A Brazilian producer of frozen concentrated orange juice (FCOJ), Citrosuco Paulista, S.A. (Citrosuco), moves pursuant to Rule 56.1 of the Rules of this Court to contest the final determination of the International Trade Administration of the United States Department of Commerce (Commerce) that Citrosuco was dumping FCOJ in the United States, Frozen Concentrated Orange Juice From Brazil: Final Determination of Sales at Less Than Fair Value, 52 Fed. Reg. 8324 (Mar. 17, 1987), and the final determination of the United States International Trade Commission (Commission) that an industry in the United States has been materially injured or threatened with material injury by reason of the less than fair value imports of FCOJ. Frozen Concentrated Orange Juice from Brazil, Inv. No. 731-TA-326 (Final), USITC Pub. 1970 (Apr. 1987).
The Court has jurisdiction under 28 U.S.C. § 1581(c) (1982). The Court holds that Florida Citrus Mutual (FCM) does not qualify as an organization which represents producers of a "like product,” but that growers of round oranges are includable in the domestic industry which produces frozen concentrated orange juice. The Court finds that since Commerce has statutory discretion to allow amendment of a dumping petition at any time, and since Commerce may self-initiate a dumping petition, any defect in a petition filed by an orange growers’ voluntary marketing association was cured when domestic producers of the like product were added as competition-ers and Commerce was not required to start a new investigation. The Court also finds that the Commission’s majority views on like product, material injury, and threat of material injury are according to law and supported by substantial evidence on the record as a whole, but remands to the entire Commission to (1) explain how or whether the Commission considered certain fair value inventories in Brazil, and (2) reconsider the significance of inventories in the United States in light of evidence that those inventories were decreasing rather than remaining stable.
*1198Background
FCM is a voluntary cooperative marketing association of "growers of citrus fruit for processing and processors of citrus fruit.” R. 28. On May 9, 1986, FCM filed antidumping petitions with Commerce and the Commission "on behalf of the U.S. industry producing FCOJ, including growers and processors.” R. 28.
A. Proceedings at Commerce:
Ten days after the petition was filed, the Florida Citrus Processors Association, whose members accounted for approximately 85 percent of all FCOJ consumed in the United States, informed Commerce that "FCM does not represent the entire industry” and that FCM’s petition had been filed "without the knowledge of [the] Processors Association.” R. 73. This letter did not express opposition to the petition, but only informed Commerce that this particular association had not been consulted before the petition was filed. Nineteen days after the filing, the National Juice Products Association, another juice products association, sent a letter to Commerce expressing opposition to the petition on behalf of all but three of its members. R. 77, 79.
Commerce has 20 days to determine whether a petition sets forth the allegations necessary to impose antidumping duties. 19 U.S.C. § 1673a(c) (1982). On the twentieth day, Commerce determined that FCM’s petition met the requirements of 19 U.S.C. § 1673a(b)(l) (Supp. IV 1986) to initiate an antidumping investigation of FCOJ from Brazil pursuant to 19 U.S.C. § 1673a(c) (1982). Frozen Concentrated Orange Juice From Brazil: Initiation of Antidumping Duty Investigation, 51 Fed. Reg. 20,321 (June 4, 1986). Commerce identified the imports under investigation as
FCOJ in a highly concentrated form for transport and further processing, sometimes referred to as frozen concentrated orange juice for manufacturing, currently provided for under the Tariff Schedules of the United States (TSUS) item number 165.29.
Id. at 20,321.
After Commerce initiated the investigation, various domestic producers of FCOJ which owned groves that produce oranges for processing into FCOJ requested that they be joined with FCM as a "co-petitioners” because they were producers of a like product. R. 900, 929, 1065, 1068, 1144. Commerce stated in its preliminary determination that because these co-petitioners were "producers of the like product, and thus interested parties, we need not address whether [FCM], an association of orange growers, has standing in this case.” Frozen Concentrated Orange Juice From Brazil: Preliminary Determination of Sales at Less Than Fair Value, 51 Fed. Reg. 37,618, 37,619 (Oct. 23, 1986). Commerce did not determine whether FCM had standing to initiate the investigation. Frozen Concentrated *1199 Orange Juice From Brazil: Final Determination of Sales at Less Than Fair Value, 52 Fed. Reg. 8324, 8324-26 (Mar. 17, 1987).
B. The Commission’s Determination:
In the Commission’s final determination, one commissioner found material injury and two commissioners found a threat of material injury. Two dissenting commissioners found neither an injury nor a threat of injury. Frozen Concentrated Orange Juice from Brazil, Inv. No. 731-TA-326 (Final), USITC Pub. 1970 (Apr. 1987). The an-tidumping duty order was issued following the Commission majority’s findings of material injury and threat of material injury. An-tidumping Duty Order; Frozen Concentrated Orange Juice from Brazil, 52 Fed. Reg. 16,426 (May 5, 1987).
Discussion
The issues presented for decision concern (A) FCM’s motion to intervene, (B) Citrosuco’s arguments against Commerce, and (C) Ci-trosuco’s arguments against the Commission’s determinations of material injury and threatened material injury. The standard of review of determinations of Commerce and the Commission is set forth in Negev Phosphates, Ltd. v. United States, 12 CIT 1074, Slip Op. 88-156, at 6-7 (Nov. 8, 1988).
A. Motion to Intervene:
On February 8, 1988, the Court granted leave to intervene to six domestic producers of FCOJ. The Court reserved ruling on FCM’s motion to intervene but allowed FCM to participate in the proceedings as amicus curiae.
The right to intervene in an action before the Court of International Trade is limited to "an interested party who was a party to the proceeding in connection with which the matter arose.” 28 U.S.C. § 2631(j)(l)(B) (1982). The statute does not allow for permissive intervention. Special Commodity Group on Non-Rubber Footwear From Brazil v. United States, 9 CIT 481, 482-83, 620 F. Supp. 719, 721 (1985). FCM was a party to the administrative proceedings since it filed the antidumping petition to initiate an antidumping investigation. See RSI India (Pvt.) Ltd. v. United States, 12 CIT 84, 678 F. Supp. 304, 306-07 (1988); RSI India (Pvt.) Ltd. v. United States, 12 CIT 1059, Slip Op. 88-153, at 8 (Nov. 3, 1988) (importers considered parties to the proceeding where they had filed comments one month before Commerce published its administrative review results, and Commerce had stated that it considered all of the comments received).
FCM’s motion to intervene presents the issue of whether growers of round oranges used to produce FCOJ may be considered an "interested party.” The statutory definitions of the term "interested party” include:
*1200(C) a manufacturer, producer, or wholesaler in the United States of a like product,
(D) a certified trade union or recognized union or group of workers which is representative of an indudustry engaged in the manufacture, production, or wholesale in the United States of a like product,
(E) a trade or business association a majority of whose members manufacture, produce, or wholesale a like product in the United States, and
(F) an association, a majority of whose members is composed of interested parties described in subparagraph (C), (D), or (E) with respect to a like product.
19 U.S.C. § 1677(9) (1982 & Supp. IV 1986) (emphasis added). It appears that Commerce’s regulations have not yet incorporated paragraph (F) of 19 U.S.C. § 1677(9) (Supp. IV 1986). See e.g„ 19 C.F.R. §§ 353.36(a), 353.38, 355.7(c), 355.26(a), 355.27, and 355.31 (1988).
Whether FCM may be considered an "interested party” depends on whether round oranges can be considered a "like product” to frozen concentrated orange juice.
A "like product” is defined as "a product which is like, or in the absence of like, most similar in characteristics and uses with, the article subject to an investigation * * *.” 19 U.S.C. § 1677(10) (1982). See also Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 634, 693 F. Supp. 1165 (1988); S. Rep. No. 249, 96th Cong., 1st Sess. 90-91, reprinted in 1979 U.S. Code Cong. & Admin. News 381, 476-77. In previous cases the court has denied intervention to trade associations where a majority of members did not produce a product like the product under investigation. Matsushita Elec. Indus. Co. v. United States, 2 CIT 254, 255-56, 529 F. Supp. 664, 667-68 (1981) (intervention denied to an organization in which only 4 of 14 members produced a like product). See also Special Commodity Group on Non-Rubber Footwear From Brazil, 9 CIT at 483-84, 620 F. Supp. at 721-22 (trade organization allowed to intervene when a majority of members during administrative proceeding were producers or wholesalers of the like product, but less than a majority was when case brought before the Court of International Trade); American Grape Growers v. United States, 9 CIT 103, 106, 604 F. Supp. 1245, 1249 (1985) (trade association with 5 of 10 members that produced a like product allowed to intervene).
The antidumping petition reveals that FCM is an association of 12,960 Florida growers of oranges that are processed into FCOJ or fresh orange juice, and that approximately 85% of the round oranges grown in Florida are used to produce FCOJ. R. 28; Frozen Concentrated Orange Juice From Brazil, USITC Pub. 1873, at A-10. In American Grape Growers v. United States, 9 CIT 103, 104, 604 F. Supp. 1245, 1247 (1985), the court found that grape growers did not produce a "like product” in an antidumping proceeding involving table wine, because grapes and table wine were obviously different products. Id. at 103, 604 F. Supp. at 1247. Congress later amended *1201the definition of a "domestic industry” to include grape growers in the case of wine and grape products. See 19 U.S.C. § 1677(4)(A) (Supp. IV 1986); H.R. 1091, 98th Cong., 2d Sess. 15-16, reprinted in 1984 U.S. Code Cong. & Admin. News 5206, 5217-18. Congress did not amend the definition of "like product.”
As with grapes and wine, the Court holds that round oranges and frozen concentrated orange juice are different products for a "like product” determination. The Court finds that FCM is not an association which has a majority of members that produce a like product as compared to FCOJ, and that this is not a situation where there is an "absence of like” which would allow the Court to analyze the characteristics and uses of round oranges. FCM’s motion to intervene is denied. The Court thus turns to the merits of this action.
B. Proceedings at Commerce:
Antidumping proceedings are usually initiated by an "interested party” who files a petition with Commerce and the Commission "on behalf of an industry.” 19 U.S.C. § 1673a(b) (Supp. IV 1986); 19 C.F.R. § 353.36 (1988); E.T. Rossides, U.S. Import Regulation 207 (1986). Citrosuco argues that Commerce should have dismissed FCM’s antidumping petition because it did not represent producers, manufacturers, or wholesalers of a "like product” and did not file "on behalf of’ the domestic industry. See Kassinger, Antidumping Duty Investigations 12 & 53 n.23A, in Law & Practice of United States Regulation of International Trade (C.R. Johnston, Jr., ed., 1987).
1. Like Product:
The Court has already found in ruling on FCM’s motion to intervene that FCM does not represent producers of a like product. Ci-trosuco urges that Commerce should have dismissed FCM’s petition, rather than adding Alconia and other producers of FCOJ as "copeti-tioners.” Citrosuco argues that Commerce’s continuation of the investigation without ruling on the challenge to FCM’s standing was unlawful and unfair. Citrosuco contends that the late addition of the co-petitioning producers neither vested FCM as an interested party nor empowered Commerce to continue an investigation which Citrosuco sees as being originally invalid.
No presumption of standing exists in an area where Congress has provided explicit instructions on the subject. Zenith Radio Corp. v. United States, 5 CIT 155, 156 (1983). Legislative history to the Trade Agreements Act of 1979 affirms that:
The standing requirements in section 732(b)(1) for filing a petition implement the requirements of Article 5(a) of the Agreement [on Implementation of Article VI of the General Agreement on Tariffs and Trade (the GATT Antidumping Code)]. The [Senate Committee on Finance] intends that the standing requirements be administered to provide an opportunity for relief *1202for an adversely affected industry and to prohibit petitions filed by persons with no stake in the result of the investigation.”
S. Rep. No. 249, 96th Cong., 1st Sess. 63, reprinted in 1979 U.S. Code Cong. & Admin. News 381, 449.
The GATT Antidumping Code identified in the legislative history provides in relevant part that "[a]n investigation to determine the existence, degree and effect of any alleged dumping shall normally be instituted upon a written request by or on behalf of the industry affected.” Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, Apr. 12, 1979, art. 5(1), 31 U.S.T. 4919, 4928, T.I.A.S. No. 9650. The GATT Antidumping Code also interprets the term "domestic industry” as referring to "the domestic producers as a whole of the like products or to those of them whose collective output of the products constitutes a major proportion of the total domestic production of those products * * *.” Id. art. 4(1), 31 U.S.T. at 4927.
Commerce contends that it was unnecessary to decide whether FCM had standing to initiate the antidumping investigation because any defect in FCM’s petition was cured by adding the co-petitioners. Commerce asserts that allowing the addition of the co-petitioners was an action within its statutory discretion.
Even without the addition of the co-petitioners, Commerce could have self-initiated an antidumping investigation in the absence of a proper written request if Commerce had sufficient evidence to warrant formal investigation into whether there exists (a) dumping, (b) material injury and (c) a causal link between the dumped imports and the injury. 19 U.S.C. § 1673a(a)(l) (Supp. IV 1986); 19 C.F.R. § 353.35 (1988); Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, Apr. 12,1979, art. 5(1), 31 U.S.T. 4919, 4928, T.I.A.S. No. 9650. See also Kassinger, Antidump-ing Duty Investigations 11, in Law & Practice of United States Regulation of International Trade (C.R. Johnston, Jr., ed., 1987).
Commerce’s ability to self-initiate an antidumping investigation contrasts with situations where a properly-filed document is an absolute or jurisdictional prerequisite to an agency’s power to act. See, e.g., Leedom v. International Union of Mine, Mill & Smelter Workers, 352 U.S. 145, 148-49 (1956) (NLRB was "under a duty to determine whether a filing has been made by each person” required at that time to affirm non-membership in the Communist party, "since its power to act on union charges is conditioned on the filing of necessary affidavits”).
Commerce has discretion to permit the addition of co-petitioners who produce the domestic product like the imported product under investigation. A petition may be amended at such time and upon such conditions as Commerce and the Commission may permit. 19 U.S.C. § 1673(b)(1) (Supp. IV 1986). Commerce has permitted petition amendments on matters of standing in previous cases. In Final Affirmative Countevailing Duty Determination; Live Swine and *1203 Fresh, Chilled and Frozen Pork Products from Canada, 50 Fed. Reg. 25,097 (June 17, 1985), aff’d in part, Alberta Pork Producers’ Mktg. Bd. v. United States, 11 CIT 563, 669 F. Supp. 445 (1987), aff’d following remand, 12 CIT 262, 683 F. Supp. 1398 (1988), a countervailing duty petition against pork products was filed by an association of hog growers. Commerce initiated the investigation and the Canadian respondents challenged the growers’ standing. Subsequently, seven United States pork packers entered the investigation as co-petitioners, and standing was no longer disputed. In Final Negative Countervailing Duty Determination; Certain Textile Mill Products and Apparel from Malaysia, 50 Fed. Reg. 9852 (Mar. 12, 1985), various associations of textile manufacturers and workers filed a countervailing duty petition against certain textile products and apparel. The Malaysian respondents challenged the petitioners’ standing. Commerce determined that while the association of manufacturers lacked standing, eight individual United States manufacturers who were substituted as parties, as well as the unions which had joined in filing the petition, did have standing.
The Court finds the additon of the co-petitioners who produced a "like product” cured any defect in FCM’s petition, and that Commerce had discretion to allow the petition to be amended. For Commerce to have dismissed FCM’s petition for lack of standing, and to have required the processors to file a new petition several months later, would have elevated form over substance and fruitlessly delayed the antidumping investigation against Brazilian FCOJ when Congress clearly intended these cases to proceed expeditiously. See S. Rep. No. 249, 1st Sess. 37, reprinted in 1979 U.S. Code Cong. & Admin. News 381, 423.
It would be unreasonable to impose a judicial requirement that Commerce terminate pending antidumping investigations and begin a new investigation. As the court has already stated:
¡Since Commerce is authorized to commence an antidumping duty investigation sua sponte whenever it determines that an investigation is warranted based upon available information, 19 U.S.C. § 1673a(a) (1982), it would be unreasonable to require that Commerce terminate an investigation commenced after the filing of a petition by an interested party when, despite inaccuracies in the petition, it finds further evidence of sales at less than fair value.
Luciano Pisoni Fabrica Accessori v. United States, 10 CIT 424, 427, 640 F. Supp. 255, 258 (1986). Accordingly, had Commerce determined to self-initiate, it would have possessed the authority to continue the FCOJ investigation on the same timetable followed in this case.
The Court holds that since Commerce has statutory discretion to allow amendment of a dumping petition at any time, FCM’s petition was cured when domestic producers of the like product were added *1204as co-petitioners and since Commerce may self-initiate a dumping petition, Commerce was not required to start a new investigation.
2. "On behalf of” a Domestic Industry:
Citrosuco also maintains that Commerce should not have initiated the dumping investigation because the petition was not filed "on behalf of an industry” under 19 U.S.C. § 1673a(b)(l) (Supp. IV 1986).
(a)
Early in the investigation Commerce received two letters stating that FCM did not represent a majority of domestic industry producing FCOJ. Commerce states that it did not consider these letters because they were sent within twenty days of the petition’s filing when Commerce is prohibited from considering outside communications while it decides whether to initiate an investigation. The Court finds that Commerce’s refusal to consider outside communications during the twenty-day period was correct. See United States v. Roses, Inc., 1 Fed. Cir. (T) 39, 46, 706 F.2d 1563, 1568-69 (1983); Luciano Pisoni Fabrica Accessori v. United States, 10 CIT 424, 426, 640 F. Supp. 255, 257 (1986).
(b)
At oral argument, Citrosuco conceded that Commerce had acted reasonably in the first twenty days of the investigation. Citrosuco argues, however, that Commerce should have dismissed the petition after that time, perhaps on the twenty-second day of the investigation, because an alleged majority of the United States processing industry opposed the petition. Citrosuco argues that companies opposing the petition produced 52.6 percent of all FCOJ or FCOJM, processed 57.1 percent of all Florida oranges processed, and accounted for 52.4 percent of all oranges processed nationwide. R. 1869. Citrosuco argues that when these figures are added to the responses of companies that took no position on the petition, the record is even clearer that FCM did not petition "on behalf of’ the domestic industry because "the total percentages of companies opposing the petition become 57.8 percent of all FCOJ or FCOJM produced, 64.5 percent of Florida oranges processed, and 59.1 [percent] of all oranges processed nationally.” Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 36-37.
Commerce has discretion to rescind an investigation when it determines that the petition is not supported by a majority of the domestic industry. Oregon Steel Mills Inc. v. United States, Nos. 88-1242 and 1243 (Fed. Cir. Nov. 23, 1988); Gilmore Steel Corp. v. United States, 7 CIT 219, 585 F. Supp. 670 (1984).
In Oregon Steel Mills, the Court of Appeals for the Federal Circuit held that Commerce has discretion in an administrative review to revoke an outstanding antidumping duty order in the absence of industry support to continue the order. The court did not hold that Commerce is required to dismiss petitions that are not proven to be *1205affirmatively supported by the domestic industry. Similarly, in Gilmore Steel Corp., the court found that Commerce had discretion to dismiss a petition that was not supported by a majority of the national domestic industry, and was thus not "on behalf of’ a United States industry. Gilmore stated that the
"interested party” criterion of section 1673a(b) is thus the first in a two-step shifting process — not only must a petitioner be a member of the affected industry, i.e., be an "interested party,” it must also show that a majority of that industry backs its petition.
Id. at 226, 585 F. Supp. at 676.
Citrosuco interprets Gilmore to impose a requirement that Commerce must dismiss petitions that are not proven to be affirmatively supported by a majority of the domestic industry. In the final determination, however, Commerce recognized that
there is nothing in the statute, its legislative history, or our regulations which requires that petitioners establish affirmatively that they have the support of a majority of their industries. In many cases, such a requirement would be so onerous as to preclude access to import relief under the antidumping and countervailing duty laws.
52 Fed. Reg. at 8325. See also Final Affirmative Countervailing Duty Determination; Certain Stainless Steel Hollow Products from Sweden, 52 Fed. Reg. 5794 (Feb. 26, 1987); Final Affirmative Countervailing Duty Determination; Certain Fresh Atlantic Groundfish from Canada, 51 Fed. Reg. 10,041 (Mar. 24, 1986).
The Court finds that Gilmore and Oregon Steel Mills stand for the proposition that Commerce has discretion to dismiss, but is not required to dismiss, petitions that are not shown to be actively supported by a majority of the domestic industry. Neither the statute nor Commerce’s regulations require a petitioner to establish affirmatively that it has the support of a majority of a particular industry, and the Court declines to impose such a requirement.
(c)
Commerce excluded opposition from parties that were either related to the Brazilian exporters or "those firms whose imports of FCOJ from Brazil exceed 50 percent of their total production.” 52 Fed. Reg. at 8326. Commerce premised the exclusions on the "related parties” statute:
When some producers are related to the exporters or importers, or are themselves importers of the allegedly subsidized or dumped merchandise, the term "industry” may be applied in appropriate circumstances by excluding such producers from those included in that industry.
19 U.S.C. § 1677(4)(B) (1982).
*1206Citrosuco argues that Commerce acted ultra vires in using this statute to exclude producers who import more than 50 percent of their total production from Brazil, because the legislative history shows that the provision was intended for the Commission rather than Commerce. S. Rep. No. 249, 96th Cong., 1st Sess. 82-83, reprinted in 1979 U.S. Code Cong. & Admin. News 381, 468-69.
The Court is unwilling to upset Commerce’s interpretation of the statute based only on this legislative history. See Brechteen Co. v. United States, 854 F.2d 1301, 1304 (Fed. Cir. 1988). Commerce has taken the position that firms with large imports of the allegedly dumped or subsidized merchandise may be excluded from the definition of the domestic industry, because they inherently lack the stake in the final investigation being pursued by the petitioner. See, e.g., Preliminary Affirmative Countervailing Duty Determination: Certain Fresh Atlantic Groundfish from Canada, 51 Fed. Reg. 1010, 1011 (Jan. 9, 1986); Preliminary Negative Countervailing Duty Determination; Low-Fuming Brazing Copper Wire and Rod from South Africa, 50 Fed. Reg. 21,328, 21,329 (May 23, 1985); Final Affirmative Countervailing Duty Determination and Countervailing Duty Order; Fabricated Automotive Glass From Mexico, 50 Fed. Reg. 1906, 1910 (Jan. 14, 1985).
Congress entrusted Commerce with discretion to administer the international trade laws, see Oregon Steel Mills Inc. v. United States, Nos. 88-1242 and 1243, at 9 (Fed. Cir. Nov. 23, 1988), although the discretion is not "unbounded.” Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 167 (1962); Freeport Minerals Co. v. United States, 4 Fed. Cir. (T) 16, 18, 776 F.2d 1029, 1031 (1985). The related parties statute appears in a list of definitions which apply to both Commerce and the Commission. In measuring the extent of industry support for an antidumping petition, the Court finds it reasonable for Commerce to apply 19 U.S.C. § 1677(4)(B) (1982) to exclude producers who derive a majority of their production from the imports under investigation.
(d)
Applying the related parties provision to exclude those producers whose imports of Brazilian FCOJ equaled fifty percent of their total production from the definition of industry for the purposes of determining industry support, Commerce found that the processors opposed to the antidumping petition accounted for only 38.64 percent of domestic FCOJ production. 52 Fed. Reg. at 8325-26. Citrosuco does not challenge Commerce’s calculations of this figure. Conf. R. 1718. After finding that the processors opposing the case did not represent a majority of the processing sector, Commerce found that it did not need to address whether the industry should also be defined for standing purposes to include growers. 52 Fed. Reg. at 8326.
*1207Commerce properly calculated that if the domestic industry was defined to include only domestic production, a majority of the domestic industry did not oppose the amended petition. The Court finds that Commerce’s interpretation and application of the related parties statute to be reasonable under the facts of this investigation, see Oregon Steel Mills Inc. v. United States, Nos. 88-1242 and 1243, at 13 (Fed. Cir. Nov. 23, 1988), and Commerce’s finding that the petition was filed on behalf of the domestic industry to be supported by substantial evidence on the record as a whole and according to law. The Court thus turns to the Commission’s determination.
c. The Commission’s Determination
In the final determination, Commissioners Eckes and Lodwick found a threat of material injury and Commissioner Rohr found material injury.
1. Definitions of the Like Product:
In their joint opinion finding a threat of material injury, Commissioners Eckes and Lodwick defined the product like the imported product under investigation as frozen concentrated orange juice for manufacturing (FCOJM). In his opinion finding material injury, Commissioner Rohr defined the like product as including frozen concentrated orange juice for retail (FCOJR) and single strength orange juice (SSOJ), as well as FCOJM. Citrosuco argues the Commission’s determination cannot stand because the majority did not agree on one definition of the domestic product that was like the imports under investigation.
(a)
Commissioners Eckes and Lodwick defined the "like product” as frozen concentrated orange juice for manufacturing (FCOJM):
FCOJM is highly concentrated form of FCOJ. Domestic extractors manufacture FCOJM by extracting orange juice from oranges, removing water from the orange juice, and then freezing the remaining concentrate. The resulting concentrate can be reconstituted into orange juice by adding water. To form reconstituted orange juice, between six to seven units of water must be added to each unit of FCOJM. [By comparison, retail FCOJ only requires that three units of water be added to it in order to form reconstituted orange juice.] FCOJM is stored in bulk, either in 55 gallon drums or in tanks which can hold up to 100,000 gallons or more.
USTIC Pub. 1970 at 4 & n.7. See also id. at RIO, R13, R14.
In defining the like product as FCOJM, Commissioners Eckes and Lodwick relied upon the fact that the imported article subject to investigation is FCOJM and "[djomestic FCOJM is thus the product which is 'identical’ to the imported product.” USITC Pub. 1970 at 6; 51 Fed. Reg. at 20,321. While identical in concentration levels, a *1208given quantity of either domestic or Brazilian FCOJM might differ in color, flavor (brix acid ratio), or defect characteristics. USITC Pub. 1970 at R15-16. The commissioners considered but discounted the significance of these variations because they did not appear to have a measureable effect on prices. USITC Pub. 1970 at 6 n.17.
The commissioners also analyzed the characteristics of FCOJM to determine whether FCOJR and SSOJ should be included in the like product definition:
Extractors make very little SSOJ or FCOJR from oranges, while all FCOJM is made directly from oranges. FCOJM is easier to store and transport than SSOJ or FCOJR. FCOJM can be stored for up to three years; SSOJ cannot be stored for extremely long periods of time. FCOJM is traded on the future’s market; SSOJ and FCOJR are not. FCOJM is imported from Brazil; SSOJ and FCOJR are not. FCOJM is sold in bulk; SSOJ and FCOJR are sold at the retail level.
USITC Pub. 1970 at 6-7 (footnotes omitted). In addition, FCOJM cannot be consumed directly, as can SSOJ, but must have considerable quantities of water added before consumption. Id. at RIO. As "FCOJM is the only product imported from Brazil” and because "FCOJM differs in many respects from FCOJR and SSOJ,” Commissioners Eckes and Lodwick reasoned that "domestic FCOJM is the product which is 'like’ the imported product” and thus identified the like product as FCOJM. USITC Pub. 1970 at 7.
Commissioner Rohr had a more expanded definition of the like product, comprising not only FCOJM but also frozen concentrated orange juice for retail (FCOJR) and single strength orange juice (SSOJ):
FCOJM is an intermediate stage in the production of orange juice. It is in essence a semifinished product and should be analyzed as such. The Commission’s analysis of semifinished products is to look at the product itself as the "like product” and include the semifinished form of the product within that definition. Within this framework, the most appropriate definition of the like product would be orange juice, including FCOJM, FCOJR, and SSOJ.
USITC Pub. 1970 at 34. Commissioner Rohr supported his conclusion by noting that many processors of orange juice could not segregate data on FCOJM production in response to the Commission’s questionnaires, thus indicating that the domestic orange juice industry is not really divided among FCOJM, FCOJR, and SSOJ producers. USITC Pub. 1970 at 34-35.
Citrosuco argues that the Commission should have followed earlier determinations that the like product was simply FCOJ. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 88-89, 97-99. See Frozen Concentrated Orange Juice from Brazil, Inv. No. 751-TA-10, USITC Pub. 1623 at 11, 28, 44 (Dec. 1984); Frozen Concentrated Orange Juice from Brazil, Inv. No. 701-TA-184 *1209(Preliminary), USITC Pub. 1283 at 4 (Sept. 1982), and (Final) USITC Pub. 1406 at 3, 18 (July 1983).
As the United States Customs Court explained when discussing the Commission’s injury determinations,
each injury investigation is sui generis, involving a unique combination and interaction of many economic variables; and consequently, a particular circumstance in a prior investigation cannot be regarded by the Commission as dispositive of the determination in a later investigation.
Armstrong Bros. Tool Co. v. United States, 84 Cust. Ct. 102, 115, C.D. 4848, 489 F. Supp. 269, 279 (1980). Thus, the Commission’s determinations must be based upon an independent evaluation of the factors with respect to the unique economic situation of each product and industry under investigation. Alberta Pork Producers’ Mktg. Bd. v. United States, 11 CIT 563, 669 F. Supp. 445, 461 (1987); Maine Potato Council v. United States, 9 CIT 293, 300 n.7, 613 F. Supp. 1237, 1244 n.7 (1985). Furthermore, "each finding as to like product must be based on the particular record at issue including the arguments raised by the parties.” Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 634, 638, n.5, 693 F. Supp. 1165, 1169 n.5 (1988) (emphasis added).
Prior Commission determinations had defined the like product simply as FCOJ. In this determination, all the parties argued that the Commission should change that definition of the like product. USITC Pub. 1970 at 5. In the prior investigations the Commission had apparently not been asked to differentiate between FCOJM and FCOJR. Id. at 7.
While the Commission is not obligated to follow prior decisions if new arguments or facts are presented that support a different conclusion, see Armstrong Bros. Tool Co. v. United States, 84 Cust. Ct. 16, 36, C.D. 4838, 483 F. Supp. 312, 328 (1980), aff’d, 67 CCPA 94, C.A.D. 1252, 626 F.2d 168 (1980), this does not permit the Commission to act arbitrarily. Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 1174, Slip Op. 88-172, at 6 (Dec. 27, 1988). This is because it is also a general rule that an agency must either conform itself to its prior decisions or explain the reasons for its departure. Secretary of Agric. v. United States, 347 U.S. 645 (1954); National Coalition Against the Misuse of Pesticides v. Thomas, 258 D.C. App. 34, 809 F.2d 875, 876 (D.C. Cir. 1987); Alhambra Foundry Co. v. United States, 12 CIT 343, 685 F. Supp. 1252, 1258 (1988), aff’d following remand, 12 CIT 1110, Slip Op. 88-160, at 7-8 (Nov. 22, 1988). This rule is not designed to restrict an agency’s consideration of the facts from one case to the next, but rather it is to insure consistency in an agency’s administration of a statute.
*1210Commissioners Eckes and Lodwick did explain the reasons for departing from prior like-product definitions of FCOJ. First, the administrative record was developed in a way that enabled the Commissioners to examine the like-product definition in a manner never raised in earlier investigations. The prior investigations did not include (a) data regarding the percentages of oranges processed directly into FCOJM, FCOJR, and SSOJ, respectively, (b) information regarding the FCOJM futures market, (c) a detailed explanation of the manufacturing and reconstituting processes, or (d) certain specific details regarding the differences between FCOJM, FCOJR, and SSOJ, such as the ability to store FCOJM for long periods of time as compared to the brief shelf life of SSOJ. Second, all of the parties, including Citrosuco, asked for the definition to be refined. Third, the country under investigation only exported FCOJM, which Commissioners Eckes and Lodwick found to be the like product. The Court finds the like-product definition of Commissioners Eckes and Lodwick is according to law and supported by substantial evidence on the record as a whole.
Commissioner Rohr’s like product definition is also according to law and supported by substantial evidence on the record as a whole. Under the facts of this investigation, i.e., where Brazil exports only FCOJM, the differing definitions of the domestic industry which both include FCOJM do not render the affirmative determination contrary to law. Although the Court and the domestic industries themselves would prefer that the Commission agree on the like product, there is no statutory requirement for each of the commissioners to agree on the same like-product definition.
The possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence. Matsushita Elec. Indus. Co. v. United States, 3 Fed Cir. (T) 44, 51, 750 F.2d 927, 933 (1984). In a recent decision sustaining a determination in which four Commissioners wrote three decisions to comprise a majority determination, the court stated that in order to affirm the Commission’s determination, "the court must be able to discern from the determination that a majority of the Commission has based its conclusions upon legally sufficient reasoning.” BMT Commodity Corp. v. United States, 11 CIT 524, 667 F. Supp. 880, 882 (1987), aff’d, 852 F.2d 1285 (Fed. Cir. 1988), motion to supplement petition for writ of certiorari under seal granted, 57 U.S.L.W. 3375 (S. Ct. Nov. 28, 1988). Additionally, an expectation that Commissioners would file concurring opinions is expressed in the House Ways and Means Committee’s Report on the Trade Agreements Act of 1979: "The Committee intends that the [Commission] determination as well as any dissenting or separate views of individual Commissioners, be specific in its statement of findings of fact and conclusions of law.” H.R. Rep. No. 317, 96th Cong., 1st Sess. 46 (1979) (emphasis added). This desire for detailed findings of concurring Commissioners would be pointless if *1211the existence of differing views precluded courts from sustaining Commission determinations.
2. Domestic Industry:
The Commission is required to determine whether a "domestic industry” is materially injured or threatened with material injury. 19 U.S.C. § 1673 (1982 & Supp. IV 1986). In this case, the Commission faced the issue of whether growers of round oranges, a raw agricultural product, should be regarded as part of the industry that produces the processed product under investigation.
a. The Two-Factor Test for Agricultural Products:
The Commissioners examined (1) the extent to which the raw product enters into a single, continuous line of production resulting in the processed product, and (2) the degree of commonality of economic interest between growers and producers. The Commission has applied this two-part analysis in previous determinations. See, e.g., Certain Fresh Atlantic Groundfish from Canada, Inv. No. 701-TA-257 (Final), USITC Pub. 1844 (May 1986); Certain Tomato Products from Greece, Inv. No. 104-TAA-23, USITC Pub. 1594 (Oct. 1984); Certain Red Raspberries from Canada, Inv. No. 731-TA-196 (Preliminary), USITC Pub. 1565 (Aug. 1984); Certain Table Wine from France and Italy, Inv. Nos. 701-TA-210 et al. (Preliminary), USITC Pub. 1502 (Mar. 1984); Lamb Meat from New Zealand, Inv. No. 701-TA-80 (Preliminary), USITC Pub. 1911 (Nov. 1981).
The court in American Grape Growers v. United States, 9 CIT 103, 604 F. Supp. 1245 (1985), cited with approval a number of Commission determinations holding that growers met this first factor, which is alternatively referred to as the "substantially all” factor because it examines whether substantially all of the raw agricultural product is processed into a single end product. The court indicated that percentages between 69 and 100 percent satisfy this factor. Conversely, the court upheld the Commission’s finding that 55 percent was insufficient in the circumstances of the case of grape growers to meet this first factor. Id. at 104, 604 F. Supp. at 1248.
The Commission’s second factor, the economic integration analysis, was approved in National Pork Producers’ Council v. United States, 11 CIT 398, 661 F. Supp. 633, 638 (1987), where the court found that the Commission’s determination of insufficient economic integration was supported by substantial evidence and was in accordance with law. The court found that the Commission
did not exceed its discretion in determining the relevant domestic industry for investigation by requiring a showing of economic integration as a prerequisite to determining whether swine growers should be included in the domestic industry * * *.
Id. at -, 661 F. Supp. at 641.
In this action the Commission identifies "sound reasons” underlying its two-part analysis:
*1212Both factors provide the framework for analyzing whether growers and processors comprise a single industry producing the processed product, that is, whether the production of growers is so devoted to the processed good and the economic interests of the growers are so enmeshed with that of the processors that growers should be considered producers of the processed like product. The Commission uses the two factors to determine whether there is merely a simple buyer/supplier relationship between growers and processors or an inter-relationship which involves inextricably intertwined economic interests. If substantially all of the raw product goes to production of the processed good, growers may be fairly regarded as the producers of the processed good, since those growers are essentially operating as producers of the processed product. If, however, growers may supply their product to processors of several different processed products, then their economic interests will be less intertwined with the interests of the processors of any particular product. The Commission also finds evidence of a single industry of interdependent participants when, in addition, there is common ownership of companies which grow a raw agricultural product and then process that product, or contractual arrangements that require the sharing of risks between growers and processors. When there is sufficient evidence of intertwined interests, the relief requested will go to growers as well as processors, as producers of the like product, and their interests should be considered in determining whether a domestic industry is sufficiently injured or threatened with material injury to justify relief under the statute. In such cases, the Commission includes all of the inter-related participants in a single definition of domestic industry.
Memorandum of United States International Trade Commission in Response to Plaintiffs Motion for Review Upon the Agency Record, at 35-36.
Additionally, the Omnibus Trade and Competitiveness Act of 1988 amended the definition in 19 U.S.C. § 1677(4) of an industry producing processed agricultural products contained by adding a new subparagraph (E). This additional subparagraph provides that unless the United States Trade Representative informs Commerce and the Commission that the subparagraph is inconsistent with international obligations of the United States, producers or growers of a raw agricultural product may be considered part of the industry producing a processed agricultural product under investigation if:
(I) the processed agricultural product is produced from the raw agricultural product through a single continuous line of production; and
(II) there is a substantial coincidence of economic interest between the producers or growers of the raw agricultural product and the processors of the processed agricultural product based upon relevant economic factors, which may, in the discretion of the Commission, include price, added market value, or other economic interrelationships (regardless of whether such coinci*1213dence of economic interest is based upon any legal relationship).
Omnibus Trade and Competitiveness Act of 1988, P.L. 100-418, § 1326(a), 102 Stat. 1107, 1203 (1988). This language conforms to the practice that the Commission employed in this determination, although the amended statute applies only to investigations which are initiated after August 23, 1988. Id.§ 1337(b), 102 Stat. at 1211.
i. Single, Continuous Line of Production:
All of the Commissioners, including those in dissent, found that the first criterion for including growers of round oranges was satisfied. Commissioners Eckes and Lodwick found that an average 73 percent of all United States round oranges are processed into some form of juice, that 96 of all oranges processed are round oranges, and that 84 percent of all juice oranges (defined as all round oranges except navel oranges) are processed into some type of orange juice. USITC Pub. 1970 at 11. The remainder of these oranges are sold on the fresh fruit market. Id. at R7. The Commissioners also found that of the oranges processed in 1985/86, the vast majority were processed into FCOJM. Id. at 11. Commissioners Eckes and Lodwick thus found that the growers of both juice oranges and round oranges satisfy the criterion of whether substantially all of the raw agricultural product enters a single, continuous line of production. The Commissioners excluded from the definition of the domestic industry those growers of other types of specialty oranges cultivated primarily for fresh fruit markets, because inclusion of those growers would divert attention away from growers who are primarily engaged in producing oranges for FCOJM. Id. at n.35.
Commissioner Rohr also articulated his conclusions on the question of "what percentage of oranges grown by growers are dedicated to the particular end use represented by the product of the processors, orange juice?” Id. at 37. Commissioner Rohr noted that approximately 72 percent of all fruit within the United States Department of Agriculture definition is processed into orange juice. Id. He further observed that the percentage rises to 73 if specialty fruit, such as tangerines and tángelos, are excluded. Id. Commissioner Rohr also found that approximately 84 percent of juice oranges were processed into juice, and determined that 84 percent was sufficiently high to meet the continuous line of production requirement and include growers in the definition of the domestic industry. Id. at 37-38. He also observed that the percentage of oranges used to produce FCOJM was higher than the 69 percent of tomatoes used in processing which the Commission found to be sufficient in Certain Tomato Products from Greece, Inv. No. 104-TAA-23, USITC Pub. 1594 (Oct. 1984). Id. at 38 n.12.
*1214Another relevant consideration in determining whether the Commission’s first factor has been satisfied is whether there is in fact a single line of production, so that production from the raw product is devoted to the particular like product, rather than adapted to the creation of multiple products. Cf. Lamb Meat from New Zealand, Inv. No. 701-TA-80 (Preliminary), USITC Pub. 1191, at 8-9 (Nov. 1981) (describing the need for only one end product from one raw agricultural product). See also Sugar Content of Certain Articles from Australia, Inv. No. 104-TAA-26, USITC Pub. 1748 (Sept. 1985) (Commission found there was no single, continuous line of production for clingstone peaches that were processsed not only into canned clingstone peaches with added sugar but also into both canned mixed fruit with added sugar and canned peaches without added sugar). In American Grape Growers, the Commission excluded grape growers from the domestic wine industry because the raw agricultural product was processed into other products, such as raisins and grape juice. See American Grape Growers v. United States, 9 CIT 103, 604 F. Supp. 1245 (1985). In response to the Commission’s determination underlying the court’s decision in American Grape Growers, Congress found that the Commission’s application of its analysis for inclusion of grape growers had been applied too restrictively and accordingly amended the definition of domestic industry specifically to protect domestic grape growers:
The term "industry” means the domestic producers as a whole of the like product, or those producers whose collective output of the like product constitutes a major proportion of the total domestic production of that product; except that in the case of wine and grape products subject to investigation under this subtitle, the term also means the domestic producers of the principle raw agricultural product (determined on either a volume or value basis) which is included in the like domestic product, if those producers allege material injury, or threat of material injury, as a result of imports of such wine and grape products.
19 U.S.C.§ 1677(4)(A) (Supp. IV. 1986). See also H.R. 1091, 98th Cong., 2d Sess. 15-16, reprinted in 1984 U.S. Code Cong. & Admin. News 5206, 5217-18.
The Court holds that the Commissioners’ finding that substantially all of the oranges produced entered a single, continuous line of production resulting in the processed FCOJ is supported by substantial evidence on the record as a whole, is according to law, and is consistent with prior Commission determinations. The percentages of round and juice oranges processed into the like product in this case, whether it is defined as FCOJM or all orange juice, are higher than the 55 percent found to be inadequate in American Grape Growers, 9 CIT at 104-05, 604 F. Supp. at 1248, and higher than the 69 percent figure for tomato products noted by Commissioner Rohr and cited with approval in American Grape Growers, 9 CIT at 104 n.5, 604 F. Supp. at 1248 n.5.
*1215ii. Economic Integration of Growers and Processors:
In examining the second criterion, sufficient economic integration, the Commission examined the sales arrangements between the growers and processors, the relationship between orange prices and orange juice prices, and correlations between the finanical performance of growers and processors.
With respect to the sales arrangements between growers and processors, the Commission found that the vast majority of sales arrangements were not arms-length cash sales. Instead, the Commission found that the growers and processors used cooperatives, full and partial participation plans, and intracompany transfers to sell oranges for processing. Commissioners Eckes and Lodwick described in detail the economic nature of the relationships, based upon information collected in the Commission’s investigation:
Growers that are members of a cooperative deliver their oranges to a cooperative-owned extracting plant for processing and marketing. As payment for their oranges, the grower-members receive the net proceeds obtained by the cooperative from the FCOJM. The share of the proceeds obtained by each grower-member is determined by the amount of oranges delivered to the cooperative. In a cooperative, the return received by the grower is thus directly connected with the cooperative-extractor’s return from the FCOJM produced.
Under a participation plan a grower agrees to sell his oranges to an extractor in exchange for a return based upon the final amount received by the extractor from the FCOJM manufactured from the grower’s oranges. There are two types of * * * plans, "full” * * * and "partial” participation plans. The return received by growers involved in a full participation plan is determined almost solely by the final price received by the returns generated from the FCOJM produced from the grower’s oranges. Growers in a partial plan receive a guaranteed "floor” price for their oranges, and receive at least part of any additional amount received by the extractors from the FCOJM. In both full and partial participation plans, the growers[’] return is tied to the extractors’ return, and both growers and extractors share some of the risk involved in the manufacture and sale of the FCOJM.
Additionally, many of the large extractors own their own groves. These extractors not only purchase oranges through * * * cooperative arrangements or through participation plans from other growers, but also process the oranges grown in their own groves. In the case of extractor-owned groves, there is again a direct connection between the extractor and the grower.
USITC Pub. 1970, at 12-13. In the vast majority of instances, growers shared the risk with the processors, either by being "owners” of the processors in cooperatives, being owned by processors, or by contractual arrangements which directly link a growers’ return on his *1216fruit to the processor’s return on orange juice. Commissioners Eckes and Lodwick also found that from information on how oranges were purchased, over 90 percent of the sales of large orange growers were made using some non-cash arrangement. Id. at 13-14.
All three Commissioners noted that the percentage of cash sales was higher in the three growing seasons preceding 1985/86, but attributed these higher percentages to a series of freezes in Florida during those years. Id. at 14; 39 n.13. As Commissioners Eckes and Lodwick explained, the freezes "caused extractors to buy a large amount of the oranges on the cash market to guarantee continued sources of supply and to maintain the capacity utilization of their extracting equipment.” Id. at 14. Therefore, Commissioners Eckes and Lodwick concluded that the lower cash sales figures for 1985/ 86 indicated that the processors and growers are returning to closer economic links. Id.
In addition to viewing sales arrangements, the Commissioners also examined the correlation between orange prices and orange juice prices to determine whether there was a commonality of economic interest between growers and processors. Id. at 14-15; 39-40. The figures examined establish similar patterns of increases and decreases over the last decade. See id. at R81. Commissioner Rohr also identified a clear relationship between key financial indicators of processors and growers, including similar trends in the unit value of shipments between growers, corporate processors, cooperative processors between 1983 and 1986. See id. at 39-40, R45, R47, R57, R60.
Citrosuco argues that if the Commission had considered the existence of some opposition by processors of orange juice to the an-tidumping petition, it would have been "impossible” for the Commission to make a finding of sufficient economic integration. Brief in Support of Plaintiff’s Motion for Judgment on the Agency Record, at 118.
The determination shows that Commissioners Eckes and Lodwick did consider this opposition but concluded that it did not alter their decision to include growers as part of the domestic industry. USITC Pub. 1970, at 15. The Commissioners explained that the processors in opposition are more dependent upon Brazilian imports than those supporting the petition, and that those in opposition to the petition sell most of their production as FCOJR and SSOJ rather than FCOJM (the like product according to Commissioners Eckes and Lodwick). On the basis of these observations, Commissioners Eckes and Lodwick determined that "the extractors in opposition do not adequately reflect the economic interests of all the extractors and so their opposition should not be given undue emphasis in deciding the issue of the commonality of economic interests between growers and extractors.” Id. at 15 n.49.
*1217The Court finds that the Commission considered and discounted the opposition to the petition. The Court finds Citrosuco’s argument that the Commission did not consider industry opposition to be without merit. Citrosuco may wish to have a different result from this consideration, but it is not this Court’s function on review to reweigh evidence. See Alberta Pork Producers’ Mktg. Bd. v. United States, 11 CIT 563, 669 F. Supp. 445, 449 (1987). Based upon the facts and analyses in the Commission’s determination, Commissioners Eckes, Lodwick, and Rohr determined that the Commission’s second factor in the two-part analysis supported inclusion of the growers in the domestic industry.
The Court finds that the Commission’s determination to include growers of round oranges in the definition of the domestic industry producing frozen concentrated orange juice is supported by substantial evidence on the record as a whole and according to law.
3. Threat of Material Injury:
Commissioners Eckes and Lodwick determined that a domestic industry was threatened with material injury by reason of dumped imports of FCOJ from Brazil. USITC Pub. 1970 at 3.
(a)
Citrosuco first attempts to discredit the Commissioners’ finding of threatened material injury by arguing that previous findings of threat in other investigations of FCOJ were incorrect and have not been substantiated by subsequent findings of material injury. See Frozen Concentrated Orange Juice from Brazil, Inv. No. 701-TA-184 (Final), USITC Pub. No. 1406, (July 1983); Frozen Concentrated Orange Juice from Brazil, Inv. No. 751-TA-10, USITC Pub. No. 1623, (Dec. 1984).
Each threat determination is based upon the facts of the particular case because each case involves a unique set of circumstances and a different period of investigation. Findings in prior, related determinations regarding threat or material injury are generally not dispositive on subsequent determinations of threatened material injury. See Armstrong Bros. Tool Co. v. United States, 84 Cust. Ct. 16, C.D. 4838, 483 F. Supp. 312 (1980), aff’d, 67 CCPA 94, C.A.D. 1252, 626 F.2d 168 (1980).
Citrosuco’s argument that the findings of threat in this investigation are improper because earlier findings of threat have not materialized is irrelevant. The Court is not charged with determining whether earlier findings of threatened material injury have, in fact, ripened into material injury of the domestic industry.
(b)
Commissioners Eckes and Lodwick considered the relevant statutory factors and provided a detailed analysis of the relevant facts and trends.
*121819 U.S.C. § 1677(7)(F) does not define "threat” but instructs the Commission to consider several economic factors in assessing the threat of material injury. In determining whether an industry in the United States is threatened with material injury by reason of imports (or sales for importation) of any merchandise, the Commission is directed to consider, among other relevant economic factors:
(II) any increase in production capacity or existing unused capacity in the exporting country likely to result in a significant increase in imports of the merchandise to the United States,
(III) any rapid increase in United States market penetration and the likelihood that the penetration will increase to an injurious level,
(IV) the probability that imports of the merchandise will enter the United States at prices that will have a depressing or suppressing effect on domestic prices of the merchandise,
(V) any substantial increase in inventories of the merchandise in the United States,
(VI) the presence of underutilized capacity for producing the merchandise in the exporting country,
(VIII) the potential for product-shifting if production facilities owned or controlled by foreign manufacturers, which can be used to produce products subject to investigation(s) under section 1671 or 1673 of this title or to find orders under section 1671e or 1673e of this title, are also used to produce the merchandise under investigation.
19 U.S.C. § 1677(7)(F)(i) (Supp. IV. 1986). While the Commission must always consider each of these factors, it is not always necessary for the Commission to discuss every statutory injury or threat of injury factor. Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 1174, Slip Op. 88-172, at 10 (Dec. 27, 1988).
i. Production Capacity:
Commissioners Eckes and Lodwick found that Brazilian extracting companies increased their production capacity between 1984 and 1987 both in terms of the quantity of oranges they can process per hour and in terms of the pounds of solids they can extract per hour. USITC Pub. 1970 at 26. The Commissioners noted that this increase in production capacity coincided with major increases in the number of orange trees planted in Brazil, which the Commission found gave the Brazilian extractors the ability to increase their FCOJM production as a result of more efficient processing methods. Id.
Citrosuco argues that the Commissioners’ reliance on evidence of new tree plantings is improper or unpersuasive (1) because some trees may die due to unforeseen natural disasters and (2) because the existence of newly planted trees does not support a finding of "imminent” threat, due to the delay between planting and production while the new trees grow. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 53-54.
*1219The statute provides that the Commission’s determinations of threatened material injury
shall be made on the basis of evidence that the threat of material injury is real and that actual injury is imminent. Such a determination may not be made on the basis of mere conjecture or supposition.
19 U.S.C. § 1677(7)(F)(ii) (Supp. IV 1986) (emphasis added). The question here is whether the Commissioners could find that the planting of new trees can threaten "imminent” material injury before the trees have fully matured.
As the Commissioners noted, there have been major increases in the number of orange trees planted in Brazil. Id. at R28-29. In the state of Sao Paulo, which produces approximately 96 percent of Brazil’s FCOJM, the number of orange trees increased from 117 million in 1984/85 to 125 million in 1985/86, and again to 137 million in 1986/87. USITC Pub. at R28. The number of bearing trees increased from 100 million in 1984/85 to 107 million in 1985/86, and again to 112 million in 1986/87. Id. Approximately 80-90 percent of the Sao Paulo crop is normally used to produce FCOJM. Id. at R29.
Despite this evidence of increased production capacity, Citrosuco argues that the record includes only evidence of "proposed expansion” and then cites Alberta Gas Chemicals Inc. v. United States, 1 CIT 312, 515 F. Supp. 780, 789-91 (1981), for the proposition that the proposed expansion is immaterial. Brief in Support of Plaintiff’s Motion for Judgment on the Agency Record, at 48.
The issue in Alberta Gas was whether a foreign supplier who was operating at nearly 100 percent capacity was going to construct new production facilities. The Commission concluded that there was no threat of material injury, in part because the expansion plans were too uncertain and depended upon too many volatile contingencies. Unlike Alberta Gas, where financing had not even been secured for the proposed expansion, the Commissions found that the exporters in this case have already improved their capacity to extract orange juice by improving their extraction process.
The Commission states that Citrosuco’s arguments ignore evidence that the number of trees actually bearing fruit increased significantly during the period of investigation from 88 million in 1982/83 to 112 million in 1986/87. USITC Pub. 1970 at R28. These figures are not simply of new plantings but of trees that have grown to maturity and are still producing. Actual production of oranges is already increasing, reflecting the availability of additional oranges. The latest estimate for Brazil’s 1985/86 production was 230 million boxes, an increase of 21 percent from the previous season. Id. at R99.
The Commission states that Citrosuco’s argument that the threat is not imminent also ignores the special nature of the domestic orange juice industry. Unlike other industries that can alter produc*1220tion quickly in response to a variety of market forces, the orange juice industry is restricted by the growing process of orange trees. This reality does not mean that the planting of millions of new trees, as well as increases in the number of bearing trees, cannot pose an imminent threat, however, because trees planted in the United States also require a maturation period. The Commission argues that the domestic industry should not be required, as Citrosuco advocates, to wait until all newly planted trees bear fruit in order to prevail where there is also evidence that the Brazilian importers have improved their ability to process more oranges efficiently and have actually increased the supply of oranges.
Having noted the Brazilian’s increased production capacity, the Commissioners found that the United States is the major market for the Brazilian companies under investigation who produce FCOJM, and that very little FCOJM is shipped to the Brazilian home market. USITC Pub. 1970 at 26. They observed that there has been an increase of exports to Europe and other areas, but concluded that this increase is not significant when compared to the amount of FCOJM shipped to the United States. They also determined that increases in European demand were less than the expected increase in Brazilian capacity and, thus, that the European market will not be able to absorb the increase in Brazilian production without a further price decrease. The Commissioners determined that such a price decrease would affect United States prices and thereby harm the domestic industry. Id. at n. 101.
With respect to the issue of increased production, Citrosuco also argues the Commissioners should have accorded greater weight to evidence that production of FCOJM would decline in 1986/87 because of drought and insect infestation in Brazil. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 51. Ci-trosuco states that the impact of natural disasters or acts of God on an exporter’s ability to increase production is a material factor in the Commission’s injury analysis, because in Operators For Jalousie and Awning Windows from El Salvador, Inv. Nos. 701-TA-272 (Final) and 731-TA-319 (Final), USITC Pub. No. 1934 (Jan. 1987), the Commission considered the effect of an earthquake upon an exporter’s capacity to export windows when reaching a negative threat determination. In that determination the Commission did not find that natural disasters will always constitute compelling or sufficient evidence to support a negative threat determination, but rather that the window exporter’s production capacity would be diverted to meet the needs of San Salvador and that exports would be disrupted as a consequence of the natural disaster. There was also evidence in that case that the disruption was not temporary, but long-term. Here, by contrast, the Commission states it is unlikely that the drought in Brazil will substantially reduce exports beyond 1986/87. See USITC Pub. 1970 at R99. The Commissioners could thus reasonably conclude that this predicted slight decline in pro*1221duction was not likely to reflect a future trend. In support of this conclusion, the Commissioners noted a recent increase in new plantings. Id. at 26 n.99.
Citrosuco also criticizes the Commissioners’ findings for failing to emphasize unofficial projections that Brazilian consumption of fresh oranges will increase and thus reduce the numbér of oranges, available for processing into FCOJM. Brief in Support of Plaintiff’s Motion for Judgment on the Agency Record, at 50. Commissioners Ecks and Lodwick considered these unofficial projections but concluded that they were not persuasive evidence that Brazilian exporters will be unable to increase FCOJM production and exports in the foreseeable future. The Commissioners were unpersuaded by this evidence because it was based upon speculations that orange prices in Brazil would drop dramatically under the new monetary regulation adopted by the Brazilian government (i.e., the "cruzado plan”), which would then increase demand by Brazilian consumers for fresh oranges. The Commissioners expressed doubt about the likely success of the cruzado plan and, as a consequence, had reservations about the reliability of these uncharacteristically low predictions of FCOJM production. USITC Pub. 1970 at 26 n.99 (citing Hopes Fade as Brazil’s Economy Falters, Cruzado Plan Not Worth Much After Initial Spending Spree, Washington Post, Mar. 12,1987, at El).
The Court finds that the Commissioners did consider unofficial predictions of decreased production in 1986/87, but reasonably concluded that they were not reliable indicators of production in the foreseeable future.
With respect to the statements of Commissioners Eckes and Lodwick about the likely increase in exports to the United States market, Citrosuco argues that any additional production will be sent to Europe. The record supports the Commissioners’ conclusions that many of the additional exports will be directed to the United States. Citrosuco focuses only on a very recent increase in exports to Europe, not a constant, long-term trend. According to data from the Bank of Brazil and Foodnews/Eurostat, exports of Brazilian FCOJM to Europe were at a record high in 1981, then declined dramatically in 1982, increased in 1983 and 1984, and then declined sharply in 1985. USITC Pub. 1970 at R86. Brazilian producers reported increases in exports to Europe from July 1982/June 1983 to July 1983/June 1984 but decreases the next two years. Conf. ITC Doc. 26 at R130. During January-June 1986, exports to Europe increased as compared to the prior interim period. USITC Pub. 1970 at R101. These increases did not result in corresponding decreases in exports to the United States. When exports to Europe increased in 1984, exports to the United States rose by an even greater pér-centage. Id. at R101. Similarly, when exports to Europe increased *1222between January/June 1985 and the corresponding period in 1986, exports to the United States also increased. Id.
The Court finds that the record supports the Commissioners’ determination that increased Brazilian production will result in greater exports to the United States, and that the conclusion is according to law. Given the sizeable increase in production capacity due to increased extracting capacity and the planting of millions of trees, it was reasonable for the Commission to find this increase in exports to the United States to be a significant factor in its determination that Brazilian imports threaten the United States industry with material injury.
ii. Market Penetration:
Citrosuco argues that import penetration is not likely to increase because Brazilian imports only "supplement” domestic production, and that Brazilian imports will decrease when United States production increases in the future. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 58-59.
With respect to the statutory threat factor of market penetration, 19 U.S.C. § 1677(7)(F)(i)(III) (Supp. IV 1986), Commissioners Eckes and Lodwick noted that the volume and value of dumped imports rose from 1982/83 to 1983/84, fell in 1984/85 and rose again in 1985/86. USITC Pub. 1970 at 25. The increase from 1984/85 to 1985/86 corresponded to a decrease in overall Brazilian imports. USITC Pub. 1970 at 25. Both the rapid increase in market penetration and the inverse relationship between dumped and fairly traded imports support the Commissioners’ conclusion that market penetration will likely increase to an injurious level.
When Florida’s production declined dramatically from 1982/83 to 1983/84, dumped imports from Brazil increased substantially. Less than fair value imports did not decline in any of the recent crop years in response to increases in United States production. From 1983/84 to 1984/85 both less than fair value imports and Florida production declined. Then from 1984/85 to 1985/86, dumped imports increased, as did Florida production. As Commissioners Eckes and Lodwick also noted, Brazilian imports have increasingly entered United States ports outside of Florida in order to bypass a 3 percent Florida Equalization Tax and to reduce inland transportation costs. USITC Pub. 1970 at 28 and n.109. In addition, the Commission found evidence that the Brazilian product is being used not only for blending with Florida juice, but is also being sold directly to reconstituters in the United States. Id. at 28. The Commission also noted that the role of Brazilian imports as supplementing domestic supply did not negate the Commission’s conclusion that those imports are injuring or threatening injury to the domestic industry as defined in this investigation. See USITC Pub. 1970 at 28 n.105.
The Court finds the Commissioners’ findings on market penetration to be reasonable and supported by the record as a whole.
*1223iii. Price Depression or Suppression:
Citrosuco attempts to dismiss the Commissioners’ findings of price depression or suppression by focusing on Brazilian import prices as compared to prices in the domestic futures market. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 66-67.
The record shows that the price of FCOJM sold in the United States dropped substantially from the beginning of 1985 through to the middle of 1986. USITC Pub. 1970 at R93, figure 8. As Commissioners Eckes and Lodwick noted, the price for Brazilian FCOJM was less than the price for domestic FCOJM for many months during that time period. Id. at 27. A comparison of monthly sales prices reveals that Brazilian prices were lower or equal to domestic prices in 16 of 24 months during the 1985/86 period. Id. at R89, table 49. A comparison of purchase prices reveals that the weighted average price for Brazilian FCOJM was equal to or lower than the price of domestic juice in 14 of 24 months during 1985/86. Id. at R90, table 50. The record includes additional evidence that the sharp drop in domestic FCOJM prices and corresponding decline in orange prices coincided with increasing volumes of dumped imports. Id. at R81, figure 5; R93, figure 8; Conf. Staff Report at R100, table 48. These facts support the conclusion of Commissioners Eckes and Lodwick that Brazilian FCOJM prices were responsible, at least in part, for the decrease in domestic FCOJM prices. USITC Pub. 1970 at 27.
The Commission argues that Citrosuco’s comparison to the futures market is not a reliable indicator of whether Brazilian prices follow or lead domestic prices since it is possible for Brazilian prices to follow the futures market and yet lead domestic prices. This possibility is likely because the futures market is predictive of future price trends and is likely to have a price setting effect upon the FCOJM market. As the Commission’s staff report states:
As in other futures markets, this one is also influenced by speculations. The FCOJM futures market is also used as a pricing mechanism; although, to a lesser extent currently than in the past. It is uniformly used and accepted as such a mechanism * * *. The futures price for FCOJM has been important in the determination of contract prices for FCOJM in recent years * * *. In addition, some sources indicate that spot-market prices are also being tied to the futures price for FCOJM.
USITC Pub. 1970 at R35 n.2. See also id. at R82. The Commission argues that Citrosuco’s speculations are not as reliable as comparisons of actual Brazilian and domestic purchase and sales prices, upon which commissioners Eckes and Lodwick relied. Citrosuco also relies on anecdotal evidence about prices, instead of the more reliable evidence gathered from responses to the Commission’s questionnaires. Brief in Support of Plaintiffs Motion for Judgment on the Agency Record, at 67.
*1224Citrosuco’s arguments regarding price suppression and depression assume the Commission must find that the only price depressing or suppressing effect on domestic prices will be dumped imports. The statute does not impose this as the relevant test. Instead, it requires consideration of "the probability that imports of the merchandise will enter the United States at prices that will have a depressing or suppressing effect on domestic prices of the merchandise.” 19 U.S.C. § 1677(7)(F)(i)(IV) (1982).
The statute provides that the Commission’s determinations of threatened material injury may not be made on the basis of mere conjecture or supposition. 19 U.S.C. § 1677(7)(F)(ii) (Supp. IV 1986). The Court finds that the Commission could properly reject the futures market evidence found to be speculative, and that the Commissioners’ findings that dumped imports will be one of several factors depressing or suppressing prices supports the determination of threatened material injury.
iv. Inventories in Brazil:
In considering increased inventories, the Commissioners looked to inventories in both the United States and Brazil. Commissioners Eckes and Lodwick found that total inventories of dumped Brazilian FCOJM in the United States decreased from 1984 to 1985, but remained constant in 1986. USITC Pub. 1970 at 29. They also found that
LTFV inventories in Brazil rose in 1985/86, but fell in interim 1986/87. The combined inventory figures, however, remain significant and are greater than the inventories of domestic FCOJM.
USITC Pub. 1970 at 29. Citrosuco argues that the Commissioners acted (a) unlawfully and (b) unreasonably in considering Brazilian inventories.
(a)
Citrosuco argues that the Commissioners acted unlawfully in considering inventories in Brazil, because the Commission is directed to consider "any substantial increase in inventories of the merchandise in the United States.” 19 U.S.C. § 1677(7)(F)(i)(V) (Supp. IV 1986) (emphasis added).
The Commission states that Brazilian inventories are relevant because Brazilian extractors use large tankers to ship FCOJM from Brazil to the United States, which permits Brazilian extractors to store FCOJM in Brazil without significantly affecting their ability to deliver FCOJM to customers in the United States as it is ordered. USITC Pub. 1970 at 29, R26. Commissioners Eckes and Lodwick found that these large tankers reduced the need of the Brazilian producers to store FCOJM inventories in the United States. Id. at 29.
*1225The Court finds that the Commissioners acted reasonably and according to law in considering inventories in Brazil because large storage tankers obviated the need to store inventories in the United States. The provision directing the Commission to consider inventories in the United States does not preclude consideration of "other relevant economic factors.” 19 U.S.C. § 1677(7)(F)(i) (Supp. IV 1986).
(b)
Citrosuco argues that the Commissioners’ acted unreasonably in considering Brazilian inventories (1) already committed under customer contracts worldwide, (2) sold at fair value, and (3) unfit for sale in the United States because of non-conformity with standards of the United States Department of Agriculture.
1. Committed Inventories:
Citrosuco asserts that the Commissioners acted unreasonably in failing to deduct the Citrosuco inventory in Brazil already committed under customer contracts worldwide. Plaintiff’s Reply to Defendants’ and Defendant-Intervenors’ Oppositions to Plaintiff’s Motion for Judgment on the Agency Record, at 23. Citrosuco argues that it does not overstate the "significance” of these sales commitments, because the "existence” of committed inventories was confirmed in Commerce’s final determination of sales at less than fair value. Id. at 23 n.9. In its final determination, Commerce stated that long-term contracts with futures-based price contingencies
constituted binding commitments under which all key elements were firm. The price terms of these contracts, pegged to futures prices, were definite and determinable.
Frozen Concentrated Orange Juice From Brazil: Final Determination of Sales at Less Than Fair Value, 52 Fed. Reg. 8324, 8328 (Mar. 17, 1987). Citrosuco argues that because there was no uncommitted supply of Brazilian FCOJ available for the United States market, pressure for the Brazilians to dump was low and the Commission should have reached a negative threat determination. Plaintiff’s Motion for Judgment on the Agency Record, at 75.
The Commission may determine that although a fact exists, its significance may be controverted by other evidence in the record. Here the Commission staff report states that the
trading, companies’ inventories are largely unpriced (75 percent or more) and are kept in foreign trade zones or bonded warehouses * * *. They are not withdrawn for consumption until the purchaser’s tanker truck loads it. Quantity commitments are voluntary; the business is not based on legal commitments but governed by commercial practices by regularly buying at a price established at the time of purchase (delivery.) Customers generally contact the importers about 1-2 weeks in advance of actual intended pick up of the FCOJM. At that time, price is negotiated; if the price asked by the importer is not suitable to *1226the purchaser, the purchaser will not take delivery and will seek FCOJM from other sources for a more favorable price. Purchasers may provide a trading company with an intended annual purchase quantity to be taken any time during taken any time during a 12-month period, but need not make legal commitments to purchase.
USITC Pub. 1970 at R27. The Commission’s staff report thus emphasizes commercial realities of how contracts were filled over the findings of how contracts were formed. The Court may not weigh the evidence concerning specific factual findings, nor may the Court substitute its judgment for that of the Commission. American Spring Wire Corp. v. United States, 8 CIT 20, 22, 590 F. Supp. 1273, 1276 (1984), aff’d sub nom. Armco, Inc. v. United States, 3 Fed. Cir. (T) 123, 760 F.2d 249 (1985). The Court cannot state that the Commissioners acted unreasonably in not deducting inventories that Ci-trosuco claims were destined for other ports, because the commercial realities of the sales shows a greater flexibility than the legal commitments that Citrosuco urges are more significant. The significance of evidence is for the Commission to determine. See id. at 23, 590 F. Supp. at 1277.
2. Inventories Sold at Fair Value:
Citrosuco also contends that the Commissioners should have deducted inventories held by Cutrale, Brazil’s largest producer, which sold its product at fair value and accounts for [a confidential percentage] of Brazilian exports to the United States. Plaintiffs Motion for Judgment on the Agency Record, at 72; Plaintiff’s Reply to Defendants’ and Defendant-Intervenors’ Oppositions to Plaintiff’s Motion for Judgment on the Agency Record, at 22-23.
In other circumstances this Court has approved the discounting of fair value sales in antidumping investigations. See Floral Trade Council of Davis, California v. United States, 12 CIT 1163, Slip Op. 88-170, at 11 n.7 (Dec. 27, 1988); Serampore Indus. v. United States, 11 CIT 866, 675 F. Supp. 1354, 1360 (1987). In this determination, however, the Court is unable to discern whether the Commission even considered excluding Cutrale’s inventories from its analysis of the threat emanating from inventories in Brazil. The testimony before the Commission does not present the Court with sufficient information to discern how the Commission treated this issue. ITC Doc. 96, at 163.
The Court finds it appropriate to remand this portion of the Commission’s determination for the Commission to explain whether or how it considered Cutrale’s fair value inventories in evaluating the threat from inventories in Brazil.
3. Inventories Below USDA Standards:
Citrosuco also states that the Commissioners should have subtracted inventories that would not meet United States Department of Agriculture (USDA) standards, since these products could alleg*1227edly go only to markets other than the United States. Plaintiffs Reply to Defendants’ and Defendant-intervenors’ Oppositions to Plaintiffs Motion for Judgment on the Agency Record, at 22-23.
Testimony before the Commission suggested, however, that products which did not meet USDA standards can be blended with other juices and be brought to satisfy the USDA standards for FCOJ. ITC Doc. 96, at 143. Absent evidence that inventories in Brazil could not be brought to conform to USDA standards, the Court finds that the Commissioners did not err in considering those inventories in analyzing the threat of material injury to a United States industry.
v. Inventories in the United States:
Commissioners Eckes and Lodwick also stated that total inventories of dumped Brazilian FCOJM in the United States remained constant in 1986. USITC Pub. 1970 at 29.
The Commission concedes that this summary of evidence was based upon incorrect total figures in the staff report. Memorandum of the United States International Trade Commission in Response to Plaintiffs Motion for Review Upon the Agency Record, at 71. Rather than remaining constant, United States inventories actually declined in 1986. Memorandum of the United States International Trade Commission in Response to Plaintiff’s Motion for Review Upon the Agency Record, at 71.
In its brief the Commission contends that this error with respect to the size of inventories in the United States is "harmless” because the Commissioners also relied on correct figures which reflected total inventories in Brazil. Id. The Commission implies that the finding is subsidiary because Commissioners Eckes and Lodwick also noted that the non-extractor importers of Brazilian FCOJM increased their bulk storage capacity in the United States, which indicated that importers of the dumped FCOJM expected to receive growing amounts of Brazilian FCOJM in the future. USITC Pub. 1970 at 29 and n.115. The Commission cites Delta Air Lines, Inc. v. Civil Aeronautics Bd., 184 D.C. App. 107, 564 F.2d 592, 598 (D.C. Cir. 1977), for the proposition that reversal is inappropriate where a factual error is of a subsidiary nature.
By arguing that the finding that United States inventories remained constant when they actually declined is merely a "harmless error,” the Commission is asking the Court to weigh the significance of a statutory threat factor. 19 U.S.C. § 1677(7)(F)(i)(V) (Supp. IV 1986). Citrosuco argues that the fact that United States inventories of Brazilian FCOJM decreased may indicate that there is less of a threat to the domestic industry. Defendant-intervenors and ami-cus curiae argue that the status of United States inventories is of no consequence to the final determination because the Commissioners "clearly relied on the importance of the combined U.S. and Brazilian inventory levels, which they found to be greater than inventories *1228of domestic FCOJM." Intervenors’ and Amicus Curiae’s Opposition to Plaintiff’s Motion for Judgment on the Agency Record, at 61 n.11.
This Court does not substitute its own judgment for the Commission’s or reweigh evidence on the record in reviewing the Commission’s determination. Alberta Pork Producers’ Mktg. Bd. v. United States, 11 CIT 563, 669 F. Supp. 445, 449 and 460 (1987). Rather, the Court must review the Commission’s determination to establish whether it is supported by substantial evidence on the record. 19 U.S.C. § 1516a(b)(l)(B) (1982). "The Court is loathe to affirm a determination that might be based on a questionable record.” Serampore Indus. v. United States, 12 CIT 825, 669 F. Supp. 665, 673 (1988). The Court finds it appropriate to remand this portion of the Commission’s determination of threatened material injury for reconsideration of the significance of inventories in the United States.
4. Material Injury:
Commissioner Rohr found that the domestic industry had suffered material injury by reason of the less than fair value imports from Brazil. USITC Pub. 1970 at 31.
The Commission’s task in determining whether there has been material injury by reason of dumped imports does not involve the weighing of causes. Instead, the Commission must determine whether those imports are a cause of material injury. As explained in British Steel Corp. v. United States, 8 CIT 86, 96, 593 F. Supp. 405, 413 (1984), "[t]he statute’s causation prerequisite to an affirmative injury determination is satisfied if the * * * imports contribute, even minimally, to the conditions of the domestic industry, and the Commission is precluded from weighing the causes of injury.”
In this case, the domestic industry suffered from several freezes during the investigation. Citrosuco seeks to treat the domestic industry’s injury as only caused by freezes in Florida. Commissioner Rohr acknowledged that the freezes left the industry in a weakened state:
Finally, while it would be unfair to attribute to the Brazilian imports the adverse impact that the recent series of freezes has had on the industry, it would similarly be unfair if I did not recognize that the freezes have left the industry in a more precarious and vulnerable position than it might otherwise have been in. I must analyze the industry as I find it, with both its strengths and weaknesses as they currently exist.
USITC Pub. 1970 at 31. Upon this framework, Commissioner Rohr noted the following indications of material injury:
1. Acreage under production by Florida orange growers declined by approximately 10 percent during each year under investigation.
2. Total orange production declined by more than 20 percent over the entire period and estimated production of orange juice solids declined 26 percent over the period of investigation.
*12293. Total grower proceeds dropped dramatically in 1985/86, as did the net income margin for growers.
4. Production of orange juice from domestic oranges declined at 26 percent over the period.
5. The number of workers involved in the processing of orange juice declined slightly over the period of investigation, as did hours worked.
6. Net sales of processors rose from 1982 to 1985 but dropped in 1986 to a level slightly below 1983 sales.
7. The cost of goods sold (COGS) increased slightly between 1982 and 1984, before dropping drastically in 1985 and 1986.
8. The operating margin of processors declined between 1982 and 1984, increased in 1985, and then dropped to only .1 percent in 1986.
Upon these facts, Commissioner Rohr found that the domestic industry was materially injured by reason of the dumped imports from Brazil. First, he discussed increases in Brazilian production of oranges and less than fair value imports that corresponded to the period of injury. He noted that total Brazilian production of oranges increased steadily throughout the investigation, from 195 million boxes to 329 million boxes, and that dumped imports increased by more than 50 percent after the first year, and increased again in the third year. He also observed the corresponding fluctuations in the market penetration of imports made at less than fair value.
Commissioner Rohr then examined the effect of the Brazilian imports upon United States prices, while acknowledging that the analysis was "somewhat clouded by fluctuations caused by the recent freezes.” USITC Pub. 1970 at 45. He observed that Brazilian prices seemed to have led both upward and downward trends in domestic prices. Id. He also noted reports by purchasers that Brazilian FCOJM prices were below domestic FCOJM in 1984 and 1986. Id. at 46. The Commission staffs report reveals that the weighted average f.o.b. purchase price for Brazilian FCOJM was less than the corresponding domestic price throughout 1984 and less than or equal to the domestic price, during a period of at least 10 months in 1986. Id. at R90. Commissioner Rohr found this price differential to be particularly significant because the orange juice market is highly price sensitive, with buyers making purchase decisions principally upon the basis of price. Id. at 46; R26-28. With respect to the price effect issue, Commissioner Rohr also relied upon evidence that on-tree orange prices and spot orange prices sharply dropped in 1986. As he explained, some decline would have been expected as Florida groves recovered from the freezes, but the severity of the decline was out of proportion to the limited recovery of domestic production that occurred during the same period.
Commissioner Rohr considered figures regarding growers who were significantly affected by the freezes to determine whether declines in the industry could be attributed only to the freezes. He ob*1230served that the financial performance of these freeze-affected growers was similar to overall grower financial performance. Although the pre-tax net margins in 1985/86 of the growers with smaller yields was less than the pre-tax net margins of the growers with greater yields, both margins declined between 1984/85 and 1985/86. See USITC Pub. 1970 at R43-44. The decline in the pre-tax margins of the growers with larger yields was from a net income before taxes of 41.0 percent in 1985 to a net income of 26.5 percent in 1986. This constituted a 35 percent decline and corresponded to a growing season when there were no significant freezes. During this period, however, there was an increase in dumped imports. Commissioner Rohr found these declines in the post-freeze year (1985/86) to be particularly important.
Commissioner Rohr also considered other evidence supporting his conclusion that dumped imports were causally related to the financial condition of the domestic industry. He found particularly persuasive the fact that growers and processors achieved their best financial operating results during the only year in which dumped imports declined, i.e., 1984/85. He also noted that acreage under production declined by approximately 10 percent in the 1985/86 crop year, even though there were no significant freezes during that time. Id. at 41.
Finally, Commissioner Rohr analyzed whether dumped imports were simply "supplementing” the domestic market by, for example, increasing when domestic production declined and decreasing as domestic production increased. Id. at 46. In this regard, he stated: "The volume of the [less than fair value] imports does not appear to be positively related to the increases and decreases in domestic production in recent years. Rather, their importation appears to be occurring despite changes in domestic production.” Id. In further support of his conclusion that less than fair value imports are an integral part of the domestic market, he also noted that an increasing amount of the Brazilian product is being marketed directly to purchasers in the United States, thereby increasing their impact upon the domestic industry. Commissioner Rohr thus concluded that the Brazilian dumped imports are a cause of material injury to the domestic industry.
The Court, concludes that Commissioner Rohr’s determination of material injury is accordiang to law and supported by substantial evidence on the record.
Conclusion
The Court finds that the determination of Commerce is supported by substantial evidence on the record as a whole and according to law. See 19 U.S.C. § 1516a(b)(l)(B) (1982); Washington Red Raspberry Comm’n v. United States, 859 F.2d 898, 902 (Fed. Cir. 1988).
*1231As to the Commission’s determination, the Court remands to the Commission for (1) explanation of how or whether the Commission considered Cutrale’s inventories in Brazil, and (2) reconsideration of the significance of inventories in the United States in light of the evidence that those inventories were decreasing rather than remaining stable. This remand is directed to the entire Commission, and not just individual commissioners. See Asociacion Colombiana de Exportadores de Flores v. United States, 12 CIT 1174, Slip Op. 88-172, at 3 n.2 (Dec. 27, 1988).
The Commission shall file its remand determination with the Court within 30 days. Citrosuco, the defendant-intervenors, and amicus curiae are granted 20 days to file comments on the remand determination. The Commission may respond to any comments filed within 10 days.