American Seafood Distributors Association
February 7, 2007

Rep. Charles Rangel, Chairman
Committee on Ways & Means
U.S. House of Representatives
1102 Longworth House Office Building
Washington, D.C. 20515

Dear Mr. Chairman:

The American Seafood Distributors Association (“ASDA”) is an association of U.S. importers of seafood, including importers of shrimp from the six countries that are now subject to an antidumping duty order issued by the Commerce Department on February 1, 2005.  Shrimp imports are absolutely vital to our members since domestic shrimp suppliers produce only 10 percent of the shrimp needed to supply fast growing demand for this valuable and relatively inexpensive protein source.

On ASDA’s behalf, I hereby respond to your January 31, 2007 request for comments from the public about what action the Committee should take with respect to the Department of Commerce (DOC) proposal concerning:  (1) which methods of comparing normal values and U.S. prices can or should be used in an investigation; and (2) whether the DOC should “zero” negative antidumping margins when it uses any of the permissible comparison methods.  ASDA strongly believes that the Committee should accept ASDA’s view that the DOC is required to adopt the following policy regarding the calculation of the weighted average dumping margin in each antidumping investigation:

1.         Under U.S. law, the Department must “normally” employ the average-to-average comparison method.  Moreover, as required by the consistent rulings of the WTO’s Dispute Settlement Body, including very recent decisions issued in cases brought by Japan and Ecuador, when it uses the average-to-average comparison method in an investigation, the Department must provide “offsets” for non-dumped comparisons, i.e., it must not zero negative dumping margins.  Furthermore, the Department must calculate average normal values and average U.S. prices over the entire period of investigation except in those very limited circumstances where using POI averages would yield unfair comparisons.  Zeroing is not permitted in these limited circumstances, and offsets must be provided here, as well.

2.         The Department may employ the transaction-to-transaction comparison method only in the types of “unusual situations” contemplated by Congress.   The Department must also provide offsets for non-dumped comparisons when it uses

the transaction-to-transaction methodology, as the WTO’s Appellate Body recently held in Japan’s WTO case.

3.         The Department may employ the average-to-transaction comparison method only in those comparatively rare situations where targeted dumping is alleged to have occurred.  The Department must also provide offsets for non-dumped comparisons when it uses the average-to-transaction method.  The WTO has affirmed that zeroing is not permitted with this comparison methodology.

Disregard of the WTO’s rulings through the Committee’s disapproval of the DOC proposal could have serious and far-reaching consequences for American interests.  On the other hand, approval of the DOC proposal would fully comply with domestic and international law, as we now explain in detail.

I. THE DEPARTMENT MUST NORMALLY USE THE AVERAGE-TO-AVERAGE COMPARISON METHOD IN INVESTIGATIONS AND MUST PROVIDE OFFSETS FOR NON-DUMPED COMPARISONS

A. Relevant Legal Background

Article 2.4.2 of the WTO Antidumping Agreement provides that:

the existence of margins of dumping during the investigation phase shall normally 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 by a comparison of normal value and export prices on a transaction-to-transaction basis.  A normal value established on a weighted average basis may be compared to prices of individual export transactions if the authorities find a pattern of export prices which differ significantly among different purchasers, regions or time periods, and if an explanation is provided as to why such differences cannot be taken into account appropriately by the use of a weighted average-to-weighted average or transaction-to-transaction comparison.

(Emphasis added.)  Section 777A(d)(1) of the Uruguay Round Agreements Act (URAA), 19 U.S.C. § 1677f-1(d)(1), implements Article 2.4.2 as follows:

(A) In general.  In an investigation under part II of this subtitle, the administering authority shall determine whether the subject merchandise is being sold in the United States at less than fair value --

(i) by comparing the weighted average of the normal values to the weighted average of the export prices (and constructed export prices) for comparable merchandise, or

(ii) by comparing the normal values of individual transactions to the  export prices (or constructed export prices) of individual transactions for comparable merchandise.

(B) Exception.  The administering authority may determine whether the subject merchandise is being sold in the United States at less than fair value by comparing

the weighted average of the normal values to the export prices (or constructed export prices) of individual transactions for comparable merchandise, if—

(i) there is a pattern of export prices (or constructed export prices) for comparable merchandise that differ significantly among purchasers, regions, or periods of time, and

(ii) the administering authority explains why such differences cannot be taken into account using a method described in paragraph (1)(A)(i) or (ii).

Thus, both the Antidumping Agreement and the implementing provisions of the URAA established three separate comparison methods for calculating dumping margins in investigations:  (1) average-to-average; (2) transaction-to-transaction; and (3) average-to-

transaction.  As we now discuss, both the Antidumping Agreement and the URAA established a strong preference for use of the average-to-average method, stating that it would be the “normal” method.[1]  Thus, by definition, the transaction-to-transaction and average-to-transaction methods are not the normal methods.  Rather, they are exceptional methods, and our review of the investigations that the Department has conducted since passage of the URAA indicates use of the transaction-to-transaction method on only three occasions in twelve years.  

Moreover, the average-to-transaction method can be used only in instances of “targeted dumping,” which the Antidumping Agreement and the URAA define narrowly to mean a “pattern” of export prices that “differ significantly among different purchasers, regions or time periods.”  To the best of the ASDA’s knowledge, the targeted dumping method has never been used in an investigation since passage of the URAA.

ASDA urges the Committee to advise the Department of its view that it should continue using its longstanding normal practice of making average-to-average comparisons in antidumping investigations.  Switching to the transaction-to-transaction method as the normal method, for example, would flatly contradict U.S. law and Article 2.4.2 of the Antidumping Agreement.  Moreover, regardless of the method it uses, the Department is obligated to make offsets in all investigations for non-dumped comparisons.  In other words, to comply with the WTO rulings, it cannot “zero” any negative dumping margins.

B. The Legislative History Demonstrates the Strong Congressional Preference for Average-to-Average Comparisons in Investigations

Although the text of 19 U.S.C. § 1677f-1(d)(1) does not expressly state a preference among the three comparison methods, the legislative history reveals the indisputable intention of Congress that the Department use the average-to-average comparison method except in carefully limited instances.  Specifically, the SAA on the URAA states that:

Consistent with the {WTO Antidumping} Agreement, new section 777A(d)(1)(A)(i) provides that in an investigation, Commerce normally will establish and measure dumping margins on the basis of a comparison of a weighted-average of normal values with a weighted-average of export prices or constructed export prices.  To ensure that these averages are meaningful, Commerce will calculate averages for comparable sales of subject merchandise to the U.S. and sales of foreign like products.  In determining the comparability of sales for purposes of inclusion in a particular average, Commerce will consider factors it deems appropriate, such as the physical characteristics of the merchandise, the region of the country in which the merchandise is sold, the time period, and the class of customer involved.

In addition to the use of averages, section 777A(d)(1)(A)(ii) also permits the calculation of dumping margins on a transaction-by-transaction basis.  Such a methodology 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 Commerce will use this methodology far less frequently than the average-to-average methodology.

H.R. Doc. No. 103-465 (1994), Vol. 1, at 842-43 (emphasis added).  Under 19 U.S.C. § 3512(d), the SAA constitutes the “authoritative expression by the United States concerning the interpretation and application of the Uruguay Round Agreements and this Act in any judicial proceeding in which a question arises concerning such interpretation or application.”  In addition: 

As is the case with earlier Statements of Administrative Action submitted to the Congress in connection with fast-track trade bills, this Statement represents an authoritative expression by the Administration concerning its views regarding the interpretation and application of the Uruguay Round agreements, both for purposes of U.S. international obligations and domestic law.  Furthermore, the Administration understands that it is the expectation of the Congress that future Administrations will observe and apply the interpretations and commitments set out in this Statement.  Moreover, since this Statement will be approved by the Congress at the time it implements the Uruguay Round agreements, the interpretations of those agreements included in this Statement carry particular authority.

SAA at 656.

After Congress expressed its strong preference for use of the average-to-average method, as well as its view that the transaction-to-transaction method would be used “far less frequently,” the Department adopted 19 C.F.R. § 351.414(c)(1), which provides that:  “In an investigation, the Secretary normally will use the average-to-average method.  The Secretary will use the transaction-to-transaction method only in unusual situations, such as when there are very few sales of subject merchandise and the merchandise sold in each market is identical or very similar

or is custom-made.”  The preamble to the Department’s proposed regulation referred to the average-to-average method as the “preferred method in an antidumping investigation.”[2]  Antidumping Duties; Countervailing Duties; Notice of Proposed Rulemaking and Request for Public Comments, 61 Fed. Reg. 7308, 7348 (Feb. 27, 1996).  Nothing in the WTO rulings on

zeroing affects this “preference.”  See also Antidumping Duties; Countervailing Duties; Final Rule, 62 Fed. Reg. 27295, 27373 (May 19, 1997).

Thus, in light of unambiguous Congressional intent, as implemented in the regulations, the Department must continue to make average-to-average comparisons in investigations unless “unusual” facts warrant the use of the transaction-to-transaction method.[3] 

C. The Department Must Continue Its Longstanding Practice of Calculating Average Normal Values and Average U.S. Prices Over the Entire Period of Investigation

The Department must also continue its normal practice of calculating the average normal value and the average U.S. price over the entire period of investigation, rather than calculating them over shorter time periods (e.g., on a monthly or semi-annual basis).  This is because 19 C.F.R. § 351.414(d)(3) provides that the Department will calculate averages over shorter time periods in only exceptional circumstances:  “When applying the average-to-average method, the Secretary normally will calculate weighted averages for the entire period of investigation or review, as the case may be.”  In fact, the Department may only calculate average normal values or U.S. prices for shorter periods when they “differ significantly” over the course of the POI.  Id.  As the Department noted in the preamble to this regulation:  

In the Department’s view, price averaging means establishing an average price for all comparable sales.  In general, we believe it is appropriate to average prices across the period of investigation, though we recognize that there are circumstances in which other averaging periods are more appropriate.  Accordingly, the proposed rule is designed to ensure that the time periods over which price averages and comparisons are made comports with the circumstances of the case, while maintaining a preference for period-wide averaging.

62 Fed. Reg. at 27373 (May 19, 1997) (emphasis added).

In accordance with this statement of intent, the Department’s longstanding and consistent practice has been to calculate average normal values and U.S. prices for all sales made over the entire POI.  The few instances in which the Department has calculated averages over a shorter period involved unique circumstances.  For example, the Department has used shorter averaging periods in cases involving “high inflation” economies.  See, e.g., Certain Cold-Rolled Flat-Rolled Carbon-Quality Steel Products from Turkey, 65 Fed. Reg. 1127, 1132-33 (Jan. 7, 2000) (preliminary determination), aff’d, 65 Fed. Reg. 15123 (Mar. 21, 2000) (final determination) (the

Department calculated normal values and U.S. prices on a monthly basis because Turkey experienced significant inflation during the POI).

The Department has also used shorter averaging periods where the use of a POI averaging period would distort the margin calculation because price levels changed significantly during the POI.  For example, in Stainless Steel Sheet and Strip in Coils from the Republic of Korea, 64 Fed. Reg. 30664, 30676 (Jun. 8, 1999) (final determination), the Department

determined that the significant and precipitous devaluation of the Korean won during the Asian financial crisis would not yield appropriate comparisons.  Therefore, it used two averaging periods.  In Static Random Access Memory Semiconductors from Taiwan, 63 Fed. Reg. 8909, 8925 (Feb. 23, 1998) (final determination), the Department utilized quarterly averaging periods because the comparison and U.S. markets had experienced significant and consistent price declines during the POI.  Similarly, in Dynamic Random Access Memory Semiconductors of One Megabit and Above from the Republic of Korea, 58 Fed. Reg. 15467, 15476 (Mar. 23, 1993) (final determination), the Department found that monthly averaging periods were more representative because prices in both markets had consistently declined throughout the POI.

In contrast, the Department has declined to use shorter averaging periods where no evidence existed that distortions would result from averaging prices over the entire POI.  See, e.g., Live Swine from Canada, 70 Fed. Reg. 12181 (Mar. 11, 2005), Issues and Decision Memorandum at Comment 5 (the Department did not depart from its “normal practice” of using annual averaging periods because respondent did not demonstrate that distortions in the dumping calculations would result unless the Department utilized shorter averaging periods); Certain Polyester Staple Fiber from the Republic of Korea, 65 Fed. Reg. 16880 (Mar. 30, 2000), Issues and Decision Memorandum at Comment 3 (the Department declined to use shorter averaging periods because “we were seeking to examine whether there was an overall general trend in prices of all subject merchandise sold by the respondents” and “price changes during the POI were neither significant nor consistent”).

All of these cases confirm that the Department’s longstanding and consistent practice has been to calculate average prices for the entire POI except in unusual situations.  No legal basis exists to change that practice for four reasons:  (1) Article 2.4.2 of the Antidumping Agreement states that the average-to-average method will “normally” be applied to “all comparable export transactions;” (2) 19 U.S.C. § 1677f-1(d)(1)(A)(i) similarly requires a calculation of “the weighted average of the normal values to the weighted average of the export prices (and constructed export prices) for comparable merchandise” without limitation of the time period for averaging; (3) the SAA states that the statutory provision is intended to implement the Antidumping Agreement, which as just noted, normally requires averaging of “all comparable export transactions;” and (4) 19 C.F.R. § 351.414(d)(3) states that the Department “normally will calculate weighted averages for the entire period of investigation,” except where “prices differ significantly over the course of the period of investigation.”  Accordingly, the Department should continue to make average-to-average comparisons in investigations over the entire POI, except in unusual circumstances of the type discussed above.

D. Whenever It Makes Average-To-Average Comparisons, the Department Must Provide Offsets for Non-Dumped Comparisons

The Department cannot employ zeroing when it uses any version of the average-to-average comparison method, according to the WTO Appellate Body’s finding in United States – Final Dumping Determination on Softwood Lumber from Canada, WT/DS264/AB/R, Aug. 11, 2004, at para. 108.[4]  The AB reached the same result in cases brought by the EU and Japan.  See WT/DS294/AB/R (17 January 2006) and WT/DS322/AB/R (9 January 2007).

As a result of the WTO’s definitive and repeated invalidation of zeroing whenever it uses the average-to-average method, we agree with the Department’s statement in its December 27, 2006 Federal Register Notice that it is required to abandon the use of average-to-average comparisons without offsets because it “is necessary to implement the recommendations of the World Trade Organization Dispute Settlement Body.”  71 Fed. Reg. at 77722.  In other words, the Department must terminate its practice of “zeroing” in investigations and, instead, grant offsets for non-dumped comparisons in order to bring its practice into conformity with its obligations under Article 2.4.2 of the Antidumping Agreement.  Moreover, because the Appellate Body’s holdings expressly extend to all variations of the average-to-average method, offsets for negative margins must be granted in investigations in those limited situations in which it uses shorter averaging periods than the entire POI.

II. THE DEPARTMENT SHOULD CONTINUE TO MAKE TRANSACTION-TO-TRANSACTION COMPARISONS ONLY IN EXCEPTIONAL SITUATIONS, AND IT MUST ELIMINATE ZEROING AND GRANT OFFSETS WHEN IT DOES

The Department has rarely exercised its authority to make transaction-to-transaction comparisons in investigations.  For example, it did so in Large Newspaper Printing Presses because there were only a few sales of subject merchandise and the merchandise itself was highly customized, which is exactly the type of situation contemplated by the SAA and 19 C.F.R. § 351.414(c)(1).[5]  See Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled, from Germany, 61 FR 38166 (Jul. 23, 1996) (final determination); Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled, from Japan, 61 Fed. Reg. 38139 (Jul. 23, 1996) (final determination).

More recently, the Department made transaction-to-transaction comparisons in its Section 129 determination in Certain Softwood Lumber Products from Canada in which the Department implemented the WTO Appellate Body’s finding with regard to zeroing when making average-to-average comparisons in investigations.  See Softwood Lumber – Section 129 Determination, 70 Fed. Reg. 22636 (May 2, 2005).  There, the Department explained that transaction-to-transaction comparisons were necessary because of the “high level of price volatility” in the U.S. and Canadian markets that could distort the results of the dumping calculations if average-to-average comparisons were made.  Id. at 22637-39; see also Preliminary Determination under

Section 129 of the Uruguay Round Agreements Act:  Antidumping Measures on Certain Softwood Lumber Products from Canada at 10, available at http://www.ia.ita.doc.gov/download/ section129/Canada-Lumber-129-Prelim-013105.pdf (“By applying the transaction-to-transaction analysis in this case, we are not intending to implement a practice 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.”)  Thus, the Department premised its use of the transaction-to-transaction methodology on a very “unusual situation.”[6] 

When making these transaction-to-transaction comparisons, the Department erroneously did not provide offsets for non-dumped comparisons on the ground that the Appellate Body’s finding was limited to average-to-average comparisons.  However, the Appellate Body in Japan’s case expressly found that the DOC’s use of zeroing in connection with the transaction-to-transaction method in original investigations was illegal.  See AB Report at par. 190.  And, even if it had not reached this result, the URAA limits the use of the transaction-to-transaction method to exceptional circumstances.  As noted in the preamble to the final rule:

In the Department’s view, the SAA makes clear that Congress did not contemplate broad application of the transaction-to-transaction method.  SAA at 842.  Specifically, the SAA recognizes the difficulties the agency has encountered in the past with respect to this methodology and suggests that even in situation where there are very few sales, the merchandise in both markets should also be identical or very similar before the agency would make transaction-to-transaction comparisons.  Accordingly, we continue to maintain that the transaction-to-transaction methodology should only be applied in unusual situations.

62 Fed. Reg. at 27374 (May 19, 1997) (emphasis added).

III. CONCLUSION

For the foregoing reasons, the ASDA strongly urges the Committee to inform the Department that it supports discontinuation of zeroing in all investigations.

Respectfully submitted,

Wally Stevens
President

[1] Neither the WTO Antidumping Agreement nor the URAA defines the words “normal” or “normally.”  However, the dictionary meaning is “standard,” “common,” or “typical.”  Thus, any application of a different method than the average-to-average method must necessarily be “non-standard,” “uncommon,” or “atypical.”  In other words, it is the exception, not the rule.  In fact, the Statement of Administrative Action (SAA) expressly refers to the transaction-to-transaction method and the average-to-transaction method as “exceptions.”

[2] In contrast, this preamble stated that the “transaction-to-transaction method will only be used in unusual circumstances.”  61 Fed. Reg. at 7349 (Feb. 27, 1996).

[3] The average-to-transaction method was expressly intended for use in the rarest of situations involving targeted dumping.

[4] See also Notice of Determination under Section 129 of the Uruguay Round Agreements Act:  Antidumping Measures on Certain Softwood Lumber Products from Canada, 70 Fed. Reg. 22636, 22640 (May 2, 2005) (Softwood Lumber – Section 129 Determination).

[5] As noted in Chapter 6 of the Antidumping Manual (at 7), “We may also establish dumping margins by comparing NV and EP or CEP on a transaction-to-transaction basis.  This is normally done only for large capital goods made to order, such as transformers.  The difference between these custom-made products render average prices meaningless.”

[6] In Softwood Lumber – Section 129 Determination, the Department also stated that its computer capabilities had improved since the passage of the SAA, which meant that it could now perform transaction-to-transaction comparisons on large databases.  However, nothing in the SAA linked the strong preference of Congress for average-to-average comparisons to the Department’s computer capabilities.  See 70 Fed. Reg. at 22641.  Thus, improvements in the Department’s technical capabilities are irrelevant to the choice of a comparison method.