PUBLIC VERSION

FINAL RESULTS OF REDETERMINATION

PURSUANT TO COURT REMAND

Queen's Flowers de Colombia, et al

v.

United States

Slip Op. 97-120 (August 25, 1997)

CIT No. 96-08-01921

SUMMARY

On August 25, 1997, the Court of International Trade (CIT) issued a remand regarding Certain Fresh Cut Flowes from Colombia; Final Results of Antidumping Duty Administrative Reviews, 61 FR 42833 (August 19, 1996) (Final Results), covering three periods of review from March 1, 1991 through February 28, 1994 (Queen's Flowers de Colombia et al v. United States, Slip Op. 97-120) (Queen's). On December 15, 1997, the Department of Commerce (the Department) released to concerned parties its draft results of redetermination pursuant to the CIT's remand (Draft Results). Subsequently, we received comments on our draft results from the "Queen's Flowers Group," (the respondents), and rebuttal comments from the Floral Trade Council (the petitioner). With consideration of these comments, we have prepared these final results of redetermination pursuant to the above remand order of the CIT.

The Final Results covered a total of 348 producers and/or exporters and their shipments of certain fresh cut flowers, i.e., standard carnations, miniature (spray) carnations, standard chrysanthemums, and pompon chrysanthemums, from Colombia to the United States during at least one of the following periods of review (PORs): March 1, 1991 through February 29, 1992 (5th POR); March 1, 1992 through February 28, 1993 (6th POR); and March 1, 1993 through February 28, 1994 (7th POR). This remand pertains only to the following 20 companies: MG Consultores, Flores Canelon, Flores la Valvanera, Flores del Hato, Agroindustrial del Riofrio, Jardines de Chia, Queen's Flowers de Colombia, Jardines Fredonia, Flores Jayvana, Flores el Cacique, Flores Calima, Flores la Mana, Flores el Cipres, Flores el Roble, Flores del Bojaca, Flores el Tandil, Flores el Aljibe, Flores Atlas, Floranova, and Cultivos Generales (collectively "the Queen's Flowers Group").(1) The CIT ordered the Department to reconsider the following aspects of the Final Results: 1) that these 20 firms are related, pursuant to section 771(13) of the Tariff Act of 1930, as amended (the Act); 2) that these firms were sufficiently related to warrant collapsing into a single entity; 3) that margins based upon best information available (BIA) were appropriate for the Queen's Flowers Group during each of the three PORs; and 4) that use of first-tier BIA margins was appropriate.

The Department has addressed each of these issues on remand. We have confirmed our previous determinations, and we have determined that the antidumping duty margins for the Queen's Flowers Group during the relevant periods of review remain the same. For ease of reference, all citations to the CIT's Queen's decision are to the confidential Slip Opinion version of August 25, 1997.

Comments Regarding General Issues

In their comments on our draft results, the respondents argue that the Department (1) is resisting certain findings and rulings of the CIT, (2) has based its conclusions upon a distorted presentation of the facts of the case, and (3) lacks objectivity and fairness in its judgment. With respect to the last point, the respondents argue that the Department's actions involved in the case appear to be biased against the respondents. The respondents note that the original LOTUS spreadsheet the Department provided to the companies for reporting data was protected by an unfortunate password that indicated neither fair nor objective treatment of the respondents.

The petitioner responds that the draft remand results are responsive to the CIT's concerns and show that the Department's final determination with respect to the Queen's Flowers Group was supported by substantial evidence and was otherwise in accordance with the law. Furthermore, the petitioner asserts that the password the Department used to protect the LOTUS spreadsheet does not demonstrate a clear "bias" against the respondents versus the petitioner because it is not apparently directed to either the respondents' counsel or the petitioner's counsel.

However, the petitioner contends that the discovery and possible disclosure of the password by the respondents' counsel may have compromised the diskette data submissions for any and all respondents involved in the underlying administrative reviews. The petitioner claims that the respondents' counsel, as legal counsel and officers of the court, had no legitimate reason to uncover or disclose the password.

Department's Response Regarding General Issues

We disagree with the respondents that, in our draft results, we resisted certain findings and rulings of the CIT, based our conclusions upon a distorted presentation of the facts of the case, and lacked objectivity and fairness in our judgment. Our findings, which are based on record evidence, are detailed below. Although one of the passwords used to protect the LOTUS spreadsheets was offensive and inexcusable, the choice of the password did not affect our fair and impartial analysis of the facts and information on the record.

Like the petitioner in this case, we are equally disturbed that the respondents' counsel uncovered our various passwords even though there were no legitimate reasons for counsel to have done so. However, to the best of our knowledge, these passwords were not uncovered by counsel until all diskette submissions were made by respondents. Also, since we have not used any diskette submissions to calculate margins in this redetermination, the issue is moot with respect to the Queen's Flowers Group companies.

DISCUSSION

Issue 1: Whether the 20 companies in the Queen's Flowers Group are Related

1. Background

In the Final Results, we determined that all 20 companies included within the Queen's Flowers Group were sufficiently related within the meaning of section 771(13) of the Act to warrant collapsing into a single entity. First, we found nine companies sufficiently related to warrant collapsing into the [ ] sub-group: MG Consultores, Flores Canelon, Flores la Valvanera, Flores del Hato, Agroindustrial del Riofrio, Jardines de Chia, Flores Jayvana, Flores el Cacique and Flores Calima. Likewise, we collapsed 14 companies into the [ ] sub-group: Queen's Flowers de Colombia, Jardines Fredonia, Agroindustrial del Riofrio, Jardines de Chia, Flores Calima, Flores la Mana, Flores el Cipres, Flores el Roble, Flores del Bojaca, Flores el Tandil, Flores el Ajibe, Flores Atlas, Floranova and Cultivos Generales.

In the second step of our analysis, we determined that each of the two sub-groups constituted "an enterprise composed of related parties and, thus, a 'person'" within the meaning of 19 C.F.R. 353.2(p) (1994). Memorandum, "Collapsing Related Parties within the Queen's Flowers Group," Feb. 1, 1996, at 4-5 (hereinafter February Memo). Having determined that the two sub-groups constituted legal "persons," the Department then determined that they were sufficiently related within the meaning of section 771(13) to warrant collapsing. See February Memo at 4-10. We based this determination upon an analysis of both financial and non-financial factors. Id.

In reconsidering our collapsing determination, we have clarified our analysis by ensuring that we examined each POR separately. This analysis shows that not all 20 companies should be included for purposes of the 5th and 6th PORs. Flores el Cacique, Flores Calima, Flores el Roble, and Flores el Ajibe were not incorporated until the 7th review, and so they cannot be included in the Queen's Flowers Group for the 5th and 6th reviews. Additionally, because [ ] did not acquire any equity ownership in Floranova until the 6th POR, Floranova should not be included in the Queen's Flowers Group for the 5th review. Three other firms, MG Consultores, Flores la Valvanera, and Flores del Bojaca, reported a change in their equity structures between the 5th and the 6th PORs. However, these changes were minor and did not reduce common ownership below the five-percent or the 20-percent threshold necessary to find parties related.

During the reviews and subsequent litigation, the Queen's Flowers Group argued that section 771(13) of the Act was not the appropriate definition of related parties to use in a collapsing analysis. Moreover, the Queen's Flowers Group argued that, even under the section 771(13) definition of related parties, the Department's determination that all 20 firms were related was not supported by substantial evidence on the record.

The CIT sustained the Department's use in its collapsing analysis of the related-party definition in section 771(13) as a reasonable construction of the statute. Slip Op. 97-120 at 13. The CIT also agreed that seven of the nine firms within the [ ] sub-group were related, id. at 14-15, and that the 14 firms included in the [ ] sub-group were related. Id. at 20.

Regarding the [ ] sub-group, the CIT agreed with the Department that, based on the record, MG Consultores owned between [ ] and [ ] percent in Flores Canelon, Flores la Valvanera, Flores del Hato, Agroindustria del Riofrio, Flores Jayvana and Flores el Cacique. Thus, each of these six companies is related to MG Consultores as subsidiaries pursuant to section 771(13)(B), and all six are related to each other pursuant to section 771(13)(D). Slip Op. 97-120 at 14-15.

However, while recognizing that Flores Calima and Jardines de Chia are members of the [ ] sub-group, the CIT found that the two companies were not related to the other members of the [ ] sub-group. Id. at 15. The CIT rejected the analysis of Jardines de Chia's relationship as a post-hoc rationalization and found that the Department had not explained its determination concerning Flores Calima. Id. at 15-16.

In addition, the Department relied in its Final Results upon non-financial factors to support its finding that Jardines de Chia and Calima were related to the other members of the [ ] group. The CIT found that, in so doing, the Department had departed without explanation from its position in two previous determinations, Final Determination of Sales at Less Than Fair Value: Fresh Cut Roses from Ecuador, 60 FR 7019, 7040 (Feb. 8, 1995) (Roses), and Notice of Final Determination of Sales at Less Than Fair Value and Final Negative Critical Circumstances Determination: Disposable Pocket Lighters From Thailand, 60 FR 14263, 14268 (March 16, 1995) (Lighters). See Slip Op. 97-120 at 16-18.

The CIT also ordered the Department to explain how the [ ] and [ ] sub-groups are related within the meaning of section 771(13) of the Act, noting that the Department's analysis suffered from the similar flaw of relying upon non-financial factors inconsistent with prior practice. Slip Op. 97-120 at 21-22. We address each of these issues below.

2. Membership of The [ ] Sub-Group

Analysis

a) Jardines de Chia

The CIT agreed that the Department properly found MG Consultores related to Jardines de Chia under section 771(13)(D) of the Act. Slip Op. 97-120 at 15. The CIT also agreed that MG Consultores has six subsidiaries, each of which is related to one another pursuant to the same statutory provision. Given the extent of the vertical (parent-subsidiary) and horizontal relationship between and among MG Consultores and these six subsidiaries, the Department has determined that they should be collapsed and treated as a single entity for purposes of examining relationships with other parties under section 771(13).(2)

Regarding Chia, we respectfully disagree with the CIT that no person or persons owns directly or indirectly at least 20 percent of Chia and the other members of the [ ] sub-group. It is true, as the CIT stated, that "[ ] indirect ownership interests in the other [ ] sub-group companies, through this interest in MG Consultores, are less than 20 percent." Id. However, [ ] owns, directly or indirectly, greater than 20 percent of Chia and each of the other members of the [ ] sub-group.

We first note that we based our original related-party and collapsing analyses in large part upon the existence of "familial relationships" between the various Queen's Flowers Group companies, rather than purely upon the holdings of individuals such as [ ]. See, e.g., February Memo at 10. As we stated, "[o]ur conclusion is that this pattern is indicative of a coordinated effort and familial control over the two [ ]." Id.; see also id. at 3 (describing the "direct" holdings of [ ] in MG and their "indirect" holdings in MG's subsidiaries). Additionally, the Department based its analysis of the [ ] sub-group under section 771(13)(D) largely upon the joint holdings of the [ ] brothers, [ ]. Id. at 4.

The broad language and intent of subsection (D), and section 771(13) as a whole, fully support this family-based approach. Subsection (D) provides for "sister company" relationships. Under this provision, parties are related if "any person or persons" own or control, "jointly or severally, directly or indirectly, through stock ownership or control or otherwise, . . . in the aggregate 20 percent or more of the voting power or control in the business" of each party. Section 771(13)(D) of the Act (emphasis added). Thus, by its express terms, the statute permits the Department to evaluate, in the aggregate, the ownership or control of two or more parties by one or more "persons," i.e., a family or other group of shareholders.(3)

The CIT in Queen's endorsed this family-based approach to section 771(13)(D) as well. In upholding the Department's determination that the [ ] sub-group companies were related, the CIT reasoned that the Department's "decision to aggregate the interests of two brothers, both producing the same product, with common interests in several businesses, was a reasonable application of the statute." Slip Op. 97-120 at 20. The same reasoning holds for the [ ] family and the relationship between MG's subsidiaries and Jardines de Chia.

(i) Stock-Ownership

During each of the three review periods, [ ] owned [ ] percent of MG Consultores in equal [ ] percent shares. Indeed, MG described itself as a "family enterprise." See MG Consultores' section A responses for each POR. As demonstrated above, MG owned equity interests of greater than 20 percent in each of its six subsidiaries. Accordingly, through their [ ] percent ownership of MG, [the Mojica family] owned "in the aggregate" between [ ] and [ ] percent in each of MG's subsidiaries. The [ ] also owned [ ] percent of Jardines de Chia, all in the name of [ ]. Because the [ ] owned at least 20 percent in MG and each of these six subsidiaries, as well as in Chia, we find that Chia and MG's subsidiaries are all related pursuant to section 771(13)(D).

(ii) Non-financial Factors

This overlapping [ ] stock ownership is alone sufficient to find the MG subsidiaries related to Chia. However, in our draft results we gave further support for our decision under subsection (D) by evidence in the record relating to non-financial factors which established common control. Draft Results at 8-11.(4) Although we continue to take the position that non-financial factors are reasonable indicia of relatedness under section 771(13), we have not relied upon such evidence in these final results.

b) Flores Calima

The CIT found that Flores Calima is not related to MG Consultores pursuant to section 771(13)(D), Slip Op. 97-120 at 16 n. 5, and that the Department had not explained its determination that Calima is related to the remaining members of the [ ] sub-group. Id. at 16.

As a preliminary matter, in reexamining the relationship of Calima to MG Consultores and the other companies in the 7th POR (Calima did not exist in the 5th and 6th PORs), we have determined that the record contains contradictory information. According to MG Consultores' section A response for the 7th POR, MG owns a [ ] percent interest in Calima. See May 16, 1994 section A response, at 3. However, Calima's section A response, dated June 13, 1995, states at page 8 that [ ] owns a [ ] percent interest in Calima. By adding up the remaining interests in Calima, we conclude that MG and [ ] cannot both own [ ] percent of Calima. Thus, one of the companies erred in identifying the holder of this [ ] percent interest.

In our draft results we noted that, to resolve this issue, it was not necessary to gather additional information or to determine which of the two original responses was correct, whether that of Calima or MG Consultores.(5) In our draft results we explained that Calima is related to the remaining [ ] sub-group companies during the 7th POR, regardless of whether the [ ] percent interest in Calima is held by [ ] or MG Consultores. In reaching this conclusion, we also relied upon record evidence of certain non-financial factors to establish common control between Calima and the other members of the [ ] sub-group through [ ]. Draft Results at 13.(6)

However, in these final results, we have determined that, while Calima is a member of the [ ] sub-group, there is insufficient record evidence to find Calima related to other members of the [ ] sub-group under section 771(13)(D) through the common control of [ ].

Comments Regarding the [ ] Sub-Group

The respondents argue that the Department, in continuing to find Chia and Calima related to other members of the [ ] sub-group in its draft remand results, disregarded the CIT's rulings on the issues. The respondents contend that the CIT ruled expressly that Jardines de Chia and Calima were not related to other members of the [ ] sub-group.

The respondents dispute the Department's use of aggregated equity interests of [ ] in Chia, Calima and members of the [ ] sub-group. The respondents argue that section 771(13) of the Act neither defines family members as related parties nor defines a family as a legal "person." Because Congress explicitly defined family members as related parties under section 773(b)(4) but omitted this definition in section 771(13), the respondents contend that the Department is precluded from adopting such an interpretation, citing Ad Hoc Committee of AZ-NM-TX-FL Producers of Gray Portland Cement v. United States, 13 F.3d 398, 401 (Fed. Cir. 1994), cert. denied 115 S. Ct. 67 (1994) ("Ad Hoc Committee"). Furthermore, the respondents argue that the Department's regulations at section 353.2(p) do not define the term "person" to include a family. Thus, the respondents contend that the Department's analysis has no basis in the statute or regulations.

In their comments on the draft remand results, the respondents confirm that the correct identity of the owner of certain shares of Calima was [ ], but they argue that the Department's analysis does not address the CIT's concern on this issue. The respondents claim that there is no statutory basis for finding Calima related to other members of the [ ] sub-group. The respondents argue that the Department's analysis, which relies on the finding of the [ ] sub-group as a "single entity," is unsupported by factual evidence. Because there are no persons who own both 20 percent of the shares of other members of the [ ] sub-group and Calima, the respondents assert, the statutory test is not met.

The respondents also claim that, in relying upon non-financial factors, the Department misinterpreted section 771(13)(D) of the Act. The respondents contend that, for Chia and Calima to be related to other members of the [ ] sub-group under this provision, there must be a person or persons that own or control 20 percent of the voting power or control of Chia and Calima and 20 percent or more of the voting power or control of each one of the other members of the [ ] sub-group.

Respondents argue that the Department, in conducting the related-party analysis, has relied upon non-financial factors which are inconsistent with the statute. The statute, respondents contend, speaks exclusively of "voting power or control in the business." Respondents argue that this voting power or control is exclusively that of the shareholders which appoint the board members, rather than the board members themselves. For this reason, respondents contend, the Department cannot legally find ownership or control by virtue of membership of the Board of Directors independent of share ownership. Respondents cite the definition of affiliated party included in the Statement of Administrative Action (SAA) at 838, as exemplary of Congress' understanding that control under the statute prior to the Uruguay Round Agreements Act (URAA) is dictated by stock ownership. Furthermore, respondents disagree with the Department's assertion that section 771(13) of the Act "is phrased broadly to allow the Department flexibility in defining relationships" and, instead, assert that the CIT has held that the statute contains "bright line" tests, citing Torrington Co. v. United States, 1997 WL 108993, at *9 (Ct. Int'l Trade Mar. 7, 1997). Finally, according to respondents, the CIT noted that the Department's reliance upon non-financial factors was inconsistent with its previous determinations in Fresh Cut Roses From Ecuador and Disposable Pocket Lighters From Thailand to rely exclusively upon ownership interests in determining whether parties were related for purposes of a collapsing analysis. Respondents contend that, while the CIT ordered the Department to either conform to these previous decisions or explain the basis for its change in a manner that is in accordance with law and supported by substantial evidence, the Department has not sufficiently explained its reliance upon non-financial factors.

The respondents assert that the Department's analysis, which relies on the fact that Chia and MG Consultores each own certain interests in a third firm and that Chia and other companies own certain interests of a U.S. importer, does not satisfy the requirements of section 771(13)(D). The respondents claim further that, because the level of ownership in these instances is well below the 20-percent threshold, these facts are irrelevant under section 771(13)(D).

In analyzing the relatedness of Chia and Calima, the respondents assert, the Department also relied incorrectly upon the voting power of the Board of Directors. The respondents argue that the statute expressly requires 20 percent or more voting power or control in the companies. Here, according to the respondents, control is meant to be the power to appoint board members, not the control vested in the board members themselves. The respondents assert that the record evidence demonstrates that voting power is proportionate to shareholdings and not the number of Board members. Citing Sugiyama Chain Co. v. United States, 852 F. Supp. 1103, 1112 (CIT 1994)(Sugiyama), the respondents argue that the issue under the statute is not who the board members are but, rather, who has the power to appoint the Board members.

In rebuttal, the petitioner asserts that the Department's related-party analysis is responsive to the CIT's concerns. The petitioner argues that the Department has a legal basis for finding that family members are related parties within the statute and, therefore, the Department was correct in aggregating the interests of those family members when determining relatedness.

Regarding the Department's consideration of family members as a "person" under section 771(13)(D) and the aggregation of the family members' ownership interests, the petitioner argues that the CIT, in this case, approved specifically the Department's decision to aggregate the interests of two brothers both producing the same product with common interests in several businesses. The petitioner cites the use of the word "persons" in section 771(13)(D) as evidence that the statute directs the Department to contemplate common ownership by a group of persons when examining relatedness. The petitioner argues further that, although the statute does not define the word "person," the regulations define "person" as including "any interested party as well as any individual, enterprise, or entity, as appropriate." The petitioner asserts that nothing in the statute precludes this interpretation of the term "person or persons" and, therefore, the Department's interpretations of the statute and its own regulations are entitled to great deference.

Furthermore, the petitioner claims, the respondents were on notice that the Department would consider familial relationships under section 773(e)(4)(A) for purposes of determining whether to rely on sales prices of inputs to calculate constructed value. The petitioner reasons that the Department's decision to recognize family relationships in applying section 771(13) is consistent with the statutory scheme and the Department's responsibility to prevent circumvention of antidumping orders. The petitioner argues that the Department cannot ignore potentially manipulative family relationships which could allow respondents to avoid the antidumping duty order. The petitioner cites the Department's recognition of this "coordinated effort and familial control" in the February Memo.

The petitioner also claims that the Department properly considered non-financial factors in its related-party analysis. The petitioner contends that the statutory 20-percent threshold is not controlling because the non-financial-factors analysis is derived from the Department's practice rather than from the statute. The petitioner argues that the Department's analysis is reasonable because the Department properly considered [ ] in Jardines de Chia and Calima. The petitioner contends that this analysis is supported by record evidence demonstrating that the boards of directors have responsibility for the management of the farms.

Finally, the petitioner contends that the Department has given sufficient explanation for the change in practice because section 771(13)(D) permits a finding of indirect ownership or control. The petitioner argues that this flexibility to change its position with sufficient explanation is permitted to the Department. For these reasons, the petitioner argues that, given the facts of the case, the Department's interpretation of the statute was reasonable and has been affirmed by the CIT in Sugiyama.

Analysis with Regard to Comments on [ ] Sub-Group

We first note that the CIT's ruling that Jardines de Chia is not related to the [ ] sub-group under section 771(13) of the Act was based upon a erroneous conclusion that no person or persons, indirectly or directly, owned 20 percent or more of interests in both Chia and the other members of the [ ] Group. While true that [ ] individual equity interest in Chia is not sufficient to find Chia related to the other members of the [ ] Group, the record evidence demonstrates that the [ ] family owns [ ] percent of Chia. Through their [ ] percent ownership in MG Consultores, the [ ] family owns an aggregate interest between [ ] and [ ] percent in each of MG's subsidiaries. The aggregate interests held by the [ ] family in Chia and other members of the [ ] sub-group establish a sister-company relationship between Chia and these members under section 771(13)(D) of the Act. For this reason, when as instructed by the CIT, we reconsidered our related-party analysis, we clarified these facts in the draft remand for purposes of the CIT's review.

Our redetermination that Chia is related to other members of the [ ] Group is based upon our analysis of the aggregate equity interests held by the [ ] family in each member of the [ ] Group. As explained in the draft remand results, we interpret the statutory language "own or control in the aggregate 20 percent or more" as authorizing us to evaluate ownership or control held by "any person or persons" on an aggregate basis for purposes of identifying related parties under section 771(13)(D) of the Act. In this case, similar to our analysis of the [ ] Group, we analyzed the aggregate interests held by members of the [ ] family as "persons" within the meaning of section 771(13)(D). We also noted that the CIT explicitly endorsed this family-based approach as a reasonable interpretation of the statute. See Slip Op. 97-120 at 20.

Furthermore, by its express terms, sub-section (D) permits the Department, in performing this aggregation, to include ownership interest of the entire group of investors held either "jointly or severally" (emphasis added). We interpret this to mean that, since we consider the [ ] family as "persons" under section 771(13)(D), the interests held in the name of individuals within the group are considered to be interests of the group as a whole. The "persons" do not have to hold an ownership interest jointly for the Department to view that interest as belonging to the investment group considered as "persons" within the meaning of subsection (D). The CIT's acceptance of the aggregation of the interests of [ ] family members accepts our inclusion of interests owned "severally" since, in our Final Results, we aggregated interests of the [ ] family even in companies where only one member of the family held an interest.(7)

The same reasoning applies to the interests held severally by the [ ] family and the relationship between MG Consultores' subsidiaries and Jardines de Chia. To establish common ownership between Jardines de Chia and the other members of the Mojica sub-group in our draft results, we relied upon the fact that [ ] owned [ ] percent of Jardines de Chia, all in the name of [ ]. Though not held jointly, the statute clearly permits, and the CIT has accepted, the addition of these shares to the other interests of the [ ] family. Because the [ ] owned at least 20 percent in MG Consultores and each of these six subsidiaries, as well as in Jardines de Chia, we continue to find that Jardines de Chia, MG Consultores, and MG Consultores' six subsidiaries are all related pursuant to section 771(13)(D).

This analysis is not flawed by the lack of an explicit directive in section 771(13)(D) of the Act that defines family members as related parties because there is no statutory requirement that the holders of the equity or control interests be related. Likewise, there is no statutory requirement that restricts an analysis of "any person or persons" to legal persons as argued by the respondents. Rather, section 771(13)(D) of the Act casts a broad net over the identity of the owners, e.g., "any person or persons," which reflects the intent of Congress to capture a broad variety of complex business relationships under the antidumping statute. See, e.g., Emergency Tariff and Antidumping: Hearing on H.R. 2435 Before the Senate Committee on Finance, 67th Congress, 1st Session, 13-14 (1921). Moreover, the definition of family members as related parties in section 773(b)(4) of the Act does not preclude our interpretation that family members may be analyzed reasonably as "persons" under section 771(13). The two statutory definitions of related parties serve two distinct purposes - one for identifying related parties for purposes of U.S. and home market sales (section 771(13)) and the other for examining transactions between related parties for purposes of constructed value (section 773(b)(4)). As such, they are not analogous to the functionally equivalent statutory provisions that the Court of Appeals for the Federal Circuit (Federal Circuit) reviewed in the Ad Hoc Committee case, which reversed the Department's interpretation of the statute to authorize the deduction of pre-sale transportation expenses from foreign market value.

Respondents have misconstrued our analysis as impermissibly injecting a "family test" in the definition of related parties under section 771(13). However, as detailed in the draft remand results, we examined the aggregate interests held by "persons" in each member of the [ ] sub-group. The equity interests held in each member company by members of the same family demonstrate the closely held nature of these corporate entities. The fact that these "persons" are members of the same family only strengthens the reasonableness of our analysis and demonstrates the degree of inter-relatedness within the sub-group.

In sum, the record evidence supports our finding that Chia is related to the other members of the [ ] sub-group under section 771(13)(D) of the Act based upon financial factors.

We have also re-evaluated the record evidence in light of comments we received from both the respondents and the petitioner and find that Calima is not related to the other members of the [ ] sub-group through the ownership or control of the [ ] family under section 771(13). Although the record evidence demonstrates overlapping stock-ownership held by the [ ] family in Calima and the other members of the [ ] sub-group, this overlapping stock-ownership alone is not sufficient to find Calima related to [ ] pursuant to section 771(13)(D). Because we have not found Calima to be related to other members of the [ ] sub-group through the common control of [ ], the issue regarding our reliance upon non-financial factors in its draft results is moot.

3. The Relationship Between the [ ] and [ ] Sub-groups

Analysis

The CIT ordered the Department to explain how its factual findings support determining that the [ ] and [ ] sub-groups are related and to identify the relevant provisions of section 771(13) of the Act. In accordance with the CIT's order, we analyzed the relationship between the two sub-groups further. Our determination remains the same. The [ ] and [ ] sub-groups are related pursuant to section 771(13)(D). In reaching this determination, we have considered both the level of direct overlapping ownership in the aggregate as well as indirect ownership or control. Moreover, we have made a separate determination for each POR.

Given that, in the previous section of this redetermination, we have changed our finding regarding the membership of the [ ] sub-group by finding that Calima is not a member of that sub-group, we have analyzed the relatedness of the two sub-groups in consideration of this new finding.(8)

To better highlight the level of overlapping ownership interests outlined in the February Memo, we recalculated the ownership structure for the companies by multiplying the paid-in capital for each firm by the stockholders' percentage ownership.(9) Second, we aggregated each individual's resulting peso investment in each firm. We also aggregated the total paid-in capital for each sub-group as a whole. Third, we divided each owner's peso-valued equity in a sub-group by the total equity for each sub-group to determine the percentage ownership for each individual in that sub-group.

In our draft remand results, to determine the pool of common owners whose common interests we would aggregate under section 771(13(D), we eliminated the ownership of individuals who held stock in only one of the sub-groups, leaving only a group of common investors. Finally, we aggregated the remaining ownership percentages of these common investors.

We have recalculated these overlapping ownership percentages for these final results of redetermination in order to reflect the revised sub-group memberships we discussed above (i.e., we have eliminated Flores Calima from the [ ] sub-group). For the 5th POR, the same [ ] individuals owned [ ] percent of the [ ] sub-group and [ ] percent of the [ ] sub-group. For the 6th POR, the same [ ] individuals owned [ ] percent of the [ ] sub-group and [ ] percent of the [ ] sub-group. For the 7th POR, the same [ ](10) individuals owned [ ] percent of the [ ] sub-group and [ ] percent of the [ ] sub-group.(11) We note that the percentage of common ownership during each POR greatly exceeds the 20-percent threshold required by section 771(13)(D) of the Act, i.e., a group of person owns or controls, in the aggregate, at least 20 percent of each sub-group. See revised and expanded "Master Chart" for the 5th, 6th, and 7th PORs (Exhibits 1, 2, and 3). Thus, the record supports our determination that the two sub-groups are related under section 771(13)(D).

However, in the draft results we did not articulate clearly our rationale for aggregating the interests of these [ ] individuals as "persons" under section 771(13)(D). For these final results of review, we have revised this aggregation of ownership interests in order to clarify the standard by which we have defined "persons" under sub-section (D) for purposes of our related-party analysis.

We based our original related-party and collapsing analysis in large part upon the existence of a pattern of common investments among family members and other individuals in the various Queen's Flowers Group companies, rather than purely upon the holdings of individuals. See February Memo at 5-6, n.1. There we laid out the repeated pattern of joint investment by a group of individuals. As we stated on page 10, "[o]ur conclusion is that this pattern is indicative of a coordinated effort and familial control over the two [sub-groups]."

The broad language and intent of subsection (D), and section 771(13) as a whole, support this family-based approach fully. Subsection (D) provides for "sister-company" relationships. Under this provision, parties are related if "any person or persons" own or control, "jointly or severally, directly or indirectly, through stock ownership or control or otherwise, . . . in the aggregate 20 percent or more of the voting power or control in the business" of each party. Section 771(13)(D) of the Act (emphasis added). Thus, by its express terms, the statute permits the Department to evaluate, in the aggregate, the ownership or control of two or more parties by one or more "persons," i.e., a family or other group of shareholders.(12)

As stated above in reference to our discussion concerning Jardines de Chia, the CIT in Queen's endorsed our aggregation of family interests under section 771(13)(D) as well. In upholding the Department's determination that the [ ] sub-group companies were related, the CIT reasoned that the Department's "decision to aggregate the interests of two brothers, both producing the same product, with common interests in several businesses, was a reasonable application of the statute." Slip Op. 97-120 at 20. The same principal holds true for a group of investors who share common interests in several firms despite the fact that some of the individual investors are not members of the same family. The CIT has recognized this methodology in the case where the ownership interests of several, otherwise unrelated, persons have a pattern of common interests in several firms. See Asociacion Colombiana de Exportadores de Flores, et al. v. United States, Consol. Court No. 96-09-02209, (Slip. Op. 98-33 at 63 (CIT Mar. 25, 1998) "Asocoflores") . In this case, the Department demonstrated a clear pattern of joint investment by a group of individuals, a pattern which illustrates that, where a member of the group invested individually, this individual interest reflects the interest of the investment group.

Based on this framework of analysis, we continue to find that, in each POR, the two sub-groups are related under section 771(13)(D). Furthermore, each member of the Queen's Flowers Group(13) is related to every other member of the Group under section 771(13)(D) through the same common owner. This common owner consists of a group of investors whose ownership interests in these firms can be aggregated reasonably as "persons" under 771(13)(D) due to familial relationships and a consistent pattern of common interests in several firms. This investment group consists of the following persons: [ ], [ ], [ ] and [ ].(14) We first note the common interests of [ ] as well as the common interests of [ ]. As we explained above in reference to our analysis involving Jardines de Chia, in both cases, the familial relationships demonstrate that each family should be considered as a single unit for purposes of examining their ownership interests. Therefore, in examining the pattern of common interests in the Group's members, we are essentially analyzing the interests of four persons: [ ], [ ], [ ], and [ ], hereinafter referred to as the [ ] investor group.

As described earlier, for the draft results we recalculated the ownership structure for each of the 20 members of the Queen's Flowers Group, the two U.S. importers and Grupo Chia, thereby determining the ultimate individual ownership and amount of paid-in capital for each company involved in these three reviews. This ownership information demonstrates the pattern of common interests among these four persons and highlights the level of overlapping ownership interests outlined in the February Memo more clearly.

The evidence on the record regarding the ownership, both direct and indirect, of the firms involved in these reviews demonstrates that these four persons invest consistently in a pattern which justifies the aggregation of their individual ownership interests. See "Pattern of Common Interests Chart" for the 5th, 6th, and 7th PORs (Exhibits 4, 5, and 6).(15) [ ], [ ], [ ], and [ ] each has an interest in Grupo Chia, Queen's Flowers Corporation, Golden Flowers and Queen's Flowers de Colombia during all three PORs.(16) The four investors added to their joint investment in the 7th POR by investing in a new company, Calima, in which each of the four took a financial interest.

The importance of these five firms in which each of the four investors has a common interest demonstrates the strength of this partnership. Grupo Chia is a real-estate company which provides office space and/or administrative services to several members of the Queen's Group. Queen's Flowers Corporation and Golden Flowers are two U.S. importers to which approximately [ ] percent of the U.S. sales of the Group were directed during the three PORs.(17) Queen's Flowers de Colombia is a large Colombian exporter of subject merchandise.(18) Finally, Calima, as a new company, represents a joint decision by the four persons to invest in a new producer of subject merchandise which exported a significant volume of subject merchandise in its first year of existence (the 7th POR).(19)

The common interests of these four investors are not limited to the above companies. Overall and in various combinations of partnership, two or three of the four investors have common interests in eight companies that exported subject merchandise in one or all of the PORs and four other producers of non-subject flowers. For example, [ ], [ ], and [ ] wholly own Jardines de Chia, an exporter of subject merchandise in all three PORs, each having [ ]-percent ownership. Additionally, [ ] and [ ] each has a significant interest in Riofrio, a company which exported the second-largest volume of subject merchandise after Queen's de Colombia for the three PORs.

Furthermore, where the [ ] and [ ] families do not invest together, each frequently has common interests with the other partners. [ ], [ ], and [ ] each has an interest in Tandil, a producer of subject merchandise, and Bojaca, a producer of non-subject flowers. [ ] and [ ] each has an interest in Mana, Cipres and Roble, three exporters of subject merchandise during one or all of the three PORs, and Fredonia, a producer of non-subject flowers during the PORs. [ ] and [ ] each has an interest in Aljibe, a producer of non-subject flowers during the PORs. [ ] also has several partnerships with [ ], each with an interest in three companies: Hato and Valvanera, exporters of subject merchandise during each of the PORs, and Cacique, a producer of non-subject flowers. These examples demonstrate that the common interests of the four go beyond merely the five firms in which each of the four invests. Partnerships among two or three of the four investors are repeated and various, thereby demonstrating that, where all four do not invest, there remains a pattern of common interests.

In light of this pattern of common interests, we find that the interests of the individual members of this investment group should be aggregated as "persons" under section 771(13)(D). Therefore where these persons invest "severally", we should consider those interests as the interests of the investment group as a whole.

To test the relatedness between the two sub-groups, we aggregated the interests held in each sub-group by this common investor group. We aggregated the collective interest in pesos of the four persons in each of the two sub-groups. We also aggregated the total paid-in capital for each sub-group as a whole. After we divided the collective interest of the investor group by the paid-in capital for the group as a whole, it is evident that, together, these four persons own directly or indirectly, in the aggregate, well over 20 percent of both the [ ] sub-group and the [ ] sub-group. The four members of the investments group own collectively [ ] percent and [ ] percent of the [ ] and [ ] sub-groups, respectively, for the 5th POR, [ ] percent and [ ] percent of the [ ] and [ ] sub-groups, respectively, for the 6th POR, and [ ] percent and [ ] percent of the [ ] and [ ] sub-groups, respectively, for the 7th POR. The percentage of common ownership during each POR greatly exceeds the 20-percent threshold required by section 771(13)(D) of the Act. See revised and expanded "Partnership Chart" for the 5th, 6th, and 7th PORs (Exhibits 1, 2, and 3). Thus, the record supports our determination that the two sub-groups are related under section 771(13)(D). Further, the record evidence demonstrates that the investor group owns collectively between 28 percent and 100 percent of each of the 20 companies, thereby establishing a sister-company relationship under 771(13)(D) among each and every member of the Queen's Flowers Group.

Comments Regarding the Relatedness of the Two Sub-Groups

The respondents argue that the Department has collapsed the [ ] and [ ] sub-groups improperly because not all 20 companies are related. The respondents claim that, while the Department has analyzed the [ ] and [ ] sub-groups correctly to determine whether the members of each group are related to other members of the same group, it has collapsed the two groups together without determining whether all members of both groups are related to all other members of the Queen's Group. The respondents argue that collapsing unrelated companies is inconsistent with the Department's new regulations at section 351.401(f)(1997), which provides for the treatment of two or more related parties as a single entity. The respondents claim that, because the stated practice of the Department is not to collapse related parties except in relatively unusual situations where the type and degree of relationship is so significant that there is a strong possibility of price manipulation, then ipso facto there cannot be such a situation among unrelated parties, citing Nihon Cement v. United States, 17 CIT 400, 426-27 (1993), and Antifriction Bearings and Parts Thereof From Germany, 54 FR 18992, 19069 (1989) (comment 25).

The respondents contend that the Department has constructed a fictional "enterprise" improperly to support its collapsing analysis which is inconsistent with the statute and factual evidence. The respondents claim that there is no legal or operational "person" corresponding to these "enterprises," as required by section 771(13)(D) of the Act. The respondents argue further that, because these enterprises do not exist, there is no "voting power or control" in any [ ] or [ ] "enterprise" to allow for a related-party analysis. The respondents assert that, while the Department in its draft results has attempted to analyze the ownership of each of these enterprises according to the proportion of paid-in capital for each member, using these ownership percentages to find common ownership or control under section 771(13)(D) is impossible because one cannot perform this test on a fictional "enterprise." Instead, the respondents suggest, the Department must find common ownership or control of legally established entities -- in this case, the 20 companies under examination in this remand. Furthermore, the respondents assert that the Department has resorted to the creation of these "enterprises" because the 20 producers involved are not all related according to the statute or past precedent. Rather, the respondents assert, many of the companies in the [ ] sub-group are not related under any statutory provision to any companies in the [ ] sub-group.

The petitioner responds that the Department's "enterprise" theory is in accordance with law. The petitioner contends that neither the statute nor the regulations preclude the Department from collapsing related companies on the basis of an "enterprise" theory. The petitioner observes that the Department's analysis relies upon the regulatory definition of "person" which includes enterprise, an undefined term. The petitioner contends that the respondents have cited to no judicial or administrative precedent to support their claim that a "person" must be an individual company and that the Department's interpretation of the statute is controlling.

Analysis with Regard to Comments on Relatedness of Two Sub-Groups

After considering the comments by both the petitioner and the respondents, we continue to find that the [ ] and [ ] sub-groups are related under section 771(13)(D). Furthermore, we have addressed the respondents' concern that we have "resorted" to the creation of fictional enterprises because we are unable to demonstrate that the 20 companies are all related to each other. By examining the common ownership or control in each of the 20 companies, we also demonstrate that each of the individual members of the Queen's Flowers Group is related to each of the other individual companies of the Group under section 771(13)(D).

As discussed above, we continue to find the [ ] and [ ] sub-groups related under section 771(13)(D) by virtue of sister-company relationships established through aggregated interests of a common investor group, namely the [ ], the [ ], [ ], and [ ]. The record evidence demonstrates that this investor group's aggregated interests in the [ ] and [ ] sub-groups exceed 20 percent in each POR. We consider the [ ] and [ ] sub-groups as two enterprises for purposes of this analysis. As explained above, based upon the record evidence, we found seven companies, the [ ] sub-group, to be sufficiently related to treat them as a single entity under our collapsing practice. Likewise, we conducted a similar analysis of the 14 companies comprising the [ ] sub-group. When analyzing the relationship of these two sub-groups, each sub-group constitutes a "person" within the meaning of 19 C.F.R. 353.2(p). This analysis is consistent with the Department's authority under the statute to ignore the otherwise separate legal existence of certain companies. See Queen's at 7-9. Once we determined that each sub-group constitutes a single enterprise, we reasonably considered the relatedness of these two "persons" under section 771(13) of the Act. Neither the statute nor the regulations restrict the Department's analysis under section 771(13)(D) to "legal or operational" persons, as respondents contend. Rather, consistent with the regulatory definition of "persons" and our collapsing authority, we have examined the relatedness of these two sub-groups as "persons" within the meaning of section 771(13)(D).

We disagree with the respondents that we have collapsed the Queen's Flower's Group using two fictional enterprises. In our final results, it is clear that we first collapsed each sub-group, thereby finding that each is a single entity. We have simply chosen to call each of these "single entities" an enterprise. Because we first found that each sub-group should be collapsed before we looked at the relatedness of the two sub-groups, each of the two sub-groups under consideration is a statutory "person" under section 771(13)(D).

Issue 2: Whether these 20 Firms Should be Collapsed Into a Single Entity

1. Background

For the Final Results, we collapsed each sub-group into a single entity. We also collapsed the two sub-groups into a single entity, the Queen's Flowers Group. In the Final Results, we stated that we make three inquiries in determining whether to collapse companies. First, we examine whether the firms are related. Second, we examine whether the firms have similar production facilities. Finally, we examine whether there exists other evidence indicating a significant potential for price or production manipulation. In determining whether there is a significant potential for the manipulation of prices or production, the types of evidence we examine include: 1) the level of common ownership; 2) the existence of interlocking officers or directors; and 3) the existence of intertwined operations. See Nihon Cement Co., Ltd. v. United States, 17 CIT 400, 425-26 (1993) (Nihon).

The CIT in Queen's noted that, although the collapsing factors listed in the Final Results were not identical to those listed in Nihon,(20) they represented a reasonable set of inquiries and, therefore, are in accordance with law. Queen's, Slip Op. 97-120 at 23. However, the CIT stated that the factors identified in the Final Results differed from those in the November Memo. The CIT ordered us to explain the set of factors upon which we relied upon in our collapsing analysis of the Queen's Flowers Group. Slip Op. 97-120 at 24. In addition, the CIT ordered us to reconsider our determination to collapse these 20 firms in light of reconsidering the relatedness issue under section 771(13). Slip Op. 97-120 at 29.

Analysis

2. The Appropriate Collapsing Criteria

The factors upon which we relied in the November Memo were very similar to those we identified in the Final Results.(21) However, the Department's collapsing analysis evolved somewhat through practice and judicial rulings from the time of the November Memo (1994) until the Final Results (1996). For instance, the November Memo referred only briefly to the threshold requirement that parties must be found to be related in order to be collapsed and this memorandum was based upon limited data provided in the responses of the original eight companies that the Department considered part of the Queen's Flowers Group. The Department later clarified, in the Final Results, that parties must be found related under section 771(13) to be collapsed. Thus, although the analyses contained in the November Memo and the Department's other memoranda remain valid, we relied upon the factors listed in the Final Results in collapsing the 20 related firms in the Queen's Flowers Group for the Final Results.(22)

3. Application of the Collapsing Criteria

Regarding the first factor in our collapsing analysis, our determination that the firms are related during each POR within the meaning of section 771(13)(D) is set forth fully above in our related-party analysis.

We examined the second factor listed in the Final Results, similar production facilities, initially in the November Memo, finding that the seven original respondents did have similar production facilities due to the production of similar flowers types and the existence of inter-company transactions. We updated our analysis of this factor in the February Memo (at page 10) to explain that members of the Group which did not produce subject flowers during one or more of the PORs were also considered to have similar production facilities due to the similarity of production processes across subject and non-subject flower types, the low cost of clearing non-subject flowers from greenhouses and replanting them with subject flowers, and the commonly owned sales infrastructure. Furthermore, the February Memo also accounts for comments raised by the Queen's Flowers Group after issuance of Certain Fresh Cut Flowers from Colombia; Preliminary Results of Antidumping Duty Administrative Reviews, 60 FR 30270 (June 8, 1995) (Preliminary Results). Except where indicated otherwise in those memoranda, we continue to find that the various firms all had similar production facilities such that retooling would not have been necessary during any review period covered by the Final Results.

Finally, regarding whether there is a significant potential for the manipulation of prices or production, the third factor, we indicated in the Final Results that the type of evidence we examine includes: 1) the level of common ownership; 2) the existence of interlocking officers or directors; and 3) the existence of intertwined operations. We examined the level of common ownership on pages five and six of the February Memo. We have also explained further this common ownership analysis above in our redetermination regarding the relatedness of the Group members for each of the three PORs.

We also examined the existence of interlocking officers and directors on pages seven and eight of the February Memo. Although we may not have stated it clearly in the narrative, in a footnote we explained that we were analyzing the common ownership and interlocking boards for the purpose of collapsing as well as determining whether the parties were related:

Although a collapsing factor, interlocking boards is also indicative of relationship, much like ownership, which is considered both in evaluating relatedness and in collapsing. In the one instance we look for the existence of relationships and in the other, the nature and extent of that relationship.

February Memo at 7 n. 2.

Thus, we considered the same ownership and management information in determining that the firms were related and in determining that these related firms should be treated as a single entity.(23)

Finally, we determined that there were intertwined operations between the various companies. We first made this determination in the November Memo. We reexamined the existence of intertwined operations on pages eight through ten of the February Memo. For this latter inquiry, we reexamined inter-company sales, joint U.S. sales efforts, and the sharing of office space and costs. We also examined new evidence of intertwined operations such as joint ventures, related-party inputs, financial transactions, and the pattern of familial relationships. Having reconsidered our determination to collapse these related parties, we confirm that the relevant firms had a significant potential for the manipulation of price or production during all three review periods. As such, we also confirm that the relevant firms should be collapsed into a single entity during each POR.

Comments Regarding Collapsing of the Group

The respondents assert that the Department's collapsing analysis is not in accordance with law and is not supported by substantial evidence. The respondents claim that the Department did not perform this collapsing analysis separately for each period of review and separately for each company as required by law. While the Department in its draft results did analyze the general related-party issue separately for each review period, the respondents argue, it has not applied the collapsing factors separately to each period and separately to each producer, citing Certain Residential Door Locks and Parts Thereof From Taiwan, 54 FR 53153, 53161 (1989), and Coated Groundwood Paper From Finland, 56 FR 56363, 56369 (1991). Rather, the respondents contend, the Department determined whether each of the collapsing factors occurred within the group of 20 companies as a whole. The respondents assert that a company-specific analysis for each review period is necessary because the facts change from review to review. As an example, the respondents point out that the number of transactions between certain companies changed in each POR.

The respondents also assert that the value of these sales was de minimis in relation to their total sales during the period and were made at arm's length. The respondents argue that the Department's analysis fails to take these facts into consideration.

Furthermore, according to the respondents, the ostensible need to collapse the companies under a single cash deposit rate to avoid future price manipulation no longer exists. Because the CIT enjoined this single rate and the Department has completed the subsequent 95/96 review, the respondents claim, the results of this remand will never serve as a cash deposit rate.

The petitioner responds that, while the respondents request that the Department conduct its collapsing analysis separately for each POR and separately for each company, the Department is under no obligation to apply its collapsing analysis in such a manner since the respondents already raised this issue in their brief before the CIT and the CIT's remand instructions did not specify on which level of specificity the Department was to perform its collapsing analysis.

The petitioner contends that the respondents are incorrect in arguing that the need for the Department's collapsing analysis has expired because the results of the remand will never serve as a cash deposit rate. According to the petitioner, the CIT must consider whether the Department's collapsing analysis was appropriate at the time it was made. Furthermore, the decision regarding the collapsing analysis affects whether companies are collapsed in subsequent reviews.

Analysis Regarding Comments on Collapsing

We disagree with the respondents' claim that our collapsing analysis is not in accordance with law or unsupported by substantial evidence. As demonstrated in the February Memo, our collapsing analysis is based upon the record evidence for each POR, and we analyzed the collapsing factors both within each collapsed sub-group and between the two sub-groups. The record evidence demonstrates that collapsing factors such as common ownership and similar production facilities apply to the relationship of each individual member of the Group with all other members. Where the collapsing factor does not apply to each member of the Group, we have adequately explained on the record those instances where the collapsing factor applies to members within each collapsed sub-group and between the two sub-group enterprises. That is, as in the second step of our related-party analysis (relationship between the two-subgroups), we applied collapsing factors to each sub-group enterprise as the relevant "company" for purposes of this analysis because this approach is consistent with our findings that each sub-group warrants treatment as a single entity. We consider this approach to be a reasonable exercise of our discretion under our authority to collapse related producers.

Furthermore, the February Memo demonstrates that we have evaluated our collapsing analysis for each POR. As we have explained previously in this redetermination, we looked to common investors during each POR to determine whether the members of the Queen's Flowers Group which existed during each POR were related to each other. The record evidence establishes conclusively that, through the [ ] investment group, members of the [ ] and [ ] sub-groups were related to each other during each of the three PORs. The record evidence also establishes that the firms in existence during each POR and, therefore, the two sub-groups all had similar production facilities and could switch flower production easily without substantial retooling. Therefore, firms which were not currently producing subject merchandise could switch to the production of subject merchandise without much effort. Finally, the record evidence establishes that there was a significant potential for the manipulation of prices or production due to common owners, interlocking officers or directors and the interplay of operations of various firms during each POR.

As established above, evidence regarding common owners applies to all members of the Group in existence during each POR. Similarly, the evidence of a strong web of interlocking officers and directors, as discussed in the February Memo, demonstrates that members of each sub-group as well as the two sub-groups are linked during each POR. While respondents assert accurately that the number of flower or input transactions may have been small during any one POR, the record evidence illustrates that such transactions took place within each sub-group and between the two sub-groups. These transactions are indicia of the potential for price and cost manipulation within each sub-group and between the two sub-groups.(24) The volume and value of such transactions do not outweigh the other evidence of intertwined operations which support our finding of a significant potential for manipulation.

The alleged arm's-length nature of these transactions is irrelevant for considering whether a significant potential for price or production manipulation exists. In making this determination, we consider the existence of such transactions, not the actual prices paid between related producers; whether related-party transactions are at arm's length becomes relevant only in our calculation of the dumping margin for the collapsed entity.(25)

We also disagree that the rationale of treating the Group as a single entity is no longer relevant. Preventing anticircumvention of the antidumping order is merely one purpose of collapsing related producers. Our collapsing practice also ensures that we calculate accurate margins for each administrative segment. See Queen's at 8. Subsequent intervening reviews that establish new cash deposit rates do not render our determination of appropriate dumping margins during each POR moot. In accordance with section 751(a)(2) of the Act, our determination also serves the purpose of determining the amount of antidumping duties to assess on all entries to the United States exported by the respondents during the PORs.

For all the above reasons, we determine that our decision to collapse these companies into a single entity and to assign a single rate for each POR to the Group's members is consistent with the statute, judicial precedent and our practice.

Issue 3: Reconsider the Use of BIA

1. Background

In the Final Results, we determined that the Queen's Flowers Group (1) failed to identify related parties, (2) failed to respond to certain factual questions, and (3) improperly reported certain cost items and failed to change those items when requested to do so. We determined that the original responding companies(26) presented a pattern of incomplete responses, misleading information, and contradictory statements, which prevented us from identifying and gathering necessary sales and constructed value (CV) data from the entire group prior to the Preliminary Results. Without this data, we were unable to calculate dumping margins for the Group during the three relevant PORs. For these reasons, we concluded that the Group significantly impeded our reviews, and we resorted to margins based on BIA.

The CIT noted that, because the Department's BIA determination was, in part, based on its determination that the Queen's Flowers Group members were related and on the failure of the respondent members of the Group to provide related-party information, the Department must reconsider its decision to rely on BIA when reconsidering the related-party determinations. The CIT also instructed the Department to address the following issues which the CIT had identified in the preliminary-injunction ruling: 1) the CIT's observation that the Department's initial questionnaires did not request complete shareholder information for companies that did not produce the subject merchandise; 2) the CIT's observation that the Department had all related-party information by June 22, 1995, a year before the Final Results; 3) the apparent failure by the Department to send the initial questionnaire for the 7th POR to Cultivos Generales' correct address; and 4) whether the deficiencies in Flores del Hato's supplemental cost response which the Department cited as a reason for BIA may have resulted from the firm receiving a questionnaire meant for Agroindustrial del Riofrio.

Analysis

In accordance with the CIT's instructions, we have reconsidered our decision to resort to BIA for all three review periods. Based on our examination of the record evidence, we continue to find that the responding companies provided incomplete responses to our specific requests for information concerning the identity, ownership, and management of related producers, suppliers, and customers of the subject merchandise.

As our collapsing analysis demonstrates, the Queen's Flowers Group is a single entity composed of 20 companies with a shared business interest. The extent of the common ownership and management of the [ ] families alone is enough to raise at least a serious question that respondents chose to ignore our attempts to investigate this web of related flowers companies. Given the extent of common ownership by the [ ] investor group in each of the 20 members of the Group, it is reasonable to conclude that the original respondents were in control of the information that we were seeking. Yet the record evidence for the three reviews demonstrates that, at the outset of each review, the Queens Flower Group was not forthcoming in providing the factual information necessary to establish the existence of this entity. Instead, because the respondents chose to respond selectively to our requests for specific information and submitted deficient, confusing, and conflicting information, we were compelled to engage in a lengthy and tedious fact-finding exercise to piece together enough information to determine that the 20 corporations could be considered a single entity under a reasonable interpretation of the statute. The ultimate impact of this uncooperative behavior was that we were unable to collect complete U.S. sales and cost information necessary for the calculation of a dumping margin. This is the very sort of conduct by respondents that the BIA rule seeks to deter. See Rhone Poulenc, Inc. v. United States, 899 F.2d 1185, 1191 (Fed. Cir. 1990); Pistachio Group of Ass'n of Food Indus. v. United States, 641 F. Supp. 31, 39-40 (CIT 1987).

As detailed in our BIA memorandum dated June 28, 1996 (final BIA memorandum), the record for each review demonstrates that in our initial Section A questionnaires we requested each respondent specifically to identify related parties involved with the subject merchandise, to describe the nature of the relationship with the identified related parties, to report relevant stock-ownership information, and to identify common board members and management. In response to this question, the respondents should have provided all relevant information for companies involved in the subject merchandise related under section 771(13) of the Act for U.S. and home market sales transactions and related under section 773(e)(4) of the Act for elements of value included in the CV calculations.

Our instructions in this regard were clear and consistent with our standard practice. Four of the original respondent members had participated previously in the fourth annual review and were familiar with the Department's practice.(27) Of these four, Jardines de Chia, Queen's Flowers de Colombia, and Jardines Fredonia were collapsed and treated as a single entity for the 4th POR. Moreover, by the time the questionnaire responses were due for each review period, each of the respondents was represented by counsel with expertise in the antidumping law. In light of these factors, the failure to provide complete and accurate responses to our initial questionnaires may not reasonably be attributed to ignorance of the Department's practice.

In examining the data submitted by the Queen's Flowers Group members, we note that the most profound deficiency in their initial responses was the failure to identify related producers of subject merchandise. Three of the firms on which we initiated administrative reviews, Queen's Flowers De Colombia Ltda., Jardines De Chia Ltda., and Agrodindustrial del Rio Frio, had knowledge of six related producers of subject merchandise, Flores la Mana, Flores el Cipres, Flores el Tandil, Flores Atlas, Flores Calima, and Flores el Roble. Each of these six producers exported subject merchandise to the United States in at least one of the three PORs. The failure to provide that crucial information denied us the opportunity to develop a complete record of sales and CV data necessary for the calculation of an accurate dumping margin. The impact of these missing data grew with each succeeding POR. In the 5th POR the sales of the undisclosed firms represented 5.05 percent of the total U.S. sales volume of the Queen's Flowers Group; in the 6th POR those sales accounted for 17.41 percent of the total U.S. sales volume of the Group; in the 7th POR the undisclosed sales amount to 24.84 percent of the total U.S. sales volume of the Group(28). The record as it developed during the course of the review proceedings demonstrates that, had each respondent member provided complete and accurate responses to our initial questionnaires and to our July 22, 1994 supplemental questionnaires, we would have been able to analyze this information, initiate a collapsing analysis, and seek out complete responses from the Group prior to our Preliminary Results. However, the records demonstrate that the orderly course of our review proceedings was constrained because we were compelled repeatedly to seek additional information from the respondents that they should have disclosed previously(29). As analyzed in detail in the final BIA memorandum, these firms provided only partial and, at times, contradictory answers to our inquiries. In addition to the deficiencies on related parties there were numerous other deficiencies in factual information necessary for calculating CV. Therefore, we determine that the record evidence continues to support our determination that it is appropriate to assign margins based upon BIA to the Queen's Flowers Group for all three review periods.

In sum, our decision to resort to BIA for all three reviews is supported by record evidence showing that the respondents failed to provide complete and accurate responses to our requests for information.

2. Deficiencies Concerning Related-Party Information for Non-Producers of Subject Merchandise

The CIT noted that one of the primary deficiencies upon which the Department relied in its BIA analysis was the failure of the original respondent members to provide complete shareholder information for other related parties that did not produce the subject merchandise. The CIT reasoned that, because the Department asked for shareholder information only for companies that produced subject merchandise in the initial Section A questionnaires and in the July 22, 1994 supplemental questionnaire, the original respondents cannot be faulted for failing to provide information that the Department never requested. Moreover, the CIT observed that, when the Department did request complete shareholder information about the companies that did not produce the subject merchandise, the information was provided.

We agree with the CIT that the original respondents of the Queen's Flowers Group should not be faulted for failing to provide information that we did not request properly. We respectfully disagree, however, that our examination of the respondents' submissions contained such analytical errors. As explained in our final BIA memorandum, we re-analyzed our preliminary BIA analysis and limited our deficiency findings to those instances in which the respondents failed to specify:

. . . . (2) parties related under the Sales Price definition that were involved in sales of the subject merchandise; or (3) parties related under the CV Input definition that supplied production inputs or other elements of CV.

Final BIA memorandum at 4 (emphasis added).(30)

In other words, with respect to the disclosure of related parties and relevant ownership and management information, we limited our final BIA analysis to the failure of the respondents to provide such information only for related parties involved in the sale and production of subject merchandise in response to our initial questionnaires.

The record evidence demonstrates that nine members of the Queen's Flowers Group, [ (31)], did not produce subject merchandise during all three review periods. However, [ ] made some sales of subject merchandise during the 5th POR, and [ ] and [ ] sold subject merchandise during the 7th POR. As detailed in section A of our final BIA memorandum ("Analysis of the Failure to Identify Related Parties"), when we analyzed the responses to our initial and supplemental questionnaires for all three PORs, we identified the extent to which the original respondents provided incomplete responses. With the exception of [ ], all of these companies ([ ]) produced the subject merchandise during all three review periods. Our analysis revealed a similar pattern of incomplete responses by the original respondents to these initial questionnaires, and, in each instance, we identified areas where the respondents failed to provide requested information for producers of subject merchandise.

Where we did in fact find deficient responses concerning ownership information for related companies that did not produce the subject merchandise, we compared the companies' responses not to our initial questionnaires but to supplemental questionnaires we issued in September 1994. In these supplemental questionnaires, we requested specifically that the respondents provide ownership and management information for [ ], the two related U.S. importers. These requests for information were not contingent upon whether the owners were producers of subject merchandise. As we noted in Item 1 of the Final BIA memorandum, all respondents provided incomplete responses to these supplemental questionnaires. Thus, contrary to the arguments respondents presented to the CIT during the hearing on the preliminary injunction, our BIA analysis did not fault erroneously the respondents for failing to provide information not requested in our initial questionnaires for the three review periods.

We also note that, in reevaluating our preliminary BIA analysis, we amended our listing of deficiencies with respect to the related-party information concerning companies that did not produce the subject merchandise. For example, in Item 4 of our December 5, 1994 memorandum, we preliminarily faulted Flores del Hato in the 7th POR for failing to identify 13 members of the Queen's Flowers Group. However, in our final BIA analysis, we limited this deficiency to a failure to identify only one related firm ([ ]) which purchased the subject merchandise from Hato. Also in Item 4 of the Final BIA memorandum, we had preliminarily found that Flores del Hato failed to provide complete information for 11 related firms and continued to fail to identify two additional firms in response to our July 22, 1994 supplemental questionnaire. In our final BIA analysis, we limited the deficiency to Flores del Hato's failure to identify six firms ([ ]) that produced the subject merchandise and for which Flores del Hato, or any other Group member, failed to provide any ownership information. Likewise, in our final BIA analysis we amended our listings of deficiencies for the following: Flores del Hato for the first part of Item 3; Agroindustrial el Riofrio for Items 6, 7, and 8; Flores Canelon for Item 11; Flores la Valvanera for Item 13; Queen's Flowers de Colombia for Item 17; and Jardines de Chia for Item 20.(32) In each of these items, we limited our listing of deficiency to a failure to identify related parties that produced the subject merchandise, eliminating those instances where a firm did not produce the subject merchandise in a particular POR. We clarify that the BIA memorandum dated June 28, 1996 represents our final BIA analysis where we had already addressed the CIT's concern.

In sum, our final BIA analysis reflects a careful comparison of the specific requests for information and the corresponding responses. Where appropriate, we corrected our preliminary deficiency findings to ensure that the respondents were not faulted for failing to provide information we did not request Our reconsideration of the record evidence in the context of this remand proceeding confirms that no such analytical flaws remain in our BIA analysis.

3. If A Complete Record Existed by June 28, 1995, Why Did The Department Use BIA?

The fact that the record was developed fully concerning this related-party information by June 28, 1995, a year before the Final Results,(33) does not detract from our BIA determination. We underscore the fact that respondents failed at the outset of the reviews to furnish information for related producers of subject merchandise. Because those producers were sufficiently related to warrant treatment as a single entity, the omission of information resulted in missing sales and CV data in each POR essential to the calculation of accurate dumping margins. Under such circumstances it is our standard practice to resort to margins based on BIA.(34) Our decision to employ BIA in the Final Results noted deficiencies regarding the failure to provide factual information pertinent to a collapsing analysis. In reconsidering our BIA decision we have not relied on those deficiencies. Rather, we have focused on the missing data for the six related producers of subject merchandise.

At the time of the Preliminary Results, we had already determined that, notwithstanding our requests for specific information, the responding companies had failed to submit adequate responses and were not cooperating with our investigation. Therefore, BIA margins were warranted. In this case, however, because we did not receive complete and accurate responses to our requests for information, yet we were confronted with serious questions concerning the inclusion of additional companies in the review proceedings, we issued questionnaires to the 12 additional companies for the sole purpose of determining whether we should include them in the Queen's Flowers Group. The post-preliminary information we received through the June 13, 1995 and July 28, 1995 responses of these additional companies did not absolve the deficient responses of the original respondents who had already had the opportunity to provide this information prior to our Preliminary Results. We never received timely information about six of these firms (Flores la Mana, Flores el Cipres, Flores el Tandil, Flores Atlas, Flores Calima, and Flores el Roble) that produced subject merchandise. Yet three of the original respondents, Queen's Flowers De Colombia Ltda., Jardines De Chia Ltda., and Agrodindustrial del Rio Frio, had access to information about these producers that were affiliated under even a conservative definition of relatedness. To allow the responses of these companies to rectify the previously identified deficiencies would reward the Group for not cooperating prior to our preliminary results and would encourage future respondents to report selectively.(35)

Comments Regarding Timeliness of Submitted Information

The respondents assert the Department did not act properly upon information it received identifying related producers. According to the respondents, in the initial questionnaire responses for the 5th and 6th PORs submitted in November 1993, the seven original respondents identified the entire "Mojica Group." Specifically, the respondents contend, MG Consultores disclosed that it owned [ ] percent of the shares of Flores Jayvana, one of the 12 companies the Department claims the respondents failed to disclose. MG Consultores, the respondents argue, similarly disclosed its relationship with Flores el Cacique early in the 7th review. The respondents assert that the fact that the Department did not request data from these related parties and did not request the Group to resubmit its responses on a consolidated basis undermines its contention that, if it had had related-party information early in the proceeding, it would have requested sales and cost data from the other related parties.

The respondents assert further that, once the Department requested disclosure of all companies in which shareholders held shares, the companies provided the complete information which was available to the responding companies by shareholder (i.e., the names of thirteen companies linked to the original seven through one or more common shareholders). The respondents assert that the Department could have requested full sales and cost responses from these 13 companies in August 1994 and would have had time to analyze and incorporate this information in the preliminary or final results but, instead, requested only Section A information (rather than full sales and cost data) in May and July 1995, over nine months later. The respondents argue that if the Department had a sufficient basis for issuing such questionnaires in May and July 1995 then it also had sufficient basis for issuing the same questionnaires in August 1994. The fact that the Department did not pursue the information in a timely manner, the respondents contend, again undermines the Department's claim that, if it had had related-party information early in the reviews, it would have requested sales and cost data from these related parties.

The respondents argue that the reasons for the delay in obtaining complete related-party information upon which the Department now relies were the Department's own delays in conducting the reviews, the complicated factual relationships at issue, and the Department's expansion of the statutory definition of related parties into unprecedented areas. Furthermore, according to the respondents, it is clear that all information upon which the Department has relied has been provided by the companies themselves.

The petitioner argues that, as early as the 5th review (for the 1990-91 period), the companies in the Queen's Flowers Group knew that the Department was concerned about collapsing related parties in the Queen's Flowers Group. Also, the petitioner argues that the [ .] The petitioner notes that, while the ownership structure of the companies changed during the periods of review, [ .] The petitioner concludes that, for these reasons, it was not sufficient that the respondents, in the end after the preliminary results of review, supplied all the data necessary to apply the Department's collapsing analysis

Analysis in Response to Comments.

We acknowledge that MG Consultores reported its ownership interest in Flores Jayvana in its November 29, 1993, Section A responses for the 5th and 6th PORs and, through our oversight, we did not pursue additional information. Therefore, we have not considered this factor in our decision to apply first-tier BIA to the Queen's Flowers Group. Even so, the decision to apply first-tier BIA to the Queen's Flowers Group is a well-founded action on our part. Our collapsing analysis treated the Queen's Flowers Group properly as the single producer and exporter in each POR. For each POR we are missing significant amounts of sales and CV data. Consistent with our standard practice, where we do not have complete sales and CV data for a respondent company, we have reasonably assigned a single margin to the Group in each POR based on BIA.

We disagree with the respondents that we had sufficient information on the record and could have requested full sales and cost responses from these 13 companies in August 1994 with sufficient time to analyze and incorporate this information in the preliminary or final results. Because the respondents failed to provide the detailed ownership and management information which we requested clearly and unambiguously in our supplemental questionnaires, we did not have sufficient information in August 1994 to do our collapsing analysis and subsequently issue sales and cost questionnaires to the remaining members of the Queen's Flowers Group. Contrary to the respondents' assertion, the production of this information was not dependent on the respondents' understanding "of the Department's expansion of the statutory definition of related parties into unprecedented areas." Regarding our July 22, 1994 questionnaire for all three PORs, it is inconceivable that clear questions could have been misunderstood or that the requested information was unavailable to this close-knit investor group. This questionnaire requested the respondents to report whether any member of their boards of directors or any stockholder was also a board member or stockholder of any other producer, reseller, bouquet converter, or U.S. importer of the subject merchandise and the respondents were to provide full ownership information for those companies. This request for information did not require the respondents to apply either statutory definition of related parties.

However, in response to the question regarding overlapping owners and board members, the respondents failed to provide full ownership information for six producers of subject merchandise: Cipres, Mana, Tandil, Calima, Atlas and Roble. While members of the Group mentioned that their shareholders had shares in Cipres, Mana, Tandil, and Calima, they failed to furnish complete ownership information for these companies, preventing the Department from conducting a collapsing analysis and seeking out the missing sales and CV data. Moreover, the respondents failed to even list two members of the 20-member Queen's Group which produced subject merchandise, Flores Atlas and Flores el Roble. We could not have sought out information from firms that we did not even know existed.

Also, the fact that we had to issue still another supplemental questionnaire in September 1994 further counters the respondents' claim that, if we had sufficient information to issue questionnaires in May 1995, then we also had a sufficient basis for issuing questionnaires in August 1994. It was not until after we received the September responses from the respondents that we learned, for the first time, the names of the six additional firms mentioned above. In addition, we learned of the strong familial ties between shareholders, managers, and board members of the original eight firms and the joint U.S. sales efforts of these 20 firms through their related U.S. importers. It was only at this point in our investigation that we were able to piece together the information necessary to determine that the application of BIA was appropriate for these firms. Therefore, our failure to act upon this information in a timely manner was attributable directly to the selective reporting of critical information by the respondents.

4. Flores Generales/Cultivos Generales

As noted by the CIT, one factor in our BIA determination was Flores Generales' failure to respond to our questionnaire for the 7th POR. See Final Results, 61 FR at 42855. The CIT was persuaded that this aspect of our BIA analysis was in error because we allegedly sent the questionnaire to the incorrect address.

In our draft remand results we stated our belief that, because we had taken all reasonably necessary steps to mail the questionnaire to Flores Generales' correct address and had a receipt from the carrier service indicating receipt, we had therefore treated Flores Generales appropriately as a non-responsive party.

Comments Regarding Flores Generales/Cultivos Generales

In response to our draft remand results the respondents indicated that an appearance had been entered on behalf of Cultivos Generales (Flores Generales) by current counsel for the 5th and 6th PORs, and that this submission had also indicated that the firm made no shipments of subject merchandise for those reviews. In the 7th POR, respondents state no appearance was entered on behalf of either Flores Generales or Cultivos Generales (the successor firm to Flores Generales). For that review the Department mailed a questionnaire directly to the firm at an address that respondent explains is in the wrong city and a street address at which the firm did not reside. Consequently, Cultivos Generales claims that it did not receive the questionnaire. Cultivos Generales indicates that its correct address is on the record in its responses to questionnaires for the 5th and 6th PORs. Therefore, Cultivos Generales contends there is no factual basis for the Department's position that it sent a questionnaire to the correct address.

In its rebuttal the FTC claims that the appearance of the counsel on behalf of Cultivos Generales (Flores Generales) in the 5th and 6th reviews was ambiguous and, therefore, the Department cannot be required to assume that Cultivos Generales was the successor to Flores Generales.

Analysis in Response to Comments

After reexamining the record evidence and comments from the parties, the Department cannot conclusively determine that it sent the questionnaire to Cultivos Generales at the correct address or that the firm ever actually received it. Therefore, since there is reasonable doubt that Cultivos Generales received this questionnaire, it would be inappropriate to characterize Cultivos Generales as a non-responsive firm and to include this factor in our reasoning to apply BIA to the Queen's Group. Notwithstanding this finding, however, we note that, because Cultivos Generales is properly considered a member of the Queen's Flowers Group, the factual question of whether the company received the questionnaire is irrelevant because it is assigned the single margin applicable to the entire Group for each POR.

5. Deficiencies Attributable to Flores del Hato

The CIT noted that some of the deficiencies in Flores del Hato's September 20, 1994 supplemental response may have been caused by the fact that the Department mistakenly sent Flores del Hato a questionnaire meant for Agroindustrial del Riofrio. Although it appears that Flores del Hato answered questions that were directed to Agroindustrial del Riofrio, we respectfully disagree that we committed any error in transmitting the supplemental questionnaire to Flores del Hato.

The record demonstrates that the questions we asked of Flores del Hato and Agroindustrial del Riofrio correspond to the deficiencies we found in their previous responses. That is, when comparing the July questionnaire responses for each company to the company-specific supplemental questionnaires, there is a direct link between the information provided and the cited deficiency. For example, Agrioindustrial del Riofrio reported purchases of other materials from unrelated suppliers on line 95 of its Section D response but did not provide the name, addresses, and facsimile numbers of the suppliers as requested. See July 22, 1994 questionnaire response of Agroindustrial del Riofrio at 22. In our September 1994 supplemental questionnaire directed to Agroindustrial el Riofrio, at question 18, we asked the company to provide this information for the suppliers of other materials identified in line 95 of its response. See September 6, 1994 supplemental questionnaire to Agroindustrial del Riofrio. Thus, there is no record evidence that the Department inadvertently asked the wrong question to the wrong respondent. Further, the company-specific supplemental questionnaires were directed properly to the correct respondent and addressed properly in care of counsel for these firms. While it is possible that the companies themselves did not receive the correct questionnaire in Colombia, there is no record evidence that the Department caused this alleged error.

We stated in our draft results that, assuming arguendo that Flores del Hato received the incorrect supplemental questionnaire, the respondent still failed to provide a proper response. Citing our final BIA memorandum, we noted that Flores del Hato reported purchases of "other materials" but failed to provide the requested identifying information and that the specific line number in which the respondent reported these material costs is irrelevant. We also concluded that the firm reported incorrect cutting information as well.

Comments Regarding Hato's Supplemental Questionnaire

The respondents contend that Hato and Riofrio did not respond properly to the questionnaires they received because the questionnaires were mixed up and the respondents received questionnaires which did not match their previous responses. The respondents argue that, at the time, they noted that the questionnaires received did not match their previous responses but were instructed to simply answer the questionnaires and that any problem would be corrected in supplemental questionnaires. While stating that the mix-up could have occurred within the Department or at the offices of Queen's counsel, the respondents contend that, since both counsel and the companies notified the Department of the apparent discrepancy and the Department continued to instruct the respondents to answer the questionnaire, it is inappropriate to use first-tier BIA against either company.

Additionally, the respondents argue that the Department is incorrect in asserting that Hato and Riofrio did not respond properly to the questionnaires which they did receive. The respondents assert that, in the case of Hato, since a certain request was limited to data concerning a certain line number and the respondent included no data in that line number, the Department cannot claim that the line-number reference is irrelevant. Furthermore, the respondents claim that the Department cited another instance concerning "cull revenue" in which the respondent purportedly did not answer the question when, in fact, the respondent did answer the question correctly.

The petitioner asserts the record is clear that the Department did not rely on answers to the September 1994 questionnaire of Hato and Riofrio alone to arrive at its determination to use first-tier BIA, but that use of BIA is perfectly acceptable where the respondent's failure to respond is not attributable to Department error.

The petitioner notes that the questionnaire requested the names, addresses and facsimile numbers of all suppliers of "other materials" purchased and reported in line 95 and that the Department was clearly asking for the information concerning the suppliers of "other materials." The petitioner also notes that Hato's response that it did not report any data on line 95 was far from forthcoming or responsive. The petitioner concludes that the Department was correct in finding Hato unresponsive with respect to that question.

Analysis in Response to Comments

We continue to find that there is no record evidence that the Department inadvertently directed the wrong supplemental questionnaire to Hato. After considering the comments from interested parties carefully, we withdraw our remarks concerning the failure of Hato and Riofrio to respond properly to the questionnaires that they did claim to have received. While Hato failed to respond to the spirit of the question concerning suppliers of "other materials," we have concluded that the absence of that information does not have a significant impact on the determination of margins. However, we respectfully submit that any mix-up of questionnaires was not attributable to error on our part.

Issue 4: Explain the Appropriateness of First-Tier BIA

In the Final Results, we concluded that the deficiencies contained in the responses of the respondents derived from a pattern of unresponsive and insufficient information. On this basis, we determined that the Queen's Flowers Group had significantly impeded the reviews and was, therefore, an uncooperative respondent, warranting assignment of first-tier BIA margins for all three review periods. In accordance with the CIT's instructions, we have reconsidered the selection of the BIA margins under the applicable two-tiered system. As we detail below, because the record evidence supports a finding that the Queen's Flowers Group significantly impeded each of the three administrative reviews, we continue to find that the assignment of first-tier BIA margins is appropriate.

In the Preliminary Results we emphasized a record replete with incomplete and contradictory factual information regarding management, ownership, and affiliations. However, in reconsidering our BIA analysis for the Final Results, we conclude that the initial deficiencies concerning undisclosed related producers of subject merchandise far outweigh the severity of the respondents' omissions of other data. While we still consider these deficiencies serious, we find that the assignment of first-tier BIA margins on evidence of the respondents' uncooperative participation in all three reviews is based on the undisclosed related-party producers, which resulted in missing sales and CV data that severely hindered our ability to calculate accurate margins for the Queen's Flowers Group.

The lack of a fully developed record for each review is appropriately attributed to the respondents who were in possession and control of the requested information. See Tianjin Mach. Import & Export Corp. v. United States, 806 F. Supp. 1008, 1015 (CIT 1992) (ruling that the burden rests on the respondents to create an adequate record). In other words, had the respondents provided complete and accurate responses at the outset of the reviews, the Department would have had necessary factual information to evaluate the nature of the interrelatedness among the entire Queen's Flowers Group, conduct a collapsing analysis, and request and receive the necessary U.S. sales and CV data so as to calculate dumping margins for the Group prior to the Preliminary Results. Instead, the record was not developed fully prior to the publication of the Preliminary Results, and the Department was compelled to take the unusual action of opening the record for submission of additional factual information after the Preliminary Results in order to complete its collapsing analysis. Under these circumstances, we concluded properly that the Queen's Flowers Group significantly impeded the timely completion of the reviews.

In this instance, evidence that the respondents significantly impeded the reviews also demonstrates a refusal to cooperate. Unlike circumstances where a refusal to cooperate is demonstrated by non-response to an entire questionnaire, a respondent is reasonably treated as noncooperative when it fails to provide the Department with information within its possession and control. See Allied- Signal Aerospace Company, Garrett Engine Division and Garrett Auxiliary Power Division v. United States, Slip. Op. 93-200 (October 14, 1993). In that case, the Federal Circuit found that the respondent's failure to respond to the Department's questionnaire did not constitute a refusal to cooperate because the firm was unable to respond due to limitations in its cost-accounting system. In that case, the respondent had notified the Department of these difficulties prior to initiation of the review and offered to provide information using a more simplified process.

Here, there is no record evidence that the original respondents confronted similar practical difficulties in providing related-party information in response to our questionnaires. To the contrary, as demonstrated by the fact that they provided information in piecemeal fashion in their supplemental responses, this information could have and should have been provided in response to our initial questionnaires. Moreover, given the extent of common ownership and management, it is entirely reasonable to expect the respondents to provide at a minimum the percentage of ownership and management positions in other companies. We therefore conclude that the pattern of incomplete and inaccurate responses constitutes evidence that the Queen's Flowers Group as a whole was uncooperative. To conclude otherwise would be contrary to the purposes of the BIA rule, as recognized by the Federal Circuit when it stated:

. . . . the ITA cannot be left merely to the largesse of the parties at their discretion to supply the ITA with information. This is particularly the case when the ITA is attempting to obtain information to conduct statutorily mandated administrative reviews because unlike the ITC, the ITA has no subpoena power.

Olympic Adhesives, Inc. v. United States, 899 F.2d 1565, 1571 (Fed. Cir. 1990).

Comments Regarding the Use of First-Tier BIA

The respondents argue that the Department's allegations that inaccurate information was provided are either incorrect or immaterial. For example, the respondents argue that the fact that Canelon denied incorrectly that it had sales of subject merchandise to [ ] and [ ] is immaterial because the small transactions were reported fully by both [ ] and [ ]. The respondents conclude that, if they are to be collapsed and treated as a single entity, the responses should be viewed together. Also, the respondents note that the Department's assertion that Queen's Flowers de Colombia failed to report that its shareholders held management positions in other related flower producers is incorrect. The respondents note that the Department relies upon notations in the certificates of existence and legal representation identifying [ ] as a legal representative of [ ]. However, the respondents note that these certificates are dated after the period of review and that, apart from his role as a legal representative of the company, there is no evidence that [ ] occupied any managerial position at either Tandil or Atlas.

Queen's arguments are paradoxical. On the one hand Queen's argues that we should consider the record as a whole. Yet, when we look at the record as a whole we find contradictory evidence. For example, in our July 22, 1994 supplemental questionnaire we asked Canelon the following question:

"6. Does your firm sell subject flowers to or buy subject flowers from any of the following firms:

[

.]"

To this question Canelon, in its August 1, 1994 response, answered that "Flores Canelon Ltda did not sell to or buy subject flowers from any of the firms listed in this point." This is inconsistent with the responses of [MG Consultores and Riofrio] in which these companies indicate that they purchased flowers from Canelon. See, e.g., [MG]'s September 19, 1994 response and [Riofrio]'s September 20, 1994 response. Because the existence of transactions between related parties is an important factor in our collapsing analysis, Canelon's response hindered our ability to conduct a collapsing analysis.

Respondents claim that the Department has relied on certificates of existence which are dated after the POR in determining that Queen's Flowers de Colombia failed to report shared management. However, according to the certificates of existence, [Erwin Jordan] was appointed as assistant manager of Flores el Tandil on January 23, 1992 through Minute No. 3 of the Board of Directors meeting of January 9, 1992 and was appointed assistant manager of Flores Atlas on May 3, 1989 through Minute No. 11 of the Board of Directors meeting of January 20, 1989. Thus, [Erwin Jordan] held a management position in Flores Atlas during each POR and in Flores el Tandil for the 6th and 7th PORs. Shared management is an important factor in our collapsing analysis and both of these firms produced subject merchandise in those PORs. Because Queen's Flowers de Colombia denied having shared management, it hindered our efforts to determine whether we needed additional information from those firms.

The respondents argue that, while the decision to apply BIA necessarily must be based on the alleged failure of the respondents to respond to specific questions asked by the Department, the Department in its draft remand results rarely quotes direct evidence of these alleged failures but, instead, relies upon inaccurate and misleading characterizations.

When we evaluated the completeness and accuracy of the responses in their entirety, we observed a pattern of unresponsiveness to our requests for information. This pattern was characterized by numerous omissions and contradictions concerning the extent of common ownership, management, and existence of related-party transactions -- both of subject merchandise and of services -- between each member of the Queen's Flowers Group. As a result of these deficiencies, we were unable to identify all entities comprising the Queen's Flowers Group completely and we were unable to request necessary CV and U.S. sales data from or otherwise conduct reviews of all related entities prior to the Preliminary Results. Although, for the 5th review, a significant portion of the Group's volume of U.S. sales existed on the record, for the 6th and 7th reviews, only 78.37 percent and 71.67 percent, respectfully, existed on the record for these reviews. Moreover, the CV data for seven of the 14 members of the Queen's Flowers Group that exported subject merchandise to the United States is missing from the record for each review.

Analysis in Response to Comments

We agree with the respondents that the decision to apply BIA must be based on the failure of the respondents to respond to specific questions we asked while conducting the reviews. However, based on the record evidence for each review period, we conclude that the Queen's Flowers Group was an uncooperative respondent and significantly impeded the conduct of our reviews. In this respect the record demonstrates that on at least three separate occasions prior to our preliminary results we attempted to obtain complete relationship information from the respondents concerning other members of the Queen's Flowers Group. Instead, the respondents elected to report selectively or even submitted misleading information. While the respondents attempt to excuse away their failure to provide information essential to our investigation, the record indicates that the questionnaires were clear with respect to the information we sought and that the respondents did not undertake to receive clarification or refinement of issues they found confusing.

As the CIT stated in Asocoflores, once Commerce has "fairly requested" the data, "it possesses the 'discretion to determine whether a respondent has complied with an information request,'" quoting Daido Corp v. United States, 893 F. Supp. 43, 49-50 (CIT 1995). Asocoflores, Slip. Op. 98-33 at 36. As detailed above, we asked the respondents specifically to disclose information about related parties involved in the production of subject merchandise. The respondents failed to furnish that information, thereby impairing our ability to calculate accurate dumping margins.

In summary, when a respondent 'picks and chooses' which data it wants the Department to see, the respondent is not attempting to cooperate with the investigation. Instead the respondent is attempting to control the results of an investigation through its selective reporting, which represents uncooperative participation by that party. Accordingly, it is appropriate to assign the Queen's Flowers Companies margins based upon first-tier BIA.

FINAL RESULTS OF REDETERMINATION

As a result of our reconsideration of the issues on remand from the CIT, we have determined that the following margins apply to all members of the Queen's Flowers Group for the 5th, 6th, and 7th PORs:

Producer/Exporter 90/91 91/92 92/93


Queen's Flowers Group 76.60 (1) 76.60 (1) 76.60 (1)

Queen's Flowers De Colombia Ltda.

Jardines De Chia Ltda.

Jardines Fredonia Ltda.

Agrodindustrial del Rio Frio

Flores Canelon

Flores del Hato

Flores La Valvanera Ltda.

M.G. Consultores Ltda.

Flores Jayvana

Flores el Cacique (7th POR only)

Flores Calima (7th POR only)

Flores la Mana

Flores el Cipres

Flores el Roble (7th POR only)

Flores del Bojaca

Flores el Tandil

Flores el Ajibe (7th POR only)

Flores Atlas

Floranova (6th and 7th PORs only)

Cultivos Generales


(1) No change to the original margin as a result of remand.

These final results of redetermination are pursuant to the order of the CIT in Queen's Flowers.

________________________

Robert S. LaRussa

Assistant Secretary

for Import Administration

April 6, 1998

(Date)

1. During the course of the reviews, the Department determined that these companies were sufficiently related to warrant collapsing into one entity for the purpose of calculating a single antidumping margin during each of the relevant periods of review. Thus, the Department refers to the companies collectively as the Queen's Flowers Group. When the Department refers to one of the companies separately, it is referring to specific data provided by that company.

2. Specifically, we determine that these firms are related, they have similar production facilities, and they have a significant potential for the manipulation of prices and production. These are the factors the Department relied upon in the Final Results, as well as for collapsing the entire Queen's Flowers Group in this remand determination. See Issue 2, below. The evidence supporting this last factor with respect to MG Consultores and the remainder of the [ ] sub-group consists of a high degree of common ownership, overlapping management and boards of directors, and additional closely intertwined operations, such as joint U.S. sales efforts, shared office space, and inter-company transactions. A complete description of the factual and legal bases for the Department's determination with respect to the entire Group is contained in the administrative record in the following memoranda: Memorandum, "Collapsing Related Parties," November 17, 1994 (November Memo); Memorandum, "Amendment to the Nov. 17, 1994 Collapsing Memo," January 10, 1995; Memorandum, "Collapsing Related Parties Within the Queen's Flowers Group," August 3, 1995 (August Memo); and February Memo, supra, as well as the accompanying exhibits. Therefore, for purposes of this remand, we have not conducted an additional collapsing analysis with respect to the [ ] sub-group.

3. The Department rejects the respondents' argument that, for a company to be related to the component-companies of a collapsed entity, such as MG and its subsidiaries, there must be a direct statutory relationship between the "outside" company and each component-company. Similarly, we reject the argument that MG's subsidiaries cannot be related to Chia because Chia is related to MG through the ownership of [ ], a "person" or "persons," rather than a separate "company." Each provision of section 771(13) uses the word "person," not "company," to describe who or what may "own or control" another party. The statute does not define the term "person." The Department has done so, in its regulations, as "any interested party as well as any other individual, enterprise, or entity, as appropriate." 19 C.F.R. § 353.2(p) (emphasis in original). Thus, the Department has reasonably interpreted "person" to include individuals as well as companies when determining whether companies are related within the meaning of section 771(13).

4. Also in our draft results, in answer to the CIT's concern, we explained how, in relying upon non-financial factors, we had acted consistently with earlier determinations in Roses and Lighters. Id. at 8-11. However, we are not relying upon non-financial factors to determine relatedness in this redetermination and, therefore, we are not incorporating an explanation regarding our consistency with Roses and Lighters.

5. In its decision, the CIT did not indicate that the Department should solicit or gather additional information during the course of this remand.

6. See Footnote 6, above, regarding the explanation of our consistency with Roses and Lighters.

7. Specifically, the CIT accepted that the [ ] held in the aggregate over 20 percent in each member of the [ ] even though in several cases the entire interest in a certain company was held severally by either [ ] or [ ]. For instance, [ ] had interests in Jardines de Chia, Riofrio, and Floranova while [ ] did not. Similarly, [ ] had an interest in Cultivos Generales while [ ] did not. In the CIT's opinion it is reasonable to consider these interests held severally as the interests of the [ ] as "persons" under section 771(13)(D).

8. In order to clarify that we examined each POR separately, we should state that, for the 7th POR, the [ ] sub-group consists of eight members while the [ ] sub-group consists of 14 members. As reviewed above in Footnote 2, not all 20 firms should be included in this analysis for purposes of the 5th and 6th PORs. Therefore, for the 5th POR, the [ ] sub-group consists of seven members, while the [ ] sub-group consists of ten members. For the 6th POR, the [ ] sub-group consists of seven members, while the [ ] sub-group consists of eleven members.

9. We find that paid-in capital is more appropriate than net worth because net worth is subject to manipulation (e.g., treating the firm as a profit or a loss center for tax considerations) and paid-in-capital is more reflective of the relationship of each investor to both one another and to other commonly owned firms at their point of incorporation.

10. In our draft results, we examined the common ownership between the two sub-groups for the 7th POR by aggregating the interests of [fourteen] individuals who held interests in each of the two sub-groups during that time. Due to our finding in these final results that Calima is not related to the [Mojica] sub-group through the common ownership of the [Mojica] family, there are now only eight individuals who had interests in each of the sub-groups during the 7th POR.

11. We do not know the paid-in capital for Jardines Fredonia. However, our analysis revealed that, regardless of the value of paid-in capital, an amount could not be found that would reduce the common ownership below 20 percent. Therefore, the missing information is not critical to this analysis.

12. The Department rejects the respondents' argument that, for a company to be related to the component-companies of a collapsed entity, such as MG and its subsidiaries, there must be a direct statutory relationship between the "outside" company and each component-company. Similarly, we reject the argument that MG's subsidiaries cannot be related to Chia because Chia is related to MG through the ownership of [ ], a "person" or "persons," rather than a separate "company." Each provision of section 771(13) uses the word "person," not "company," to describe who or what may "own or control" another party. The statute does not define the term "person." The Department has done so, in its regulations, as "any interested party as well as any other individual, enterprise, or entity, as appropriate." 19 C.F.R. § 353.2(p) (emphasis in original). Thus, the Department has reasonably interpreted "person" to include individuals as well as companies when determining whether companies are related within the meaning of section 771(13).

13. See Footnote 2 concerning membership of the Queen's Flowers Group for the 5th, 6th, and 7th PORs.

14. In establishing the standard by which we have aggregated the ownership interests of these individuals as "persons" under section 771(13)(D), we have changed the aggregated investor group only slightly from our draft results. For the 5th, 6th, and 7th PORs, we have eliminated [ ] from our aggregation. We determined that the record evidence does not support that [ ] shares in the pattern of common interests with the other members of the investor group. For the 7th POR, our determination in these final results that Calima is not a member of the [ ] sub-group eliminated six other investors from our consideration since, after Calima was eliminated from the [ ] sub-group, these six no longer held interests in each of the two sub-groups. None of these shared a pattern of common interests with the investor group identified in these final results.

15. This is a summary of information included in the "Master Charts" (Exhibits 1, 2, and 3). The "Master Charts," in turn, reflect data included in the February Memo.

16. In considering the record evidence of a pattern of common interests among a group of investors, we considered only those cases where an investor had a five-percent or greater ownership interest in a company. In practice, the Department has determined that this five-percent threshold qualifies as "any interest" and establishes a parent-subsidiary relationship between two entities under section 771(13)(B) and (C). See Antidumping; Final Determination of Sales at Not Less Than Fair Value: Certain Forged Steel Crankshafts from Japan, 52 FR 36984, 36985 (October 2, 1987). Therefore, in explaining the pattern of common interests among members of this investor group, where we state that an investor had an "interest" in a company, we mean that they had an interest sufficient to establish that they are related to that company under section 771(13)(B) and (C).

17. In each of the three PORs, the group members sold collectively between [ ] and [ ] percent of their U.S. imports by volume through the related importers. See February Memo at 9.

18. Queen's Flowers de Colombia has paid-in capital of Colombian pesos [ ] and exported [ ] to the United States of any of the Queen's Group members. See Exhibit 1 for paid-in capital and "U.S. Sales" Chart for sales of subject merchandise (Exhibit VII).

19. In the 7th POR, Calima exported [ ] of subject flowers with a total value of $[ ]. See Section A (Quantity and Value Chart) of the company's response.

20. In Nihon, the CIT upheld the following factors for collapsing related parties: 1) the companies are closely intertwined; 2) transactions take place between the companies; 3) the companies have similar types of production equipment; and 4) the companies are capable of manipulating prices or affecting production decisions.

21. The November 17, 1994 collapsing memorandum analyzed the following criteria:

Do the related manufacturers have interlocking boards of directors?

Do they have similar production processes, facilities or equipment so as to facilitate shifting of production between facilities?

Do they operate as separate and distinct entities?

Do they share marketing and sales information or offices?

Are they involved in the pricing or production decisions of the other entity?

22. As noted above, in reconsidering our collapsing determination, we have determined that a separate analysis should be made for each period of review.

23. Because we are no longer relying upon non-financial factors in determining relatedness among the members of the Group, the existence of interlocking board members and management is only a collapsing factor in this redetermination.

24. See February Memo at 8-10, Footnote 7, and Attachments 1-4. In this memorandum, we reviewed numerous transactions and intertwined operations which linked the two related sub-group enterprises. For instance, [ ], [ ], and companies in each sub-group had interests in [ ], a company which rented office space to and/or provided a mail drop for members of each sub-group. We reviewed the investments in each of the two U.S. importers by all members of each sub-group. Flores Atlas, a member of the [ ] sub-group, rented its land from [ ], a stockholder of [ ] members of the [ ] sub-group. Finally, the four attachments illustrate that, in numerous instances and combinations, a member of one sub-group sold subject merchandise, non-subject merchandise and/or cuttings to a company that was either a member of both sub-groups (e.g., MG Consultores' sales of subject merchandise to Riofrio) or the other sub-group (e.g., MG sales of cuttings to Mana and Calima).

25. We also disagree that the inter-company transactions were made at arm's length. We do not know that the sales of flowers or the rental of office space or the sharing of secretarial salaries were made at arm's length. The proper evaluation of these transactions was prevented by the non-compliance of the original respondents to our questionnaire and their failure to provide data justifying the arm's-length nature of the transactions. Of the one type of inter-company transaction noted in the post-preliminary responses, sales of cuttings, we were provided data and we determined that these sales were not made at arm's-length prices. See Footnote 6 of the February Memo.

26. In the 5th POR, these companies were Queen's Flowers de Colombia, Agroindustrial del Riofrio, M.G. Consultores, Jardines de Chia, Flores la Valvanera, Flores Canelon, and Jardines Fredonia. For the 6th and the 7th PORs, in addition to these firms, Flores del Hato was a respondent. We clarify that, when referencing "respondents," we mean those member companies whose responses we examined for each review period.

27. Two additional firms provided Section A responses in the 4th POR. However, because these two firms were not selected as "sample" respondents in that review, further responses were not required.

28. These sales volumes are calculated without considering sales made by Flores Jayvana.

29. For example, the August 1, 1994 and the September 1994 supplemental questionnaire responses represented second and third chances for these firms to provide complete and accurate information concerning relevant ownership and management information for related parties.

30. We clarify that the third component, failure to identify parties related in the CV Input definition, applied to producers of subject merchandise.

31. As discussed in the preceding analysis, [ ] is not part of the Queen's Flowers Group for the 5th POR.

32. The Queen's Flowers Group acknowledged the Section A deficiencies discussed in Items 10 and 14. We did not change our position for Items 16, 18, 21, 22, and 23 as a result of our reexamination in the context of this remand proceeding.

33. We respectfully clarify that this date, rather than June 22, 1995, which the CIT cited in its opinion, is the accurate date on which the records were developed fully concerning the related-party information.

34. See, e.g., Persico Pizzamiglio, S.A. v. United States, Slip Op. 94-61 (CIT 1994)(upholding the Department's resort to BIA where a respondent was unable to demonstrate the completeness of its U.S. sales). This opinion upholds our standard practice of employing BIA in the absence of complete sales or cost information. See, e.g., Tapered Roller Bearings from Japan: Final Results of Administrative Review, 56 FR 65228 (December 16, 1991), in which we stated that reliance on an incomplete home market sales database would result in an inaccurate foreign market value calculation and ultimately an unreliable dumping margin, thus necessitating the use of BIA; Grain-Oriented Electrical Steel from Italy: Final Determination of Sales at Less Than Fair Value, 59 FR 33953-54 (July 1, 1994), in which we determined to use BIA when we were unable to make price-to-price comparisons because we were unable to verify the accuracy of costs of production.

35. We also clarify that our BIA analysis is independent of our decision to collapse the Group. Through the November 17, 1994 and January 10, 1995 collapsing memoranda, we initially explored collapsing 20 firms into the Queen's Flowers Group. However, on March 24, 1995, the 20 firms filed comments objecting to our decision to collapse because, among other reasons, 12 of the companies never received a questionnaire. By seeking additional information and reducing our collapsing analysis to the eight original respondents, we were recognizing the 12 companies' claim of independence and preserving their rights of "due process."