Dongguan Sunrise Furniture Co., Ltd. v. United States

865 F. Supp. 2d 1216, 2012 CIT 79, 34 I.T.R.D. (BNA) 1613, 2012 Ct. Intl. Trade LEXIS 80, 2012 WL 2045753

• Court: United States Court of International Trade
• Decided: June 6, 2012
• Docket: Consol. 10-00254
• Status: Published

Opinion

OPINION AND ORDER

RESTANI, Judge:

This action challenges the Department of Commerce’s (“Commerce”) final results rendered in the fourth antidumping (“AD”) duty review of certain wooden bedroom furniture (“WBF”) from the People’s Republic of China (“PRC”). See Wooden Bedroom Furniture From the People’s Republic of China: Final Results and Final Rescission in Part, 75 Fed.Reg. 50,992, 50,992 (Dep’t Commerce Aug. 18, 2010) (“Final Results ”). Plaintiffs Dongguan Sunrise Furniture Co., Ltd., Taicang Sunrise Wood Industry Co., Ltd., Taicang Fairmont Designs Furniture Co., Ltd., and Meizhou Sunrise Furniture Co., Ltd. (collectively “Fairmont”) moved for judgment on the agency record. See Mem. of Points and Auths. in Support of Rule 56.2 Mot. for J. Upon the Agency R. by PI. Fairmont Designs et al. (“Fairmont Br.”). Intervenor Plaintiffs Coaster Company of America, COE Ltd., Langfang Taincheng Furniture Co., Ltd. and Trade Masters of Texas, Inc. (collectively “Coaster”) moved for judgment on the agency record. 2 Mot. for J. Upon the Agency R. Pursuant to Rule 56.2 by Consolidated Pis. Coaster Co. of Am., COE Ltd., Langfang Tiancheng Furniture Co., Ltd. and Trade Masters of Texas, Inc. (“Coaster Br.”). Intervenor Defendants American Furniture Manufacturers Committee for Legal Trade and Vaughan-Bassett Furniture Company, Inc., (“AFMC”) also moved for judgment on the agency record. The AFMC’s Rule 56.2 Brief in Support of its Mot. for J. on the Agency R. (“AFMC Br.”). For the reasons stated below, the court remands in part and sustains in part the Final Results. The Government’s request for a remand on the issue or zeroing is granted. 3 See Def.’s Resp. to Pis.’ Rule 56.2 Mots. (“Def.’s Br.”).

BACKGROUND

In January 2005, Commerce published an AD duty order on WBF from the PRC. Notice of Amended Final Determination of Sales at Less Than Fair Value and Antidumping Duty Order: Wooden Bedroom Furniture From the People’s Republic of China, 70 Fed.Reg. 329, 329 (Dep’t Commerce Jan. 4, 2005). In January 2009, AFMC and others requested an administrative review of certain companies exporting WBF to the United States between January I, 2008 and December 31, 2008, thereby triggering the fourth administrative review of WBF. 4 Wooden Bedroom Furniture From, the People’s Republic of China: Preliminary Results of Anti-dumping Duty Administrative Review and Intent To Rescind Review in Part, 75 FecLReg. 5952, 5952-58 (Dep’t Commerce Feb. 5, 2010) (“Preliminary Results”). After publishing a notice of initiation and receiving questionnaire responses and comments, Commerce selected three mandatory respondents: Dalian Huafeng Furniture Co., Ltd. (“Huafeng”), Guangdong Yihua Timber Industry Co., Ltd. (“Yihua”), and Shanghai Aosen Furniture Co., Ltd. (“Aosen”). Preliminary Results, 75 Fed.Reg. at 5953.

On April 20 and 21, 2009, Commerce issued the antidumping questionnaire to the three mandatory respondents and made it available to voluntary respondents, including Fairmont. Id. During April and May 2009, AFMC and all other interested parties withdrew their request for review for two of the mandatory respondents, Huafeng and Yihua, and several other companies. Id. As a result, Commerce named Fairmont as an additional mandatory respondent on May 29, 2009. Id. Aosen withdrew from participation in the review, leaving Fairmont as the only cooperating mandatory respondent. Id. Fairmont responded to Commerce’s questionnaires and supplemental questionnaires between April 2009 and January 2010. Id.; Fairmont Br. 44 n. 155. In October and November 2009, Commerce verified Fairmont’s responses and found that Fairmont had failed to report sales of more than twenty in-scope product models. Id. at 5954; Preliminary Analysis Memorandum (Feb. 1, 2010), C.R. 356 at 30-32.

In the February 2010 Preliminary Results, Commerce calculated the wage rate using the now invalidated regression based methodology. Preliminary Results, 75 Fed.Reg. at 5962; see 19 C.F.R. § 351.408(c)(3), abrogated by Dorbest Ltd. v. United States, 604 F.3d 1363, 1377 (Fed. Cir.2010) (“Dorbest IV”). In response to Dorbest IV, Commerce, in this proceeding, placed additional labor data on the record and requested comments from interested parties. Labor Wage Rate (July 14, 2010), P.R. 916 at 1. AFMC, Coaster, and Fairmont submitted timely comments. See P.R. 919, 920, 921, 925, 926.

In the Final Results, Commerce revised the surrogate wage rate, the brokerage and handling surrogate value, and used the financial statements of various Philippine companies to calculate surrogate values. Final Results, 75 Fed.Reg. at 50,993-94; Issues and Decision Memorandum for the Final Results of the Administrative Review of the Antidumping Duty Order on Wooden Bedroom Furniture from the People’s Republic of China, A-570-890, POR 1/1/08-12/31/08, at 72 (Aug. 11, 2010) (“Issues and Decision Memorandum ”), available at http://ia.ita.doc.gov/frn/summary/ PRC/2010-20499-l.pdf (last visited June 5, 2012). Commerce assigned Fairmont a separate rate of 43.23%, which included the rate of 216.01% applicable to the PRC-wide entity for the unreported sales as adverse facts available (“AFA”). Final Results, 75 Fed.Reg. at 50,998.

JURISDICTION AND STANDARD OF REVIEW

The court has jurisdiction pursuant to 28 U.S.C. § 1581(c). The court will not uphold Commerce’s final determination in an AD review if it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law....” 19 U.S.C. § 1516a(b)(l)(B)(i).

DISCUSSION

I. Partial Adverse Facts Available

Fairmont argues its partial AFA rate is contrary to law because six products are not subject merchandise, Fairmont acted to the best of its ability, and the AFA rate applied is aberrational and not a reasonably accurate estimate of Fairmont’s actual dumping margin. Fairmont Br. 36-58.

A. Subject merchandise

Fairmont argues that six of the twenty-three products are not subject merchandise and thus, Commerce never requested information relating to these products and the information cannot be considered missing from the record. 5 Fairmont Br. 36-43. Fairmont’s argument lacks merit.

Commerce may use facts otherwise available if “an interested party ... withholds information that has been requested by the administering authority or the Commission under this subtitle.... ” 19 U.S.C. § 1677e(a)(2). Commerce’s Section C Questionnaire requested Fairmont to report sales data for all of its subject merchandise. Questionnaire Resp. (June 15, 2009) , Confidential App. to Def.’s Resp. to Pis. Rule 56.2 Mots. (“Def.’s Confidential App.”) Tab 2 at 2. Thus, whether Fairmont withheld or failed to provide requested information relating to these six products depends on whether the products are subject merchandise.

1. The dresser and custom cabinet products

Fairmont argues the dresser and custom cabinet products 6 are not subject merchandise because they are multifunction combination units designed for the living room of a hotel suite. Fairmont Br. 37, 40. Fairmont does not dispute that the dresser products contain a dresser, but argues the products are not subject merchandise because the dresser and custom cabinets are combined with a minibar and a TV panel and such combination units are not included in the scope language. Id. at 37-40.

Commerce found the products were subject merchandise because the customer expectations, end-uses, and manner of display were consistent with WBF. Issues and Decision Memorandum for the Final Results of the Administrative Review of the Antidumping Duty Order on Wooden Bedroom Furniture from the People’s Republic of China: Whether Certain Unreported Sales Determined to be Subject Merchandise in the Preliminary Results are Subject Merchandise, A-570-890, POR: 1/1/08-12/31/08, at 4-5 (Aug. 11, 2010) , C.R. 388 (“Confidential Issues and Decision Memorandum ”). Commerce relied on the products’ size, functionality, Fairmont’s descriptions of the products, Fairmont’s invoices, photographs of the products in hotel suites, and the fact that a hotel was the exclusive purchaser of the products. Id. Commerce also relied on the scope definition relating to combination units, and concluded the scope included items not specifically mentioned if the items were consistent with the scope language. Id.

Here, the shape and functionality of the products suggests that they are dressers. 7 Exhibit 13: FDUSA Unreported Sales Product Information (Dec. 10, 2009), C.R. 430 (“Exhibit 13 ”) at 47, 50, 66, 71 (diagrams showing dimensions of products). The products are combined either with a minibar or with a minibar and a TV panel, which does not fit within the scope language referring to combination units of desks and computer stands. See Final Results, 75 Fed.Reg. at 50,994-95 (stating the scope includes “desks, computer stands, filing cabinets, bookcases, or writing tables that are attached to or incorporated in the subject merchandise”) (“section 6”). The products also do not fit the “typical” definition of a door chest 8 or armoire 9 which are the two types of WBF the scope specifically mentions may be combined with a TV or other electronics. Id. at 50,994. Thus, a narrow reading of the scope definition suggests the products are not subject merchandise. The scope definition, however, provides only the “typical” physical descriptions of subject WBF. Moreover, the scope states that it includes “other bedroom furniture consistent with the above list,” id. at 50,995 (“section 7”), which suggests a product need not exactly match the listed products in order to be subject merchandise.

Fairmont argues that interpreting the scope language to include any subject merchandise that is combined with non-subject merchandise elements improperly expands the scope. PI. Fairmont Reply Brief (“Fairmont Reply”) 26-27. Fairmont argues section 6 of the scope should be read as the exclusive type of combination units and that if section 7 is interpreted to include any combination unit as long as some part is subject merchandise, then section 6 has no meaning. Id. at 27. Defendants argue that section 7 includes combination dresser units, even if they are not specified in the scope, as long as the combination units are otherwise consistent with the scope language. Def.’s Br. 15-16; The AFMC’s Resp. in Opp’n to Resp’ts’ 56.2 Mots, for J. on the Agency R. 5-6 (“AFMC’s Opp. Br.”).

Commerce’s finding that the products are subject merchandise is supported by substantial evidence on the record. The scope’s reliance on “typical” descriptions of products and the inclusion of any WBF consistent with the scope language demonstrates that a product may be subject merchandise even if it does not match the listed scope items. Final Results, 75 Fed.Reg. at 50,955 (including within the scope “other bedroom furniture consistent with the above list”). Thus, Commerce has not impermissibly expanded the scope merely by recognizing that not every type of WBF will exactly match the physical description of the listed products. Commerce’s interpretation does not render section 6 meaningless because only combination units consistent with the scope language, including section 6, could be considered subject merchandise under section 7. Inclusion of dresser space for a minibar is not very different from including space for a desk, bookcase, electronics, or filing cabinets, all of which are covered by scope language relating to combination units. Because it is undisputed that the products include characteristics consistent with scope merchandise and because the scope includes products consistent with the scope language, Commerce’s finding is supported by the record.

2. Nightstand back panel 10

Fairmont argues the nightstand back panel is an unfinished back panel that is excluded from the scope because it is a part that lacks the essential character of WBF. Fairmont Br. 41-43.

Commerce’s decision that the back panel is subject merchandise is supported by substantial evidence. Although Fairmont argues the back panel is a separate product that can be used for a variety of furniture pieces, there is no evidence on the record to show that the back panel was used as anything other than part of a nightstand. 11 Thus, it was reasonable for Commerce to find that the back panel and other portion of the nightstand were two pieces of an unassembled nightstand.

Even if Fairmont is correct that the back panel should be considered a spare part (and not a component), the back panel would be subject merchandise. Parts are included in the scope if they “possess the essential character of wooden bedroom furniture in an unassembled, incomplete, or unfinished form.” 12 Final Results, 75 Fed.Reg. at 50,995 n. 29. Even if the back panel itself lacks the essential character of WBF, it does possess the essential character of WBF in an “unassembled, incomplete, or unfinished form” because it can be combined with another part to create a nightstand. Thus, Commerce’s finding that the back panel was one half of a two-piece nightstand that was sold unassembled, and thus, was a component of a nightstand, is supported by substantial evidence on the record.

3. Wall-Mounted Product 13

Fairmont argues the wall-mounted product lacks the essential character of WBF because it is mounted to the wall and does not stand with any legs, and therefore, cannot be a night “stand.” Fairmont Br. 43. Fairmont argues the product is excluded under the scope language excluding “end tables, occasional tables, [and] wall systems” because it is a wall system. Id. at 43 & n. 151 (quoting Final Results, 75 Fed.Reg. at 50,995). Commerce found there was no reason to treat a table mounted to the wall differently from a table with legs. Confidential Issues and Decision Memorandum 10.

Commerce’s finding that the product is consistent with a nightstand is supported by substantial evidence. The scope includes “night tables, night stands” and “other bedroom furniture consistent with the above list.” Final Results, 75 Fed. Reg. at 50,994-95. Although it lacks legs, the product is consistent with the description and use of a bedside table. 14 Thus, all of the disputed products were properly found to be subject merchandise.

B. Fairmont failed to act to the best of its ability

Fairmont disputes that it failed to act to the best of its ability when responding to Commerce’s questionnaires and supplemental questionnaires. Fairmont Br. 44-45. Commerce found Fairmont failed to act to the best of its ability because it failed to report all of its sales and insisted that certain sales were out of scope, as discussed above, despite repeated questions from Commerce. Preliminary Results, 75 Fed.Reg. at 5960; Issues and Decision Memorandum 123-25. Fairmont’s argument lacks merit.

Commerce is permitted to use adverse inferences when selecting from among the facts available if it finds that a party “has failed to cooperate by not acting to the best of its ability to comply with a request for information.... ” 19 U.S.C. § 1677e(b). Commerce must do more than merely find the party has failed to provide the information. Nippon Steel Corp. v. United States, 337 F.3d 1373, 1382 (Fed.Cir.2003). Commerce must make an objective finding that a reasonable importer would have known the requested information should be in its records. Id. at 1382-83. Commerce must also make a subjective finding that the lack of cooperation is a result of “(a) failing to keep and maintain all required records, or (b) failing to put forth its maximum efforts to investigate and obtain the requested information from its records.” Id. Moreover,

[I]t is presumed that respondents are familiar with their own records. It is not an excuse that the employee assigned to prepare a response does not know what files exist, or where they are kept, or did not think — through inadvertence, neglect, or otherwise to look beyond the files immediately available.

Id. at 1383.

Here, Commerce found that given the explicit instructions of the Section C Questionnaire, Fairmont should have been aware of the need to provide a complete sales listing of its subject merchandise. Issues and Decision Memorandum 122. The Section C Questionnaire notified Fairmont that other companies had failed to report all sales of subject merchandise and asked Fairmont to confirm that it properly reported all of its sales of subject merchandise. Questionnaire Resp., Def.’s Confidential App. Tab 2 at 2. Fairmont confirmed that it had reported all sales of subject merchandise. Id. Thus, Fairmont knew that it should keep records of its sales and report all of its sales of subject merchandise.

Commerce’s subjective finding that Fairmont failed to put forth its maximum effort because it performed a perfunctory identification of in-scope sales is also supported by the record. Fairmont does not dispute that it directed an individual clerk to identify all in-scope sales. This clerk identified subject merchandise by searching for the terms “night tables” and “night stands” but did not look for similar words like “bedside tables.” Issues and Decision Memorandum 123; Fairmont Cmts. (Nov. 23, 2009), P.R. 749 at 5-6. Similar to the hypothetical in Nippon, it is not an excuse that an employee did not know how to identify in-scope merchandise. See Nippon, 337 F.3d at 1383. Many of the unreported products are easily identifiable as subject merchandise from the product descriptions. 15 Preliminary Analysis Memorandum, C.R. 356 at 30-32. And, for the majority of the products, Fairmont did not challenge Commerce’s finding that the in-scope merchandise were easily identified as subject merchandise from Fairmont’s sales listing. Issues and Decision Memorandum 123. Thus, had Fairmont properly instructed its employee, it could have readily identified all of the unreported subject merchandise sales. Because Fairmont performed a perfunctory evaluation of its own records and a reasonable amount of effort would have uncovered the disputed sales, Commerce did not err in finding that Fairmont failed to put forth its maximum effort in investigating its records.

Moreover, there was nothing unusual or truncated about the review process that could excuse Fairmont’s failure to accurately report its sales. Fairmont asked to be reviewed and had access to Commerce’s antidumping questionnaire on April 21, 2009. Preliminary Results, 75 Fed.Reg. at 5,953. Fairmont requested and received an extension to file its Section C/D questionnaire response and filed its response on June 15, 2009, two months after the questionnaire was issued. Issues and Decision Memorandum 122; see Questionnaire Resp., Def.’s Confidential App. Tab 2 at 1. Commerce then extended the deadline for the Preliminary Results from October 5, 2009 to February 1, 2010. Wooden Bedroom Furniture from the People’s Republic of China: Extension of the Time Limit for the Preliminary Results of the Antidumping Duty Administrative Review, 74 Fed.Reg. 47,919, 47,919 (Dep’t Commerce Sept. 18, 2009). Commerce conducted verification between October 26, 2009 and November 11, 2009, approximately six months after Fairmont first received the antidumping questionnaires. Preliminary Results, 75 Fed.Reg. at 5954. Fairmont never informed Commerce that it had difficulty responding to the questionnaires and Fairmont, particularly as a volunteer, should have been prepared to provide a full sales listing on June 15, 2009. At this time, Fairmont had only received one supplemental questionnaire on June 10, 2009. See Fairmont Br. 44 n. 155 (listing all supplemental questionnaires and number of questions). Thus, the multiple supplemental questionnaires and questions facing Fairmont in June through October cannot excuse its failure to provide an accurate sales listing on June 15. 16 Commerce therefore, did not err in finding that Fairmont had failed to act to the best of its ability. 17

C. 19 U.S.C. § 1677m(d)

Fairmont argues that even if it did not act to the best of its ability, Commerce can not apply adverse inferences because Commerce did not identify the deficiency in its response or give Fairmont an opportunity to remedy it. Fairmont Br. 47-48. Defendant argues that Commerce notified Fairmont of the deficiency in the June 26 and July 30 supplemental questionnaires. Def.’s Br. 22-25. Fairmont’s argument lacks merit.

If Commerce determines that a response to a request for information does not comply with the request, Commerce “shall promptly inform the person submitting the response of the nature of the deficiency....” 19 U.S.C. § 1677m(d). Once Commerce identifies a deficient submission, it must, “to the extent practicable, provide that person with an opportunity to remedy or explain the deficiency in light of the time limits established for the completion of investigations or reviews under this subtitle.” Id.

Here, the record shows that Commerce did not identify the deficiencies in Fairmont’s sales list until verification. In the June 26 supplemental questionnaire, Commerce simply stated that the products “may” be in scope. Questionnaire Resp. Letter, Def.’s Confidential App. Tab 3 at 2. Commerce did not identify the thirteen products not listed either by product code or description in the June 26 or July 30 supplemental questionnaires. Moreover, the July 30 supplemental questionnaire merely requested a clarification of a prior response by Fairmont, and did not suggest that Fairmont’s response was faulty. Supplemental Questionnaire Resp., Defi’s Confidential App. Tab 6 at 2 ¶ 94. Because the evidence does not suggest that Commerce was aware of specific deficiencies earlier, the court analyzes whether Commerce was required to provide Fairmont an opportunity to remedy the deficient submissions found at verification.

Commerce is required to provide an importer the opportunity to remedy a deficient submission only “to the extent practicable” in light of the time limits for reviews. 19 U.S.C. § 1677m(d). If the submission is not received within the applicable time limits, Commerce may disregard the submission if the information cannot be verified or the party has not acted to the best of its ability in providing the information. Id. § 1677m(d)(2).

Here, in response to the deficiencies found at verification, Fairmont offered to submit the missing sales and FOP information relating to the unreported products. Fairmont Cmts. (Nov. 23, 2009), P.R. 749 at 9. Commerce declined to accept this information, stating that it lacked time to consider the information, issue any supplemental questions, and verify the information. Issues and Decision Memorandum 115. Commerce also noted that allowing a party to wait until Commerce discovers an omission would allow the party to game the system. Id.

Fairmont argues Commerce’s reason for declining to consider information relating to the unreported sales is unsupported by substantial evidence because Commerce continued to accept new information, issue supplemental questionnaires, and conduct verification with another respondent through March 2010 and accepted new information. Fairmont Br. 50-51. The other respondent to which Fairmont refers, Nanjing Nanmu Furniture Co., Ltd. (“Nanjing Nanmu”), did not participate in the fourth administrative review. Preliminary Results, 75 Fed.Reg. at 5957. Instead, Nanjing Nanmu reported that it did not have any sales of subject merchandise during the period of review. Id. at 5958-57. In the Preliminary Results, Commerce stated that it lacked the information necessary to confirm the lack of Nanjing Nanmu shipments and stated it would continue to investigate. Id. at 5955. On March 1, 2010, Commerce issued a supplemental questionnaire to Nanjing Nanmu requesting information on its third-party relationships and sales. Supplemental Questionnaire (March 1, 2010), P.R. 825. Also in March 2010, Commerce provided a verification agenda to Nanjing Nanmu and placed the verification report on the record. Verification Agenda (March 4, 2010), P.R. 829; Verification of the Questionnaire Resp. of Nanjing Nanmu (March 31, 2010), P.R. 878. Fairmont cites to this supplemental questionnaire and verification as evidence that Commerce continued to issue supplemental questionnaires, conduct verification, and place additional evidence on the record through March 2010. 18 Whether Nanjing Nanmu accurately reported its sales determined whether it would receive a PRC-wide rate or its own rate. In contrast, Fairmont’s information, as the only cooperating mandatory respondent, determined the rate applied to all respondents qualifying for a rate separate from that of the PRC-wide entity. Thus, Commerce’s adherence to its deadlines during Fairmont’s verification, in order to issue the Preliminary Results on time, while delaying investigation of Nanjing Nanmu’s sales is not arbitrary. Although Commerce could have allowed the submission of the data, it was not required to do so.

D. Reasonableness of the AFA rate selected

Fairmont argues the partial AFA rate of 216.01% is not a reasonably accurate estimate of Fairmont’s actual dumping margin, albeit with a built in increase, because the selected AFA rate (1) is aberrational and (2) was not adjusted based on verified record evidence. Fairmont Br. 52-56. Fairmont also challenges the rate on procedural grounds, arguing that using a rate calculated from a different respondent, as opposed to Fairmont’s own data, was a change of policy without justification and Fairmont did not have an opportunity to comment on whether the other respondent was similar to Fairmont. Id. at 57. Defendants argue the AFA rate is reasonable and Fairmont had notice of the 216.01% rate and the opportunity to raise concerns in its case brief. Def.’s Br. 32; AFMC Opp. Br. 25-26. Fairmont’s argument has merit.

When Commerce applies an AFA rate that was calculated for a different respondent in a prior review, as it did here, Commerce must corroborate that rate with secondary information. See 19 U.S.C. § 1677e(c). The corroboration requirement tempers Commerce’s otherwise wide discretion in selecting AFA rates by “blocking] any temptation by Commerce to overreach reality in seeking to maximize deterrence.” F.lli De Ceceo Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027, 1032 (Fed.Cir.2000). Corroboration requires that Commerce use reliable facts to tie the AFA rate to the commercial reality of a particular respondent under review. Gallant Ocean (Thai.) Co. v. United States, 602 F.3d 1319, 1323-24 (Fed.Cir.2010) (finding an AFA rate must be both reliable and relevant to the particular respondent at issue). The goal is to calculate a reasonably accurate estimate of respondent’s rate, albeit with a built-in increase, as a deterrent for noncompliance. F.lli De Ceceo, 216 F.3d at 1032 (invalidating the 46.67% AFA rate imposed by Commerce because, inter alia, it “was many times higher than [respondent’s] actual dumping margin”).

Here, Commerce found the selected partial AFA rate was reliable because it is a company-specific rate calculated in the 2004-2005 new shipper review of Shenyang Kunyu Wood Industry Co., Ltd. (“Kunyu”) for WBF. Issues and Decision Memorandum 130; See Wooden Bedroom Furniture from the People’s Republic of China: Preliminary Results of 2001-2005 Semi-Annual New Shipper Reviews and Notice of Final Rescission of One New Shipper Review, 71 Fed.Reg. 38,373, 38,-378 (Dep’t Commerce July 6, 2006). Commerce must do more however, than merely proceed with the assumption that prior calculated margins are ipso facto reliable. See Lifestyle Enter. v. United States, 844 F.Supp.2d 1283, 1287-89, n. 5 (CIT 2012) (citing Ferro Union, Inc. v. United States, 23 CIT 178, 203, 44 F.Supp.2d 1310, 1334 (1999)). 19 Instead, Commerce must provide some justification for finding that the 2004-2005 rate for a relatively small shipper is relevant and reliable for this respondent in this time period. Commerce has failed to do so here.

Commerce found the Kunyu rate was relevant to Fairmont because a percentage of Fairmont’s sales were dumped at margins above 216%. 20 Issues and Decision Memorandum 129; Fairmont Br. 53 n. 191 (calculating that 1.21% of sales by value were dumped at margins above 216%). While the use of a respondent’s own sales data for corroboration may mollify commercial reality concerns, it is not necessarily sufficient in every case, especially in light of conflicting record evidence. See Lifestyle Enter., 844 F.Supp.2d at 1298. 21 Here, Commerce has not explained why a small percentage of Fairmont’s sales can be considered relevant and reliable for Fairmont’s unreported sales. See Gallant Ocean, 602 F.3d at 1324 (finding transaction-specific margins insufficient for corroboration where “Commerce did not identify any relationship between the small number of unusually high dumping transactions with [respon dent’s] actual rate”). Moreover, Commerce found that because the variety and amount of Fairmont’s sales and product line are so divergent, it was too difficult to use Fairmont’s own sales data to calculate a partial AFA rate. If this is the case, further doubt is cast on the use of a small percentage of sales for corroboration that the rate represents Fairmont’s commercial reality as a whole. Rather, the large diversity in Fairmont’s sales suggests using some broader base to derive a partial AFA rate or to corroborate it.

This is not a total AFA case where the record is devoid of all sales data, making it difficult for Commerce to determine a relevant and reliable rate for a respondent. Nor is it a case where the record contains demonstrably untrustworthy information. In these types of cases, Commerce has greater discretion in attempting to determine a relevant and reliable rate. See 19 U.S.C. § 1677e(c) (requiring Commerce to corroborate “to the extent practicable”). Here, Commerce has obtained and verified approximately 97% of Fairmont’s sales data and found a rate for these sales which was in the range of 34%. Commerce has not provided record evidence to justify the difference between the rate for the reported sales and 216.01% for the unreported sales. Fairmont placed a great deal of relevant information on the record and there is no evidence that it omitted the relatively small amount of sales data for strategic reasons. 22 Commerce is directed to calculate a new partial AFA rate which is corroborated or exercise its discretion to permit late filing of more data from Fairmont. 23

II. Surrogate Value for Labor 24

Coaster argues Commerce erred in not using wage data exclusively from the Philippines, as opposed to multiple countries, or if multiple country data are used, that Commerce erred in not using industry-specific data as opposed to economy-wide (i.e. manufacturing-sector) data. 25 Coaster Br. 7-10. AFMC argues Commerce erred in selecting the range of economically com parable countries based on absolute differences, instead of relative differences, in gross national income (“GNI”) and that Commerce should not have relied on data from India. AFMC Br. 24-30. Defendant argues Commerce applied an interim wage rate methodology previously approved by the court that is consistent with Dorbest TV and the statute. Def.’s Br. 58-61.

A. Country-specific data

Coaster argues that Commerce should have used labor data from the primary surrogate country, 26 the Philippines, and did not adequately explain why Philippine labor data were not the best available information compared to multi-country labor data. Coaster Br. 7-9. Coaster’s argument lacks merit.

In determining normal value for non-market economies, Commerce must use “the best available information regarding the values of such factors in a market economy country or countries considered to be appropriate by the administering authority.” 19 U.S.C. § 1677b(c)(l)(B) (emphasis added). When valuing the factors of production, Commerce “shall utilize, to the extent possible, the prices or costs of factors of production in one or more market economy countries .... ” 19 U.S.C. § 1677b(c)(4) (emphasis added). Thus, the statute permits the use of multicountry data if Commerce finds this is possible and appropriate.

In determining that multi-country data were preferable to data from a single country, Commerce found that because there is significant variation among the wage rates of comparable market economies, reliance on wage data from a single country was unreliable and arbitrary. Issues and Decision Memorandum 152. Commerce explained that “[t]here are many socio-economic, political and institutional factors, such as labor laws and policies unrelated to the size or strength of an economy, that cause significant variances in wage levels between countries.” Id. at 153. 27 In order to minimize the effects of these variations, Commerce used data from multiple countries. Id. Because the statute permits Commerce to use data from multiple countries and because it provided a reasonable explanation as to why it is preferable to do so, the court concludes that Commerce’s selection of multi-country data as the best available data is consistent with the governing statute.

B. Industry-specific data

Coaster argues Commerce erred in using aggregated manufacturing sector data, instead of data from a category more specific to the furniture industry. 28 Coast er Br. 10-11. Specifically, Coaster argues that Commerce should have relied on the International Labor Organization (“ILO”) classification 36 “manufacture of furniture; manufacture [not elsewhere classified]” submitted by Fairmont on July 19, 2010. Coaster Br. 10-12; Fairmont’s Wage Rate Cmts. (July 19, 2010), P.R. 921 at Ex. 2. Coaster’s argument has merit.

The statute requires that Commerce value the factors of production, to the extent possible, by using data from economically comparable countries that are significant producers of comparable merchandise. See 19 U.S.C. § 1677b(e)(4). The statute is silent as to how the wage rate should be calculated, but Commerce must be reasonable in calculating the wage rate from the best available information under the framework of the statute. See id. § 1677b(c)(l).

In response to multiple remands during the Dorbest litigation, Commerce developed a multi-step process to calculate the wage rate. The first three steps of the Dorbest methodology do not differ from the procedures Commerce used in the instant proceedings. 29 Commerce here, however, did not take the additional step of determining whether there is available industry-specific earnings data. Here, Commerce used economy-wide data even when industry-specific earnings or wage data was available for the same year. See Fairmont’s Wage Rate Cmts., P.R. 921 at 57-87. Thus, unlike the process described in the 2010 and 2011 Dorbest Remand Results, Commerce used economy-wide earnings data instead of further narrowing the data to industry-specific data. Commerce cannot, therefore, argue that Commerce has merely applied the interim methodology that it developed following Dorbest TV and which the court ultimately sustained. See Def.’s Br. 65.

Commerce attempted to justify its use of economy-wide data by stating it did not have time to adequately research this data. Coaster submitted industry-specific wage data from the Philippines in January 2010, seven months before the Final Results, and Coaster submitted industry-specific Philippines and multi-country wage data on July 19, 2010, which was the deadline set by Commerce for comments on its interim methodology. Fairmont Cmts. (Jan. 25, 2010), P.R. 803 at Ex. 14; Coaster Wage Rate Cmts. (July 19, 2010), P.R. 919 at Attach. 1; Fairmont’s Wage Rate Cmts., P.R. 921 at 67-87. Commerce’s argument that it did not have time to research the industry-specific data is un persuasive because it received the information within its own deadline.

Commerce also identified several distortions in the data submitted by Fairmont. These distortions, however, do not adequately explain why Fairmont’s data were not the best available information because similar distortions exist in the data used by Commerce and industry-specific data were far closer to the “comparable merchandise” data required by the statute. See 19 U.S.C. § 1677b(c)(4)(B). 30 For example, Commerce noted that the industry-specific ILO category included the manufacturing of furniture as well as other “not elsewhere classified” industries. Issues and Decision Memorandum 157. Commerce noted this would present a distortion because the furniture wages could not be separated from those of other industries and because each country may have a different definition of the furniture industry that would result in variations among countries. Id. at 157-58. Commerce’s data include all the manufacturing industries of a country. See id. at 158. Thus, Commerce’s data were already distorted to a greater degree because the wages relating to furniture cannot be separated from the other manufacturing industries. It is also unclear from Commerce’s data how each industry is weighted when calculating a manufacturing-wide wage rate and how differences in industry definitions among countries may affect how the various industry rates are averaged together to create a manufacturing sector rate. Thus, Commerce has not provided a sufficient explanation as to why economy-wide data are preferable to industry-specific data, even if the more specific data include some distortions.

Finally, Commerce found that the use of industry-specific data would result in narrowing the number of countries with available information from twenty-two to thirteen and that more data were preferable to less. Id. Although the number of countries is a relevant consideration, Commerce has not adequately explained why the decrease in countries trumps the advantages of using industry-specific data, especially in view of its shift to single country data. In the 2010 and 2011 Dorbest Remand Results Commerce relied on data from three and five countries respectively, which suggests data from thirteen countries is not too small a data set from which to calculate an accurate wage rate. Thus, the court remands for Commerce to calculate a wage rate using industry-specific data, or provide substantial evidence for why manufacturing data is preferable to a more industry-specific category, given the statutory requirement to value the factors of production based on data related to comparable merchandise.

C. Selection of bookend countries 31

AFMC argues that Commerce should have used relative differences in GNI when selecting the range of countries considered to be economically comparable to China and that by using absolute differences, Commerce’s selection is biased towards low-income countries. AFMC Br. 25-26. AFMC requests that Commerce recalculate the wage rate with a top bookend country that has a GNI 2.48 times greater than China’s GNI, which equals the relative difference between China’s GNI and the lowest bookend country. 32 AFMC Br. 26-28. AFMC’s argument lacks merit.

In calculating the wage rate, Commerce must use data from countries “at a level of economic development comparable to that of the nonmarket economy country....” 19 U.S.C. § 1677b(c)(4)(A). It is Commerce’s practice to rely on GNI when determining whether a country is economically comparable. See Dorbest V, 755 F.Supp.2d at 1295. 33 AFMC does not challenge Commerce’s reliance on GNI, but only the range of GNI that should be considered economically comparable to China’s.

Here, Commerce selected bookend countries that were evenly distributed around China’s GNI, with twenty-seven countries below China’s GNI and twenty-five countries above China’s GNI.' Labor Wage Rate, P.R. 916 at 4. Although AFMC has pointed out an alternative method for determining which countries are economically comparable that would result in a more preferable rate for AFMC, it has not shown that Commerce’s methodology or the use of absolute differences is unreasonable or unsupported.

D. India data 34

AFMC argues Commerce should not have used data from India in calculating the surrogate value for labor because the Indian data are distorted. AFMC Br. 28. Specifically, AFMC argues the data reported by India to the ILO include only those workers earning less than 1,600 rupees per month, or 6,500 rupees per month after 2005, as required by the Indian Payment of Wages Act. Id. at 28-29. AFMC argues that the data points that exceeded the reporting cap may be errors, or may include wages plus gratuities and bonuses, and that there is no explanation for why reported Indian wages nearly tripled from 2005 to 2006 other than the change in the cap in 2005. Id. at 29.

Commerce found the record evidence was insufficient to establish that the survey data were distorted by the wage cap. Issues and Decision Memorandum 159. Commerce noted that for some years before 2005, India reported a national average wage rate for manufacturing or an industry-specific wage rate that was above the cap. Id. (citing a 2006 industry specific wage of 6,678 rupees per month when the cap was 6,500 rupees per month and a 2004 national wage of 1,732 rupees per month when the cap was 1,600 rupees per month). Commerce noted that the notes to the Indian data cited by AFMC had not been updated since 1995 and that the ILO had expressed concern over relying on the methodology notes. AFMC Wage Rate Cmts. (July 19, 2010), P.R. 920 at Ex. 7 (“[T]his methodological description was made available to us in 1995 and no update has been received from India, nor effected by the ILO Dpt. of Statistics hence care should be exercised when using the information.”). Finally, Commerce speculated that it may be that the majority of wage earners make below the cap and thus, would be included in the survey even if there were no cap. Issues and Decision Memorandum 159.

Commerce’s explanations are speculative and not persuasive. India information is used in numerous cases and Commerce should be able to determine if actual wage data are being reported or whether the data are artificially capped. A statute that requires a cap would taint the data. Commerce should explain whether there is such a statute and what effect it has and make a judgment accordingly without guessing as to what might explain the discrepancies.

III. Other Normal Value Issues 35

Fairmont argues Commerce improperly relied on certain financial statements and double counted items when calculating financial ratios. Fairmont Br. 2-13. Fairmont’s argument lacks merit.

A. Financial statement selections

Fairmont argues that Commerce should not rely on the financial statements of Diretso Design Furnitures, Inc. (“Diretso”), APY Cane, Inc. (“APY Cane”), and Interior Crafts of the Islands, Inc. (“Interior Crafts”) because they are not producers of comparable merchandise. 36 Fairmont Br. 2-12. Fairmont also argues Commerce erred in excluding the financial statement of Tequesta International, Inc. (“Tequesta”). Id. at 12. AFMC argues Commerce should not have relied on the financial statements of Insular Rattan and Native Products Corp. (“Insular Rattan”) because it does not include a line item for taxes, which is necessary to determine whether the company received tax subsidies. AFMC Br. 11-14. AFMC’s argument has merit although Fairmont’s arguments do not.

In valuing the factors of production, Commerce must rely on the best available information from the surrogate country. 19 U.S.C. § 1677b(c)(l)(B). Commerce stated that in selecting the best available information, it uses data based on the “specificity, contemporaneity, and quality of the data.” Issues and Decision Memorandum 72. Commerce selected financial statements based on whether the company:

(1) manufactured wooden bedroom furniture; (2) had contemporaneous financial statements on the record; (3) received no subsidies found by the Department to be countervailable; (4) did not maintain significant retail operations outside the factory; (5) provided sufficient data for the Department to calculate surrogate factory overhead, SG & A and profit ratios; and (6) had an operating profit in 2008.

Id. at 83-84.

1. Diretso

Fairmont argues that Diretso is an interior design contractor and not a furniture producer. Fairmont Br. 7. Fairmont argues Diretso’s production machinery and equipment costs are too small for a furniture producer, that the 98% depreciation of machinery and equipment is too high for an operating manufacturer, and that 70% of labor costs are devoted to design services. 37 Id. at 7-8. Defendant argues Fairmont did not raise some of these issues before the agency and thus, the court should not consider them. 38 Def.’s Br. 34-35.

Commerce found Diretso was a producer of WBF because its website advertises “what appear to be wooden beds and bedside tables” and described the company as a manufacturer. Issues and Decision Memorandum 95.

Commerce’s finding is supported by substantial evidence on the record. Although Fairmont notes that Diretso’s promotional materials state that it furnishes the houses of upper class society, Fairmont ignores the preceding statement that Diretso’s “principal activity is to manufacturing [sic] furniture and furniture accessories.” AFMC Cmts. (Sept. 11, 2009), P.R. 677 at Attach. 13 at 135. Additionally, although Fairmont notes that promotional materials state that “Diretso’s new take in home furnishings made them a popular name in the interior design industry,” Fairmont does not define “take in home” or show why this would prove Diretso does not make the furniture that is being used by the interior design industry or other customers. Fairmont Rebuttal Submission (March 15, 2010), P.R. 848 at Ex. 1 at 5. Fairmont also ignores other sections of Diretso’s promotional materials, which refer to Diretso as “[m]akers of quality furniture.” Id. at Ex. 1 at 4. Thus, Fairmont’s arguments have not cast significant doubt on Commerce’s conclusion that Diretso is a furniture producer.

2. APY Cane

Fairmont argues APY Cane is not a furniture producer because its raw material costs were under 10% of its sales income and its work-in-process and finished goods inventories had virtually no movement in 2008. Fairmont Br. 8. Fairmont also argues APY Cane was in financial collapse during 2008, demonstrated by a 55% drop in sales, and should therefore not be used. Id. at 8-9.

Commerce’s conclusion to use APY Cane’s financial statement is supported by substantial evidence on the record. APY Cane’s financial statement describes its primary current business operation as manufacturing and exporting furniture and accessories. AFMC Surrogate Value Submission (March 4, 2010), P.R. 826 at 90. Commerce found that APY Cane had the type of costs indicative of a manufacturer, such as work-in-process and finished goods balances, raw materials costs, depreciation costs only for machinery and transportation, and no depreciation for an office building. Issues and Decision Memorandum 100-01. Although there was a 55% drop in sales in 2008, Commerce explained this should be viewed in the context of the international recession, and despite the drop, APY Cane had a profit in 2008. Id. at 100. Although APY Cane’s statements showed low raw material costs and little movement of work-in-process and finished goods, there is an expense line for freight and handling. AFMC Surrogate Value Submission, P.R. 826 at Attach. 5-A at 97. Thus, there is not sufficient contrary evidence here to render Commerce’s selection unsupported by substantial evidence.

3. Interior Crafts

Fairmont argues that Interior Crafts is a retailer and not a producer. Fairmont Br. 10. Fairmont relies on a statement in Interior Crafts’ financial statement that states its “primary purpose is to engage in manufacturing, selling of furniture & fixtures to distribute whether wholesale or retail household furniture made of wood.” Id. (emphasis in original) (quoting Interior Crafts’ financial statement, P.R. 826 at 230). Fairmont argues this statement should be interpreted to mean Interior Crafts is a retailer. Id. Fairmont also argues that Interior Crafts cannot be a manufacturer because its assets are largely comprised of “Leasehold Improvements,” which Fairmont states are not related to production activities. Id. at 10-11 & n. 40.

Although ambiguous, it was reasonable for Commerce to interpret the above sentence in Interior Crafts’ financial statement as demonstrating that it is a manufacturer of furniture and that it distributes to both wholesalers and retailers. Commerce also relied on Interior Crafts’ financial statement, its website, its product line of bedroom furniture, and its brochures to support its finding that Interior Crafts is a manufacturer. Issues and Decision Memorandum 108-09. Commerce also noted that Interior Crafts’ financial statement included beginning and ending work-in-process balances and that the depreciation of machinery and tools were 60% of total depreciation costs. Id. at 109. Fairmont has not disputed this evidence and thus, there is substantial evidence on the record to support Commerce’s finding. Furthermore, even if Fairmont’s argument and inferences relating to leaseholds are correct, this single line-item does not cast significant doubt on the substantial evidence supporting Commerce’s finding that Interior Crafts is a manufacturer.

4. Tequesta

Fairmont argues Commerce should have included Tequesta’s financial statement in its calculation of surrogate financial ratios. Fairmont Br. 12. Commerce excluded Tequesta’s statement because of the fifty-five manifests of shipments to the United States, only fourteen identify the Philippines as the country of origin and thus, Commerce found that Tequesta is primarily a reseller. Issues and Decision Memorandum 86. Fairmont argues the country of origin is irrelevant because it is not known what percentage of sales are represented by these fifty-five manifests, Tequesta’s financial statements say they are a manufacturer, and because some of the entries have a different country of origin but are marked “Made in Philippines.” Fairmont Br. 13. Because only some of the shipments are made in the Philippines or designate the Philippines as the country of origin, Commerce was justified in excluding Tequesta’s financial statement. 39

5. Insular Rattan

AFMC argues that the financial statement of Insular Rattan should not have been used because it is missing a line item for income taxes and because the auditor was the same auditor on a financial statement that Commerce rejected in the third administrative review. AFMC Br. 5-6, 11-14. Defendant argues that it is Commerce’s policy to reject incomplete financial statements only when the statement is missing critical information. Def.’s Br. 37-38.

Here, Commerce stated that it does not rely on taxes in calculating financial ratios and thus, the lack of a tax line alone does not render the statement unreliable. Issues and Decision Memorandum 88. Commerce also concluded that the lack of a tax line did not prevent it from finding that there were no subsidies because “[t]here is a large amount of information on the record regarding Insular Rattan and Petitioners have not cited to any evidence of subsidies received by Insular Rattan.” Id.

The court agrees with AFMC that the missing tax line is a relevant consideration that must be explained by Commerce. Although Commerce does not use taxes directly when calculating surrogate values, Commerce sometimes relies on notes to the tax line to determine whether the entity received disqualifying subsidies. See id. at 83-84 (selecting financial statements based on whether company receives subsi dies). Commerce also requires complete financial statements for general reliability reasons and improper omission of a tax line may mean that other items Commerce does use, such as profit, are not usable. 40 Commerce’s explanation that there may have been a tax holiday or some other explanation for the lack of tax line is speculation and cannot serve as substantial evidence. Without further information, the court cannot determine whether Commerce has decided unreasonably to use a dubious financial statement. Thus, the court remands for an explanation as to why Commerce finds that Insular Rattan’s financial statement is generally reliable and also unaffected by subsidies.

B. Double counting

1. Indirect Materials/Factory Overhead

Fairmont argues that four companies’ financial statements should not be used (Las Palmas Furniture, Inc. (“Las Palmas”), Clear Export, Heritage Muebles Mirabile Export Inc. (“Heritage”), and Interior Crafts) because these companies include certain raw materials in their factory supplies or indirect material line items, which Commerce improperly classifies as overhead, resulting in double counting. 41 Fairmont Br. 2-4, 6-10. Specifically, Fairmont argues FOP items including “glue, nails, paints, dowels, rivets, tacks, hardware, sandpaper, paper, oil, fasteners {e.g. steel screws, nut steel, bolt steel, washer steel, hinge steel, pin steel, steel hook, steel bar, screw/bolt pack, steel nails, etc.)” were included as “factory supplies” or “indirect materials” in the financial statements of these surrogates, resulting in double counting for these items. Fairmont Br. 2-4. In support, Fairmont cites the financial statements of two companies, Celloom Furniture Corporation and Bodega Arts & Designs, Inc. Id. at 3 n. 6.

The court finds Fairmont’s evidence insufficient to establish that any of the listed items are included both in FOP and overhead. First, the financial statements cited are not for the same companies used by Commerce, and thus, cannot establish double counting. 42 Second, the financial statements do not explain in sufficient detail what is included in each line item, and thus, cannot establish that the listed items are included. For example, the Bodega Arts & Designs statement refers only to “pre-fabricated materials” and “supplies.” Fairmont Rebuttal Cmts., P.R. 848 at 535. The Celloom Furniture statement merely refers to “supplies and packing materials.” Id. at 522. No explanation is given as to what these line items refer, and thus, they cannot provide evidence of double counting. Thus, Commerce’s finding that Fairmont had failed to demonstrate double counting actually occurred is supported by substantial evidence.

2. Berbenwood third-party services

Fairmont argues Commerce double counted by including third-party services received by Fairmont, such as subcontractor’s materials, labor, and energy, in factory overhead. Fairmont Br. 6. Defendant argues Commerce merely followed its practice and there is no evidence that the third party expenses included subcontractor costs. Def.’s Br. 46.

Commerce’s decision to classify “Third Party Services” as factory overhead is reasonable and supported by the evidence. Commerce stated that, consistent with its practice, because Berbenwood’s financial statements account for direct labor, materials, and energy as separate line items, the third-party expenses are classified as overhead costs. Issues and Decision Memorandum 92. There is no explanation on Berbenwood’s financial statement to demonstrate what is included in the “Third Party services” line and Fairmont presents no evidence to suggest that this line item includes labor costs. AFMC Rebuttal Cmts. (Jan. 14, 2010), P.R. 793 at 28. Thus, Fairmont has not shown that third-party costs are double counted.

3. Selling Costs

Fairmont argues that because it did not incur any selling costs when selling to its non-Chinese affiliate, the line items relating to selling costs should be excluded from the financial statements of Diretso, APY Cane, Heritage, Interior Crafts, Clear Export, and Coast Pacific Manufacturing Corp. (“Coast Pacific”). Fairmont Br. 5.

Commerce stated it did not exclude selling costs because it is Commerce’s longstanding practice to not attempt to adjust financial statements on a line-by-line basis, as this may lead to unintended distortions in the data rather than achieving greater accuracy. Issues and Decision Memorandum 77. Fairmont has not demonstrated that Commerce’s policy is unlawful. Thus, it was reasonable for Commerce to decline to adjust the financial statements to match the exact expenses of Fairmont. See Magnesium Corp. of Am. v. United States, 166 F.3d 1364, 1372 (Fed.Cir.1999) (noting statutory mandate directing Commerce to use “to the extent possible” prices in comparable market economies did not require an item-by-item accounting in calculating factory overhead).

4. Clear Export’s Import and Export Expenses

Fairmont argues Clear Export’s “Import and Export Expenses,” “Insurance” and “Fumigation” line items should not be classified as SG & A because these costs are already counted elsewhere in Commerce’s calculations. Fairmont Br. 10. Commerce found there was nothing on the record to show that the “import and export expenses” had already been included elsewhere in the normal value calculation, and thus, classified this line item as overhead. Issues and Decision Memorandum 104. Fairmont does not provide any citation to demonstrate how these particular line items are already included in Commerce’s calculations as either freight costs or brokerage and handling expenses. Thus, Fairmont has not provided sufficient evidence to show that Commerce has double counted in its calculations. 43

IV. Brokerage and Handling

Fairmont argues that Commerce’s use of data from the World Bank’s Doing Business in the Philippines: Trading Across Borders (“World Bank report”) to calculate a surrogate value for brokerage and handling is contrary to law. 44 Fairmont Br. 14. Specifically, Fairmont argues (1) the report uses price quotes, and not actual prices, (2) Commerce, in a different proceeding, found a similar World Bank report to be unreliable, (3) the report includes costs not incurred by Fairmont, and (4) the report uses costs for a 20-foot container, while Fairmont uses a 40-foot container, and Commerce’s assumption that the per-cubic foot costs are the same for both sizes of containers is unsupported by the record. Fairmont Br. 14-19. Fairmont argues that Commerce should have considered data from the Philippine Ports Authority to value brokerage fees and data from three Philippine brokerage companies, including two price quotes and one invoiee, to value handling. Fairmont Br. 19; Fairmont Surrogate Value Cmts. (March 5, 2010), C.R. 362 at Ex. 6A-6G. Fairmont’s argument lacks merit.

Here, Commerce considered the deficiencies in the data, as raised by Fairmont, and reasonably concluded that the deficiencies did not outweigh the benefits of relying on the World Bank data. See Issues and Decision Memorandum 48, 50. Commerce found the World Bank data were reliable because the data were published publicly, the data were based on six sources, instead of the three sources proposed by Fairmont, the World Bank is a reputable source of data, and the data were contemporaneous, specific to the costs in question, and tax exclusive. Issues and Decision Memorandum 47-52. In contrast, the data proposed by Fairmont to calculate brokerage costs were not published publicly, included one invoice and two price quotes, 45 the data did not represent a wider range of data than the World Bank report, and Commerce could not tell how the quotes were solicited or whether they were self-selected from a broader range of quotes. Id. at 51. Given the benefits found by Commerce of using the World Bank data versus the difficulties and unreliability of using the data supplied by Fairmont, the court concludes that the decision to use World Bank data is supported by substantial evidence.

Additionally, the flaws raised by Fairmont relating to the World Bank data are not sufficient to render the data unreliable. Fairmont argues the World Bank report is unreliable because in a different proceeding, Commerce found that the verified rates of three Indian companies were far lower than the rates reported in the World Bank data for India. Fairmont Br. 14 & n. 57. Fairmont argues that because the same methodology is used in all World Bank Doing Business reports, the court should conclude that the methodology of the World Bank Doing Business in the Philippines report is invalid. Id. at 14-15. The fact that three Indian companies report different data than the World Bank report, which aggregates data from multiple companies, does not cast significant doubt on the World Bank report’s methodology. 46 Moreover, Commerce has consistently found the World Bank to be a reliable source for data. Issues and Decision Memorandum 47 & n. 218.

Fairmont also argues that the surrogate value here differs substantially from surrogate values calculated for other Chinese companies. As a result, Fairmont argues that Commerce’s calculation is unlawful because it is inconsistent and unpredictable. Fairmont notes that Commerce has previously calculated surrogate brokerage and handling values for Chinese companies of $0.0083/kg, $0.0039/kg, and $0.0074/kg, as opposed to Fairmont’s value of $0.0297/ kg. 47 Fairmont Br. 15. None of these other reviews involved WBF or covered the same period of review at issue here. See id. 15 nn. 58-60; Fairmont Final Cmts. (Dec. 10, 2009), P.R. 757 at Ex. 1 at 5; App. To Memorandum of Points and Auths. in Support of R. 56.2 Mot. for J. upon the Agency R. by PI. Fairmont Designs Furniture Co., Ltd. (“PI. Fairmont App.”) Tab 74 at 17, Tab 76 at 10. These other reviews also used a simple average of three companies, instead of aggregated data from the entire country. PI. Fairmont App. Tab 74 at 17, Tab 76 at 10. Surrogate values calculated with a different methodology for a different period of review do not render Commerce’s calculations invalid. Additionally, the record shows that the surrogate value calculated here is not aberrational. Commerce compared the surrogate value to data on the record. This comparison showed that the per-volume cost of brokerage and handling in the World Bank report was less than in the World Bank India report, suggesting that the data here are not unreasonably high. Issties and Decision Memorandum 48-49; Fairmont Surrogate Value Cmts., C.R. 362 at Ex. 6A. And, the price was in line with the only data on the record for the same period of review. 48 Issues and Decision Memorandum 48-49. Based on these comparisons, the court cannot find that the World Bank Philippine data are unusually high. This is true, even if in other proceedings, with different records, Commerce calculated different surrogate values for different Chinese companies.

Fairmont also argues the World Bank data include costs that are not incurred by Fairmont, and thus, it would be contrary to law to apply those costs to Fairmont. Fairmont Br. 16-17. Fairmont notes that it had lower costs related to obtaining letters of credit and document services than the costs used in the World Bank report. Id. This argument is unavailing as Fairmont did not raise this argument before the agency. See Issues and Decision Memorandum 44-47. Regardless, the court does not find this sufficient to require a remand or recalculation because Commerce is not required to calculate an exact surrogate value for each respondent. See Nation Ford Chem. Co. v. United States, 166 F.3d 1373, 1377 (Fed.Cir.1999) (surrogate value process is “difficult and necessarily imprecise”) (citing Sigma Corp. v. United States, 117 F.3d 1401, 1407 (Fed.Cir.1997)). Instead, Commerce must use the best available information to estimate what a Chinese company’s costs would be in a market economy. Fairmont’s argument that it, as a non-market participant, did not incur some costs included in the World Bank report does not suggest that the World Bank report is not the best available information.

Finally, Fairmont argues that the surrogate value is not accurate because, even though it is a per-cubic foot value, it was calculated using a twenty-foot container and then applied to Fairmont’s forty-foot containers. Fairmont Br. 17-18. Fairmont argues that Commerce’s assumption that the per-cubic costs do not change depending on the size of the container is erroneous because the predominate portion of brokerage and handling is professional services, which will not change based on the size of the container. Id. at 18. Fairmont argues that converting the total cost of a twenty-foot container into a per-cubic foot ratio creates a distorted value because brokerage costs are based on value, not volume, and thus, do not increase proportionally with the number of cubic feet. Id. at 18-19. This argument fails because Fairmont has not presented evidence that brokerage costs are based on value, not volume, and do not increase proportionally with the number of cubic feet.

Thus, although World Bank Doing Business in Philippines data may not be perfect, Fairmont has not shown that the World Bank data are not the best available information when compared to Fairmont’s price quote, invoice, and Philippine Ports Authority data.

Y. Freight Revenue

A. 19 U.S.C. § 1677a

Fairmont argues that Commerce erred by treating freight revenue as an offset to freight expenses, capping freight revenue at freight expenses. 49 Fairmont Br. 20. Fairmont argues that freight revenue is part of the price of the subject merchandise under 19 U.S.C. § 1677a(a) and (b) and should not be treated as an adjustment under § 1677a(c)(2)(A). Id. at 20-23. Fairmont’s arguments lack merit.

In calculating export price and constructed export price (“U.S. price”), Commerce, consistent with § 1677a(c)(2)(A), deducts respondent’s freight expenses from that price. 19 U.S.C. § 1677a(c)(2)(A). 50 Although § 1677a(e)(2) does not expressly address freight revenue, Commerce then offsets respondent’s freight expenses with related freight revenues, resulting in a net freight expense. Issues and Decision Memorandum 67, 68. An example may help clarify this methodology: An exporter sells subject merchandise to an unaffiliated purchaser in the United States for a price of “X” dollars (United States price). It costs the exporter “Y” dollars (freight expenses) to ship the merchandise to the place of delivery in the United States.' The unaffiliated purchaser pays the exporter “Z” dollars (freight revenue) to provide the necessary shipping. Under these circumstances, Commerce makes the following adjustment: Export Price = X - (Y - Z). “Z,” however, is capped at “Y” so that any freight revenue that is in excess of freight expenses is not equated for. 51

Commerce’s approach is reasonable under the statute. It accords with the statutory language, allows Commerce to accurately account for freight expenses that a respondent actually incurred, and ensures that a respondent’s U.S. price is not overstated by profit earned from freight services.

Section 1677a(c)(2)(A) does not specify whether the “costs, charges, or expenses” incident to moving the subject merchandise should be calculated based on net or gross expenses. Given the silent statute, Commerce may apply its own reasonable methodology. United States v. Eurodif S.A., 555 U.S. 305, 316, 129 S.Ct. 878, 172 L.Ed.2d 679 (2009); see Fla. Citrus Mut. v. United States, 550 F.3d 1105, 1110 (Fed. Cir.2008) (applying the same reasoning to the term “import duties,” which is also found in § 1677a(c)(2)(A)). Commerce has elected to adjust for net expenses. Issues and Decision Memorandum 69 (“[T]he Department now consistently applies a policy of treating freight revenue as an offset to freight costs and capping freight revenue by the amount of corresponding freight costs.”). Commerce’s net revenue methodology is reasonable because it more accurately reflects the actual costs incurred by Fairmont in moving its subject merchandise and prevents Fairmont from inflating its export price with freight revenues. The plain language of § 1677a(c)(2) deals exclusively with downward adjustments to U.S. price. 19 U.S.C. § 1677a(c) (“The price used to establish export price and constructed export price shall be .... reduced by....”). If Commerce were to alter its methodology as Fairmont proposes and not cap freight-related revenue by the amount of related freight expenses, adjustments under § 1677a(c)(2) could potentially increase the export price or constructed export price (i.e. Commerce would “reduce” export price by subtracting a negative number). This would contradict the plain import of the statute.

Fairmont also takes issue with Commerce’s more general statutory construction, contending that because freight revenue is integral to the price of its subject merchandise, it should be included in U.S. price under § 1677a(a) or (b) rather than accounted for as an adjustment under § 1677a(c)(2)(A). Fairmont Br. 20. This argument overlooks the statutory require-, ment to adjust export price or constructed export price to permit an “apples-to-apples” comparison. In determining export or constructed export price, ,19 U.S.C. § 1677a requires Commerce to make adjustments in order to properly assess the amount by which normal value exceeds the United States price. 19 U.S.C. §§ 1673, 1677a(c)-(d). Adjustments are necessary because the reported prices “represent prices in different markets affected by a variety of differences in the chain of commerce” and must be adjusted to “reconstruct the price at a specific, ‘common’ point ..., so that value can be fairly compared on an equivalent basis.” SKF USA Inc. v. INA Walzlager Schaeffler KG, 180 F.3d 1370, 1373 (Fed.Cir.1999) (quoting Smith-Corona Grp. v. United States, 713 F.2d 1568, 1572-73 (Fed.Cir.1983)). Thus, it was reasonable for Commerce not to consider freight revenue as part of the price of the subject merchandise.

The record also casts doubt on Fairmont’s claim that freight revenue is inherently part of the price at which the subject merchandise is sold. The record shows that Fairmont charges separately for freight and classifies freight revenue under its own accounting code. Exhibit 6, C.R. 423 at Ex. 6 at 2, 4 (“Inland Freight” charged separately from price of merchandise on invoice); Verification Report of Fairmont Designs (Jan. 4, 2010), P.R. 773 at 6 (listing Fairmont’s accounting code for “freight revenue”). Fairmont further argues that with regard to its hospitality sales, merchandise price and freight revenue are intertwined because Fairmont is able to adjust the price of the goods and freight revenue based on the customer’s need. Fairmont Br. 26-27. Record evidence indicates that Fairmont rarely made this type of adjustment and that when Fairmont calculated the amount to charge for freight, it was largely concerned with freight costs. Verification Report of Fairmont Designs, P.R. 773 at 17. 52 Although Fairmont has put forth evidence to sug gest that the freight revenue it generated was more than a simple reimbursement for freight expenses, a proper “apples-to-apples” comparison should not include profit earned from the sale of a service (freight) as opposed to profit earned from the sale of the subject merchandise (furniture).

B. 19 C.F.R. § 351.401(c)

Fairmont argues that freight payments are “reasonably attributable” to the subject merchandise and thus should be considered adjustments to U.S. price under 19 C.F.R. § 351.401(c). Fairmont Br. 24; Fairmont Reply 16. This claim lacks merit.

Section 351.401(c) directs Commerce to use a price in the calculation of export or constructed export price which is “net of any price adjustment, as defined in § 351.102(b), that is reasonably attributable to the subject merchandise.” 19 C.F.R. § 351.401(c). Price adjustments are defined as “any change in the price charged for subject merchandise or the foreign like product, such as discounts, rebates and post-sale price adjustments, that are reflected in the purchaser’s net outlay.” 19 C.F.R. § 351.102(b)(38). Although the definition contains the phrase “such as” and is therefore illustrative, the purpose of the price adjustment provision is to account for any changes to the actual starting price of the subject merchandise and not to reflect any related expenses. Luoyang Bearing Corp. v. United States, 28 CIT 733, 770, 347 F.Supp.2d 1326, 1360 (2004). Thus, it was reasonable for Commerce to interpret the definition of price adjustment to not include the related freight expense.

C. Capping freight revenue versus capping other values

Fairmont argues that its freight revenue is the same as selling products with cost and freight (“C & F”) or delivered term sales. 53 Fairmont Reply 17. Because Commerce treats the price of C & F or delivered term sales as an integrated price, Fairmont insists that it is inconsistent and unlawful for Commerce to treat Fairmont’s pricing scheme differently. Id. This claim lacks merit.

Section 1677a(c)(l) lists the upward adjustments that are to be added to export price when not included in that price. This list does not include freight revenue. 19 U.S.C. § 1677a(c)(l). Thus, it is reasonable for Commerce to read the list of upward adjustments in § 1677a(c)(l) as an exclusive list that does not include freight revenue.

Further, as discussed above, record evidence indicates that Fairmont’s freight-revenue is separate and unique from the revenue generated from the sale of its goods. Fairmont’s approach of charging separately for freight demonstrates separate strategies for collecting payment for freight and the goods sold. Unlike Fairmont’s separate treatment of freight revenue, C & F prices are designed to link transportation costs with the price of the subject merchandise. Issues and Decision Memorandum 70. The purchaser makes a single payment and the exporter pays all expenses incurred in transporting the subject merchandise to the United States. The distinction between the two methods of pricing is sufficient to warrant different treatment by Commerce. If Commerce were to ignore this distinction and integrate freight revenue into U.S. price, Fairmont would be able to use profits from the sale of a freight service to increase the export price and thereby decrease the amount of antidumping duties imposed on the subject merchandise.

D. Inland freight offset

In the Preliminary Results, because Fairmont stated that freight revenue was provided based on the amount of U.S. inland freight only, Commerce capped freight revenue by the amount of U.S. inland transportation costs. Issues and Decision Memorandum 69. Fairmont argues that if Commerce continues to find that freight revenue should be capped at freight expense, then for those Hospitality Division sales for which Fairmont reported U.S. inland freight expenses in the ocean freight field, Commerce should include an estimate of the amount of the purported imbedded inland freight expenses. Fairmont Br. 30. This claim lacks merit.

Fairmont alleges that its motivation for imbedding inland freight expenses in the ocean freight field, and reporting zero in the U.S. inland freight column, was a Commerce questionnaire instruction that stated:

[I]t is not uncommon for certain of these transport expenses to be combined in a single fee paid [to] a transport company (e.g., combined ocean transport and U.S. internal transport to the customer’s place of delivery). If amounts are combined, do not attempt to separate them but report them in a single field and explain in your narrative response.

Antidumping Questionnaire (April 21, 2009), P.R. 445 at C-21.

Fairmont argues that by doing as Commerce requested, it has been left with an erroneous freight expense cap of zero for certain hospitality sales in which it incurred inland freight costs. Fairmont Br. 32. Although, Fairmont has put forth two methods for estimating the U.S. land freight expenses reported in the ocean freight field, Fairmont’s methodology requires Commerce to estimate U.S. land freight. Id. at 32. Fairmont has failed to show that any information exists on the record to accurately disaggregate U.S. inland freight amounts from ocean freight amounts and thus, it was reasonable for Commerce to decline to make such an estimation. See 19 C.F.R. § 351.401(b)(1) (“The interested party that is in possession of the relevant information has the burden of establishing to the satisfaction of the Secretary the amount and nature of a particular adjustment....”).

E. Procedural errors

Fairmont asserts that Commerce failed to provide an opportunity to comment upon the decision to cap the reported freight revenue because Commerce did not specifically notify Fairmont of the issue until the Preliminary Results and because Commerce rejected factual information submitted by Fairmont. Fairmont Br. 32-33. This claim lacks merit.

Commerce has established department precedent for its practice of capping freight revenue. Issues and Decision Memorandum 68; see Issues and Decision Memorandum for the Antidumping Duty Administrative Review of Polyethylene Retail Carrier Bags from the People’s Republic of China, A-570-886, AR: 8/01/06-07/31/07, at 12-14 (Feb. 4, 2009), available at http://ia.ita.doc.gov/frn/ summary/PRC/E9-2930-l.pdf (last visited June 5, 2012). It was foreseeable that Commerce would continue to follow this practice in the present case.

Fairmont’s rebuttal information was untimely new factual information. Pursuant to 19 C.F.R. § 351.301(b)(2), Commerce established a regulatory deadline for the submission of new factual information. See Letter Regarding New Factual Information (Mar. 25, 2010), P.R. 866 at 1. Fairmont’s rebuttal information was submitted beyond this deadline. Id. Although 19 C.F.R. § 351.301(e)(1) allows a party to submit certain rebuttal evidence in response to new evidence placed on the record by interested parties, it does not permit parties to submit new factual information in response to administrative determinations. Id. § 351.301(c)(1). While strict adherence to this provision might result in an unfair procedure in some cases, based on Commerce’s prior practice of capping freight revenue, Fairmont should have expected Commerce to continue this practice in the present case. In fact, Fairmont commented on the practice in its administrative case brief to Commerce. Fairmont Case Brief, P.R. 892, Vol. I at 24-34. Thus, there is no procedural defect to correct here.

VI. Combination Rates

AFMC argues that Commerce abused its discretion by refusing to develop the record with necessary information from respondents and by requiring AFMC to supply conclusive proof of circumvention. AFMC Br. 16, 24. AFMC relies on the increase in shipments by Aosen, 54 the similarity of invoices by Nanjing Nanmu and other companies, 55 and an article from Furniture Today that found circumvention was occurring to argue that Commerce cannot decline to investigate in light of such prima facie evidence of circumvention. 56 AFMC Br. 14,17,19.

The application of combination rates is left to the discretion of Commerce. See 19 C.F.R. § 351.107(b)(1) (Commerce “may establish” combination cash deposit rates for exporters and its supplying producers); see also Tung Mung Dev. Co. v. United States, 354 F.3d 1371, 1381 (Fed.Cir.2004) (affirming Commerce’s decision to apply combination rates in light of Commerce’s discretion). An agency’s failure to collect pertinent data, however, in some situations may constitute an abuse of discretion. See U.S. Steel Grp. v. United States, 18 CIT 1190, 1202, 873 F.Supp. 673, 687 (1994), aff'd 96 F.3d 1352 (Fed.Cir.1996) (holding failure to investigate not an abuse of discretion when evidence of production capability did not prove company produced a certain product).

Commerce examined and verified the sales of Nanjing Nanmu and determined that the sales were actually from Nanjing Nanmu. Issues and Decision Memorandum 37. Commerce found the increase in Aosen’s shipments and other shipments from producers through lower-rate companies may be explained by legitimate busi ness reasons, such as not having an export license and thus, did not mandate further investigation. Id. Although the inference could be drawn that circumvention did occur because shipments through other companies occurred, AFMC’s evidence does not show that the shipments through other companies were unlawful. See U.S. Steel Grp., 18 CIT at 1202, 873 F.Supp. at 687 (finding no abuse of discretion when inference could be drawn from evidence but evidence was not definitive). Moreover, Commerce applied the 216.01% PRC-wide rate to both Aosen and Nanjing Nanmu, which should prevent any circumvention by these companies.

The broader issue is whether Commerce should in its short form questionnaire, which focuses on whether a respondent is to get a rate other than that of the PRC-entity, ask about shipments of subject merchandise for or by another company. Apparently this type of inquiry was included previously. The court is concerned that Commerce’s answer that it cannot act because it has no circumvention data and the fact that it does not ask for the data creates a familiar geometric object. The court declines to order a new investigation here because AFMC’s evidence of circumvention is largely based on its own client’s general statements to a magazine. This is a troubling area, however, and Commerce should be prepared to alter its investigation techniques or explain its actions carefully in the future. It is also not a satisfactory answer that Commerce does attend to these problems in new shipper reviews.

VII. Separate Rates of Other Respondents

AFMC argues that if Fairmont’s separate-rate dumping margin of 43.23% is increased, Commerce should adjust the weighted-average separate-rate of the other respondents, which were calculated based on Fairmont’s rate. AFMC Br. 30. Defendant argues only those respondents subject to an injunction of liquidation (Fairmont, Coaster, and Longrange) may have their rates adjusted based on any changes to Fairmont’s rates. Def.’s Br. 72. Defendant’s argument has merit. Domestic parties may have liquidation of entries enjoined. Zenith Radio Corp. v. United States, 710 F.2d 806, 811-12 (Fed.Cir.1983). Thus, AFMC could have preserved this relief. Further new deposit rates have already been set for the future in other reviews. In this circumstance, the court will not change rates for entries that have already been liquidated.

CONCLUSION

For the foregoing reasons, the court remands the matter for Commerce to redetermine Fairmont’s AFA rate if an AFA rate is used, redetermine Fairmont’s separate rate using industry-specific wage data or provide substantial evidence as to why manufacturing sector data is preferable, explain whether the Indian wage data are distorted by a reporting cap, explain why Insular Rattan’s financial statement is generally reliable and usable, and provide an explanation of its zeroing practice. If Commerce calculates a different separate rate for Fairmont, Commerce shall make appropriate adjustments to the separate rates of the parties before the court in this litigation. Commerce’s determination is sustained in all other respects.

Commerce shall file its remand determination with the court within 60 days of this date. The parties shall have 30 days thereafter to file objections, and the Government will have 15 days thereafter to file its response.

2 . Fairmont adopted the arguments made by Coaster and Coaster adopted the arguments made by Fairmont. Fairmont Br. 36; Coaster Br. 16-17.

3 . The government requested remand on the issue of zeroing in order for Commerce to provide the explanation requested by the Federal Circuit in JTEKT Corp. v. United States, 642 F.3d 1378, 1384-85 (Fed.Cir.2011). Because both judicial and agency developments have occurred since the Final Results and because Commerce did not provide the explanation here that was sustained by the court in Union Steel v. United States, 823 F.Supp.2d 1346, 1347-48 (CIT 2012), the court grants the request.

4 . The subject merchandise includes the following items:

(1) Wooden beds such as loft beds, bunk beds, and other beds; (2) wooden headboards for beds (whether stand-alone or attached to side rails), wooden footboards for beds, wooden side rails for beds, and wooden canopies for beds; (3) night tables, night stands, dressers, commodes, bureaus, mule chests, gentlemen’s chests, bachelor’s chests, lingerie chests, wardrobes, vanities, chessers, chifforobes, and wardrobe-type cabinets; (4) dressers with framed glass mirrors that are attached to, incorporated in, sit on, or hang over the dresser; (5) chests-on-chests, highboys, lowboys, chests of drawers, chests, door chests, chiffoniers, hutches, and armoires; (6) desks, computer stands, filing cabinets, bookcases, or writing tables that are attached to or incorporated in the subject merchandise; and (7) other bedroom furniture consistent with the above list.

The scope of the order excludes the following items: (1) Seats, chairs, benches, couches, sofas, sofa beds, stools, and other seating furniture; (2) mattresses, mattress supports (including box springs), infant cribs, water beds, and futon frames; (3) office furniture, such as desks, stand-up desks, computer cabinets, filing cabinets, credenzas, and bookcases; (4) dining room or kitchen furniture such as dining tables, chairs, servers, sideboards, buffets, corner cabinets, china cabinets, and china hutches; (5) other non-bedroom furniture, such as television cabinets, cocktail tables, end tables, occasional tables, wall systems, book cases, and entertainment systems; (6) bedroom furniture made primarily of wicker, cane, osier, bamboo or rattan; (7) side rails for beds made of metal if sold separately from the headboard and footboard; (8) bedroom furniture in which bentwood parts predominate; (9) jewelry armoires; (10) cheval mirrors; (11) certain metal parts; (12) mirrors that do not attach to, incorporate in, sit on, or hang over a dresser if they are not designed and marketed to be sold in conjunction with a dresser as part of a dresser-mirror set; (13) upholstered beds and (14) toy boxes.

Final Results, 75 Fed.Reg. at 50,994-95 (footnotes containing definitions omitted).

5 . Fairmont challenges Commerce's determination that these products are in scope for six of the twenty-three products. Thus, for the remaining seventeen products, the issue of whether Commerce may apply facts available is not before the court.

6 . The four products are [[Confidential Data Deleted]], [[Confidential Data Deleted]], [[Confidential Data Deleted ]], and [[Confidential Data Deleted]]. Preliminary Analysis Memorandum, C.R. 356 at 30-32. Although only two of the products contain a TV panel, the court analyzes the four products together because the analysis of the scope language relating to combination units is equally applicable.

7 . All four products [[Confidential Data Delet ed]]. Exhibit 13, C.R. 430 at 47, 50, 66, 71.

8 . "A door chest, which is typically a chest with hinged doors to store clothing, whether •or not containing drawers. The piece may also include shelves for televisions and other entertainment electronics.” Final Results, 75 Fed.Reg. at 50,994 n. 22.

9 . “An armoire is typically a tall cabinet or wardrobe (typically 50 inches or taller), with doors, and with one or more drawers shelves, and/or garment rods or other apparatus for storing clothes. Bedroom armoires may also be used to hold television receivers and/or other audio-visual entertainment systems.” Final Results, 75 Fed.Reg. at 50,994 n. 25.

10 . This is product [[Confidential Data Deleted ]]. Preliminaiy Analysis Memorandum, C.R. 356 at 30-32.

11 . The record shows that [[Confidential Data Deleted]]. Exhibit 13, C.R. 430 at 60. Fairmont has not cited to evidence on the record to show that the back panel was ever [[Confidential Data Deleted]]. Id.; Confidential Issues and Decision Memorandum 8.

12 . The scope does not include "unfinished furniture parts made of wood products ... that are not otherwise specifically named in this scope ... and that do not possess the essential character of wooden bedroom furniture in an unassembled, incomplete, or unfinished form.” Final Results, 75 Fed.Reg. at 50,995 n. 29.

13 . This is product [[Confidential Data Deleted!]. Preliminaiy Analysis Memorandum, C.R. 356 at 30-32.

14 . Fairmont’s diagrams describe the item as a [[Confidential Data Deleted ]]. Exhibit 13, C.R. 430 at 78 (diagram of product).

15 . Some of the unreported products are described in Fairmont's records as [[Confidential Data Deleted ]], [[Confidential Data Delet ed]], and [[Confidential Data Deleted]]. Questionnaire Resp., Def.'s Confidential App. Tab 2; Preliminary Analysis Memorandum, C.R. 356 at 30-32.

16 . Fairmont argues the unreported sales were inadvertent reporting from being overwhelmed. Fairmont Br. 48. As noted above, Fairmont cannot claim to be overwhelmed by multiple supplemental questionnaires when it submitted its sales list of subject merchandise. Moreover, failure to report between two and three percent of sales by value is not so de minimis that it was in error for Commerce to find that the unreported sales cannot be disregarded.

17 . Commerce also relies on its post-June 15 correspondence with Fairmont as reason to impose a partial AFA rate. Issues and Decision Memorandum 122-23. Commerce’s characterization of the communications is not persuasive and the court does not rely on it.

18 . Fairmont also cites to a June 30, 2010 memorandum from Commerce that placed on the record a letter from the Chinese government expressing concern with Commerce’s actions in relation to Fairmont as well as Commerce's reply. Memorandum to File (June 30, 2010), P.R. 914. Although new to the record, this information is not new factual information relating to the calculation of dumping margins and did not need to be verified.

19 . Commerce also justified its decision by stating it has used the same AFA rate in every segment of the current proceeding since the new shipper review and has not received any information that it is not appropriate. Issues and Decision Memorandum 130. This is no longer the case because the court found in Lifestyle Enterprise that the 216.01% rate is an outlier when compared to the rates applied in previous WBF reviews. 844 F.Supp.2d at 1289-92 &n. 11.

20 . Commerce found that [[Confidential Data Deleted ]]. Proprietary Memorandum Regarding Corroboration (Aug. 11, 2010), C.R. 387 at 2. [[Confidential Data Deleted ]]. Id. at 6-7. Fairmont calculated that 1.21% of sales by value were dumped at rates above 216%. Fairmont Br. 53. AFMC does not dispute the 1.21% calculation but notes that [[Confidential Data Deleted ]] of Fairmont's sales by quantity were dumped at rates above 216%. AFMC’s Opp. Br. 22.

21 . That Lifestyle Enterprise involved corroboration with another company's data does not change the essential fact that Kunyu bears little relation to a company like Fairmont or to the mass of data that is available for Fairmont.

22 . Fairmont suggests that the margin for the unreported sales would actually be lower and it asserts that it was not permitted to demonstrate that Kunyu was not an acceptable source for a corroborating rate. As the court rejects Commerce's rate on other grounds, it is not necessary to decide if the record was unreasonably restricted or what the actual rate might be.

23 . This is not to say that Commerce must make adjustments to an AFA rate based on the respondent's own information. Commerce’s refusal to adjust its AFA rate based on Fairmont’s own indirect selling and international freight expenses or an ex-factory value is justified. Fairmont does not, after Commerce selects an AFA rate, get the opportunity to adjust that rate to its exact commercial reality. This does not, however, permit Commerce to select a rate that is not grounded in commercial reality in the first place.

24 . A dumping margin is the difference between the normal value ("NV”) of merchandise and the price for sale in the United States. See 19 U.S.C. § 1673e(a)(l); 19 U.S.C. § 1677(35). For merchandise exported from a non-market economy, such as the PRC, Commerce calculates NV "on the basis of the value of the factors of production utilized in producing the merchandise and to which shall be added an amount for general expenses and profit plus the cost of containers, coverings, and other expenses.” 19 U.S.C. § 1677b(c)(l). The factors of production include, but are not limited to, labor hours, raw materials, energy and other utilities, and representative capital cost, including depreciation. Id. § 1677b(c)(3). Surrogate values from market economy countries are used as a measure of these costs. See id. § 1677b(c)(l), (4); GPX Int’l Tire Corp. v. United States, 715 F.Supp.2d 1337, 1347 (CIT 2010), aff'd, 666 F.3d 732 (Fed.Cir.2011).

25 . The rate has been described as an economy-wide rate but the parties seem to agree it is a manufacturing-sector rate.

26 . The court notes that for future reviews, Commerce intends to calculate the wage rate using data from the surrogate country only. See Antidumping Methodologies in Proceedings Involving Non-Market Economies: Valuing the Factor of Production: Labor, 76 Fed.Reg. 36,092, 36,092-93 (Dep’t Commerce June 21, 2011). This change in policy is not retroactive and did not apply to this case.

27 . In making its finding, Commerce did not specifically mention the data from the Philippines. Coaster argues Commerce's explanation as to multiple versus single country data does not adequately explain why the data from the Philippines were not the best available information. Coaster Br. 8-9. This argument lacks merit. The Philippines is a single country and therefore, presents the same problems cited by Commerce that apply for using data from a single country, even if Commerce did not specifically name the Philippines. The court is informed however, that for future reviews Commerce did use data limited to the Philippines.

28 . Coaster also argues the court has previously found multi-industry, country-wide labor data to be unlawful. Coaster Br. 13-14 (citing Allied Pac. Food (Dalian) Co. v. United States, 32 CIT 1328, 1358-59, 587 F.Supp.2d 1330, 1357 (2008)). This argument lacks merit. Allied Food invalidated Commerce's regression based methodology and did not specifically rule on manufacturing sector data versus industry-specific data because industry-specific data were lacking. See Allied Pac. Food, 32 CIT at 1361-65, 587 F.Supp.2d at 1357-59, 1361.

29 . The Dorbest methodology is as follows. First, Commerce creates a list of economically comparable countries based on gross national income. Second, based on this list, Commerce then identifies which countries had exports of comparable merchandise during the period of review. Third, Commerce identifies which of these countries reported wage data during an applicable five-year period. Fourth, Commerce determines which countries reported industry-specific data. Finally, Commerce calculates an average wage rate from those countries found to be economically comparable that have exporters of comparable merchandise and which reported the appropriate data. In calculating the average wage rate, Commerce uses the three-digit sub-classification data when available. Dorbest Ltd. v. United States, 755 F.Supp.2d 1291, 1295-96 (CIT 2011) ("Dorbest V ") (describing methodology used in the 2010 Dorbest Remand Results); see also Dorbest Ltd. v. United States, 789 F.Supp.2d 1364, 1369 (CIT 2011) ("Dorbest VI”) (describing methodology used in the 2011 Dorbest Remand Results to include a determination as to which countries reported industry-specific wage data).

30 . The Government suggested at oral argument that 19 U.S.C. § 1677b(c)(4) concerning the valuation of factors of production (including labor) addresses the selection of surrogate countries and does not direct Commerce to use wage data for production of "comparable merchandise.” This is clearly incorrect. The factors of production are those used in “producing the merchandise” and the valuation of those factors "shall be based on the best available information regarding the values of such factors in a market economy country or comítries....” 19 U.S.C. § 1677b(c)(l)(B). One cannot divorce the valuation process from the factors of production for the comparable merchandise. The statute makes no sense if once Commerce selects proper countries it can look to uninformative information from such countries.

31 . When determining which countries are economically similar to China, Commerce uses per-capita GNI to identify countries at roughly the same level of economic develop ment as China. Here, those countries were India, the Philippines, Indonesia, Colombia, Thailand, and Peru. Issues and Decision Memorandum 153. Of these countries, Commerce selects the highest and lowest GNIs, which were India at $950 and Colombia at $4,100. Id.; see Labor Wage Rate, P.R. 916 at 4. Commerce then considers all market economy countries with a GNI that falls between India and Colombia, which resulted in fifty-two countries. Labor Wage Rate, P.R. 916 at 4. India and Colombia are referred to as the "bookend” countries.

32 . AFMC would have Commerce select an upper bookend country with a GNI of $5,852 instead of $4,100. See AFMC Br. 26.

33 . In Dorbest V, the court invalidated the selection of bookends when the lowest and highest bookend countries had GNIs below that of China. 755 F.Supp.2d at 1298. That is not the case here because Colombia and twenty-four other countries had GNIs above China.

34 . If AFMC were to prevail on this issue and India data were excluded from Commerce's calculations, the surrogate value wage rate would increase from $1.47 per hour to approximately $1.51 per hour. See Surrogate Values Memorandum (Aug. 11, 2010), P.R. 937 at Attach. IX.

35 . As stated above, when determining NV in an NME, Commerce determines values for general expenses and profit plus the cost of containers, coverings, and other expenses. 19 U.S.C. § 1677b(c)(l). Supra n. 21. Commerce usually calculates separate values for selling, general and administrative (“SG & A") expenses, manufacturing or factory overhead, and profit using ratios derived from financial statements of one or more companies that produce identical or comparable merchandise in the surrogate country. See, e.g., Shanghai Foreign Trade Enters. Co. v. United States, 28 CIT 480, 482, 318 F.Supp.2d 1339, 1341 (2004); U.S. Dep't of Commerce, 2009 Antidumping Manual, ch. 10, 17-18 (2009), available at http://ia.ita.doc.gov/ admanual/index.html (last visited June 5, 2012). To calculate the SG & A ratio, Commerce typically divides a surrogate company’s SG & A costs by its total cost of manufacturing. Shanghai Foreign Trade, 28 CIT at 482, 318 F.Supp.2d at 1341. For the manufacturing overhead ratio, Commerce divides total manufacturing overhead of the surrogate producer by the total materials, labor, and energy costs of the surrogate producer. Id. Finally, to determine a surrogate ratio for profit, Commerce divides before-tax profit of the surrogate producer by the sum of direct expenses, manufacturing overhead, and SG & A expenses of the surrogate producer. Id. These ratios are converted to percentages (“rates”) and multiplied by the surrogate values assigned by Commerce for the direct expenses, manufacturing overhead, and SG & A expenses of the exporter-respondent. Id.

36 . At oral argument, Fairmont abandoned its argument that Berbenwood Industries, Inc. is not a manufacturer of the subject merchandise.

37 . Fairmont compared the line item "Cost of Service" to the line item "Direct Labor” and found the “Cost of Service” line represented 70% of the total "Direct Labor” line. Fairmont Reply 9. There is no explanation of why "Cost of Service” equates to design services.

38 . Although Fairmont argued before the agency that Diretso was not a furniture producer, Fairmont did not raise below its specific arguments relating to equipment costs, depreciation, or labor. See Fairmont Case Brief, P.R. 904, Vol. 1 at 7. The court is reluctant to consider in the first instance inferences derived from the interpretation of specific items on financial statements, such as the depreciation and equipment costs. See Ta Chen Stainless Steel Pipe, Ltd. v. United States, 28 CIT 627, 644, 342 F.Supp.2d 1191, 1206 (2004) (noting exhaustion requirement ensures agency can use its expertise and develop the factual record). Nevertheless, here, the ability to draw the opposite inference from these particular lines in the financial statement does not defeat the otherwise substantial evidence supporting Commerce's determination, such as the promotions materials and website that describe Diretso as a wooden bedroom furniture manufacturer.

39 . Commerce also rejected Tequesta’s financial statement because it did not separately distinguish between purchased raw materials and purchased finished goods. Issues and Decision Memorandum 84 n. 379. Fairmont argues Commerce has no evidence to support its assumption that Tequesta purchased any finished goods, and thus, there is no reason for the financial statement to have distinguished between purchases of raw materials and purchases of finished goods. Fairmont Br. 12-13 & n. 44. Commerce’s finding is supported by the record. Tequesta imports products that do not have a Philippine country of origin and are not marked as "Made in the Philippines.” AFMC Cmts. (Sept. 28, 2009), P.R. 698 at 11. Because the record shows that Tequesta sells some products that it does not make but the financial statement only refers to purchases of raw materials and does not include a line item for purchased finished goods, Commerce’s concern that using Tequesta’s statement may result in a distortion is supported by substantial evidence on the record.

40 . Commerce previously rejected Insular Rattan's financial statement for defects not found here but the same kind of problem seems to continue. See Issues and Decision Memorandum for the Final Results of the 2007 Anti-dumping Duty Administrative and New Shipper Reviews, A-570-890, AR/NSR: 1/1/07— 12/31/07, at 35 (Aug. 10, 2009), available at http://ia.ita.doc.gov/frn/summaiy/PRC/E919666-1.pdf (last visited June 5, 2012), aff'd Lifestyle Enter., Inc. v. United States, 768 F.Supp.2d 1286, 1311 (CIT 2011).

41 . Fairmont also argues that the indirect materials line must include these raw materials because the ratio of factory supplies to indirect materials is so high. Fairmont Br. 4-5. Commerce properly rejected this argument, Issues and Decision Memorandum 74, because the mere fact that a company has high overhead expenses does not necessarily demonstrate that certain raw materials must have been included in the overhead calculations.

42 . Fairmont's reliance on an affidavit from a Philippine financial officer is not substantial evidence because it does not relate to any of the financial statements used by Commerce and only provides generalized statements about the Philippine economy.

43 . Fairmont argues that APY Cane's import and export shipping costs and Heritage’s export expenses should be excluded from SG & A. Fairmont Br. 9. The court rejects this argument for the same reasons as those stated in regard to Clear Export.

Fairmont also argues that several other line items from several financial statements should be excluded from SG & A either because of double counting or because Fairmont does not incur the same expenses, including APY Cane's fumigation, Clear Export’s fumigation and insurance, Interior Crafts' documentation, and Las Palmas’ delivery expenses, importation fees, and insurance. Defendant argues the court should not address these arguments because they were not brought before the agency. Def.'s Br. 48.

In relation to other line items, Commerce stated that its policy is "to not make adjustments to the financial statements data, as doing so may introduce unintended distortions into the data rather than achieving greater accuracy----In calculating overhead and SG & A, it is the Department's practice to accept data from the surrogate producer's financial statements in toto, rather than performing a line-by-line analysis of the types of expenses included in each category.” Issues and Decision Memorandum 75 (quoting previous reviews). Commerce is not required to do a line-by-line analysis in calculating factory overhead. Magnesium Corp. of Am., 166 F.3d at 1372. Thus, the court finds Commerce did not err in including these particular line items, as exclusion may have resulted in even greater distortions in the data.

44 . Commerce is to subtract from its calculations of normal value "costs, charges, and expenses incident to bringing the foreign like product from the original place of shipment to the place of delivery to the purchaser....” 19 U.S.C. § 1677b(a)(6)(B)(ii). These costs are generally called movement expenses. The subtraction of these costs from a respondent's normal value is intended to allow a fair comparison to net (or ex-factory) prices, which are not affected by the extra costs experienced by an exporter in shipping products around the world. Movement expenses include, inter alia, brokerage and handling and freight costs. In non-market economies, Commerce calculates a surrogate value for brokerage and handling.

45 . Fairmont argues that because the World Bank report relied on price quotes from lawyers and business consultants, as opposed to actual customer and supplier data, the data are not reliable, as anonymous price quotes will be overstated. Fairmont Br. 14, 16. Be cause the alternative evidence submitted by Fairmont would also require reliance on price quotes, the possible use of price quotes in the World Bank data does not subtract from the substantial evidence supporting Commerce’s finding.

46 . Fairmont argues that Commerce did not explain why the brokerage and handling expenses of a particular company would differ significantly from the expenses in a broad-based survey of market participants. Fairmont Reply 21. Although Fairmont argued the World Bank data were aberrational compared to the Indian companies, Fairmont did not argue before the agency that the difference in rates rendered the World Bank’s methodology inaccurate, see Fairmont Case Brief (April 13, 2010), P.R. 379 at vol. II at 44-48; Issues and Decision Memorandum 44-47, and thus, Commerce could not address it. This is not a grounds for finding error.

47 . Before Commerce, all the parties agreed that the surrogate value used in the Preliminary Results was flawed because the data used did not include handling expenses. Issues and Decision Memorandum 47. In the Final Results, Commerce calculated a surrogate value of $0.5520 per cubic foot, Surrogate Value Memorandum, P.R. 937 at Attach. 5, or $1,297.30 per container, Fainnont Surrogate Value Comments (March 5, 2010), P.R. 830 at 533. Fairmont proposed a value of $107.03 per container (brokerage) plus $158.60 per container (handling) for a total of $265.63 per container. Fairmont Surrogate Value Comments, P.R. 830 at 533.

48 . The only data on record for the same period of review were from an Indian company, Navneet Publications (India) Ltd. ("Navneet”). Fairmont calculated a brokerage and handling expense of $0,020, which is similar to the World Bank’s calculation of $0,029. Issues and Decision Memorandum 48 n. 229.

49 . "Freight revenue” is the amount that a purchaser pays an exporter to ship the subject merchandise to the United States or, put differently, it is the amount of revenue an exporter generates by providing shipping services. "Freight expenses” are the costs incurred by an exporter, which are incident to bringing the subject merchandise to the United States.

50 . Export price and constructed export price shall be reduced by "the amount, if any, included in such price, attributable to any additional costs, charges, or expenses ..., which are incident to bringing the subject merchandise from the original place of shipment in the exporting country to the place of delivery in the United States....” 19 U.S.C. § 1677a(c)(2)(A).

51 . Under the same circumstances, Fairmont proposes the following equation: Export Price = (X + Z) - Y. Fairmont argues that “Z" should not be subject to any cap.

52 . Commerce’s verification report stated:

We asked company officials if freight revenue was ever based on non-freight concerns .... The Hospitality division sales manager explained that for Hospitality division sales, in very rare situations, the customer may ask that adjustments to the purchase price be reflected in freight charges rather than in the selling price. The Home division sales manager said that to his knowledge FDUSA [Fairmont's U.S. subsidiary] bases all calculations of freight revenue on Home division sales on freight cost considerations and on nothing else.

Verification Report of Fairmont Designs, P.R. 773 at 17.

53 . C & F is a term of sale signifying that the price invoiced or quoted by a seller for a shipment does not include insurance charges, but does include all expenses related to freight up to a named destination.

54 . AFMC argues that because Aosen’s shipments increased from 300 containers to 2222 containers after receiving a zero cash deposits rate and Aosen refused to cooperate in the present review, Commerce should have concluded that Aosen was participating in a circumvention scheme. AFMC Br. 17-19.

55 . AFMC argues [[Confidential Data Delet edI]], who had a 216.01% rate, was using the names of exporters with lower rates when creating its invoices. AFMC Br. 19-20. For example, one invoice supposedly from Nanjing Nanmu reflects [[Confidential Data Deleted!]. According to AFMC, \[Confidential Data Deleted]]. Id. Commerce found the Nanjing Nanmu invoices had been verified as actually from Nanjing Nanmu and that AFMC had not supported its assertions relating to [[Confidential Data Deleted ]] with factual evidence. Issues and Decision Memorandum 37. Commerce further found that even if there are shipments through lower-rate companies, the shipments likely can be explained by legitimate commercial reasons. Id.

56 . In the Third Administrative review of WBF, the court rejected the Furniture Today article as insufficient proof to require Commerce to apply combination rates. Lifestyle Enter., 768 F.Supp.2d at 1313-14.