KOENIG & BAUER-ALBERT AG, et al., Plaintiffs, v. UNITED STATES, Defendant, and Goss Graphics, Inc., Defendant-Intervenor, 90 F. Supp. 1284

OPINION POGUE, Judge. On June 23, 1998, this Court remanded certain aspects of the Department of Commerce’s (“Commerce” or “the Department”) determination in Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled, From Germany, 61 Fed.Reg. 38,166 (Dep’t Commerce 1996)(final determ,)(“Germany Final ”). See Koenig & Bauer-Albert AG v. United States, 22 CIT -, 15 F.Supp.2d 834 (1998) (“KBA /”). On September 17, 1998, Commerce issued its Final Results of Redetermination Pursuant to Remand (Redetermination List, Pub. Doc. 8, Conf, Doc. 4)(Sept. 17, 1998) (“Redetermination”).

On remand, Commerce did not adequately address the Court’s concerns regarding the issues of “collapsing” and cost-averaging; thus the Court remanded these issues for a second time. See Koenig & Bauer-Albert AG v. United States, 23 CIT -, -, 44 F.Supp.2d 280, 287 (1999) (“KBA 77”). On August 10, 1999, Commerce issued its Final Results of Redetermination Pursuant to Second Court Remand (Second Re-determination List, Pub. Doc. 5)(Aug. 10, 1999) (“Second Redetermination”). The Court now reviews Commerce’s Second Redetermination.

Standard of Review The Court will uphold a Commerce determination in an antidumping investigation unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law[.l” 19 U.S.C. § 1516a(b)(l)(B)(i)(1994). Background MAN Roland Druckmaschinen AG and MAN Roland Inc. (“MAN Roland”), respondents in the underlying administrative proceeding before Commerce, produce large newspaper printing presses (“LNPPs”) at a facility in the western German city of Augsburg; MAN Roland’s wholly-owned subsidiary, MAN Plamag, produces LNPPs at a facility in the eastern German city of Plauen.

The Plauen facility incurs lower labor and overhead costs than the Augsburg facility. See MAN Roland Supp. Questionnaire Secs. C, D,E (Final List, Conf. Doc. 39)(Dec. 13, 1995)(“MAN Roland Responses”) at Sec. D, p. 54. MAN Roland alleged throughout Commerce’s investigation that it and MAN Plamag met the criteria for “collapsing,” and that therefore, in calculating the cost of production (“COP”) and constructed value (“CV”) of its LNPPs, Commerce should have averaged the labor and overhead costs of both factories. See Germany Final at 38,187-88. 1. Collapsing MAN Roland and MAN Plamag a.

Background Collapsing is a practice whereby Commerce determines that affiliated companies should be regarded as one entity, and therefore calculates a single, weighted-average dumping margin to be assessed to the collapsed entity as a whole. See AK Steel Corp. v. United States, 22 CIT-, -, 34 F.Supp.2d 756, 764 (1998), aff'd in part, rev’d in part, 203 F.3d 1330 (Fed.Cir. 2000); Asociacion Colombiana de Exportadores v. United States, 22 CIT -, -6 F.Supp.2d 865, 893 (1998)(“Asociacion Colombiana ”). Commerce initially disagreed with MAN Roland’s argument that MAN Roland and MAN Plamag should be collapsed.

See Germany Final at 38,188. Apparently because it decided not to collapse the two companies, Commerce determined that “[it] should not average costs for [MAN Roland] and MAD [sic] Plamag.” Id. In KBA I, the Court found Commerce’s explanation insufficient and directed Commerce to reconsider on remand its decision not to average costs. See 22 CIT at-, 15 F.Supp.2d at 849-50. “[T]he only context in which the discussion of whether to average the production costs of affiliated parties ... occurfs] is in the context of collapsing.” KBA II, 23 CIT at-, 44 F.Supp.2d at 287.

Yet in its Redetermination, Commerce failed entirely to address the collapsing issue, while explaining at length its decision not to average the costs of MAN Roland and MAN Plamag. See Redetermination at 2-9. Upon review, the Court remanded the collapsing and cost-averaging issues for a second time, and ordered Commerce to “apply its collapsing practice as it then existed [i.e., at the time of Germany Final] and was later codified at 19 C.F.R. § 351.401(f).” KRA II, 23 CIT at-, 44 F.Supp.2d at 287.

In its Second Redetermination, Commerce applied its collapsing regulation and decided to collapse MAN Roland and MAN Plamag for purposes of calculating COP and CY. See Second Redetermination at 1. As a result, if affirmed by the Court, the revised final dumping margin of 39.53% will be applied to subject merchandise entered by either MAN Roland or MAN Plamag. See id. at 1-2. b. Discussion The antidumping statute does not directly address collapsing.

Thus, in determining whether Commerce’s collapsing practice is in accordance with the law, “the question for the court is whether the agency’s answer is based on a permissible construction of the statute.” Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) (“Chevron”). In other words, the Court must determine whether Commerce’s collapsing practice is a reasonable

interpretation of the statute. Commerce has interpreted the statute as giving it discretion to collapse and has developed a collapsing practice. See, e.g., Certain Pasta From Italy, 61 Fed.Reg. 30,326, 30,351 (Dep’t Commerce 1996)(final determ.); Certain Hot-Rolled Carbon Steel Flat Products From, Canada, 58 Fed. Reg. 37,099, 37,107 (Dep’t Commerce 1993)(final determ.); Certain Granite Products From Spain, 53 Fed.Reg. 24,335, 24,337 (Dep’t Commerce 1988)(final de-term.). To conform with the Uruguay Round Agreements Act (“URAA”), Commerce promulgated 19 CFR § 351.401(f). See supra note 4. Commerce’s collapsing practice has been approved by the court as a reasonable

interpretation of the antidumping statute. See Asociación Colombiana 22 CIT-, --6 F.Supp.2d 865, 893; Queen’s Flowers de Colom. v. United States, 21 CIT 968, 971-72, 981 F.Supp. 617, 622-23 (1997)(“Queen’s Flowers ”). AK Steel confirmed that Commerce’s collapsing practice continues following the passage of the URAA. See 22 CIT at-, 34 F.Supp.2d at 765, aff'd, 203 F.3d 1330 (Fed.Cir.2000) at 22. Commerce’s collapsing practice is a permissible construction of the statute, and is thus in accordance with the law. Commerce’s decision to collapse MAN Roland and MAN Plamag is also supported by substantial evidence.

Commerce indicated that it collapsed MAN Roland and MAN Plamag because the two companies “satisfy all three criteria enumerated [in 19 C.F.R. § 351.401(f) ] based on the totality of the facts relevant during the [period of investigation (“POI”) ].” Second Redetermination at 5. Commerce found that MAN Roland and MAN Plamag are affiliated companies, see Second Rede-termination at 5 (citing MAN Roland Sec. A Questionnaire (Final List, Conf.

Doc. 15)(Sept. 27, 1995) at 30); that MAN Roland and MAN Plamag have “production facilities for similar or identical products that would not require substantial retooling of either facility in order to restructure manufacturing priorities,” see Second Re-determination at 5 & n. 4 (citing MAN Roland Case Brief (Final List, Conf. Doc. 97)(June 3, 1996) at 75); and that the two companies exhibit a “significant potential for the manipulation of price or production,” see Second Redetermination at 5-6 (citing MAN Roland Responses at App. D-6-A, D-6-B; MAN Roland Supp. Questionnaire Secs. A,D,E (Final List, Conf.

Doc. 50)(Jan. 31, 1996)(“MAN Roland Supp. Questionnaire Resp.”) at 56-58). The decision to collapse is not contested by MAN Roland. See Comments on Draft Remand Determination (Second Redeter-mination List, Pub. Doc. 4)(June 30, 1999)(“MAN Second Redetermination Comments”). II. Averaging MAN Roland and MAN Plamag’s Production Costs a. Background The central dispute in this case is which costs of MAN Roland and MAN Plamag Commerce should average in determining the COM component of CV for the collapsed company.

In the Second Redeter-mination, Commerce determined that it should not average the labor and overhead costs of MAN Roland and MAN Plamag, as requested by MAN Roland, because “the Department’s normal practice is to compute costs on a control-number-(CONNUM-) specific basis.” Second Redetermination at 6. All parties agree that each LNPP falling within the scope of the Germany Final investigation was a unique product. Id. at 7. Accordingly, each LNPP had been assigned a unique CONNUM. Id.

Therefore, according to Commerce, “there was no need to weight-average production costs between the two factories” of the collapsed companies, because there was no matching CONNUM— that is, no identical LNPPs — produced at both factories. Id. Commenting on Commerce’s draft Second Redetermination, MAN Roland conceded that the two facilities did not produce “identical” LNPPs having the same CONNUM, but rejected Commerce’s position that its “normal” or “established” practice is to average production costs only for identical merchandise. See MAN Second Redetermination Comments at 2.

MAN Roland has alleged from the start of this litigation that Commerce’s “established practice” is, rather, to average costs “where a respondent has the ability to produce the subject merchandise at two plants .... ” MAN Roland’s Mem. Supp. Mot. J. Agency R. at 33 (emphasis added). In KBA I, this Court could not verify Commerce’s assertion that averaging costs only for identical merchandise is its “normal practice” in the context of affiliated parties. See 22 CIT at-, 15 F.Supp.2d at 849 n. 7.

As a procedural matter, the Court found Commerce’s explanation of its cost-averaging practice to be based on a post hoc rationalization, and therefore did not consider it on the merits. See id.

The Court commented that if Commerce were to rely upon the “identical merchandise requirement” in its remand determination, it would have to explain how its stated cost-averaging practice was consistent with two earlier Commerce determinations that appeared to contradict its position: Certain Fresh Cut Flowers From Colombia, 55 Fed.Reg. 20,491, 20,497 (Dep’t Commerce 1990)(final results admin, review)(“Fresh Flowers ”); and Silicon Medal From Brazil, 59 Fed.Reg. 42,806, 42,808 (Dep’t Commerce 1994)(final results admin. review)(“Silicon Metal ”). See id.

In KBA II, the Court determined that Commerce had failed on remand to provide sufficient evidence of an identical merchandise requirement in the context of affiliated parties. See KBA II at-, 44 F.Supp.2d at 282-87. First, Commerce did not explain to the Court’s satisfaction how the “identical merchandise” requirement was consistent with Fresh Flowers and Silicon Metal. See id. at -, 44 F.Supp.2d at 282-83.

Second, the Court found that three other determinations cited by Commerce in its Redetermination “failed to demonstrate that its identical merchandise requirement is its established practice in the context of affiliated parties.” KBA II, 23 CIT at -, 44 F.Supp.2d at 283. Finally, the Court found that Commerce had not explained how its identical merchandise requirement was consistent with its collapsing practice. See id. at-, 44 F.Supp.2d at 283-87. The Court noted that the language of 19 C.F.R. § 351.401(

f) and Commerce’s collapsing practice prior to promulgation of that regulation indicated that Commerce would collapse where producers “have production facilities for similar or identical products that afford the company the ability to manipulate its manufacturing priorities.” Id. at-, 44 F.Supp.2d at 285 (emphasis added). The Court asked Commerce on remand to address its concern that “Commerce’s application of the identical merchandise requirement is inconsistent with 19 C.F.R. § 351.401 and contrary to Commerce’s collapsing practice.” Id. at --•, 44 F.Supp.2d at 287. b.

Discussion Cost-averaging for affiliated entities is not explicitly addressed in the anti-dumping statute. Thus, as with collapsing, “the question for the court is whether the agency’s answer is based on a permissible construction of the statute.” Chevron, 467 U.S. at 843, 104 S.Ct. 2778. In other words, the Court must determine whether Commerce’s identical merchandise requirement for collapsed entities is a reasonable

interpretation of the statute. While, as noted, “the only context in which the discussion of whether to average the production costs of affiliated parties ... occur[s] is in the context of collapsing,” KBA II, 23 CIT at -, 44 F.Supp.2d at 287, this Court has recognized that “Commerce’s collapsing analysis and its constructed value calculation are separate.” Queen’s Flowers, 21 CIT at 975, 981 F.Supp. at 624.

In the Second Redetermination, Commerce explained how its collapsing practice differs from its cost-averaging practice with reference to the identical merchandise requirement: The Department’s collapsing test relies, in part, upon the ability of two (or more) affiliated producers to produce identical or similar merchandise. However, the Department’s requirement for weight-averaging production costs in calculating COP and CV is actual production of identical (i.e., same CONNUM) products at both (or multiple) production facilities. Id. at 9.

This difference reflects Commerce’s contention that each practice addresses a different issue in relation to the dumping margin. As described by Commerce, “collapsing, as it relates to computing COP, is a specific rule dealing with whether the Department should include facilities owned by an affiliate in its weighted-average, CONNUM-specific COP computation.” Id. at 10. The cost-averaging issue, on the other hand, “deals with the general rule for computing a single, CONNUM-specific, weighted-average COP.” Id. Further, Commerce explained that different policy goals underlie collapsing and cost-averaging.

Commerce collapses facilities to prevent the possibility of manipulation of the antidumping law through shifting production to a less expensive, affiliated facility. See Second Redetermination at 4. As Commerce explained in a subsequent review of the Fresh Flowers determination discussed above: [O]ur concerns over shifting production refer to a longer period of time; thus, if Company A receives a lower margin than Company B, we are concerned that Company A would increase production of new flowers to take advantage of a lower margin while Company B would, over time, reduce production due to its higher margin.

Alternatively, more of the production of Company A could be shifted to the U.S. market. Certain Fresh Cut Flowers From Colombia, 61 Fed.Reg. 42,833, 42,854 (Dep’t Commerce 1996)(final results admin, reviews). To prevent such manipulation, Commerce applies a single dumping margin to “the entire producer or reseller, not merely a part of it;” that is, to the collapsed entities. Second Redetermination at 4. Thus, in the example above, if collapsed, Company A and Company B would be assigned the same dumping margin, and consequently would not be able to take advantage of Company A’s lower dumping margin.

In sum, in order to satisfy the criteria of the collapsing regulation, “there is no requirement that the companies produce the identical products (ie., the same CONNUMs), only that they produce (or have the ability to produce) similar products.” Id. at 9-10. By contrast, when calculating the dumping margin, regardless of whether that margin is applied to the entities of a collapsed company or to a single company comprised of one or more facilities, Commerce focuses on the actual costs of production. See Second Redetermination at 10.

As such, only the costs of facilities that in fact produce identical merchandise are included in the weighted average. According to Commerce, including the theoretical costs of MAN Plamag merely because it had the ability to produce the subject merchandise “fails to account for the reality of the production process.” Id. Commerce’s explanation of the differences between its collapsing and cost-averaging practices demonstrates that its identical merchandise requirement is not unreasonable.

Commerce collapses where facilities actually produce or have the ability to produce identical or similar merchandise to prevent manipulation of the antidumping laws. On the other hand, Commerce cost-averages only where facilities actually produce identical merchandise in order to arrive at the actual, rather than theoretical, costs of production. Implicit in Commerce’s explanation of its cost-averaging practice is a general policy favoring accuracy in calculating dumping margins, which the Federal Circuit has recognized as “the basic purpose of the statute.” Rhone Poulenc, Inc. v. United States, 8 Fed.

Cir. (T) 61, 67, 899 F.2d 1185, 1191 (1990). Accordingly, Commerce’s explanation on remand addresses the Court’s concern, expressed in KB A II, that Commerce’s “identical merchandise” requirement for cost-averaging was inconsistent with its collapsing regulation and contrary to its collapsing practice. See KBA II, 23 CIT at -, 44 F.Supp.2d at 282-84, 286-87. Collapsing effectively prevents manipulation by applying a single dumping margin to collapsed facilities, while the identical merchandise requirement promotes accuracy in the calculation of the dumping margin to be applied.

In this case, because each LNPP was determined to be unique, including weight-averaged labor and overhead costs for merely similar or merely theoretical LNPPs would diminish the accuracy of Commerce’s margin calculation.

Moreover, Commerce’s explanation resolves the question of whether the identical merchandise requirement was Commerce’s established practice at the time Germany Final was issued. “Commerce has the flexibility to change its position providing that it explains the basis for its change and providing that the explanation is in accordance with law and supported by substantial evidence.” KBA II, 23 CIT at -, 44 F.Supp.2d at 287 (quoting Cultivos Miramonte S.A. v. United States, 21 CIT 1059, 1064, 980 F.Supp. 1268, 1274 (1997)(footnotes omitted)); see also AK Steel, 203 F.3d 1330 (Fed.

Cir.2000) at n. 10 (“That Commerce changed its

interpretation, however, need not change the court’s analysis.”)(citing Chevron, 467 U.S. at 863, 104 S.Ct. 2778 (“An initial agency

interpretation is not instantly carved in stone.”)). As observed in KBA II, the determinations Commerce cited in its Redetermination in support of its identical merchandise requirement did not clearly indicate the existence of an established practice. 23 CIT at ——, 44 F.Supp.2d at 283-84. In the Second Redetermination, Commerce provided no further evidence of an established practice, but did adequately explain its cost-averaging practice.

Thus, whether this explanation constitutes a “change” of practice, or simply illuminates a practice that had previously been applied, but not weh documented, the Court finds that Commerce’s cost-averaging practice as explained in the Second Redetermination is a reasonable

interpretation of the statute, and is thus in accordance with the law. Accordingly, the Court finds that Commerce correctly determined not to average the overhead and labor costs of MAN Roland and MAN Plamag. III. Valuation of Transferred Inputs Though refusing to average the labor and overhead costs of MAN Roland and MAN Plamag, Commerce determined that it was required to revalue inputs transferred between the two companies at the cost of producing the input rather than the transfer price used in Germany Final. See Second Redetermination at 6,7.

Commerce explained: “Treating these affiliated companies as a single entity necessitates that inputs transferred between them also be valued based on the group as a whole.” Id. at 6; see also AK Steel, 22 CIT at-, 34 F.Supp.2d at 764-66, aff'd, 203 F.3d 1330 (Fed.Cir.2000) at 22-25. The recalculation of CV on this basis resulted in a reduction of the final dumping margin from 39.60% to 39.53%. See Second Redetermination at 1-2; IA Staff to File: Adjustment Calculations for Second Remand — MRD (Second Redetermination List, Pub. Doc. 1, Conf. Doc. l)(June 10, 1999).

In AK Steel, the court approved Commerce’s method of valuing costs of transfers between collapsed companies based on COP. See 22 CIT at-, 34 F.Supp.2d at 764-66. The court held that Commerce’s decision to value costs of transfers between collapsed companies based on COP was in accordance with the law: “Once collapsed, the POSCO Group was treated as a single entity, not a set of affiliated persons.

Commerce reasonably determined that it should act consistently with its collapsing determination and not apply inconsistent solitary provisions, thereby arbitrarily increasing respondents’ liability.” AK Steel, 22 CIT at-, 34 F.Supp.2d at 766. A recent Federal Circuit decision affirms the court’s ruling, noting that “once Commerce has decided to treat the companies as one ‘person’ for purposes of the anti-dumping analysis, it is not statutorily required to apply the [fair value] and [major input] provisions.” AK Steel, 203 F.3d 1330 (Fed.Cir.2000) at 24.

In this case, Commerce acted consistently with the practice approved in AK Steel by this court and affirmed by the Federal Circuit. Commerce collapsed MAN Roland and MAN Plamag according to its collapsing regulation. Once collapsed, it treated the two companies as a single producer rather than affiliated parties, and thus properly disregarded the “fair value” and “major input” provisions. Further, MAN Roland has raised no objection to the calculation of the revised dumping margin, and the Court can find none on independent review.

Therefore, the Court concludes that Commerce’s valuation of the cost of transfers between MAN Roland and MAN Plamag based on COP is supported by substantial evidence, and otherwise in accordance with the law. Conclusion Commerce’s revision of its determination of sales at less than fair value with respect to LNPPs from Germany is supported by substantial evidence on the administrative record and is otherwise in accordance with the law. Accordingly, Commerce’s Final Results of Redetermi-nation Pursuant to Second Court Remand are affirmed.

JUDGMENT This case having been duly submitted for decision, and the Court, after due deliberation, having rendered a decision herein; now, in conformity with said decision, it is hereby ORDERED that the U.S. Department of Commerce’s Final Results of Redeter-mination Pursuant to Second Court Remand in Large Newspaper Printing Presses and Components Thereof, Whether Assembled or Unassembled (“LNPP”) from Germany is sustained. . The administrative record in this case consists of three lists of documents.

Reference will be made to the "Final List” when citing to the record in Germany Final, to the “Rede-termination List” when citing to the record in the Redetermination, and to the “Second Re-determination List” when citing to the record in the Final Results of Redetermination Pursuant to Second Court Remand. Cites to the administrative record specify whether reference is made to a public document or to a business proprietary document. . In determining normal value ("NV”) for purposes of calculating the antidumping margin, sales made at less than COP may be disregarded. See 19 U.S.C. § 1677b(b)(l)(1994).

COP is the sum of three components: cost of manufacturing ("COM”); selling, general, and administrative ("SG & A”) expenses, and profits; and packing costs. See 19 U.S.C. § 1677b(b)(3). . Constructed value, ("CV”) is used when it is not possible to calculate NV based on the price of a foreign like product sold in the comparison market. See 19 U.S.C. § 1677b(a)(4). CV is the sum of COM, SG & A expenses and profits, and packing costs. See 19 U.S.C. § 1677b(e).

For CV purposes, COM is "the cost of materials and of fabrication or other processing of any kind employed in producing the [imported] merchandise, during a period which would ordinarily permit the production of the merchandise in the ordinary course of business.” 19 U.S.C. § 1677b(e)(l). COM in-eludes such costs as cost of materials, labor and overhead. See Steel Wire Rod From Venezuela, 63 Fed.Reg. 8,948, 8,952 (Dep’t Commerce 1998)(suspension antidumping inves-tig.) ("Venezuelan SWR ").

SG & A expenses and profits are "the actual amounts incurred and realized by the specific exporter or producer being examined in the investigation or review for [SG & A] expenses and profits, in connection with the production and sale of a foreign like product, in the ordinary course of trade, for consumption in the foreign country.” 19 U.S.C. § 1677b(e)(2)(A). 19 U.S.C. § 1677b(e)(2)(

B) provides methods for calculating SG & A expenses and profits where the "actual amounts” are not available. Packing costs include "the cost of all containers and coverings of whatever nature, and all other expenses incidental to placing the subject merchandise in condition packed ready for shipment to the United States.” 19 U.S.C. § 1677b(e)(3). . 19 C.F.R. § 351.401(1) provides:

(1) Treatment of affiliated producers in anti-dumping proceedings—

(1) In general. In an antidumping proceeding under this part, the Secretary will treat two or more affiliated producers as a single entity where those producers have production facilities for similar or identical products that would not require substantial retooling of either facility in order to restructure manufacturing priorities and the Secretary concludes that there is a significant potential for the manipulation of price or production.

(2) Significant potential for manipulation. In identifying a significant potential for the manipulation of price or production, the factors the Secretary may consider include: (

i) The common level of ownership; (ii) The extent to which managerial employees or board members of one firm sit on the board of directors on an affiliated firm; and (iii) Whether operations are intertwined, such as through the sharing of sales information, involvement in production and pricing decisions, the sharing of facilities or employees, or significant transactions between the affiliated producers. 19 C.F.R. § 351.401(10(1999).

Note that although Germany Final was issued before the regulation was promulgated, the regulation was proposed on February 27, 1996, “and had been relied upon by Commerce as instructive and consistent with Commerce's practice and policy before its effective date. Therefore, Commerce was aware of the proposed regulation when it addressed MAN Roland’s request to collapse on July 23, 1996.” KBA I, 22 CIT at-, 44 F.Supp.2d at 286 (citations omitted). .

"In determining whether to collapse related or affiliated companies, the Department must decide whether the affiliated companies are sufficiently intertwined as to permit the possibility of price manipulation. In making this decision, the Department considers factors such as:

(1) The level of common ownership; (2) interlocking boards of directors; (3) the existence of production facilities for similar or identical products that would not require retooling either plant’s facilities to implement a decision to restructure either company's manufacturing priorities; and (4) whether the operations of the companies are intertwined as evidenced by coordination in pricing decisions, shared employees or transactions between the companies.” Certain Pasta from Italy, 61 Fed.Reg. at 30,351. .

Control numbers, or “CONNUMs” are used by Commerce to designate "merchandise that is deemed 'identical' based on the Department’s model matching criteria.'’ Redetermi-nation at 3. Commerce occasionally uses internal model numbers instead of CONNUMs to identify identical merchandise. For example, in Certain Hot-Rolled. Carbon Steel Flat Products From Korea, 58 Fed.Reg. 37,176, 37,186 (Dep’t Commerce 1993) (final de-term.), Commerce used the more specific company-provided model numbers because, under the special circumstances of that case, "[fit would be distortive” to rely on the CON-NUMs. Id.

However, as Commerce itself notes, CONNUMs are used as the basis for product identification in most cases. See infra note 13, Antidumping Manual at 29. . Commerce described how it used CON-NUMs in this case to calculate the COM: "[I]n accordance with our normal practice, we next determined the CONNUM-specific, weighted-average cost of manufacturing ("COM”) for all subject merchandise produced by [MAN Roland] and MAN Plamag. Specifically, we first determined the CON-NUM-specific, weighted-average COM for each factory individually.

If the same CONNUM was produced at more than one factory, we would weight-average each factory’s actual COM for that CONNUM using each factory’s respective production quantify-” Second Redetermination at 7. . In Fresh Flowers, Commerce averaged the costs of two facilities of a collapsed company in computing a combined CV, finding that, “[a]lthough the flowers [were] somewhat different, [it] considered] spider chrysanthemums and standard chrysanthemums to be the same type and therefore calculated one CV for both.” Fresh Flowers at 20,497.

The Court responded that “somewhat different” and "same type” appeared not to be the same as “identical”. KBA II, 23 CIT at -, 44 F.Supp.2d at 283. Meanwhile, in Silicon Metal, Commerce averaged the costs incurred by several different furnaces because "other furnaces used to produce non-subject merchandise can be used to produce silicon metal.” 59 Fed.Reg. at 42,-808 (emphasis added). Commerce clarified on remand that all the furnaces in question had in fact produced the subject merchandise. See id. at-, 44 F.Supp.2d at 283.

Nonetheless, the Court understood that Commerce had cost-averaged "to prevent the respondent from being able 1o avoid dumping liability through the manipulation of production.” Id. Thus, "Silicon Metal did not require that the multiple facilities actually produce identical merchandise; rather, the decision was based on the ability to produce the subject merchandise at more than one facility.” Id. .

See Redetermination at 3 (citing Open-End Spun Rayon Singles Yarn From Austria, 62 Fed.Reg. 43,701, 43,703 (Dep’t Commerce 1997)(final determ .((“Austrian Yam ”); Canned Pineapple Fruit From Thailand, 62 Fed.Reg. 42,487, 42,491 (Dep’t Commerce 1997)(prelim. results admin, review)("Thai. Pineapple"); Antifriction Bearings (Other Than Tapered Roller Bearings) and Parts Thereof From France, et. al, 61 Fed.Reg. 66,-472, 66,497 (Dep’t Commerce 1996)(final results admin. review)(“Antifriction Bearings ”)).

The Court found that these determinations did not support Commerce's claim of an "established practice” because they were decided after the date of Germany Final. See id. "Moreover, none of Commerce’s cites specifically addressefd] the question at issue here: whether Commerce only averages production costs incurred by affiliated parties at multiple facilities where the facilities produce identical merchandise during the period of review.” Id. at-, 44 F.Supp.2d at 283-84 & n. 3. .

As Commerce notes, "the Department's long-standing practice is to calculate a separate dumping margin for each manufacturer or exporter investigated.” Second Redeter-mination at 4 (citing Certain Hot-Rolled Carbon Steel Flat Products From Japan, 58 Fed. Reg. 37,154, 37,159) (Dep’t Commerce 1993)(final determ.). . Commerce explains, Given that each factory’s results are affected by the merchandise actually produced, it would be unreasonable to adjust the actual cost of producing CONNUMs at one plant for the labor and overhead rates incurred at another plant to produce other merchandise.

It ignores the reality that had the MAN Plamag factory attempted to produce the LNPPs sold to the United States during the POI, it may have operated less efficient - ly and/or required more-highly paid workers that are more technically qualified. Second Redetermination at 10-11. . See Venezuelan SWR, 63 Fed.Reg. at 8,952 ("The cost information reported to the Department that will form the basis of the NV calculations ... must be ... [rjeflective of the actual cost of producing the product.”) .

See, e.g., Antidumping Manual, Ch. 7 at 29 (Rev. 1/98) (citation omitted): [T]he boundaries of the averaging groups are extremely important.... The items within the averaging groups should share as many common characteristics as feasible. For example, we nearly always calculate model-specific weighted-average prices.... Calculation of these 'narrower' weighted-average prices yields more accurate results than broad averages which mix sales with different characteristics which affect prices.

In a slightly different context, Commerce notes that it will compare NV to export price or constructed export price on a transaction-to-transaction basis for "made to order” goods, because "[t]he difference between these custom-made products render [sic] average prices meaningless”; that is, where the "averaging group” consists only of one product, averaging does not contribute to the accuracy of the comparison. Antidumping Manual, Ch. 6 at 7 (Rev. 1/98).

In this case, Commerce "based NV on CV because we determined that the particular market situation, which requires that the subject merchandise be built to each customer’s specifications, does not permit proper price-to-price comparisons.” Germany Final, 61 Fed.Reg. at 38,-171. Note that while the Antidumping Manual is not a binding legal document, it does give insight into the internal operating procedures of Commerce. . The "transfer price” is used to determine the price of inputs between affiliated parties, in accordance with 19 U.S.C. § 1677b(f)(2)-(3X1994):

(2) Transactions disregarded A transaction directly or indirectly between affiliated persons may be disregarded if, in the case of any element of value required to be considered, the amount representing that element does not fairly reflect the amount usually reflected in sales of merchandise under consideration in the market under consideration. If a transaction is disregarded under the preceding sentence and no other transactions are available for consideration, the determination of the amount shall be based on the information available as to what the amount would have been if the transaction had occurred between persons who are not affiliated.

(3) Major input rule If, in the case of a transaction between affiliated persons involving the production by one of such persons of a major input to the merchandise, the administering authority has reasonable grounds to believe or suspect that an amount represented as the value of such input is less than the cost of production of such input, then the administering authority may determine the value of the major input on the basis of the information available regarding such cost of production, if such cost is greater than the amount that would be determined for such input under paragraph (2). .

In AK Steel, the companies were collapsed for both sales and cost purposes; in other words, they were treated as both "a single exporter [and] a single producer for purposes of the antidumping inquiry.” See AK Steel, 22 CIT at-, 34 F.Supp.2d at 764.

AK Steel explicitly distinguished Commerce’s decision in Germany Final to apply the fair value and major input provisions on the ground that “there the companies manufactured different equipment models and were collapsed for cost purposes for selected items only, but not for all purposes, as in this case.” AK Steel, 22 CIT at -, 34 F.Supp.2d at 765-66 (citations omitted). It should be noted first that only MAN Roland exported the subject merchandise. See Second Redetermination at 6 n. 5 (citing MAN Roland’s Supp. Questionnaire Resp. at 56).

Thus, there was no reason for Commerce to collapse for sales purposes in order to treat MAN Roland and MAN Plamag as a single exporter. Moreover, AK Steel was decided before the Court’s decision in KBA I. KBA I remanded Commerce's decision not to cost-average after Commerce found that MAN Plamag was only an "affiliated party to [MAN Roland].” Germany Final at 38,188; see 22 CIT at-, 15 F.Supp.2d at 849. In hindsight., there was no need to distinguish Commerce’s decision to disregard the fair value and major input rules in AK Steel, and its decision to apply the fair value and major input rules in Germany Final.

Commerce considered MAN Roland and MAN Plamag to be affiliated, rather than collapsed, companies at the time of Germany Final, and thus applied the fair value and major input rules.