Notice of Determination Under Section 129 of the Uruguay Round Agreements Act: Antidumping Measures on Certain Softwood Lumber Products From Canada, 22636-22646 [05-8745]

Agencies

• DEPARTMENT OF COMMERCE
• International Trade Administration

[Federal Register Volume 70, Number 83 (Monday, May 2, 2005)]
[Notices]
[Pages 22636-22646]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: 05-8745]


-----------------------------------------------------------------------

DEPARTMENT OF COMMERCE

International Trade Administration

[A-122-838]


Notice of Determination Under Section 129 of the Uruguay Round 
Agreements Act: Antidumping Measures on Certain Softwood Lumber 
Products From Canada

AGENCY: Import Administration, International Trade Administration, 
Department of Commerce.
SUMMARY: Consistent with section 129 of the Uruguay Round Agreements 
Act, which governs the Department of Commerce's (the Department's) 
actions following World Trade Organization (WTO) reports, the 
Department has calculated new rates with respect to the antidumping 
duty investigation on certain softwood lumber products from Canada, in 
order to implement the recommendations of the WTO Appellate Body. On 
April 27, 2005, the U.S. Trade Representative, after consulting with 
the Department and Congress, directed the Department to implement this 
determination. The new rates apply to unliquidated entries of the 
subject merchandise that are entered, or withdrawn from warehouse, for 
consumption on or after April 27, 2005.

Effective Date: April 27, 2005.

FOR FURTHER INFORMATION CONTACT: Constance Handley or Shane Subler, at 
(202) 482-0631 or (202) 482-0189, respectively; AD/CVD Operations, 
Office 1, Import Administration, International Trade Administration, 
U.S. Department of Commerce, 14th Street & Constitution Avenue, NW., 
Washington, DC 20230.

SUPPLEMENTARY INFORMATION:

Background

    On April 2, 2002, the Department published a final determination of 
sales at less than fair value (LTFV) in the antidumping duty 
investigation on certain softwood lumber from Canada. See Notice of 
Final Determination of Sales at Less Than Fair Value: Certain Softwood 
Lumber Products from Canada, 67 FR 15539 (April 2, 2002) (Final 
Determination) and accompanying Issues and Decision Memorandum. 
Following an affirmative injury determination issued by the United 
States International Trade Commission, the Department published an 
antidumping duty order on this product on May 22, 2002. See Notice of 
Amended Final Determination of Sales at Less Than Fair Value and 
Antidumping Duty Order: Certain Softwood Lumber Products From Canada, 
67 FR 3606 (May 22, 2002).
    Subsequently, the Canadian government requested the establishment 
of a WTO dispute resolution panel (the Panel) to consider various 
aspects of the Department's final determination in this case. The Panel 
circulated its report on April 13, 2004. See United States--Final 
Dumping Determination on Softwood Lumber from Canada, WT/DS264/R (April 
13, 2004).
    On May 13, 2004, the United States and Canada appealed certain 
findings and conclusions in the Panel report. The WTO Appellate Body 
(the Appellate Body) issued its report on August 11, 2004. See United 
States--Final Dumping Determination on Softwood Lumber from Canada, WT/
DS264/AB/R (August 11, 2004) (Appellate Body Report). The Appellate 
Body Report and the Panel report, as modified by the Appellate Body 
Report, were adopted by the WTO Dispute Settlement Body (DSB) on August 
31, 2004. See Minutes of the Meeting, Dispute Settlement Body, August 
31, 2004, WT/DSB/M/175 (Sept. 24, 2004).
    On September 27, 2004, the United States indicated to the DSB that 
it intended to implement a decision consistent with the recommendations 
and rulings of the DSB. See WTO News, https://www.wto.org/english/news_
e/news04_e/dsb_27sep04_e.htm. On November 5, 2004, pursuant to 
section 129(b)(2) of the Uruguay Round Agreements Act (URAA), the 
United States Trade Representative requested that the Department issue 
a determination that would render the Department's actions in the 
investigation not inconsistent with the findings of the DSB.
    On January 31, 2005, the Department issued its Preliminary 129 
Determination.\1\ On February 22, 2005, the Department received a joint 
brief filed by the British Columbia Lumber Trade Council and its 
constituent associations; the Ontario Forest Industries Association; 
the Ontario Lumber Manufacturers Association; the Quebec Lumber 
Manufacturers Association; Abitibi Group; Canfor Corporation; Slocan 
Forest Products Ltd.; Tembec Inc.; West Fraser Mills Ltd.; and 
Weyerhaeuser Company (collectively, the Canadian Parties).\2\ On

[[Page 22637]]

March 7, 2005, the Department received rebuttal comments from the 
Coalition for Fair Lumber Imports (the Coalition), a domestic 
interested party.
---------------------------------------------------------------------------

    \1\ See Preliminary Determination Under Section 129 of the 
Uruguay Round Agreements Act: Antidumping Measures on Certain 
Softwood Lumber Products from Canada (Preliminary 129 
Determination), accessible at https://ia.ita.doc.gov/download/
section129/Canada-Lumber-129-Prelim-013105.pdf. This document is 
also on file in the Central Records Unit, Room B-099 of the main 
Commerce Building.
    \2\ See letter from the Canadian Parties to the Department, 
dated February 22, 2005 (Canadian Parties' Brief).
---------------------------------------------------------------------------

    On April 15, 2005, the Department issued its final Section 129 
Determination. See Notice of Determination Under Section 129 of the 
Uruguay Round Agreements Act: Antidumping Measures on Certain Softwood 
Lumber Products from Canada. On April 19, 2005, the Department 
forwarded its final determination to the U.S. Trade Representative. On 
April 25, 2005, the U.S. Trade Representative held consultations with 
the Department and the appropriate congressional committees with 
respect to this determination. On April 27, 2005, in accordance with 
sections 129(b)(4) and 129(c)(1)(B) of the URAA, the U.S. Trade 
Representative directed the Department to implement this determination.
    Section 129 of the URAA \3\ governs the nature and effect of 
determinations issued by the Department to implement findings by WTO 
panels and the Appellate Body. Specifically, section 129(b)(2) provides 
that ``notwithstanding any provision of the Tariff Act of 1930 * * *,'' 
within 180 days of a written request from the U.S. Trade 
Representative, the Department shall issue a determination that would 
render its actions not inconsistent with an adverse finding of a WTO 
panel or the Appellate Body. See 19 U.S.C. 3538(b)(2). The Statement of 
Administrative Action, U.R.A.A., H. Doc. 316, Vol. 1, 103d Cong. (1994) 
(SAA), variously refers to such a determination by the Department as a 
``new,'' ``second,'' and ``different'' determination. See SAA at 1025, 
1027. This determination is subject to judicial review separate and 
apart from judicial review of the Department's original determination. 
See 19 U.S.C. 1516a(a)(2)(B)(vii).
---------------------------------------------------------------------------

    \3\ Citation to ``section 129'' refers to section 129 of the 
URAA, codified at 19 U.S.C. 3538.
---------------------------------------------------------------------------

    In addition, section 129(c)(1)(B) of the URAA expressly provides 
that a determination under section 129 applies only with respect to 
unliquidated entries of merchandise entered, or withdrawn from 
warehouse, for consumption on or after the date on which the U.S. Trade 
Representative directs the Department to implement that determination. 
In other words, as the SAA clearly provides, ``such determinations have 
prospective effect only.'' SAA at 1026. Thus, ``relief available under 
subsection 129(c)(1) is distinguishable from relief in an action 
brought before a court or a {North American Free Trade 
Agreement{time} (NAFTA) binational panel, where* * * retroactive relief 
may be available.'' Id.

Appellate Body Findings and Conclusions

    Article 2.4.2 of the Agreement on Implementation of Article VI of 
the General Agreement on Tariffs and Trade 1994 (the Antidumping 
Agreement) provides that there are three means of calculating a dumping 
margin ``during the investigation phase.'' The agreement states that 
``normally'' a margin ``will be established on the basis of a 
comparison of a weighted average normal value with a weighted average 
of prices of all comparable export transactions' or that it will be 
established ``by a comparison of normal value and export prices on a 
transaction-to-transaction basis.'' The third means of comparison, a 
comparison of ``a normal value on a weighted average basis with 
individual export transactions,'' is provided for when certain criteria 
exist.
    In the investigation of softwood lumber from Canada, the Department 
calculated dumping margins for the investigated respondents using 
weighted-average-to-weighted-average comparisons. Specifically, the 
Department compared weighted-average export prices (EPs) or constructed 
export prices (CEPs) to weighted-average normal values (NV). When the 
EP or CEP was greater than the NV, the comparison showed no dumping. In 
these circumstances, the Department did not offset or reduce the amount 
of dumping found on other comparisons based on the amount by which the 
EP or CEP exceeded the normal value for distinct comparisons. When the 
EP or CEP was less than the normal value, the comparison was considered 
to have revealed dumping. In order to calculate the weighted-average 
dumping margin, the Department aggregated the amount of dumping found 
through these comparisons and divided it by the aggregate value of all 
U.S. sales (regardless of whether they were dumped) to ensure that the 
results took account of all comparisons and, thus, all U.S. sales, 
dumped and non-dumped.
    In its report, the Appellate Body rejected the United States' 
arguments (1) that the text of Article 2.4.2 of the Antidumping 
Agreement did not address the methodology at issue in this 
investigation; (2) that certain WTO members, including the United 
States, did not offset their calculations for non-dumped comparisons in 
their investigation calculations before, during, and following the 
implementation of the Antidumping Agreement, and that absent language 
addressing this methodology in the Agreement, members did not negotiate 
and agree that this methodology should be considered impermissible, and 
(3) that under Article 17.6 (ii) of the Antidumping Agreement, the 
Appellate Body was required to find that WTO members which applied this 
methodology acted in conformity with Article 2.4.2 of the Agreement. 
See paragraphs 107-108 of the Appellate Body Report.
    The Appellate Body concluded, at paragraph 108 of its decision, 
that ``based on the ordinary meaning of Article 2.4.2 read in its 
context,'' the Department's comparison methodology was ``prohibited 
when establishing the existence of margins of dumping under the 
weighted-average-to-weighted-average methodology.'' The Appellate Body 
did not address the other methodologies provided for in Article 2.4.2, 
namely * * * ``the transaction-to-transaction methodology'' or * * * 
``the weighted-average-to-individual methodology.'' See id. at 
paragraph 63 and 104-105.

Implementation

    In light of the Appellate Body's findings and recommendations, we 
have determined to apply the transaction-to-transaction methodology in 
this Section 129 Determination. Therefore, the Department is 
implementing the recommendations and rulings of the DSB as follows.
    To determine the dumping margin for each respondent, we matched 
individual transactions in the U.S. sales database with individual 
transactions in the home market database. See also Comment 7. In 
seeking to determine which specific home-market transaction would be 
the most suitable match for a given U.S. transaction, we began our 
analysis with the model-match characteristics used in our Final 
Determination. Consistent with our Final Determination, we did not 
match across product type, species, or grade group.
    Because lumber prices were extremely volatile and the market was in 
a constant state of flux during the period of investigation (POI), we 
first attempted to find an identical match at the same level of trade 
on the same day. If no identical match was found, we looked for an 
identical home-market sale the day before the U.S. sale, then the day 
after the U.S. sale, and so forth, up to seven days before or after the 
U.S. sale. We did not match U.S. sales to home market sales that 
occurred either more than seven days before or more than

[[Page 22638]]

seven days after the date of the U.S. sale. If no identical sale was 
found at the same level or trade, we looked for an identical match at a 
different level of trade. We then began to look for the most similar 
sale, based on product characteristics and level of trade, in the same 
manner.
    When sales were equally similar based on product characteristics, 
we identified the sale with the smallest difference in the variable 
cost of manufacturing as being the most similar. We did not match sales 
whose difference in variable cost exceeded 20 percent of the total cost 
of manufacturing of the U.S. sale.
    We limited the window to sales within a two-week time frame because 
we are looking for a specific sale that represents the best possible 
match. Given the high level of price volatility, we felt that a window 
period of any longer than seven days on either side of individual U.S. 
sales would result in these sales being matched to home market sales 
made under different market conditions. We note in cases where price 
volatility is not as important a consideration, it may be more 
appropriate to use another period, such as the 90/60-day window period 
used in administrative reviews.
    Within these parameters, we found a significant number of instances 
in which more than one home market sale qualified as an equally 
appropriate match. In order to identify the most appropriate match 
among the equally qualified sales, we looked for the sale that was the 
most similar in quantity to the U.S. sale. Section 773(a)(6)(C)(i) of 
the Tariff Act of 1930, as amended (the Act), contemplates that the 
sale quantity may have an effect on price. While the parties did not 
claim a quantity adjustment in this case, to the extent that the 
quantity of merchandise sold may affect the price of an individual 
transaction, we have taken that factor into account by using it as our 
first ``tie-breaker.''
    For all companies, if there was still more than one equally 
appropriate match, we took customer categories, as reported by the 
individual respondents, into account. In order to do so, we had to give 
the customer categories a numerical ranking, to reflect which 
categories would be considered the most similar. Wherever possible, we 
attempted to be consistent between companies. For example, we 
considered wholesalers to be more comparable to distributors than to 
retailers. Where there were still multiple equally comparable 
transactions, we looked for the transaction with the most comparable 
channel of distribution.
    When there remained multiple equally comparable transactions, we 
attempted to distinguish the single most appropriate match based on 
total movement expenses. Movement is the most significant expense 
related to the sale of softwood lumber. The amount of movement expenses 
can be considered indicative of the distance between the customer and 
the mill, and of the logistical coordination necessary to comply with 
the delivery terms of the sale. One company, Slocan, reported 
commissions. Accordingly, for this company, as a ``tie-breaker,'' we 
also looked at whether or not a commission was paid. We did not 
consider the total amount of the commission because the commission was 
price dependent: considering the amount of the commission would result 
in a match to the sale with the most similar price, rather than one 
made under the most similar conditions.
    The final criterion we used to distinguish among equally comparable 
transactions was the number of days between payment and shipment. We 
used the number of days that payment was outstanding rather than the 
code for terms of sale, because the former more accurately reflects 
exactly when the customer paid. We did not use indirect selling 
expenses as a tie-breaker because such expenses are strictly price-
dependent. Just as in the case of commissions, relying on indirect 
selling expenses to define the most similar sale would result in 
selecting the sale with the closest price as the match, rather than the 
sale made under the most similar conditions. After we considered these 
criteria, a small number of U.S. sales still had more than one equally 
comparable home market match. In these cases, we programmed the 
computer to select the first observation on the short list of equally 
comparable sales.
    We believe that there are particular benefits from this analysis 
which do not exist in the context of the weighted-average-to-weighted-
average comparisons. It is beyond question that the prices for lumber 
during the POI in both the United States and Canadian markets were 
volatile. See Notice of Final Determination of Sales at Less Than Fair 
Value: Certain Softwood Lumber from Canada, 67 FR 15539 (April 2, 
2002), and accompanying Issues and Decision Memorandum at Comment 4 
(Softwood Lumber Decision Memo); see also Memorandum from Constance 
Handley, Program Manager, to the File, re: Price Volatility, dated 
January 28, 2005. To the extent that the sales volume of a particular 
product varies over time and between the markets, the weighted-average 
price of any particular product could be skewed toward a period of low 
prices in one market and toward a period of high prices in the other 
market. In such a case, the weighted-average margin calculated for that 
product would not reflect the dumping, or lack of dumping, that may 
have occurred on the individual sales incorporated into the average. In 
the transaction-to-transaction analysis, however, the matching of 
identical or similar merchandise within a narrow time frame allows us 
to judge more accurately whether dumping was occurring when sales were 
made under the same market conditions.
    With respect to United States law on this issue, section 
777A(d)(1)(A)(i) and (ii) of the Act provides that in antidumping 
investigations, the Department may calculate a dumping margin using 
either weighted-average-to-weighted-average comparisons or transaction-
to-transaction comparisons, with no stated preference.
    Congress, in the SAA, stated that ``normally'' the Department will 
measure dumping margins on the basis of weighted-average-to-weighted-
average comparisons. See SAA at 842. The SAA states that a transaction-
to-transaction analysis ``would be appropriate in situations where 
there are very few sales and the merchandise sold in each market is 
identical or very similar or is custom made. However, given past 
experience with this methodology and the difficulty in selecting 
appropriate comparison transactions, the Administration expects that 
the Department will use this methodology far less frequently than the 
average-to-average methodology. Id. at 842-43.
    Section 19 CFR 351.414(c) of the Department's regulations, adopted 
shortly after the URAA came into force, adopted the SAA's preference 
for weighted-average-to-weighted-average comparisons in investigations, 
explaining that the Department will only use the transaction-to-
transaction means of comparison ``in unusual situations.'' The language 
of the regulation directly tracks the language of the SAA, and the 
Department explained in the Preamble to its final regulations that this 
provision was implemented to reflect the language of the SAA. See 
Preamble, Antidumping and Countervailing Duty Final Rule, 62 FR 27295, 
27373-7374 (May 19, 1997) (Preamble). The Department further explained 
in the Preamble that the reason for this preference was directly tied 
to difficulties the agency had in the past with regard to the 
transaction-to-

[[Page 22639]]

transaction methodology and concerns about the difficulty of 
guaranteeing that ``merchandise in both markets'' would be ``identical 
or very similar'' in order for such a comparison to work appropriately. 
Id. at 27374.
    The language of the SAA and the regulations does not prohibit the 
application of the transaction-to-transaction analysis in this case. 
First, there are no statutory or regulatory hierarchical criteria which 
govern the selection of the comparison methodology. The preferences 
expressed in the SAA and regulations merely indicate that in ``normal'' 
cases, weighted-average comparisons will be applied. However, among 
other things, the volatility of prices of subject merchandise and of 
the product sold in Canada during the POI distinguishes this case from 
the norm.
    Second, the SAA was drafted and implemented in 1994, and the 
regulations soon followed in 1997. Both of these sources explain that 
the preference for a weighted-average methodology was based upon past 
experiences and an expressed difficulty in selecting appropriate 
comparison transactions. The Department's computer resources have 
improved greatly in the last few years, and many resource and 
programming difficulties the Department faced in 1994, and even in 
1997, for conducting transaction-to-transaction matching on large 
databases no longer exist.
    Third, when the URAA was negotiated, the Department did not apply 
an offset for non-dumped sales in antidumping investigations. 
Consequently, when Congress expressed a preference for weighted-average 
comparisons and when the Department adopted its regulations, they did 
so in the context of the Department's long-standing approach of not 
applying such an offset when making such comparisons. Because the 
Department is precluded in this instance from not offsetting non-dumped 
sales after making weighted-average-to-weighted-average comparisons, it 
is not clear that the stated preferences at the time of the SAA and 
regulations should continue to apply.
    Accordingly, for all of these reasons, we have calculated dumping 
margins using the transaction-to-transaction methodology. By applying 
the transaction-to-transaction analysis in this case, we are not 
intending to implement an approach that applies to all antidumping 
investigations. As discussed above, the use of this methodology is 
premised on the combination of facts and circumstances that have led to 
and support this determination. Moreover, because the Appellate Body 
Report requires the offset for non-dumped sales only for a weighted-
average-to-weighted-average comparison, we have not applied the offset 
for non-dumped sales in our transaction-to-transaction comparison.

Interested Party Comments

Comment 1: Applicability of the Appellate Body Ruling to a Transaction-
to-Transaction Methodology

    The Canadian Parties contest the Department's decision in the 
Preliminary 129 Determination to continue to not make an offset for 
non-dumped sales under a change from a weighted-average-to-weighted-
average dumping calculation methodology to a transaction-to-transaction 
methodology. According to the Canadian Parties, the Appellate Body's 
decision on ``zeroing'' is not limited to the weighted-average-to-
weighted-average methodology, but extends equally to the transaction-
to-transaction methodology. Therefore, the Canadian Parties argue that 
the Department's determination fails to bring the United States into 
compliance with its obligations under the Antidumping Agreement.
    The Canadian Parties argue that the Department misrepresented the 
Appellate Body Report when it stated that the report required an offset 
for non-dumped sales only in the context of a weighted-average-to-
weighted-average comparison. They argue that the Appellate Body 
concluded that ``margins of dumping,'' as used in Article 2.4.2 of the 
Antidumping Agreement, must take into account the product in question 
as a whole. The Canadian Parties assert that by employing ``zeroing'' 
under the transaction-to-transaction methodology, the Department treats 
dumped transactions differently from non-dumped transactions. 
Therefore, they maintain that this does not treat the product in 
question as a whole. In addition, the Canadian Parties contend that the 
transaction-to-transaction methodology exacerbates the Department's 
failure to account for the product as a whole by applying ``zeroing'' 
at the model (control number) level.
    The Canadian Parties also maintain that the Panel in dicta already 
concluded that ``zeroing'' in the context of a transaction-to-
transaction methodology would be inconsistent with Article 2.4.2 of the 
Antidumping Agreement. As they note, the Panel stated, ``We are of the 
view that the use of zeroing when determining a margin of dumping based 
on the transaction-to-transaction methodology would not be in 
conformity with Article 2.4.2 of the AD Agreement.'' \4\ Citing the 
Appellate Body's ruling in EC--Bed Linen \5\ and its discussion of 
``zeroing'' in United States--Corrosion-Resistant Steel,\6\ the 
Canadian Parties contend that the Appellate Body has concluded that 
``zeroing'' denies a ``fair comparison'' between export price and 
normal value. They argue that the ``fair comparison'' requirement of 
Article 2.4 applies to both transaction-to-transaction comparisons and 
weighted-average comparisons.
---------------------------------------------------------------------------

    \4\ See Panel Report, United States--Final Dumping Determination 
on Softwood Lumber from Canada, WT/DS264/R, adopted as modified by 
the Appellate Body on 31 Aug. 2004, para. 7.219, n.361.
    \5\ See Report of the Appellate Body, European Communities--
Anti-Dumping Duties on Imports of Cotton-Type Bed Linen from India, 
WT/DS141/AB/R, Adopted 12 Mar. 2001, para. 55 (EC--Bed Linen).
    \6\ See Report of the Appellate Body, United States--Sunset 
Review of Anti-Dumping Duties on Corrosion-Resistant Carbon Steel 
Flat Products from Japan, WT/DS244/AB/R, adopted 9 Jan. 2004, para. 
135 (United States--Corrosion--Resistant Steel).
---------------------------------------------------------------------------

    In response to the Canadian Parties, the Coalition contends that 
the only issue before the Appellate Body concerned the use of 
``zeroing'' under the weighted-average-to-weighted-average comparison 
methodology. According to the Coalition, the Appellate Body 
acknowledged that it was not addressing the issue of ``zeroing'' under 
a transaction-to-transaction methodology or average-to-individual 
methodology. The Coalition contends that Article 2.4.2 of the 
Antidumping Agreement prescribes the transaction-to-transaction 
methodology. Contesting the Canadian Parties' argument that the 
Appellate Body's interpretation of the Antidumping Agreement ``forbids 
zeroing,'' the Coalition argues that the Appellate Body's decision 
precludes the United States from using ``zeroing'' only in conjunction 
with the weighted-average methodology. Therefore, the Coalition argues 
that the Preliminary 129 Determination is consistent with the Appellate 
Body's decision.
    The Coalition asserts that the Canadian Parties are attempting to 
broaden the scope of the Appellate Body's ruling in claiming that the 
Appellate Body's decision is instructive regarding the use of 
``zeroing'' with a transaction-to-transaction methodology. The 
Coalition contends that the Panel noted that it was not instructing any 
party when it stated, ``We are mindful that we are not called upon to 
decide whether zeroing is allowed or disallowed under the transaction-
to-transaction and weighted-average-normal-value to individual export

[[Page 22640]]

transaction methodologies.'' \7\ Further, the Coalition claims that the 
Canadian Parties ignore the portion of the Appellate Body's report that 
states, ``In this appeal, we are not required to, and do not address, 
the issue of whether zeroing can, or cannot, be used under the other 
methodologies prescribed in Article 2.4.2, namely, comparing normal 
value and export prices on a transaction-to-transaction basis (the 
``transaction to transaction methodology'') * * *'' \8\ According to 
the Coalition, this demonstrates unambiguously that the Appellate Body 
declared ``zeroing'' in conjunction with the transaction-to-transaction 
methodology to be outside of the scope of its review.
---------------------------------------------------------------------------

    \7\ See Panel Report, United States--Final Dumping Determination 
on Softwood Lumber from Canada, at para. 7,219 n.361, WT/DS264/R 
(April 13, 2004).
    \8\ See Appellate Body Report at para. 63.
---------------------------------------------------------------------------

    Department's position: We disagree with the Canadian Parties. The 
Canadian Parties have interpreted incorrectly the scope of the 
determination of the Appellate Body. The Appellate Body identified 
clearly that ``the precise scope of the appeal'' before it was the 
Department's methodology ``as applied in the anti-dumping investigation 
at issue in this case.'' See Appellate Body Report at 63 (emphasis 
included). Furthermore, the Appellate Body stated, without 
qualification, the following:

    Canada's claim before the Panel was limited to the consistency 
of zeroing when used in calculating margins of dumping on the basis 
of a comparison of weighted average normal value with a weighted 
average of prices of all comparable export transactions (the 
``weighted-average-to-weighted-average methodology'') under Article 
2.4.2 of the Antidumping Agreement. Therefore, in this appeal, we 
are not required to, and do not address, the issue of whether 
zeroing can, or cannot, be used under the other methodologies 
prescribed in Article 2.4.2, namely, comparing normal value and 
export prices on a transaction-to-transaction basis (``the 
transaction-to-transaction methodology''), or comparing a normal 
value established on a weighted average basis to prices to prices of 
individual export transactions (the ``weighted-average-to-individual 
methodology'').

Id. Thus, once it had rendered its decision, the Appellate Body stated 
that ``we have concluded, based on the ordinary meaning of Article 
2.4.2 (of the Antidumping Agreement) read in its context, that zeroing 
is prohibited when establishing the existence of margins of dumping 
under the weighted-average-to-weighted-average methodology.'' See 
Appellate Body Report at para. 108.
    The Canadian Parties' argument that the Appellate Body's decision 
applies beyond the weighted-average-to-weighted-average comparison 
methodology is unpersuasive. Indeed, the Canadian Parties acknowledge 
that the Panel's statement in a footnote of its report regarding the 
transaction-to-transaction comparison methodology was only ``in dicta'' 
and that ``the Appellate Body did not expressly forbid zeroing 
specifically in conjunction with the transaction-to-transaction 
methodology.'' See Canadian Parties' Brief at 3, 7. Nonetheless, they 
argue that the Preliminary 129 Determination ``does not comply with the 
findings of the WTO Appellate Body'' and they state that the Appellate 
Body's analysis ``extends equally to a transaction-to-transaction 
comparison methodology that incorporates the practice of zeroing.'' See 
Canadian Parties' Brief at 2-3. The Appellate Body Report contained no 
such decision. As the Appellate Body clearly indicated, the matter 
before it was the consistency of the Department's methodology with the 
United States' obligations under the Antidumping Agreement, as applied 
in the weighted-average-to-weighted-average analysis in the Final 
Determination. The Department's Section 129 Determination does not 
involve ``zeroing'' in a weighted-average-to-weighted-average 
comparison methodology, in full compliance with the Appellate Body's 
determination. Thus, this Section 129 Determination renders the 
Department's analysis ``not inconsistent with the findings of the 
Appellate Body.'' See section 129(b)(2) of the URAA.

Comment 2: Change in Comparison Methodology

    Citing the decision by the Court of International Trade (CIT) in 
Borden,\9\ the Canadian Parties assert that the principle of finality 
in U.S. administrative law prevents the Department from changing the 
comparison methodology used in the LTFV investigation, which has not 
been challenged. They argue that section 129 actions are limited to 
bringing an agency determination into conformity with WTO obligations. 
Further, they argue that the language of section 129, which authorizes 
the Department to ``take action not inconsistent with the findings of 
the panel or the Appellate Body,'' mirrors the language that Congress 
used with respect to the Department's actions for NAFTA panel remands. 
This parallel language, the Canadian Parties contend, demonstrates that 
in section 129 actions the Department may not revisit issues ``not 
necessary for WTO compliance.'' See Canadian Parties'' Brief at 11. 
Because the Department can comply with the Appellate Body's ruling 
without changing its comparison methodology, the Canadian Parties' 
position is that it must do so.
---------------------------------------------------------------------------

    \9\ See Borden, Inc. v. United States, 4 F. Supp. 2d 1221, 1242 
(Ct. Int'l Trade 1998) reversed on other grounds Borden, Inc. v. 
United States, 7 Fed. Appx. 938, 2001 WL 312232 (Fed. Cir. 2001) 
(Borden).
---------------------------------------------------------------------------

    The Coalition argues that the Appellate Body did not direct the 
Department to make its determination consistent with Article 2.4.2 in a 
specific manner. The Coalition asserts that Article 2.4.2 expressly 
authorizes a WTO member to apply the transaction-to-transaction method 
in an investigation. Therefore, contesting the Canadian Parties' 
conclusions, the Coalition asserts that the transaction-to-transaction 
method is permissible under Article 2.4.2 of the Antidumping Agreement, 
the Act, the SAA, and the Department's regulations.
    In addition, the Coalition argues that the Department acted in 
accordance with the U.S. Court of Appeals for the Federal Circuit's 
(CAFC's) decision that ``zeroing'' is permissible under sections 
771(35)(A) and (B) of the Act.\10\
---------------------------------------------------------------------------

    \10\ See Corus Staal v. United States, 283 F. Supp.2d. 1357 
(Fed. Cir. 2005).
---------------------------------------------------------------------------

    Department's position: We disagree with the Canadian Parties. Not 
granting an offset for non-dumped sales has consistently been an 
integral part of the Department's weighted-average-to-weighted-average 
analysis. In fact, Canada's challenge to the methodology relied on the 
language of Article 2.4.2 regarding comparisons between weighted-
average export prices and normal values. The Appellate Body Report also 
relied on language particular to the use of the weighted-average-to-
weighted-average comparison methodology in finding an offset 
requirement. See, e.g., Appellate Body Report at para. 63.
    The Canadian Parties' argument that the Department cannot apply a 
different methodology to implement the findings of the Appellate Body 
Report is incorrect, because such a reading of the law would, in 
effect, seriously undermine the effectiveness of section 129(b)(2) of 
the URAA. As noted above, Congress intended for a determination 
pursuant to a section 129(b) decision to be a ``new,'' ``second'' or 
``different'' determination. See SAA at 1025, 1027. Thus, the 
Department may modify its calculations or methodologies to effectuate 
its compliance with a WTO decision. Furthermore, the calculations or 
methodologies that are necessary to

[[Page 22641]]

implement this decision rest largely with the discretion of the 
Department, the United States Trade Representative and Congress. The 
Canadian Parties' interpretation of section 129 would remove discretion 
from the various governmental bodies in implementing a decision under 
section 129. Thus, we reject the Canadian Parties' claims that the 
Department may not modify its comparison methodology and employ a 
transaction-to-transaction analysis to bring its determination into 
compliance with the WTO Appellate Body Report.
    With respect to the Canadian Parties' arguments involving Borden, 
we agree that finality is an important aspect of agency proceedings. 
However, the holding of Borden does not apply in this case. First, and 
most obviously, Borden involved a remand redetermination, while this 
case, of course, involves the implementation of a section 129 decision. 
These are entirely different proceedings, with the first involving 
implementation of a specific order to the agency from a domestic court. 
Implementation of a section 129 decision involves multiple governmental 
agencies that may implement a decision in any manner ``not inconsistent 
with the recommendation'' of the WTO. Second, in Borden, the 
Department's level of trade analysis was challenged, an issue that is 
distinct from the CEP calculations in general. See Borden at 1242. In 
this case, the Department's approach of not offsetting non-dumped 
sales, only as part of its weighted-average-to-weighted-average 
comparison methodology, was challenged, and the Department has modified 
its calculations to address the Appellate Body's concerns. As stated 
previously, the language of the Antidumping Agreement that provides for 
this comparison methodology was an integral part of the Appellate 
Body's basis for finding the ``zeroing'' methodology inconsistent with 
the Antidumping Agreement. Thus, the facts of Borden do not apply in 
this case.
    Finally, the Canadian Parties argue that ``finality'' is important, 
and we agree, which is why the agency based its calculations entirely 
upon the facts of the record already before it. As the CIT explained in 
Dupont Teijin Films USA, LP, et. al. v. United States, Slip Op. 2004-70 
(June 18, 2004), once a final determination has been made, the agency 
may only reopen the record and amend its decisions in limited 
circumstances, such as an ``express granting of relief by the court.'' 
Id. at 13. In this case, Commerce was able to modify its calculations 
without adding new factual information to the record. Thus, the agency 
respected the finality of the record in making its determination.
    Accordingly, the Department has determined, pursuant to section 
129(b)(2), that the transaction-to-transaction methodology is an 
appropriate methodology to apply in this case in order to bring its 
determination into conformity with the findings of the Appellate Body.

Comment 3: Requirements under U.S. Law for Use of Transaction-to-
Transaction Methodology

    Referring to the SAA and 19 CFR 351.414(c), the Canadian Parties 
contend that both establish a strong preference for the weighted-
average-to-weighted-average methodology. The only exceptions 
established by the SAA and the Department's regulations, the Canadian 
Parties argue, are for unusual circumstances in which a respondent has 
very few sales and in which the merchandise in each market is 
identical, very similar, or custom-made. Contesting the Department's 
statement in the Preliminary 129 Determination that the transaction-to-
transaction method was disfavored in 1997 because of computer 
programming difficulties, the Canadian Parties contend that the 
Department's regulations and the SAA do not expressly state this. 
Further, the Canadian Parties argue that the Department did not address 
the substantive reason for the preference for the weighted-average 
methodology. They assert that the transaction-to-transaction method, in 
contrast to the weighted-average method, creates a bias toward dumping, 
as the CIT explained in Borden. Furthermore, they contend that the 
Department's regulations are binding upon it until they are formally 
amended.
    The Coalition contends that section 777A of the Act does not 
restrict the Department from comparing the normal values of individual 
transactions to individual export prices or constructed export prices 
of comparable merchandise. The Coalition also claims that even though 
the Department's regulations establish a preference for the weighted-
average-to-weighted-average method, the regulations do not preclude the 
Department from applying the transaction-to-transaction method in 
investigations. Furthermore, the Coalition notes that the SAA states 
specifically that section 777A(d)(1)(A)(ii) permits the calculation of 
dumping margins on a transaction-to-transaction basis. Also, according 
to the Coalition, the SAA states that the URAA establishes a preference 
for use of a weighted-average or transaction-to-transaction methodology 
in the investigation phase of an antidumping proceeding.\11\
---------------------------------------------------------------------------

    \11\ See SAA at 842-843 (emphasis added).
---------------------------------------------------------------------------

    The Coalition maintains that the Department did not act arbitrarily 
or abuse its discretion by using the transaction-to-transaction method. 
In the Preliminary 129 Determination, the Coalition notes, the 
Department explained that employing the transaction-to-transaction 
methodology allows it to determine more accurately whether dumping 
occurred. Further, the Coalition contends that the Department provided 
a reasonable explanation of why the preference for employing the 
weighted-average-to-weighted-average method, as stated in the 
Department's regulations, may no longer apply.
    Department's position: We disagree that the Department's 
application of the transaction-to-transaction methodology is ``contrary 
to U.S. law'' or that the agency is disregarding the SAA by applying 
this methodology. See Canadian Parties Brief at 12. Both of these 
statements, and the arguments which follow, are unsupported by the 
Department's analysis.
    As we explain above, sections 777A(d)(1)(A)(i) and (ii) of the Act 
allow for either weighted-average-to-weighted-average comparisons or 
transaction-to-transaction comparisons, with no stated preference. 
Furthermore, the SAA reflects the understanding shared by the 
Administration and Congress in 1994 that the Department would 
``normally'' apply weighted-average-to-weighted-average comparisons in 
investigations in light of ``past experience'' and difficulties in 
``selecting appropriate comparison transactions.'' This position was 
drafted and implemented over ten years ago, when the Department did not 
offset for non-dumped sales in its weighted-average-to-weighted-average 
comparisons in antidumping investigations and when computer technology 
was inferior to the computer technology of 2005. See SAA at 842-43. 
Thus, the concerns about ``past experiences'' do not require the 
Department to simply use a ``modified'' weighted-average-to-weighted-
average methodology championed by the Canadian Parties. Furthermore, 
earlier concerns about difficulties in selecting appropriate matching 
characteristics are addressed to a great extent through modern computer 
technology. Similarly, the Department's regulations, as reflected in 19 
CFR 351.414(c), were drafted in 1996 and implemented in 1997, and 
mirror the same concerns expressed in the SAA. It should also be

[[Page 22642]]

noted that the regulation uses the term ``normally'' in stating the 
preference for the weighted-average-to-weighted-average methodology, 
indicating that other methodologies may be used.
    Despite the Canadian Parties' claims, the Department is not 
``disregarding the SAA'' or its regulations in its Section 129 
Determination. Instead, as the agency has explained above, the normal 
presumptions of the Department's methodologies, as stated in the SAA 
and the regulations, do not apply in this case.
    The United States explained first to the Panel, and then to the 
Appellate Body, that the Department's methodology when the URAA was 
signed into law in 1994 was to not offset for non-dumped sales as part 
of the weighted-average-to-weighted-average methodology in 
investigations. The United States argued that because the Antidumping 
Agreement is silent on this issue, it did not believe that its 
methodology was inconsistent with its international obligations. The 
Appellate Body disagreed with this assessment of the United States' 
obligations. Nonetheless, this analysis is clearly relevant with 
respect to the claim under the SAA and the regulations that the 
Department ``normally'' will apply a particular methodology. What was 
``normal'' in an antidumping investigation in the United States in 1994 
is, under the Appellate Body's analysis, inconsistent with our WTO 
obligations, as applied in this case. However, absent the Department's 
``normal'' analysis, neither the SAA, nor the regulations, direct the 
Department as to the appropriate alternative methodology.
    Finally, to the extent that the Canadian Parties argue that the 
Preliminary 129 Determination, if finalized, would be inconsistent with 
other provisions of the Antidumping Agreement, we note that our 
analysis in this Section 129 Determination complies fully with United 
States law and regulations. Furthermore, the Act, the URAA, and the 
Department's regulations are consistent with United States obligations 
under the Antidumping Agreement and all other WTO Agreements. See SAA 
at 669 (speaking to the consistency of the URAA with United States 
international obligations). Accordingly, the Department has determined 
that no further analysis is warranted with respect to these arguments, 
and that this Section 129 Determination implements an analysis that is 
consistent with our domestic and international obligations, as well as 
with the Appellate Body Report.

Comment 4: Statutory Preference for Identical Matches

    The Canadian Parties assert that section 771(16) of the Act 
requires the Department to apply methodologies that maximize the number 
of identical matches of merchandise for comparison. The Canadian 
Parties contend that the weighted-average methodology used in the LTFV 
investigation produced mostly identical matches of merchandise, but 
that the transaction-to-transaction methodology used in the Preliminary 
129 Determination produced mostly non-identical comparisons. Further, 
they argue that there is no additional information on the record of the 
proceeding to justify such a change.
    The Coalition did not address this point specifically, but argued, 
as discussed above, that the use of a transaction-to-transaction 
methodology is permissible under U.S. law, and within the Department's 
discretion.
    Department's position: We disagree that the determination of which 
comparison methodology to use hinges on the number or percentage of 
identical matches obtained. While the Department has an established 
precedent for using price-to-price matches where possible,\12\ the 
transaction-to-transaction methodology is consistent with our statutory 
obligation in that it exhausts all possible identical matches within 
the two-week window of contemporaneous sales before searching for 
similar matches, and exhausts all price-to-price matches based on 
comparisons to similar merchandise before going to constructed value. 
The Canadian Parties have not suggested that the time period for 
looking for identical matches be expanded; indeed, they have stated 
that even seven days on either side of the sale may be too long a 
period to address the volatility.
---------------------------------------------------------------------------

    \12\ See Cemex S.A. v. United States, 133 F.3d 897 
(Fed.Cir.1998).
---------------------------------------------------------------------------

    Although section 771(16) of the Act lists identical matches as the 
first choice among the options for selecting a match, it does not 
address the issue of the time period over which the search for 
identical matches should be conducted in a transaction-to-transaction 
methodology. Section 773(a)(1)(A) states that the price to be used for 
normal value must be ``at a time reasonably corresponding to the time 
of the sale used to determine the export price or constructed export 
price.'' We note that in administrative reviews individual U.S. sales 
are matched to home market sales within a time frame that is less than 
the whole review period. See 19 CFR 351.414 (e)(2). In addition, in 
cases where use of a limited time period was warranted by special 
circumstances in the market, such as high inflation, the Department has 
used averaging periods shorter than the full POI.\13\ The same logic 
applies when doing transaction-to-transaction comparisons. Absent a 
specific statutory mandate on the time period to be used, the 
Department must exercise its discretion in determining the most 
appropriate period over which to search for an identical match. 
Depending on the market conditions for a given product, this time 
period could vary from case to case. For a discussion on the 
appropriate time period, see Comment 5 below.
---------------------------------------------------------------------------

    \13\ See, e.g., Certain Steel Concrete Reinforcing Bars From 
Turkey; Final Results, Rescission of Antidumping Duty Administrative 
Review in Part, and Determination Not To Revoke in Part 69 FR 64731 
(November 8, 2004); Final Determination of Sales at Less Than Fair 
Value: Polyvinyl Alcohol from Taiwan, 61 FR 14064 (March 29, 1996).
---------------------------------------------------------------------------

Comment 5: Price Volatility

    The Canadian Parties argue that there is no rational connection 
between the Department's price volatility finding and its conclusion 
that price volatility justifies a switch to the transaction-to-
transaction comparison methodology. In fact, according to the Canadian 
Parties, the Department's price volatility findings support a weighted-
average-to-weighted-average methodology because averaging smooths price 
volatility.
    The Canadian Parties argue that the Department incorrectly presumed 
that pairing each U.S. sale with a home market transaction on or around 
the same date would correct for price volatility. This presumption, the 
Canadian Parties maintain, would only be correct if there was no or 
limited price volatility within the time periods where the transactions 
were matched. The Canadian Parties contend that there is no evidence 
the volatility does not exist or is limited during the periods. They 
have provided analyses from Abitibi, Canfor, Tembec, and Weyerhaeuser 
that demonstrate price volatility within various limited time frames.
    The Canadian Parties argue that in past cases, such as Flowers from 
Colombia,\14\ the Department recognized that price averaging, not 
transaction-to-transaction comparisons, is the best methodology for 
addressing issues posed by highly volatile prices. Moreover, according 
to the Canadian Parties, in Flowers from Colombia, the Department found 
that price-to-price comparisons were particularly inappropriate in 
conjunction with the

[[Page 22643]]

treatment of non-dumped sales as having zero margins.
---------------------------------------------------------------------------

    \14\ See Final Determination of Sales at Less than Fair Value: 
Certain Fresh Cut Flowers from Colombia, 52 FR 6842,6843 (March 5, 
1987).
---------------------------------------------------------------------------

    According to the Coalition, there is ample evidence on the record 
demonstrating that the volatility of net prices would be significantly 
reduced by pairing each U.S. sale with a home market sale on or around 
the same date of sale. First, the Coalition points out, the bi-weekly 
tests used by the Canadian Parties overstate the period in which the 
Department looked for a match, because the transactions selected as the 
most appropriate match could never be more than seven days from the 
U.S. date of sale.
    Further, the Coalition argues that the Canadian Parties' use of 
standard deviations to measure the relative volatility of various price 
strings is misleading. The better statistical analysis for comparing 
the relative volatility of various different price series, the 
Coalition maintains, is the coefficient of variation, which takes into 
account the standard deviation as a percentage of the mean. Using this 
measure on high volume products for all six respondents, the Coalition 
concludes that daily prices have a lower coefficient of variation than 
weekly prices, which have a lower coefficient of variation than monthly 
or quarterly prices. Therefore, the Coalition concludes that using 
prices of transactions that occurred on the same date or, at most, not 
more than seven days from the U.S. sale, significantly reduces 
volatility.
    Finally, the Coalition argues that the Canadian Parties' arguments 
regarding price volatility are flawed because they did not take into 
account all the factors the Department used in its transaction-to-
transaction methodology. According to the Coalition, when all these 
factors are taken into account, price volatility among potential 
product matches is eliminated.
    Department's position: We disagree with the Canadian Parties that 
use of a weighted-average-to-weighted-average comparison is the only 
way to account for price volatility in the lumber market. First, we 
find that the Canadian Parties' cite to Flowers from Colombia is 
inapposite. In Flowers from Colombia, the Department rejected a 
transaction-to-transaction analysis because of (i) the administrative 
burden and (ii) the perishable nature of the product in question, which 
meant that ``end of the day'' sales were made at distress prices. The 
Department stated that because it treated non-dumped sales as having 
zero margins, the distress sales would be given a disproportionate 
weight.
    Unlike fresh cut flowers, lumber is not a highly perishable product 
that needs to be disposed of by the end of each business day regardless 
of price. Thus, there is no separate, identifiable class of sales that 
can be said a priori to give rise to a distortion in our dumping 
analysis, as was the case in Flowers from Colombia.
    With regard to the Canadian Parties' demonstration that prices can 
vary widely in a single day, large price ranges on a single day may 
indicate that the companies are reacting to fluctuations in market 
prices, but it may also indicate that they are able to sell to 
different customers at different prices. The purpose of our dumping 
analysis is to look at an individual company's selling practices to 
determine whether it is engaging in unfair price discrimination. When 
faced with a situation where there were multiple sales of the same 
product on the same day, the criteria we have selected as tie-breakers 
allow us to determine which sales were made under the most similar 
circumstances.
    With regard to the Canadian Parties' use of a standard deviation 
analysis, we agree with the Coalition that the coefficient of variation 
gives a clearer idea as to whether variability is reduced by limiting 
matches to a shorter time period. While the coefficient of variation 
analysis demonstrates that the greatest reduction in volatility can be 
achieved by matching sales made only on the same day, we have had to 
balance our desire to reduce the effect of price volatility with our 
statutory preference for price-to-price matches. Therefore, we have 
continued to look for matches within a seven-day period on either side 
of the U.S. sale.

Comment 6: Matching Hierarchy

    The Canadian Parties contend that the Department has provided no 
factual basis for its matching criteria and hierarchy used to match 
individual U.S. sales to home market transactions, and that the record 
is insufficient to implement a transaction-to-transaction comparison 
methodology. According to the Canadian Parties, the Department deployed 
a methodology never used for an investigation and has not accorded the 
parties an opportunity to submit new factual information regarding the 
process. They state that, as a result, the following problems have 
developed: (1) The Department's use of a biweekly period to control 
price volatility is unsupported by record evidence and entirely 
ineffective; (2) using the date of sale to match transactions may be 
inappropriate because the actual pricing of the merchandise took place 
on a different date; (3) the Department's methodology ignores 
differences between spot sales, which stem from market conditions at 
the time, and contract sales, which are based on pricing formulas; and 
(4) the Department's approach fails to account for ``random length'' 
\15\ sales, whose pricing effects are evened out in an averaging 
methodology but may inappropriately impact margins on a transaction-to-
transaction basis.
---------------------------------------------------------------------------

    \15\ For the purposes of this Section 129 Determination, we are 
defining a random length sale as any sale which contains multiple 
lengths, for which a blended (i.e., average) price was reported.
---------------------------------------------------------------------------

    The Coalition contends that at no point in the case did the 
respondents complain about the dates of sale being used, nor did they 
suggest that the Department should refrain from mixing and matching 
spot sales and contract sales, or refrain from matching mixed-length to 
single-length sales. According to the Coalition, all these issues were 
just as relevant when the Department was matching on a weighted-
average-to-weighted-average basis. Because none of the Canadian 
respondents raised these issues during the investigation, the Coalition 
maintains that the Department should dismiss these claims.
    Department's position: We disagree with the Canadian Parties that 
there was no factual basis for the matching criteria used in the 
transaction-to-transaction matching hierarchy. The issue of the two-
week period is discussed in detail above. With regard to date of sale, 
the Department's policy on date of sale is well established. Section 
351.401(i) states, ``In identifying the date of sale of the subject 
merchandise or foreign like product, the Secretary normally will use 
the date of invoice, as recorded in the exporter or producer's records 
kept in the ordinary course of business. However, the Secretary may use 
a date other than the date of invoice if the Secretary is satisfied 
that a different date better reflects the date on which the exporter or 
producer establishes the material terms of sale.''
    Section A of the Department's questionnaire asks numerous questions 
related to whether a date other than invoice date would better reflect 
the date on which the material terms of sale were established. After 
reviewing all of the responses, the Department stated in its 
preliminary determination, ``{W{time} e generally relied on the date of 
invoice as the date of sale. Consistent with the Department's practice, 
where the invoice was issued after the date of shipment, we relied on 
the date of shipment as the date of sale.'' \16\ Date of

[[Page 22644]]

sale was not an issue for the final determination.
---------------------------------------------------------------------------

    \16\ See Notice of Preliminary Determination of Sales at Less 
Than Fair Value and Postponement of Final Determination: Certain 
Softwood Lumber Products From Canada, 66 FR 56062 (November 6, 
2001).
---------------------------------------------------------------------------

    Further, the Canadian Parties have suggested that it may have been 
more appropriate for the Department to establish the date of sale 
differently for different types of sales made by the same company. The 
Preamble states: ``{W{time} e have retained the preference for using a 
single date of sale for each respondent, rather than a different date 
of sale for each sale.'' \17\ Nowhere in the Department's regulations 
does it imply that a different date of sale methodology should be 
employed when the Department uses a transaction-to-transaction 
methodology in calculating margins. Further, the Canadian Parties have 
suggested that the Department should consider only the date on which 
the price was set and not when all the material terms of sale were set. 
To do so would be contrary to our regulations and precedent.
---------------------------------------------------------------------------

    \17\ See the Preamble at 27348.
---------------------------------------------------------------------------

    Consistent with our regulations and precedent, our determination on 
date of sale for each respondent was based on its description of its 
selling practices overall. The respondents all reported the earlier of 
invoice or date of shipment as the date of sale. We found in reviewing 
the responses and at verification that, for the preponderance of sales, 
the invoice date most properly reflects when the material terms of 
sales (i.e., price and quantity) are set. For example, the Abitibi 
Sales Verification Report states, ``Based on our examination of the 
company's records, we noted that, generally, terms of sale, such as 
quantity ordered, may change from the order date to the invoice date, 
especially with respect to direct sales. For this reason, the invoice 
date is generally found to be the most appropriate basis for the date 
of sale.'' \18\ Consistent with the companies' responses, we used, and 
have continued to use, the earlier of invoice date or date of shipment 
as the date of sale.
---------------------------------------------------------------------------

    \18\ See Memorandum from Magd Zalok and Amber Musser, Import 
Compliance Specialists to Gary Taverman, Director, Office 5 re: 
Verification of the Sales Response of Abitibi-Consolidated Inc. in 
the Antidumping Investigation of Certain Softwood Lumber Products 
from Canada dated January 31, 2002, at page 9. (Abitibi Sales 
Verification Report)
---------------------------------------------------------------------------

    With regard to the Canadian Parties' suggestion that it may have 
been appropriate to consider whether a sale was made using spot prices 
or a contract price, we note that typically contracts are written to 
reflect market prices. For example, in its Section A questionnaire 
response, Abitibi states that prices are set by agreed upon formulas 
and that the ``pricing formulas are based on a spread above a third 
party publication pricing series, usually, Random Lengths.'' \19\ In 
other words, the contract prices are designed to move with the market. 
Although Abitibi mentions that some specialty products may have firm 
fixed-price contracts, information on the record indicates that 
reportable lumber products were not generally sold with firm fixed 
prices.\20\ To the extent the contract sales are distinguished by 
customer category or channel of distribution, they were taken into 
account in distinguishing between equally similar matches.
---------------------------------------------------------------------------

    \19\ See Abitibi Section A Questionnaire response, dated June 
22, 2001, at page A-26.
    \20\ See, e.g.,Abitibi Section A Questionnaire Response at 
Annexes A-6 and A-7.
---------------------------------------------------------------------------

    Regarding sales made on a ``random length'' basis, we acknowledge 
that the sales are not identified in the database. During the 
investigation, the Department did not, with one exception, get data 
regarding these sales because the respondents did not keep any 
information which would allow them to identify the underlying length-
specific prices. During the first administrative review of the order, 
we subsequently devised a methodology to deconstruct prices for at 
least some of the sales made on this basis. We asked the respondents 
for data to identify these sales, and they provided these data. We note 
that the respondents vociferously argued that we should be using the 
blended invoice price, despite the fact that, as here, we were matching 
individual U.S. transactions.\21\ In light of the respondents' 
inconsistent positions on this issue and the time that would have been 
necessary to collect these additional data (as compared with the time 
available to complete this determination), we have continued to use the 
reported prices for random length sales.
--------------