1994 contain any obligation regarding the methodology a Member may use in calculating the ad valorem subsidy rate. None of the provisions of the SCM Agreement that Canada cited in support of its claim establish the calculation obligations Canada suggests the United States has violated. Given that a panel�s terms of reference �establish the jurisdiction of the panel by defining the precise claims at issue in the dispute,� and that the identification of the specific provision of the covered agreements is a �minimum prerequisite� for stating the legal basis of the claim, this Panel should reject Canada�s attempts to bootstrap claims that the ad valorem rate calculation is inconsistent with other provisions of the SCM Agreement.

4.275 Article 19.4 of the SCM Agreement and Article VI:3 of GATT 1994 merely provide that the countervailing duty rate imposed may not exceed the amount of the subsidy the investigating authority has found to exist. Neither provision contains particular calculation obligations of the sort Canada asserts have been violated. Canada, therefore, has failed to present a prima facie case on this issue.

4.276 The United States� calculation of the ad valorem subsidy rate is consistent with the SCM Agreement and the evidence developed through the investigation. The United States included in the numerator the volume of provincial softwood timber entering sawmills, multiplied by the benefit per cubic meter. This method captured the total value of the subsidy provided to sawmills. The United States then allocated that benefit over all sales of products that resulted from the lumber production process. This methodology accounted for the fact that the production process yields other products as well as lumber.

4.277 The United States conducted this investigation on an aggregate basis, calculating the total benefit provided by each province, as discussed above. Canada does not dispute the United States� authority to conduct this aggregate analysis, but it continues to assert that the United States was obligated to conduct a pass-through analysis to establish the amount of any benefit received by certain producers of the subject merchandise, i.e., independent remanufacturers.

4.278 In Canada�s first response to the Panel�s questions, it acknowledged that, �[w]here the timber harvester and the producer of subject merchandise are the same �recipient� of the alleged subsidy, no pass-through analysis would be required.� This fact pattern describes the vast majority of the producers of the subject merchandise and includes both sawmills and remanufacturers. Thus, Canada�s statement acknowledges that the United States was not required to conduct a pass-through analysis for at least the vast majority of the lumber at issue.

4.279 Canada also fails to address Article 19.3 of the SCM Agreement, which specifically allows Members to apply definitive countervailing duties to exports of an uninvestigated exporter or producer as long as the exporter or producer may obtain an expedited review to establish its own rate.

4.280 Additionally, Canada overreaches when it contends that a �[s]ubsidy pass-through analysis is required in every instance where the subsidy found to exist is allegedly bestowed on one person while the countervailing duty is imposed on the products of another.� Canada�s attempt to read such an obligation into Article 19.4 of the SCM Agreement has no basis in the text. If Canada were correct, every time a Member investigates one or more companies and applies their subsidy rate to uninvestigated exporters or producers � a common practice � that Member violates the SCM Agreement. Such practices do not, however, violate the SCM Agreement, as evidenced by the last sentence of Article 19.3 of the Agreement.

4.281 Canada argues that the United States understated the denominator by failing to include the value of certain �residual products� in the denominator. Canada failed to submit any evidence from which the United States could separate the value of additional products resulting from the lumber production process from the broader residual products category. Accordingly, the United States did not include the residual products category in the denominator. Canada also disputes the value used to represent remanufactured products that the United States selected and used in the denominator. As noted above, the Panel should decline Canada�s request to engage in a de novo review and re-weigh the evidence before the administering authority.

4.282 In its questions to the parties, the Panel requested Canada to address how the conversion factors that the United States used were based on �manifestly incorrect data.� Canada failed to provide the information the Panel requested. Instead, Canada cited alternative sources of conversion factors. The existence of alternative sources, however, does not mean that the United States� selection was �manifestly incorrect.�

4.283 Article 2.1(c) of the SCM Agreement contains clear and objective criteria for determining when a subsidy is specific. Where a subsidy programme is used by a limited number of �certain enterprises� � i.e., an enterprise, industry, or group of enterprises or industries � it is specific in fact. Other than considering the extent of diversification of economic activities and the length of time the subsidy programme has been operating, Members are not obligated to conduct any further specificity analysis.

4.284 The United States met its obligation to demonstrate that provincial stumpage subsidies are specific and thus actionable under the SCM Agreement. In the Final Determination, the United States found that the users of stumpage were a �limited group of wood product industries� that included �pulp and paper mills and the saw mills and remanufacturers which are producing the subject merchandise.� The industries comprising this limited group fall squarely within the ordinary meaning of the term �industry,� which identifies industries by general product, such as automobiles or textiles, or by the type of activity engaged in, such as mining or banking. Thus, the provincial stumpage subsidies are specific under Article 2.1(c) of the SCM Agreement.

4.285 Canada responds to this finding by attempting to rewrite the specificity test, seeking to redefine terms such as �industry� and �group,� and to create exceptions that do not exist in Article 2. For example, Canada claims that the term �industry� requires a �product-based identification of industries,� such that individual industries would be distinguished on the basis of a particular end product, or set of end products, that they make. By contrast, both in its ordinary meaning, and within the context of Article 2 of the SCM Agreement, �industry� is used broadly, referring to makers of a general class of products � such as �the steel industry� � regardless of the number or diversity of end products that the industry produces.

4.286 Canada itself admits, in response to question 27 from the Panel, that a subsidy to a single large industry could be found specific, even where its producers make a vast diversity of products, as in the steel, automobile, textile, and telecommunications industries. This admission contradicts Canada�s argument that an �industry� must be defined narrowly on the basis of a particular end product or set thereof.

4.287 Canada likewise admits that a subsidy granted solely to auto and textile producers could be specific under Article 2.1, notwithstanding the dissimilarity of their end products. It nonetheless still insists that the two industries cannot form a single �group� of industries that is specific within the meaning of Article 2.1(c). Instead, Canada maintains that �the industries producing �autos� and the industries producing �textiles�� are actually �two groups of industries� that �appear to be a �limited number� of certain enterprises,� and thus specific. Notwithstanding Canada�s acceptance of diversity in the product mix of an �industry,� Canada finds such dissimilarity incompatible with its view of the term �group,� which for Canada requires �similarity and relatedness� of output products. In its ordinary meaning, and in the context of Article 2 of the SCM Agreement, the definition of �group� is far more straightforward, meaning simply �more than one� enterprise or industry, without regard to the �similarity or relatedness� of their end products.

4.288 Canada continues to argue that each industry is defined by a narrow class of end products, claiming that the �immediate users of stumpage� include �at least 23 categories of industries, and the industries are as unrelated as lumber, agricultural chemicals, paper, and furniture.� This argument is not only legally flawed, it rests on misleading evidence. Canada�s key exhibit is a survey of forest product industries that lists 201 products made by tenure holders, categorized into 23 categories that Canada claims to be different industries. In reality, Canada�s list of industries is little more than an exercise in hairsplitting � assigning multiple industries to a single sawmill based on its output.

4.289 The fact remains that the vast majority of tenures in Canada are entered into directly between the provincial governments and �wood processing facilities,� and in most instances only wood processing facilities � such as sawmills that produce lumber � are eligible to obtain a tenure contract. The record clearly demonstrates that provincial stumpage is used by an extremely limited group of industries in Canada. Changing the definition of �industry� cannot change the objective facts.

4.290 The Panel likewise should reject Canada�s attempt to create exceptions to the specificity test contained in Article 2.1(c) of the SCM Agreement. Nothing in the text of Article 2 permits an actionable subsidy to escape the disciplines of the SCM Agreement based on the intent of the granting Member or the �inherent characteristics� of a good provided at below market rates. Nor does Article 2 require the �limited number of certain enterprises� to be established relative to the �eligible users,� as Canada claims. Finally, the Panel should dismiss Canada�s contention that a finding on the �limited number� prong is not sufficient to support a determination of specificity under Article 2.1(c). Both the language and underlying logic of Article 2.1(c) make clear that it is unnecessary to make findings on all prongs of the test for a determination of specificity in fact. The provisions of the SCM Agreement are clear, and the United States has met those obligations. Canada should not be permitted to rewrite the Agreement in a manner more to its liking.

4.291 Canada claims that the United States� final decision, in response to comments from Canadian parties, to use data from Minnesota as the basis for calculating the market benchmarks for Alberta and Saskatchewan was inconsistent with Articles 12.1, 12.3, and 12.8 of the SCM Agreement. The United States� conduct of this investigation was entirely consistent with its obligations.

4.292 Canada�s claim under Article 12.1 is premised on the assertion that parties were denied the right to present evidence because the use of alternative states, such as Minnesota, had not been an issue in this investigation. The use of northern US states generally, and the issue of the benchmark state to use for Alberta and Saskatchewan specifically, were very much at issue. Moreover, all parties had ample opportunity to present information and argument on this issue, as evidenced by Saskatchewan�s proposal that the United States use Alaska instead of Montana.

4.293 Canada�s claim under Article 12.3 of the SCM Agreement is equally unfounded. Canada has failed to cite to a single piece of record information on market benchmarks to which the parties were denied access.

4.294 Finally, Canada acknowledges that nothing in Article 12 suggests that an investigating authority must engage in endless cycles of notice and comment. Nevertheless, Canada�s claim under Article 12.8 is based on the erroneous premise that Article 12.8 requires the investigating authority to provide an opportunity for notice and comment with respect to the final decision made on each issue before the determination becomes final. Nothing in Article 12.8 imposes such a requirement.

4.295 In the context of the SCM Agreement, the �essential� facts are those that are necessary to determine whether definitive measures are warranted. A market benchmark is certainly essential to a determination of adequate remuneration, but all of the facts �under consideration� with respect to the calculation of the market benchmarks were made known to the parties. Significantly, Article 12.8 of the SCM Agreement refers to the �essential facts under consideration.� Thus, by its own terms, Article 12.8 is concerned with the ongoing investigative process during which the investigating authority is still �considering� the facts. Article 12.8, therefore, cannot be interpreted to apply to the investigating authority�s final decision, at which point the issues have been decided and the facts are no longer �under consideration.�

4.296 Article 12.8 does not impose any specific method of informing the parties of the �essential facts under consideration.� Rather, it specifies that the process used must inform parties �in sufficient time for the parties to defend their interests.� The United States� procedural rules are designed to guarantee a very open and transparent process in order to accomplish this goal, which was achieved in this case.

4.297 The record establishes that the United States provided parties with ample opportunity to provide information and argument on whether the Maine stumpage price should include studwood and pulpwood. Quebec submitted considerable information and argument on this very issue. In the end, the United States agreed with Quebec and adjusted the benchmark calculation accordingly. Canada claims, however, that the United States withheld information from the parties and denied them the opportunity to prepare presentations on the basis of that information, in violation of Article 12.3 of the SCM Agreement. The information allegedly withheld is the December 20, 2001, letter from the Maine Forest Products Council (�MFPC�). As discussed in response to the Panel�s questions, this information was provided to the parties in time for them to prepare presentations on the basis of that information. Neither party was afforded an opportunity for sur-rebuttal and there is no obligation in Article 12 to provide such an opportunity.

4.298 The thrust of Canada�s claim is a wholly unsubstantiated allegation that the United States intentionally withheld the MFPC Letter. The US regulations for the filing of information require the parties submitting the information to ensure that the information is placed on the record and provided to all interested parties. The MFPC Letter was not filed in accordance with those regulations. Canada�s claim that the United States acted inconsistently with Article 12.3 of the SCM Agreement must therefore fail.

4.299 For the reasons set forth above as well as in the United States� first written submission, oral statements at the first substantive meeting of the Panel, and first response to the Panel�s questions, the United States requests that the Panel reject Canada�s claims in their entirety.

G. SECOND ORAL STATEMENT OF CANADA

4.300 The following summarizes Canada's arguments in its second oral statement.

1. Financial Contribution

4.301 The ordinary meaning of �goods� is movable, tangible items. Immovables � such as buildings, roads, mineral deposits and trees in the forest � are not goods. Intangible rights, whether concerning intellectual property or real property, are not goods. In the facts of this case, tenure agreements and licences convey a right to harvest trees in a defined area. This is a right in respect of real property, and with that right come obligations that run regardless of any harvest; the right exists even in respect of trees that have not yet been planted. This right is in respect of an immovable � a standing tree � that is not a good; and it is a right in respect of something that may not even exist at the time the right is conveyed. It is an intangible and therefore not a good.

4.302 The ordinary meaning of �goods� excludes intangibles. An intangible right does not become a �good� just because it is a factor enabling the creation of a good. An intangible right does not become a �good� just because the right holder�s objective is to produce a good. Harvesting rights permit the holders to produce logs; however, rights are distinct from the trees to which they attach. Harvesting rights are also distinct from the good, the log, which is not produced until the harvester invests the effort and incurs the significant costs of harvesting and processing standing timber into logs. The provision of rights is not the same thing as the provision of goods.

4.303 The ordinary meaning of goods also excludes immovables subject only to limited exceptions. Among these exceptions are growing crops and �timber to be cut� that is identified in a sales contract. A timber sales contract covers specific trees; it does not cover just any trees that are or may be found in an identified area. Tenure agreements do not relate to identified trees, but as conceded by the US in its Second Written Submission, they relate to an identified area. Non-identified trees, whether or not they are in an identified area, are �immovables� and because they do not fit within the exception, they are not goods. The provision of immovables is not the same thing as the provision of goods.

4.304 The definition of subsidy has two elements. Financial contribution relates to what the government does, benefit to what the recipient receives. Article 1.1(a)(1) does not require a determination as to what the recipient ends up with at the end of the day or why the recipient enters into a given relationship with a government; rather, it requires a determination as to what the government provides. The government, in this instance, does not provide �cut timber�.

4.305 Harvesters subject to a tenure or licence assume a variety of forest management obligations for the period of the tenure or licence over the area covered by the tenure or licence. These obligations run regardless of the harvest of any trees. Tenures and licences do not involve the sale of anything � they involve the conferral of harvesting rights. The United States has admitted that tenures and licences involve identified areas. They do not involve identified trees. Standing trees under tenures and licences are not �goods� for the purposes of the SCM Agreement.

4.306 A right to harvest standing timber is an intangible interest in respect of real property and is therefore not �goods�; standing trees with roots firmly in the ground are immovables and therefore not �goods�. Stumpage programmes do not provide goods and do not constitute a financial contribution.

2. Benefit

4.307 Article 14(d) requires that the adequacy of remuneration be determined and measured on the basis of �prevailing market conditions � in the country of provision�. Accordingly, benefit must be determined on the basis of existing market conditions in Canada. Article 14(d) does not permit benefit to be determined and measured using benchmarks that are outside the country of provision. It also does not permit an investigating authority to measure adequacy of remuneration by comparing the government price to a hypothetical �fair market value�.

4.308 This interpretation of Article 14(d) is consistent with the Appellate Body�s analysis of Article 1.1(b) in Canada � Aircraft. In that case, the Appellate Body directed a comparison that is to be informed by what the recipient can obtain in the market. The Article 1.1(b) �market� in a provision of goods context under Article 14(d) is the actual �market� as it exists in the country of provision.

4.309 The US reliance on an extract from the second implementation panel in Brazil � Aircraft to argue that the comparison market must be a market undistorted by the government�s financial contribution is misplaced. First, Brazil � Aircraft did concern the standard set out in Article 14(b). At issue in that case was whether the PROEX payments made by the Brazilian government to certain banks resulted in a benefit to purchasers of Brazilian aircraft. In order to determine this, the panel looked at the impact of the PROEX payments on the terms and conditions of the export credit financing available to the purchasers of these aircraft. Therefore, though unstated, this part of the decision did in fact turn on the standard set out in Article 14(b).

4.310 Second, the reference to an undistorted market should be seen in its proper context � that is, against the background of the original case. The panel determined in that case that the comparison must be to the market for commercial lending rates and not another government market. The panel did not find that a commercial or private market rate or price cannot be used as a benchmark if there is government involvement in the marketplace.

4.311 The Appellate Body�s decision in US � Countervailing Measures on Certain EC Products also provides no support for the rejection by the US of the in-country evidence submitted by Canadian provinces. First, the Appellate Body�s decision was made in a different context that did not consider the explicit language of Article 14(d). Second, while that case involved consideration of the presumption at issue in arm�s-length sales by government, the plain language of Article 14(d) governs this case. In the present case, the SCM Agreement�s plain language requires a benefit determination based on benchmarks that reflect the �prevailing market conditions � in the country of provision�. This language establishes a clear rule, not a �presumption� that might be rebutted in limited circumstances.

4.312 The central issue before this Panel with respect to benefit is whether the SCM Agreement requires investigating authorities to use in-country evidence to determine adequacy of remuneration. Canada�s purpose in presenting the in-country evidence on the record in this case is to demonstrate that (i) the record contained substantial evidence of prevailing market conditions in Canada that should have been used to determine adequacy of remuneration, and (ii) even if arguendo the US was permitted to reject this evidence, the reasons it offers fail to stand up to any objective scrutiny.

4.313 In its Second Written Submission, the US has again conceded that data on private stumpage in both Qu�bec and Ontario exist. With respect to the TDAs, this data was developed by opposing commercial interests in Alberta to establish the market value of healthy standing timber. In addition, the four primary exporting provinces all provided evidence that showed that their stumpage programmes are administered consistent with market principles. Similarly, BC provided economic evidence that reinforced this conclusion. Finally, the United States� focus on alleged flaws in how public timber prices are set is misplaced as these are irrelevant to the benchmarks required under Article 14(d).

4.314 The US rationalizes the rejection of in-country evidence by discussing certain forest management practices that allegedly demonstrate that the provinces distort private timber markets. In particular, the United States makes selective reference to appurtenancy requirements, alleged restrictions on the transfer of tenures and annual allowable cuts in specific provinces. None of these practices were the subject of the investigation. Further the Final Determination contains no analysis that demonstrates that the provinces� management of their stumpage programmes distorted the market. Rather, USDOC relied on the Preamble to presume price suppression and supported this presumption with anecdotal evidence and a flawed economic analysis. In any event, as the panel in US � Softwood Lumber III  found, Article 14(d) does not allow an investigating authority to decline to use in-country prices because they may be affected by the government�s financial contribution.

4.315 Article 14(d) requires that adequacy of remuneration be determined using prevailing market conditions in the country of provision. Market prices available on the world market can become part of prevailing market conditions �in� the country of provision where there are imports of the good into the country of provision. If there are imports at a given price, then the price is no longer an out-of-country price, but rather is part of the prevailing market conditions in the country of provision.

4.316 The US prices used by USDOC as benchmarks do not fall within this category. USDOC found that the subsidized �good� was either standing timber or the right to harvest standing timber. US stumpage is not, and cannot, be made �available� in Canada. Standing trees growing in the US cannot be harvested in Canada, and the right to harvest these trees growing in the US cannot be exercised in Canada. Logs are not timber-harvesting rights, nor are they standing timber. Standing timber in the US cannot be transported across the border, as it must be cut in the US.

4.317 Also, contrary to the United States� assertion, USDOC did not compare prevailing market conditions in Canada with those in the US. USDOC ignored a host of differences in market conditions, and as for the adjustments it did make, failed to adequately take into account differences between individual provinces and US comparison areas.

4.318 The US dismisses the provincial evidence demonstrating that stumpage programmes are operated in accordance with market principles as not relevant to a benefit analysis. This is so, it argues, because the �standard is not the cost to government, but rather benefit to the recipient.� The US thus dismisses the third benchmark under the regulations it adopted to implement the results of the Uruguay Round. There is no question that the United States believed, at the time of adoption, that this benchmark was consistent with the analysis contemplated by Article 14(d), and was not, as the US now argues, based on a �cost to government� standard.

4.319 In using this benchmark over the last seven years and as recently as two weeks ago, USDOC has examined whether the government in question has covered its costs; whether it has earned a reasonable rate of return; and whether it applied market principles in setting its rates or prices. In considering these factors, USDOC has described its analysis as examining �whether the government�s price was determined according to the same market factors that a private . . . [party] would use . . . . .�²⁰ USDOC�s current attempt to dismiss the third benchmark as irrelevant is contradicted by its consistent application of this benchmark since the regulations were promulgated.

4.320 USDOC also expressly refused to consider evidence demonstrating that provincial stumpage systems cannot confer a trade advantage. This economic evidence demonstrated that, for in situ natural resource markets, receipt of the input for a lower fee or charge does not affect the supply curve for the exported product. In this instance, USDOC assumed the existence of trade distortion rather than considering evidence that demonstrated that there was no effect on the marginal cost or supply. This evidence demonstrates that provincial stumpage programmes neither increase the production of logs and lumber, nor lower their prices relative to a private competitive market.

3. Pass-through

4.321 The pass-through issue in this dispute is not simply about calculating the amount of the alleged stumpage subsidy; it is first about determining whether a subsidy exists. This means making the required determinations of an indirect financial contribution and of a benefit conferred thereby. The amount of a subsidy cannot be calculated, aggregated or allocated until a subsidy is first determined to exist.

4.322 The United States does not contest that arm�s-length transactions exist and that no subsidy has been determined to exist in those instances. The United States impermissibly presumed subsidization in all cases other than alleged direct subsidization, asserting incorrectly that all producers of subject lumber were �direct recipients� of Crown stumpage. There is no exception in the SCM Agreement that allows a Member to presume subsidization in a countervailing duty investigation.

4.323 The United States has countervailed softwood lumber, not standing timber. The subject lumber products are a few examples of the many downstream products produced in many distinct industries that receive the alleged subsidy directly. If the subsidy to timber harvesters is presumed to have passed through to downstream producers of the subject lumber, then it must be presumed to have passed through to producers of all downstream products. The question then is how the stumpage subsidy could be limited to the mills upstream in the US specificity finding, when all products and industries downstream are presumed to have received it. US arguments on pass-through make US arguments on specificity impossible, and internally contradictory.

4. Specificity

4.324 There are two central errors in the US approach to specificity: first, it applied the wrong standard; and second, it incorrectly found that stumpage was used by a limited group of industries.

4.325 First, sole reliance on the �limited users� factor, as interpreted and applied by the United States, renders the specificity requirement meaningless. The factors set out in Article 2.1(c) do not, in and of themselves, establish specificity. Nowhere in that provision does it say, �a subsidy is de facto specific if one or more of the following factors exist�; nowhere does it say, �where the users of the subsidy are limited in number, no matter the reason, the subsidy shall be deemed to be specific�. Article 2 requires not automaticity, but clear substantiation on the basis of positive evidence.

4.326 Second, the words �industry� and �group of industries� require an objective analysis of the record evidence in terms of the products produced by the users of a programme. In this case, an objective assessment of the products produced by the thousands of actual immediate users of stumpage identifies at least 23 standard categories of industries. When the analysis is extended to include indirect users of stumpage, as the US has done in its impermissible pass-through presumption, many more and varied industries are added.

4.327 The US interpretation reads the word �group� out of Article 2 altogether. However, as the United States itself put it recently in US � Offset Act (Byrd Amendment), the issue is �how small and homogenous a group of beneficiaries must be in order to qualify as a �a group of enterprises or industries��, not whether the subsidy is used by the entire economy, or simply by more than one enterprise or industry. The products produced by stumpage users are as numerous and diverse as those produced in the agricultural sector, which in the view of the United States is neither a �single large industry� nor even a group of industries. Again, in US � Offset Act (Byrd Amendment), the United States argued that groupings �such as �all manufacturing� and �all agriculture,� are too broad to qualify as a �group of enterprises or industries� for specificity purposes.� Stumpage is used by many varied industries because it is a right of access to exploit a natural resource. That natural resource is not altered in any way by governments.

4.328 Moreover, the purpose of countervailing duties is to protect domestic producers from injury caused by subsidized imports of like products. Logically, therefore, the level of aggregation used to identify the user industries for specificity purposes (through the products they produce) should be commensurate with the level of aggregation used to identify the allegedly injured domestic producers (through the products they produce). To do otherwise impermissibly concentrates for specificity purposes an otherwise broad alleged subsidy and imputes it solely to producers of the subject merchandise � exactly what the United States did in the Final Determination.

5. Calculations

4.329 Canada�s Article 19.4 claims rest on two legal foundations. First, and as the US agrees, Article 19.4 requires that a countervailing duty rate must not exceed the subsidy per unit rate found in respect of exported and subsidized goods. Second, the subsidy per unit rate must be calculated correctly, based on a correctly determined subsidy amount, and a correctly determined subsidy. The United States disagrees. It argues that Article 19.4 requires nothing more than a reconciliation of a subsidy per unit rate, however determined, with the countervailing duty rate actually imposed. Such an outcome renders the obligation in Article 19.4 meaningless.

4.330 The Panel must determine whether the United States has met its legal obligation, under Article 19.4, to limit the amount of the countervailing duty rate to the correctly determined alleged subsidy per unit. The United States has a legal obligation to ensure that any countervailing duty imposed does not go beyond the purposes of a countervailing duty � to offset injurious subsidization. It may not escape its obligations by characterizing every issue as a matter that is within the unfettered discretion of investigating authorities.

6. Conduct of the Investigation

4.331 Canada�s claims concerning the conduct of the investigation at issue are independent from its claims regarding the illegality of the Final Determination of subsidy. In conducting this investigation, the United States violated its obligations under Article 12, and the countervailing duties imposed on the basis of this tainted investigation therefore violate Articles 10 and 32.1 of the SCM Agreement.

4.332 First, USDOC determined that its benchmark states would border the relevant provinces. The US appeared before the WTO and argued that its benchmarks were valid because they were �contiguous� to the relevant provinces. The United States switched the benchmark state to compare with Alberta and Saskatchewan from Montana � a contiguous state � to Minnesota � a non-contiguous state � without giving interested parties notice or any opportunity to comment on the specific choice of the benchmark. The United States thus violated Articles 12.1, 12.3 and 12.8.

4.333 Second, a senior USDOC official specifically requested information vital to USDOC�s use of Maine as a benchmark state for Qu�bec. The Maine Forest Products Council sent a letter in response. USDOC did not put the letter on the record for months. The petitioners then submitted new evidence � and the United States admits that new evidence was submitted � in response to USDOC�s request for comments on the letter. USDOC refused to permit interested parties to respond to the petitioners� new evidence despite the fact that USDOC relied on it in making its Final Determination. The United States failed to provide a timely opportunity to see relevant information; it failed to provide an opportunity for interested parties to prepare presentations on the basis of that relevant new information. It thus violated Article 12.3.

H. SECOND ORAL STATEMENT OF THE UNITED STATES

4.334 The following summarizes the United States' arguments in its second oral statement.

4.335 The record demonstrates that the provinces own timber and, through provincial tenures, provide timber to lumber producers. There should therefore be no question that the provinces provide a financial contribution within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement.

4.336 The ordinary meaning of �goods� includes standing timber. Canada�s argument to the contrary is reduced to the assertion that tenures are not sales of �timber to be cut,� within the meaning of the US UCC, even though the sole reason for acquiring a tenure is to harvest timber, and the tenure holder pays for and receives title to only the timber it cuts. Moreover, the UCC provides that the sale of timber to be cut is always a contract for the sale of goods, regardless of whether the timber is �identified� at the time of the contract. More to the point, however, the provinces �provide� a �good� within the meaning of Article 1.1(a)(1)(iii) of the SCM Agreement.

4.337 Canada also suggests, without citation to record evidence, that many tenures create freely transferable �rights to harvest.� The record evidence establishes, however, that the provinces retain control over tenures, which cannot be transferred without the provinces� approval. The subcontracting of harvesting operations is not the sale or transfer of a �right to harvest.� The tenure holder, not the subcontractor, remains at all times the province�s contracting party.

4.338 It is the substance of what occurs in the provincial tenure systems that is controlling, i.e., whether a provision of goods takes place. As discussed in our previous submissions, the ordinary meaning of the terms �provides� and �goods or services other than general infrastructure� sweeps broadly. No matter how Canada characterizes provincial tenures, the fact remains that the provinces are providing timber to lumber producers. A financial contribution therefore exists.

4.339 The key legal issue in this dispute is whether Article 14(d) of the SCM precludes, in all cases, the use of data from sources outside the country under investigation to determine the adequacy of remuneration. As the Appellate Body has stated, the issue is whether �the �financial contribution� makes the recipient �better off� than it would otherwise have been, absent that contribution.�²¹ In other words, the Appellate Body recognized that in order to determine whether a financial contribution confers a benefit, it is essential to compare the position of the recipient with the financial contribution to what the position of the recipient would have been absent the financial contribution. Moreover, the Appellate Body�s statement provides the context for the Appellate Body�s conclusion that the point of comparison is the �marketplace.� Thus, as the Brazil � Aircraft panel concluded, looking to the marketplace to determine whether the recipient is better off than it would otherwise have been absent the financial contribution necessarily means looking to a marketplace undistorted by the government�s financial contribution.²²

4.340 It is the view of the United States, as well as the European Communities, that Article 14(d) does not prohibit Members from relying on data from sources outside the country of provision to determine the adequacy of remuneration if reliable, market-driven pricing data does not exist in the country of provision. We will not repeat the arguments supporting that conclusion, but will comment on Canada�s flawed three-pronged analysis of Article 14(d).

4.341 First, Article 14(d) provides that the adequacy of remuneration must be determined �in relation to� prevailing market conditions in the country of provision. Canada, however, substitutes �in relation to� with �on the basis of� or �in comparison with.� The more reasonable interpretation is that the broader phrase �in relation to� was agreed to by Members because Article 14 explicitly sets out �guidelines,� i.e., general principles, not detailed rules. �In relation to� is sufficiently broad to allow for various means of performing a comparison that relates to market conditions in the country under investigation, rather than limiting Members to analyses �on the basis of,� or �in comparison with,� certain types of data. Permitted methods can include analyses that rely on data from sources outside the country, if the data is probative of the fair market value for the good in the country of provision.

4.342 Second, Canada ignores the word �market� in the phrase �prevailing market conditions,� as if any conditions are �market� conditions. The Panel, however, must give meaning to the word �market.� In that regard, the logic of the Brazil � Aircraft Panel Report is compelling. In determining whether the government�s financial contribution confers a benefit � i.e., whether the recipient is better off with the financial contribution than it would otherwise have been absent the financial contribution � the �market� conditions must be conditions that are undistorted by the government�s financial contribution.

4.343 Third, adequacy of remuneration must be determined in relation to prevailing market conditions �in the country of provision.� That begs the question, however, whether there are in fact �market� conditions for the good �in the country of provision� that provide probative evidence for determining adequacy of remuneration. Where such evidence does not exist in the country of provision, as in this case, nothing in Article 14(d) precludes a Member from relying on market data from sources outside the country that is probative of fair market value in the country of provision.

4.344 The importance of using market benchmarks, even if they are based on data from sources outside the country of provision, is underscored by Article 15(b) of the Protocol on the Accession of the People�s Republic of China²³, which is cited, but misinterpreted, by Canada. The United States negotiated Article 15(b) of the China Protocol because the United States, along with other Members, recognized that China was in transition from a state-controlled economy to a market economy. Although Article 14(d) of the SCM permits the use of market data outside China, it only applies in countervailing duty cases under Part V of the SCM Agreement. Recognizing the importance of �market� benchmarks, the Members incorporated the language that Canada references in Article 15(b) of the China Protocol to clarify that external benchmark data can be used under Article 14(d) of the SCM Agreement, and also to ensure that such data can be used in proceedings under Parts II and III of the SCM Agreement.

4.345 Other provisions in the China Protocol likewise repeat obligations already binding on WTO Members. For example, the China Protocol provides that �China shall ensure that internal taxes and charges . . . shall be in conformity with the GATT 1994.�²⁴ The inclusion of this provision does not mean that other Members need not ensure that their internal taxes conform to GATT 1994, yet this is effectively Canada�s argument.

4.346 With respect to the facts of this case, the United States has, in its prior submissions and oral presentations, demonstrated that four of the six provinces did not provide private prices for timber that could be used for market benchmark purposes. Canada has failed to refute that fact. Rather, it argues that evidence that the provinces earn a profit on timber sales is sufficient to establish that they do not provide a benefit, even though those profit calculations do not include any cost or value for the trees themselves. More importantly, however, a sale for less than adequate remuneration � even a profitable one � confers a benefit.

4.347 The record evidence also demonstrates that the overwhelming state control of timber sales in Canada distorts sales in the private sector. Canada responds to this evidence by inviting the Panel to conduct de novo review. Canada argues, for example, that the United States �ignored� evidence of market conditions in Canada. The record in this case demonstrates that the United States in fact considered and weighed all of the evidence and was persuaded by economic analyses, which are cited to in the Final Determination, and other documentary evidence that the state-controlled timber sales systems distorted the private market.

4.348 Canada also now states that it has �consistently taken the position . . . that a price that is available to purchasers in the country of provision makes that price part of the prevailing market conditions in the country of provision.�²⁵ Given that US timber prices are available to purchasers in Canada, the United States� benchmark calculation is consistent with the position Canada now endorses. Canadian lumber producers can purchase US timber, cut it (or have it cut), and transport it to their mills in Canada.

4.349 The United States has also established that the calculation of the market benchmarks included appropriate adjustments for conditions of sale in Canada. Canada�s attempts to call those adjustments into question do not withstand scrutiny.

4.350 Finally, Canada has not made a prima facie case that the United States erred in failing to conduct a market distortion analysis. Canada has failed to identify any obligation in the SCM to conduct such an analysis because no such obligation exists. The United States� benefit calculation is consistent with Article 14. Nothing further is required, and obligations that are not found in the Agreement may not be imposed on the United States.

4.351 Under Article 2.1 of the SCM Agreement, a subsidy is specific if it is used by a limited number of industries or group of industries. The unrefuted record facts demonstrate that provincial tenures are used by a very limited group of timber processing industries. To construct an argument that these provincial programmes are, nonetheless, not specific, Canada artificially inflates the very limited group of industries.

4.352 Canada interprets the term �industry� so narrowly that almost every product becomes an industry unto itself. Canada�s dissection of the timber processing industries flies in the face of the ordinary meaning of �industry� as the term is used in Article 2. Moreover, Canada�s position with respect to the timber processing industries is inconsistent with its acknowledgement that a subsidy used by a single large industry, such as automobiles or textiles, may be specific notwithstanding the diverse range of products the industry produces.

4.353 Canada also challenges the United States� specificity finding by inventing criteria that do not exist. Nothing in Article 2.1(c) requires an investigating authority to consider the government�s intent or the �nature� of the good when determining whether a subsidy is specific. It is entirely permissible to find specificity based solely on the limited number of users.

4.354 Canada cites the negotiating history of the SCM in an effort to overcome the lack of any support in the text of Article 2.1(c) for its claim that Article 2.1(c) requires consideration of factors such as the �inherent characteristics� of the good and evidence of intentional targeting. In fact, the negotiating history that Canada cites demonstrates that such concepts ultimately were not adopted by the Members. The Members abandoned language on intent and �inherent characteristics� after the second Cartland draft, and such factors may not be read into the SCM Agreement.

4.355 The only additional factors that a Member must take into account under Article 2.1(c) are the extent of diversification of economic activities within the granting authority�s jurisdiction and the length of time the subsidy programme has been in operation. No one argued that the number of subsidy recipients is limited because the programme had not been in operation long enough to be more widely distributed. The issue of economic diversification was raised only by British Colombia (�B.C.�). The evidence B.C. submitted, however, demonstrated that the timber processing industries accounted for approximately 6 per cent of B.C.�s economy. Thus, based on B.C.�s own data, 94 per cent of B.C.�s economy did not use the provincial tenure system.

4.356 The United States� views on Canada�s claims regarding the calculation of the ad valorem rate and the conduct of the investigation have been presented in our prior submissions and statements. We would like to make only a few additional points in response to Canada�s second submission.

4.357 First, Canada mischaracterizes the United States� views on Article 19.4 of the SCM Agreement. Article 19.4 provides that the countervailing duty rate must be calculated on a per-unit basis. It also provides that the countervailing duty levied may not exceed the subsidy found to exist. That is the extent of the obligations in Article 19.4. The United States� argument that Canada has failed to identify any obligation in Article 19.4 to conduct an upstream subsidy analysis or to allocate subsidies by volume rather than by value is a far cry from arguing that a Member may impose countervailing duties at any rate it wishes. The fact remains, however, that Canada has failed to identify any obligation in Article 19.4 that supports its claim.

4.358 Second, the United States calculated the ad valorem duty rate by dividing the subsidy by the value of the output of the lumber production process. Canada argues that the numerator in that ad valorem rate calculation was impermissibly inflated by the inclusion of that portion of the Crown timber that ended up as products other than lumber. Canada�s numerator argument is simply an argument that the United States was required to allocate the subsidy on the basis of volume rather than on the basis of value. As discussed previously, Article 19.4 simply obligates Members to calculate the countervailing duty rate on a per-unit basis. There is no requirement to allocate an input subsidy based on volume rather than value.

4.359 Third, Canada continues to suggest that the Minnesota Public Stumpage Price Report (�Minnesota Price Report�) specifies a conversion factor of 6.25 for converting from thousand board feet to cubic meters. It does not. The Minnesota Price Report contains sawtimber prices reported in thousand board feet and pulpwood prices reported in cords. The price report contains a factor that Minnesota uses to convert between cords and board feet. The United States, however, only used the sawtimber prices, which are bid, sold, and reported in thousand board feet, and therefore needed to convert from board feet to cubic meters. The Minnesota Price Report does not contain such a conversion factor. We note, however, that the timber sales manual of the Minnesota Department of Natural Resources (�Minnesota Timber Sales Manual�), which publishes the Minnesota Price Report, was on the record. The Minnesota Timber Sales Manual provides a conversion factor of 3.48 cubic meters per thousand board feet. Canada does not, however, advocate using that conversion factor. Rather, Canada derived its own conversion factor from selected information in the Minnesota Price Report.

4.360 There is no one conversion factor that is universally accepted. The record evidence suggested a wide range of possible conversion factors, ranging from 3.48 to 8.51. The United States considered all of that evidence and provided a reasoned explanation for its decision to rely on the factors in the report by the International Trade Commission. The choice of conversion factor is therefore entirely consistent with the SCM Agreement.

4.361 Finally, Canada suggests that, even though the provinces knew the criteria the United States was using to select the benchmark states and had all the data on the states under consideration, Alberta and Saskatchewan were not on notice of the possibility that the United States could select a state more than 1,000 kilometres away. That assertion is contradicted by the record of the investigation. Alberta and Saskatchewan both argued that the United States should not use Montana as a benchmark state because of differences in the species mix. Saskatchewan also proposed using data from Alaska, which is more than 1,000 kilometres from, and obviously not contiguous with, Saskatchewan. Thus, it is evident that the provinces knew that factors such as climate, terrain, and species mix � not proximity � were the key considerations, and that a non-contiguous state might be selected for the benchmark. The record of the investigation therefore establishes that the provinces knew the essential facts under consideration.

4.362 For the reasons discussed above and in our prior submissions and presentations to the Panel, the United States asks the Panel to dismiss Canada�s claims.