C-580-851
2nd Administrative Review
ARP: 01/01/2004-12/31/2004
Public Document
Office 1
MEMORANDUM TO: David M. Spooner
Assistant Secretary for Import Administration
FROM: Stephen J. Claeys
Deputy Assistant Secretary for Import Administration
DATE: February 7, 2007
SUBJECT: Issues and Decision Memorandum for the Final Results in
the Second Administrative Review of the Countervailing
Duty Order on Dynamic Random Access Memory Semiconductors
from the Republic of Korea Background
On August 11, 2006, the Department of Commerce (“the Department”) published the preliminary results of this administrative review. See Preliminary Results of Countervailing Duty Administrative Review: Dynamic Random Access Memory Semiconductors from the Republic of Korea, 71 FR 46192 (“Preliminary Results”). The “Analysis of Programs” and “Subsidies Valuation Information” sections, below, describe the subsidy programs and the methodologies used to calculate the benefits from these programs. We have analyzed the comments submitted by the interested parties in their case and rebuttal briefs in the “Analysis of Issues” section, below, which also contains the Department's responses to the issues raised in the briefs. n1 We recommend that you approve the positions described in this memorandum. Below is a complete list of the issues in this administrative review for which we received comments and rebuttal comments from parties:
n1 The Department received a case brief from Micron Technology, Inc. (“Micron”) and a rebuttal brief from Hynix Semiconductor, Inc (“Hynix”)
Comment 1: Benefit to Hynix of the 2004 Cash Buyout Program
Comment 2: The Department's Failure to Investigate Thoroughly the GOK's
Entrustment or Direction of Hynix's Creditors in Connection with
the CBO Components of the Non-Memory Asset Sale
Comment 3: Entrustment or Direction of Hynix's Creditors in Connection with
the Tranche A Acquisition Financing and CBO Components of the Non-
Memory Asset Sale
Comment 4. Whether the Department Should Have Investigated Hynix's Sale of
Its LCD and Non-Memory Assets
Comment 5: Uncreditworthy Benchmark Interest/Discount Rate
Changes in Ownership
Effective June 30, 2003, the Department adopted a new methodology for analyzing privatizations in the countervailing duty context. See Notice of Final Modification of Agency Practice Under Section 123 of the Uruguay Round Agreements Act, 68 FR 37125 (June 23, 2003) (“Modification Notice”). The Department's new methodology is based on a rebuttable “baseline” presumption that non-recurring, allocable subsidies continue to benefit the subsidy recipient throughout the allocation period (which normally corresponds to the average useful life (“AUL”) of the recipient's assets). However, an interested party may rebut this baseline presumption by demonstrating that, during the allocation period, a change in ownership occurred in which the former owner sold all or substantially all of a company or its assets, retaining no control of the company or its assets, and that the sale was an arm's-length transaction for fair market value.
Hynix's ownership changed during the AUL period as a result of debt-to-equity conversions in October 2001, and December 2002, and various asset sales. However, Hynix has not rebutted the Department's baseline presumption that the non-recurring, allocable subsidies received prior to the equity conversions and asset sales continue to benefit the company throughout the allocation period. See Hynix's March 30, 2006 supplemental questionnaire response (“Hynix SQNR”) at 4. See also Dynamic Random Access Memory Semiconductors from the Republic of Korea: Preliminary Results of Countervailing Duty Administrative Review, 70 FR 54523, 54524 (September 15, 2005) (“AR1 Preliminary Results”) (unchanged in final).
Subsidies Valuation Information
Allocation Period
Pursuant to 19 CFR 351.524(b), non-recurring subsidies are allocated over a period corresponding to the AUL of the renewable physical assets used to produce the subject merchandise. Section 351.524(d)(2) of the Department's regulations creates a rebuttable presumption that the AUL will be taken from the U.S. Internal Revenue Service's 1977 Class Life Asset Depreciation Range System (the “IRS Tables”), as updated by the Department of the Treasury. For dynamic random access memory semiconductors (“DRAMS”), the IRS Tables prescribe an AUL of five years. During this review, none of the interested parties disputed this allocation period. Therefore, we continue to allocate nonrecurring benefits over the five-year AUL.
Discount Rates and Benchmarks for Loans
For loans that we found countervailable in the investigation or in the first administrative review, and which continued to be outstanding during the period of review (“POR”), we have used the benchmarks used in the first administrative review (these are described below).
Long-Term Rates
For long-term, won-denominated loans originating in 1986 through 1995, we used the average interest rate for three-year corporate bonds as reported by the Bank of Korea or the International Monetary Fund (“IMF”). For long-term, won-denominated fixed-rate loans originating in 1996 through 1999, we used an annual weighted-average of the rates on Hynix's corporate bonds, which were not specifically related to any countervailable financing. We did not use the rates on Hynix's corporate bonds for 2000-2003 for any calculations because Hynix was found to be uncreditworthy in those years.
For U.S. dollar-denominated loans, we relied on the lending rates as reported in the IMF's International Financial Statistics Yearbook.
For the years in which we previously determined Hynix to be uncreditworthy (2000 through 2003), we used the formula described in 19 CFR 351.505(a)(3)(iii) to determine the benchmark interest rate. For the probability of default by an uncreditworthy company, we used the average cumulative default rates reported for the Caa-to C-rated category of companies as published in Moody's Investors Service, “Historical Default Rates of Corporate Bond Issuers, 1920-1997” (February 1998). For the probability of default by a creditworthy company, we used the cumulative default rates for investment grade bonds as published in Moody's Investor Services: “Statistical Tables of Default Rates and Recovery Rates” (February 1998). For the commercial interest rates charged to creditworthy borrowers, we used the rates for AA-won-denominated corporate bonds as reported by the Bank of Korea (“BOK”) and the U.S. dollar lending rates published by the IMF for each year. See Comment 5.
Short-Term Loans
Consistent with the methodology used in the first administrative review, we relied on the money market rates as reported in the IMF's International Financial Statistics Yearbook for short-term interest rates.
Analysis of Programs
I. Programs Determined to Not Confer Subsidies During the POR
A. GOK Entrustment or Direction of Debt Reductions
In the investigation and the first administrative review, the Department determined that Hynix received countervailable subsidies from creditors that were entrusted or directed by the GOK to provide Hynix with financial support in the form of loans, debt-to-equity conversions and debt forgiveness. We reached these determinations on the basis of a two-part test: First, we determined that the GOK had in place a governmental policy to support Hynix's financial restructuring to prevent the company's failure. Second, we found that the GOK acted upon that policy through a pattern of practices to entrust or direct Hynix's creditors to provide financial contributions to Hynix. See Final Affirmative Countervailing Duty Determination: Dynamic Random Access Memory Semiconductors from the Republic of Korea, 68 FR 37122 (June 23, 2003) and accompanying Issues and Decision Memorandum (“Investigation”) at 47-61 and Dynamic Random Access Memory Semiconductors from the Republic of Korea: Final Results of Countervailing Duty Administrative Review, 71 FR 14174 (March 21, 2006) and accompanying Issues and Decision Memorandum (“First Administrative Review”) at 5-10. We also found that “this policy and pattern of practices continued throughout the entire restructuring process through its logical conclusion.” See Investigation at 47-61. These findings covered the period through 2003.
According to Micron, the GOK's “policy to prevent Hynix's failure continued unabated beyond the original investigation into the first and second periods of review,” and the GOK acted to ensure that Hynix's corporate and financial restructurings were carried out by Hynix's creditors during 2004. See Micron's January 12, 2006 submission (“New Subsidy Allegations (NSA)” at 13-15. As such, Micron contends, the GOK entrusted or directed Hynix's creditors to facilitate the sale of Hynix's assets, such as its System IC unit, by providing acquisition financing and by forgiving portions of Hynix's debt before and after the System IC sale. The System IC unit is referred to in this memorandum as Hynix's non-memory assets.
The Department declined to investigate the alleged subsidies conferred by the sales of Hynix's assets in 2003 and 2004, but is investigating the alleged debt forgiveness that occurred before and after the sale of the non-memory assets. See Memorandum from Ryan Langan, International Trade Compliance Analyst, to Susan Kuhbach, Director, Office 1, regarding Countervailing Duty Administrative Review: Dynamic Random Access Semiconductors from Korea; New Subsidy Allegations (June 8, 2006) (“New Subsidy Allegations Memorandum”). Specifically, the alleged subsidies that we are investigating in this review involve debt that was reduced as part of the following financial transactions: 1) Tranche A of the acquisition financing for the sale of the non-memory assets to MagnaChip Semiconductor LLC (“MagnaChip”); 2) the October 2004 Cash Buyout (“CBO”); and 3) the December 2004 CBO. According to Micron, Hynix's creditors were entrusted or directed by the GOK to forgive debt as part of each of these financial transactions.
As in the investigation and the first administrative review, the question before the Department in this segment of the proceeding is whether the GOK entrusted or directed Hynix's creditors to provide financial contributions to Hynix, within the meaning of section 771(5)(B)(iii) of the Tariff Act of 1930, as amended (“the Act”). To answer that question, we applied the same two-part test that we used in the investigation and first administrative review. As such, the focus of our analysis has been to determine whether the record evidence demonstrates that the GOK maintained its policy to save Hynix and that a pattern of GOK practices to implement such a policy existed during the period of review (i.e., calendar year 2004).
The record evidence in this review demonstrates that the GOK-entrusted or -directed financial restructurings of Hynix in 2001 and 2002 largely achieved the GOK's objective of preventing Hynix's collapse by 2004. Specifically, the record evidence shows that Hynix's financial condition in 2004 improved substantially in comparison to 2001 through 2003. For instance, Hynix consistently generated significant revenue, profit, and return on equity throughout 2004. See Preliminary Results, 71 FR at 46195. Furthermore, industry analysts held favorable views of Hynix throughout the POR. See Hynix's January 27, 2006 Rebuttal Factual Information at Exhibits 3, 10, 19, 21, 22, 26, 27, 33, and 35. Although Micron provided further evidence (e.g., reports and news articles) on Hynix's financial crisis in relation to the Department's earlier findings of entrustment and direction, the information did not relate directly to the GOK's actions or Hynix's financial standing in 2004. See Preliminary Results, 71 FR at 46196, 46197. For these reasons, we find that the evidence on the record shows that Hynix was no longer at risk of failure during the POR, as it was in prior years, and, therefore, a principal motivation and basis for the GOK's past policy regarding Hynix was no longer present in 2004.
We additionally examined whether there was evidence that the GOK was still influencing Hynix's financial dealings through entrustment of direction of its creditors. In the first review, the Department found that Hynix's Creditors' Council was dominated by GOK-owned or -controlled banks, which were subject to significant GOK influence. We also found that the GOK influenced the remaining creditors through these banks. See First Administrative Review at 10 and Section B and C of Comment 1. However, the record evidence in this review demonstrates that the GOK did not maintain its dominance of the Creditors' Council in 2004, because of the change in ownership of Korea Exchange Bank (“KEB”) and the arrival of new, foreign-owned creditors on the Creditors' Council. See Preliminary Results 71 FR at 46198.
As we stated in the First Administrative Review, the GOK was the largest shareholder of KEB and held considerable influence or control over the lending decisions of the institution. We also found that Commerzbank's 23.6 percent ownership of KEB in 2002 did not immunize KEB from GOK influence or control because the GOK was KEB's single largest shareholder. See First Administrative Review at 34.
In the current POR, however, Lone Star's 51 percent ownership stake coupled with Commerzbank's approximately 15 percent ownership stake transformed KEB into a majority foreign-owned institution. See Memorandum from the Team to the File regarding Countervailing Duty Administrative Review: Dynamic Random Access Memory Semiconductors from the Republic of Korea; Calculations for the Preliminary Results for Hynix Semiconductor, Inc. (August 7, 2006) (“Preliminary Calc Memo”) at Attachment 3. With KEB's restructuring, we find that in 2004 that KEB was no longer a GOK-owned or -controlled creditor. As a result, the GOK no longer had the same ability to influence or control KEB's lending decisions as it did in prior periods.
The GOK also no longer held a controlling majority of the voting rights on Hynix's Creditors' Council. In fact, the voting rights held by GOK-owned or -controlled creditors in 2004 did not even constitute a majority of the votes on the Creditors' Council. See Preliminary Calc Memo at Attachment 3. Therefore, we find that the GOK-owned or -controlled banks no longer dominated the Creditors' Council. Thus, even if the GOK did continue to have a policy to assist Hynix in 2004 (and, as we indicated above, the record evidence does not show that it did), a key factor that permitted the GOK to effectuate such a policy -control of the Creditors' Council -was no longer in place in 2004.
In sum, Hynix's improved financial situation in 2004, the lack of evidence demonstrating a GOK policy or pattern of practices to entrust or direct Hynix's creditors to provide financial assistance to Hynix in 2004, and the GOK's lack of sufficient voting rights to dominate the Creditors' Council in 2004 lead us to conclude that the GOK did not entrust or direct Hynix's creditors to reduce or forgive Hynix's debt in 2004. We also note that, unlike prior segments of this proceeding, the record in this review contains no evidence that the GOK threatened or otherwise pressured Hynix's creditors during 2004. Therefore, we find that the debt reductions Hynix received from non-GOK entities in 2004 are not countervailable.
In prior segments of this proceeding, we have distinguished between those banks found to be “government authorities” within the meaning of section 771(5)(B) the Act, and banks found to be “entrusted or directed” by the GOK within the meaning of section 771(5)(B)(iii) of the Act. See First Administrative Review at 6-7. The record information in this review does not show any new evidence or changed circumstances that would lead us to revisit our prior determinations that the KDB and other “specialized” banks are government authorities and that the financial contributions made by these entities fall within the meaning of section 771(5)(B)(i) of the Act. Therefore, although we have determined that the GOK did not entrust or direct non-GOK entities to provide financial contributions in 2004, we must further address whether government authorities provided countervailable subsidies. For the reasons discussed below, we find that the debt reductions provided by the KDB and other GOK entities in connection with the financial transactions newly alleged and under investigation in this review do not confer countervailable subsidies.
Tranche A of the Acquisition Financing for the Sale of Hynix's Non-Memory Assets
Record information indicates that in July 2004, Hynix's Creditors' Council agreed to provide acquisition financing for MagnaChip's purchase of Hynix's non-memory assets. Concurrently, the Creditors' Council agreed to the terms for the October CBO. See Hynix SQNR at Exhibit 9. Tranche A of the acquisition financing involved the transfer of new loans received by Hynix and previously existing loans from Hynix to MagnaChip. The total debt transferred to MagnaChip under Tranche A was KRW 154.9 billion, which formed part of the purchase price MagnaChip paid for Hynix's non-memory assets. Hynix also reported that, prior to the transfer of the existing loans, Hynix's creditors reduced the original debt amount through an application process established by the Creditors' Council. Micron has alleged that the debt reduction constitutes a direct transfer of funds in the form of debt forgiveness, within the meaning of section 771(5)(D)(i) of the Act.
No GOK entities participated in Tranche A financing. Instead, the banks that agreed to discount the Hynix debt that was transferred to MagnaChip were wholly foreign-owned banks or non-GOK entities. Absent GOK entrustment or direction to participate in Tranche A financing, any debt reductions provided by these creditors do not constitute a financial contribution and, therefore, are not countervailable.
The October and December CBOs
According to Hynix, the expected cash proceeds from the sale of the non-memory assets and income from its normal business operations enabled Hynix to repay numerous outstanding loans in 2004, prior to their maturity. n2 These repayments were made under the October CBO, which occurred concurrently with the sale of the non-memory assets and Tranche A acquisition financing. Hynix also repaid its loans early and at a discount under the December CBO, which occurred after the non-memory assets sale. See Hynix SQNR at 5-8 and Exhibit 9. See also Hynix's June 30, 2006 New Subsidy Allegation questionnaire Response (“Hynix NSQR”) at Exhibit NA-9.
n2 We note that all of the loans affected by these early repayments are loans that the Department has previously found to have been provided to Hynix at the entrustment or direction of the GOK.
The terms of the October CBO included a maximum cash buyout rate of 70 percent for unsecured loans and a fixed cash buyout rate of 96 percent for secured loans. In other words, the Creditors' Council established maximum early payment discounts of 30 percent and 4 percent on unsecured and secured loans, respectively. The Creditors' Council also established a target amount for repayment for the entire CBO, limitations on the amount of secured loans that would be repaid under the CBO, and a hierarchy of loans that were eligible for the CBO. See Hynix SQNR at 5-8 and Exhibit 9. See also Hynix NSQR at Exhibit NA-9.
To carry out the early payment, the Creditors' Council established a bidding process under which each creditor would bid or apply to participate in the CBO. Therefore, the types of debt repaid under the CBO would largely depend on which creditors applied to participate in the CBO and the type of debt that they held. According to the terms set by the Creditors' Council, the discount rates for the October CBO applied equally to all participating creditors, even though some creditors offered discount rates greater than 30 percent on unsecured debt. See Hynix SQNR at 5-8 and Exhibit 9. See also Hynix NSQR at Exhibit NA-9.
Similarly, Hynix repaid existing loans prior to their maturity under the December CBO at a discount. According to Hynix, the discount rates for the December CBO were established by Hynix, not the Creditors' Council. (However, the discount rates were similar to the rates for the October CBO.) Like the October CBO, the December CBO relied upon an application process under which creditors applied to participate and identified the types of loans that they wanted repaid by Hynix. See Hynix NSQR at Exhibits NA-11 and NA-12.
We determine that the October and December CBOs were early repayment plans under which creditors could exchange loans with a maturity in 2006 for a discounted amount (i.e., cash) in 2004. As discussed in the Analysis of Comments section below at Comment 1, we also continue to find that the discounts taken by the participating creditors do not constitute loan forgiveness, as described in section 351.508 of the Department's regulations. Instead, the discounts reflect the value to Hynix of repaying the loans and the value to its creditors of obtaining repayment prior to the scheduled maturity of the loans and have been examined under section 351.505(a).
Section 771(5)(E)(a) of the Acts states that a benefit is conferred from a loan “If there is a difference between the amount the recipient of the loan pays on the loan and the amount the recipient would pay on a comparable commercial loan that the recipient could actually obtain on the market.” Under the CBOs, the amount that Hynix paid on the loans was determined by the discount rates its creditors were willing to accept. Therefore, whether a benefit was conferred on Hynix as a result of the CBOs depends on whether the repayment terms on the loans held by government authorities differed from the repayment terms on the loans held by commercial lenders.
In the current review, wholly-foreign owned creditors accounted for over 30 percent and 80 percent of the discounted debt in the October and December CBOs, respectively. On an aggregate basis, wholly-foreign owned creditors accounted for over 40 percent of the debt discounted under the two CBOs. See Preliminary Calc Memo at Attachment 3. We also determined that the government authorities and the wholly-foreign owned banks participated in the October and December CBOs on the same terms. As noted above, creditors were free to apply for early repayment, and the discount rates in the CBOs applied equally to all participants.
In the investigation and first administrative review, we found that wholly-foreign owned creditors operating in Korea, such as Citibank, were not entrusted or directed by the GOK to participate in government-led bailouts of Hynix. As such, these wholly-foreign owned banks could have been used as commercial benchmarks, although they were not used because their portion of the loans and equity infusions being reviewed was so small. See First Administrative Review at Comments 5 and 6.
With regard to Citibank, we acknowledge that in the first administrative review, we cited an additional reason for not using Citibank as a commercial benchmark: although we did not find Citibank to be entrusted or directed by the GOK per se, we found that GOK influence extended to Citibank during the POR of the first administrative review because of the GOK's dominance of the Creditors' Council. See First Administrative Review at Comment 6. However, as discussed above, the GOK no longer dominated the Creditors' Council in 2004. Consequently, a key factor we previously found to have given the GOK the ability to influence Hynix's other creditors -control of the Creditors' Council -was no longer present in 2004. Moreover, the Department finds no other record evidence in the present review indicating that Citibank's participation in the October or December 2004 CBOs was subject to GOK influence.
Therefore, we find that Hynix's early repayments of debt to GOK entities at a discount do not confer a benefit on Hynix and, consequently, are not countervailable. We further note that even if the Department were to find that the GOK entrusted or directed Hynix's creditors to participate in the CBOs, such financial contributions to Hynix would not constitute countervailable subsidies because the participation by Citibank and other wholly-foreign owned banks on identical terms means that no benefit is conferred on Hynix.
Specificity
With regard to any benefits attributable to the current POR, because we have found that the GOK did not entrust or direct Hynix's creditors to forgive debt in 2004, and that debt reductions provided by GOK entities in 2004 did not confer a benefit to Hynix, we need not address the issue of specificity with respect to those alleged subsidies.
With regard to earlier subsidies that we have previously examined, the Department determined in the investigation that the GOK entrusted or directed credit to the semiconductor industry through 1998. See Investigation at 12-21. For the period 1999 through June 30, 2002, the Department determined that the subsidies were specific to Hynix under section 771(5A)(D)(iii) of the Act because the GOK's entrustment or direction to provide financial contributions, and the benefits thereby conferred, involved current or former Hyundai Group companies, and Hynix in particular. Id. at 17-19. In the first administrative review, the Department found the December 2002 restructuring was de facto specific to Hynix within the meaning of section 771(5A)(D)(iii)(I) of the Act. See First Administrative Review at 10-11.
Nothing on the record of this review would lead us to reconsider these prior specificity findings.
II. Programs Previously Determined to Confer Subsidies
We examined the following programs determined to confer subsidies in the investigation and first administrative review, and find that Hynix continued to receive benefits under these programs during the POR.
A. GOK Entrustment or Direction Prior to 2004
In the investigation, the Department determined that the GOK entrusted or directed creditor banks to participate in financial restructuring programs, and to provide credit and other funds to Hynix, to prevent Hynix's failure. The financial assistance provided to Hynix by its creditors took various forms, including new loans, convertible and other bonds, extensions of maturities and interest rate reductions on existing debt (which we treated as new loans), Documents Against Acceptance (“D/A”) financing, usance financing, overdraft lines of credit, debt forgiveness, and debt-for-equity swaps. The Department determined that these were financial contributions that constituted countervailable subsidies during the period of investigation (“POI”).
In the first administrative review, the Department found that the GOK continued to entrust or direct Hynix's creditors to provide financial assistance to Hynix throughout 2002 and 2003. The financial assistance provided to Hynix during this period included the December 2002 debt-for-equity swaps and the extensions of maturities and/or interest rate deductions on existing debt.
In an administrative review, we do not revisit the validity of past findings unless new factual information or evidence of changed circumstances has been placed on the record of the proceeding that would compel us to reconsider those findings. See, e.g., Certain Pasta from Italy: Preliminary Results and Partial Rescission of Seventh Countervailing Duty Administrative Review, 69 FR 45676 (July 30, 2004), affirmed in Certain Pasta From Italy: Final Results of Seventh Countervailing Duty Administrative Review, 69 FR 70657 (December 7, 2004). No such new information has been presented in this review and, thus, we find that a re-examination of the Department's findings in the investigation and first administrative review is unwarranted.
Therefore, we are including in our benefit calculation the financial contributions countervailed in the investigation and in the first administrative review: bonds, debt-to-equity swaps, debt forgiveness, and long-term debt outstanding during the POR. In calculating the benefit, we have followed the same methodology used in the first administrative review.
Because we found Hynix to be unequityworthy at the time of the debt-for-equity swaps in 2001 and 2002, we have treated the full amount swapped as grants and allocated the benefit over the five-year AUL. See 19 CFR 351.507(a)(6) and (c). We used a discount rate that reflects our finding that Hynix was uncreditworthy at the time of the debt-to-equity conversions. For the loans, we have followed the methodology described at 19 CFR 351.505(c) using the benchmarks described in the “Subsidies Valuation Information” section of this notice.
We divided benefits from the various financial contributions by Hynix's POR sales to calculate a countervailable subsidy rate of 31.79 percent ad valorem for the POR.
Implemented under the Framework on Science and Technology Act, the Operation G-7/HAN Program (“G-7/HAN Program”) began in 1992 and ended in 2001. The purpose of this program was to raise the GOK's technology standards to the level of the G-7 countries. The Department found that the G7/HAN Program ended in 2001. See Investigation at 25. However, during the POR, Hynix had outstanding interest-free loans that it had previously received under this program. See Hynix' December 22, 2005 Questionnaire Response at 19 and Exhibit 12. The G-7/Han Program was found to provide countervailable subsidies in the investigation. No new evidence has been provided that would lead us to reconsider our earlier finding. Therefore, we have calculated a benefit for these loans.
To calculate the benefit of these loans during the POR, we compared the interest actually paid on the loans during the POR to what Hynix would have paid under the benchmark described in the “Subsidy Valuation Information” section of this notice. We then divided the total benefit by Hynix's total sales of subject merchandise for the POR to calculate the countervailable subsidy. On this basis, we determine that countervailable benefits of 0.07 percent ad valorem existed for Hynix.
C. 21st Century Frontier R&D Program
The 21st Century Frontier R&D Program (“21st Century Program”) was established in 1999 with a structure and governing regulatory framework similar to those of the G-7/HAN Program, and for a similar purpose, i.e., to promote greater competitiveness in science and technology. The 21st Century program provides long-term interest-free loans in the form of matching funds. Repayment of program funds is made in the form of “technology usance fees” upon completion of the project, pursuant to a schedule established under a technology execution, or implementation contract.
Hynix reported that it had loans from this program outstanding during the POR. See Hynix's December 22, 2005 Questionnaire Response at Exhibits 12 and 13.
In the investigation, we determined that this program conferred a countervailable benefit on Hynix. No new evidence has been provided that would lead us to reconsider our earlier finding.
To calculate the benefit of these loans during the POR, we compared the interest actually paid on the loans during the POR to what Hynix would have paid under the benchmark described in the “Subsidy Valuation Information” section of this notice. We then divided the total benefit by Hynix's total sales in the POR to calculate the countervailable subsidy rate. On this basis, we calculated a subsidy rate of less than 0.005 percent ad valorem for this program. Therefore, consistent with our past practice, we did not include this program in our net countervailing duty rate. See, e.g., Notice of Preliminary Results of Countervailing Duty Review: Certain Softwood Lumber Products from Canada, 70 FR 33088, 33091 (June 7, 2005).
III. Programs Previously Found Not to Have Been Used or Provided No Benefits
We find that the following programs were not used during the POR: See Hynix's December 22, 2005 Questionnaire Response at 24 and the GOK's December 22, 2005 Questionnaire Response at 13.
A. Short-term Export Financing
B. Tax Programs Under the TERCL and/or the RSTA
1. Tax Credit for Investment in Facilities for Productivity
C. Tax Reductions or Exemption on Foreign Investments under Article 9 of
D. Duty Drawback on Non-Physically Incorporated Items and Excessive Loss
E. Export Insurance
F. Electricity Discounts Under the RLA Program
G. System IC 2010 Project
Enhancement (Article 25 of RSTA/Article 25 of TERCL)
2. Tax Credit for Investment in Facilities for Special Purposes
(Article 25 of RSTA)
3. Reserve for Overseas Market Development (formerly, Article 17 of
TERCL)
4. Reserve for Export Loss (formerly, Article 16 of TERCL)
5. Tax Exemption for Foreign Technicians (Article 18 of RSTA)
6. Reduction of Tax Regarding the Movement of a Factory That Has Been
Operated for More Than Five Years (Article 71 of RSTA)
the Foreign Investment Promotion Act (“FIPA”)/FIPA (Formerly Foreign
Capital Inducement Law)
Rates
In the first administrative review, the Department found that “any benefits provided to Hynix under the System IC 2010 Project are tied to non-subject merchandise” and, therefore, that “Hynix did not receive any countervailable benefits under this program during the POR,” in accordance with 19 CFR 351.525(b)(5). See First Administrative Review at 15. No new information has been provided with respect to this program. Therefore, we find that Hynix did not receive any countervailable benefits from the System IC 2010 Project during the POR.
Analysis of Comments
Comment 1: Benefit to Hynix of the 2004 Cash Buyout
Micron's Argument:
Micron asserts that the Department erred in applying its loan methodology when assessing the 2004 Cash Buyout Plan (“2004 CBOs”) n3 that was part of the sale of Hynix's non-memory assets to MagnaChip. Instead, Micron argues, the Department should have treated the 2004 CBOs as loan forgiveness. Micron points to language in Hynix's financial statement that, “during 2004, the Company realized an extraordinary gain on forgiveness of debt amounting to W43,727 million.” Micron urges the Department to reverse its finding in the Preliminary Results and countervail the full amount of the loan forgiveness associated with the 2004 CBOs.
n3 In using the term “2004 CBOs,” we are including the discounting of debt transferred to MagnaChip as part of the Tranche A acquisition financing.
According to Micron, there is no basis under the statute or the regulations to analyze the 2004 CBOs using a loan methodology. In Micron's view, creditors withdrew their existing loans, and accepted repayment at less than the outstanding principal value. Micron argues that there is nothing in the statute at 19 USC 1677(5)(E)(ii) or in the regulations at 19 CFR 351.505(a) which permits the Department to use the loan methodology to measure the benefits where a creditor forgives a portion of loan principal. To the contrary, Micron contends, the regulations explicitly require that the benefit from loan forgiveness be addressed under 19 CFR 351.508(a).
Micron objects to a “premise” of the Department's finding in the Preliminary Results, i.e., that “the discounts reflect ... the value to [Hynix's] creditors of obtaining repayment prior to the scheduled maturity of the loans.” Micron alleges that this position contravenes a fundamental principle of the countervailing duty law. Specifically, Micron contends that by looking at the value of the 2004 CBOs to Hynix's creditors, the Department improperly measured the subsidy by reference to the cost to government and not by the benefit to Hynix.
Finally, Micron points to Certain Cut-to-Length Carbon Steel Plate from Mexico: Preliminary Results of Countervailing Duty Administrative Review, 64 FR 48796, 48799 (September 8, 1999) and Certain Cut-to-Length Carbon Steel Plate from Mexico: Final Results of Countervailing Duty Administrative Review, 65 FR 13368 (March 13, 2000) and accompanying Issues and Decision Memorandum at Comment 2 (“CTL Plate from Mexico”). In CTL Plate from Mexico, the Department addressed a situation virtually identical to that of Hynix's 2004 CBO, Micron contends, and the Department specifically rejected application of the loan methodology. According to Micron, the Mexican respondent, AHMSA, prepaid outstanding principal owed to the Government of Mexico (“GOM”) at a discount, with the amount of the discount calculated using an allegedly commercial' interest rate. Although the GOM claimed to have received the full value of the outstanding debt, the Department concluded that the discounted prepayment eliminated AHMSA's debt without AHMSA having to repay the full amount of the principal owed. Consequently, the Department found that AHMSA received a countervailable benefit in the form of debt forgiveness.
Therefore, consistent with Hynix's characterization of the 2004 CBOs as debt forgiveness, the statute, the regulations, and past practice, Micron argues that the Department should measure the benefit received by Hynix under the 2004 CBOs pursuant to 19 CFR 351.508(a).
Hynix's Rebuttal:
Hynix contends that the Department has the discretion to consider the facts of the situation and, based on the facts, to determine the appropriate valuation method to apply. In this case, Hynix claims, the Department reasonably interpreted the facts and concluded that the 2004 CBOs were essentially renegotiations of existing loans.
Hynix further claims that Micron elevates form over substance to avoid “inconvenient” facts. In particular, Hynix points out that a significant amount of the debt paid off in the 2004 CBOs was held by wholly foreign-owned creditors; government authorities and wholly foreign-owned banks participated in the 2004 CBOs on the same terms; all creditors could voluntarily apply for early repayment; and the discount rates applied equally to all participants. Thus, in Hynix's view, the Department was correct to find no benefit.
Hynix contends that treating the 2004 CBOs as debt forgiveness would contravene the statute, court decisions and the entire purpose of the CVD law. According to Hynix, this is because the Department must examine whether a benefit exists in the context of what is otherwise commercially available. Hynix claims that 19 USC 1677(5)(E) describes the conditions under which a benefit will be found, and that for every type of program enumerated the Department must find inconsistency with commercial considerations in order to find a benefit. Hynix also cites to AK Steel Corp. v. United States, 192 F.3d 1367 (1999) and to Geneva Steel v. United States, 20 CIT 7 (1996), as requiring the Department to look to whether the terms of an alleged subsidy program are the same as those available on the market.
Hynix disputes the similarity of the situation in CTL Plate from Mexico to that of Hynix and its creditors. In CTL Plate from Mexico, the GOM had previously assumed AHMSA's debt to foreign creditors and then, to calculate the amount of the discount for early repayment, the GOM created an artificial amortization schedule, according to Hynix. Moreover, Hynix contends, the GOM was the only lender that participated in AHMSA's early repayment program. In Hynix's CBOs, in contrast, a number of private, independent commercial and government creditors were able to compete to receive early repayment through a voluntary bidding process, Hynix claims, with the result that terms for the early payoffs were market determined. Moreover, Hynix states, the same terms applied to all participating creditors.
Department's Position:
Micron has argued that the Department's treatment of the 2004 CBOs under its loan methodology conflicts with the statute, the Department's regulations and Department practice. We disagree. First, contrary to Micron's claim, the Department did not employ a “cost to government” standard in determining the countervailability of the 2004 CBOs. Micron has selectively quoted from the Department's statement in the Preliminary Results in this regard, and the full quotation makes clear that the Department was describing the commercial nature of the transaction (“... the discounts reflect the value to Hynix of repaying the loans and the value to its creditors of obtaining repayment prior to the scheduled maturity of the loans ...”) and not that it was adopting a cost-to-government measure of the subsidy.
Second, in this instance, the statute and the Department's regulations do not dictate the method for valuing the benefit in this type of transaction. Although the Department's regulations provide different methods for valuing different forms of financial contributions, it is up to the Department to determine the form of the financial contribution and then measure any benefit according to the regulations. For example, in the 1993 steel cases the Department encountered various financial instruments that did not fall readily into the basic categories of “grants,” “loans,” or “equity.” The Department developed criteria for assigning these so-called “hybrid instruments” into one or another of these basic categories and then, having characterized the financial contribution, applied the prescribed methodology for measuring the benefit. See Final Affirmative Countervailing Duty Determinations: Certain Steel Products from Austria, 58 FR 37217, 37254 (July 9, 1993).
Micron advances a mechanical approach -because Hynix paid its creditors less than the face value of the outstanding debt, debt forgiveness occurred and the benefit must be measured under 19 CFR 351.508. We believe that this is an overly simplistic view. Hynix and its creditors agreed to swap amounts owed in the future (2006) for cash in 2004. Because cash today is normally worth more than the promise of payment two years from now, it is reasonable that the borrower (Hynix) would demand a discount for repaying early and that the lenders (Hynix's creditors) would agree to a discount. Such discounting occurs frequently in the market among commercial actors: a discount would be expected, for example, on early redemption of bonds. And, because cash today is worth more than the promise of payment in the future, it would be inappropriate for the Department to automatically treat early payment discounts as debt forgiveness. Instead, further analysis is necessary.
Such an analysis occurred in CTL Plate from Mexico. In that case, the respondent claimed that it repaid its debt early, at a discount, and that the rate used to calculate the discount was a commercial rate. As Micron has noted, the Department found that the discount amounted to debt forgiveness. A similar analysis, but with a different result, occurred in Final Affirmative Countervailing Duty Determinations: Certain Steel Products from Belgium, 58 FR 37273, 37181-82 (July 9, 1993) (“Belgian Certain Steel”). In that case, Belgian government agencies that were creditors of the respondent sold their loan back to the respondent for cash and shares in the respondent prior to the maturity of the loan. Using the respondent's discount rate, we determined that the respondent should have paid more than it did to repurchase its debt. Thus, instead of finding debt forgiveness, we found that the sale of the loan “was inconsistent with commercial considerations.” n4
n4 Although the Department characterized the Belgian situation as a sale of shares in CTL Plate from Mexico, a careful review of Belgian Certain Steel shows that the Department viewed the instrument as a loan.
In analyzing the 2004 CBOs, we find that they differ from the early loan repayment examined in CTL Plate from Mexico in certain important respects. In CTL Plate from Mexico, the government was the only creditor that participated in the early repayment agreement, and it created an amortization period for the loans as the basis for calculating the discounted amount to be repaid. In the instant review, the 2004 CBOs were an early repayment process that was open to all of Hynix's creditors (Korean government authorities, Korean banks, and wholly-foreign owned banks) and participation in the process was voluntary. See Hynix SQNR at 5-8, Exhibit 9 and Hynix NSQR at Exhibits NA-9, NA11 and NA-12. The debt that was repaid early through the CBO had a fixed repayment date (2006). Importantly, creditors other than the Korean government authorities chose to participate and the terms of the early repayment applied equally to the policy banks and the commercial banks. Id. The presence of these commercials actors clearly distinguishes this situation from that examined in CTL Plate from Mexico.
Finally, Micron has pointed to language in Hynix's financial statements that refers to the early repayment at a discount as “forgiveness of debt.” We acknowledge that respondents' financial statements can provide important information used in calculating countervailable subsidies (e.g., financial information used for determining a respondent's creditworthiness or equityworthiness, the number of shares issued in exchange for an equity investment, timing of subsidy events). However, while a company's characterization of a financial contribution in its financial records may inform the Department's decision, it is not dispositive. (See, e.g., Final Affirmative Countervailing Duty Determinations: Certain Steel Products from France, 58 FR 37304, 37306-7 (July 9, 1993). The French respondents argued that “Loans with Special Characteristics (PACs)” should be treated as equity because they “...would be included in shareholders equity on the balance sheet...” The Department disagreed and treated the PACs as debt.)
Therefore, for the reasons explained above, the Department has continued to evaluate the 2004 CBOs under the loan methodology described in 19 CFR 351.505(a) for the final results. See “Analysis of Programs.”
Comment 2: The Department's Failure to Investigate Thoroughly the GOK's
Entrustment or Direction of Hynix's Creditors in Connection with
the Debt Forgiveness Component of the Non-Memory Asset Sale
Micron's Argument:
In Micron's view, the Department failed to ask the hard questions that might have elicited evidence of the GOK's continued entrustment or direction of Hynix's creditors in connection with the 2004 CBOs. According to Micron, the Department did not fully discharge its investigative obligation and, instead, simply deferred to the respondents' denials of GOK involvement, even when such denials were proven false in prior reviews.
In particular, Micron notes that the Department asked no questions regarding communications or meetings between the GOK and Hynix's creditors in the initial questionnaire and there was only a single question, directed to Hynix, in the March supplemental questionnaires. Micron further contends that upon initiation of the investigation of the 2004 CBOs, the Department still asked no specific questions regarding communications or meetings between the GOK and Hynix's creditors, instead asking simple “yes or no” questions. According to Micron, Hynix responded by denying GOK involvement. Because Hynix submitted similar denials in the investigation and first administrative review, and the Department concluded otherwise, i.e., that the GOK did entrust or direct Hynix's creditors, Micron argues that the Department should not have accepted Hynix's denials in this review.
Micron contends that the GOK has taken increasingly extraordinary steps to avoid scrutiny of its actions with regard to Hynix as the case has progressed and, accordingly, the Department should have investigated the 2004 CBOs thoroughly. Its failure to do so, in Micron's view, is an abuse of discretion (citing, Bethlehem Steel Corp. v. United States, 162 F. Supp. 2d 639 (Ct. Int'l Trade 2001) (“Bethlehem II”)).
Micron concludes by pointing out that the Department's reluctance to pursue the complete facts regarding the 2004 CBOs is compounded by its failure to initiate an investigation of the BOE Hydis (LCD) and MagnaChip asset deals. Because these provide critical context to the 2004 CBOs, the Department could have gleaned additional facts directly relevant to the GOK's policies and practices pertaining to the 2004 CBOs.
Hynix's Rebuttal:
Hynix contends that because Micron cannot find fault with the Department's preliminary determination regarding entrustment and direction, Micron has instead attacked the manner in which the Department investigated the alleged subsidy. But, Hynix claims, the Department's approach in this review was virtually identical to the investigative methods used in the prior review and many other reviews of countervailing duty orders. According to Hynix, the Department issued six questionnaires and allowed Micron to submit comments before sending each round of supplemental questionnaires.
Department's Position:
We disagree with Micron that the Department failed to investigate the 2004 CBOs fully. Micron is correct that there were no questions related to the 2004 CBOs or more generally towards GOK entrustment or direction of Hynix's creditors in 2004 included in the initial, November 2005 questionnaire. However, this was because at the time the initial questionnaire was sent, no new subsidies (i.e., subsidies bestowed during 2004) had been alleged. Absent the receipt of new subsidies in 2004, there was simply no need to question whether the GOK entrusted or directed Hynix's creditors in 2004.
Micron's new subsidy allegations were filed in January 2006. Additional filings in support of the allegations were made in April and May 2006. In June, the Department agreed that Micron had provided a basis to investigate the discounting of loans transferred to MagnaChip and the 2004 CBOs but no basis for investigating the other alleged subsidies. See New Subsidy Allegations Memorandum. Thus, the debt forgiveness subsidy as alleged by Micron was not under investigation until June.
Micron objects to the questions asked by the Department once the 2004 CBOs were being formally investigated, but this overlooks the extensive information already on the record about the 2004 CBOs. First, although the Department had not yet initiated an investigation of the 2004 CBOs, the Department sought information relating to the debt elimination in the March supplemental questionnaire. Hynix included this information in its supplemental questionnaire response. See Hynix SQNR at pages 7-8. Additionally, Micron submitted pre-initiation information in support of its allegation on January 12, 2006, and April 26, 2006, and post-initiation factual information on July 13, 2006. Hynix also filed rebuttal information regarding the alleged subsidies on January 27, 2006, and May 8, 2006. Thus, the Department had ample information on which to base its determination.
Micron claims the Department should not rely on Hynix's denials of government involvement. As evidenced by the discussion above and our response to Comment 3, we did not. Our conclusion that the GOK did not entrust or direct Hynix's creditors to participate in the 2004 CBOs is based on substantial evidence; it was reached after a thorough review of the information on the record, including circumstantial evidence of the sort relied upon by the Department in the investigation and first administrative review.
We further disagree with Micron that Bethlehem II is relevant here. In that case, the CIT found that the Department had “ignored” the plaintiffs' allegations by investigating the terms of the respondent's lease rather than the allegation that government-provided infrastructure did not serve the public welfare. The CIT also found that the Department erroneously decided not to investigate the reduction of import duties on slab. In this review, we fully investigated the subsidy as alleged by Micron, i.e., GOK entrustment or direction of Hynix's creditors to provide debt relief.
Comment 3: Entrustment or Direction of Hynix's Creditors in Connection with
the Tranche A Acquisition Financing and CBO Components of the Non-
Memory Asset Sale
Micron's Argument:
Micron asserts that the Department erred in failing to countervail the subsidies conferred by the discounting of loans transferred to Magnachip and the CBOs. Micron argues that the GOK continued its policy of ensuring Hynix's survival well into 2004 when the non-memory asset sale occurred. As part of its assertions, Micron contends that the non-memory asset sale was part of the 2002 bailout n5 (which the Department countervailed), disputes that Hynix had rebounded financially by 2004, and argues that the non-memory asset sale was critical to Hynix's continued recovery.
n5 Throughout its brief, Micron refers to the December 2002 restructuring as the 2003 bailout or restructuring. Consistent with the Department's determination in the first administrative review, we will refer to this event as the 2002 restructuring and, in the interest of clarity, have changed Micron's references to 2002.
Micron maintains that the burden is on respondents to demonstrate that the GOK no longer had a practice of supporting Hynix. Nonetheless, the Department misconstrued and overlooked evidence that the GOK continued to exert lending pressure in 2004 and that GOK entities and GOK-owned or controlled banks continued to have clout in 2004.
Micron's main arguments are as follows:
The non-memory asset sale was part of the 2002 bailout. Micron alleges that the sale of the non-memory assets, which gave rise to the discounting of loans transferred to Magnachip and CBOs, was part of the 2002 bailout, an event which has already been countervailed by the Department. In support of this, Micron claims that the controlling legal authorization for the sale occurred before 2004; the MOU that approved the sale of the non-memory assets was approved by the steering committee of Hynix's Creditors' Council in January 2003, and, the Creditors' Council agreed in December 2002 to an aggressive restructuring that included the sale of non-core businesses.
The non-memory asset sale was critical to Hynix's recovery and continued improvement. According to Micron, Hynix had not fully recovered by 2004 and the GOK had reason to continue to be concerned about Hynix's competitiveness in 2004. In support of this argument, Micron claims that Hynix had $500 million in debt coming due in 2004; its debts were 4.3 trillion won (in October 2004); its non-memory division was a drain on the company, having lost money in 2003; its stock was still 80 percent owned by creditor banks; and its financial prospects were dim in the eyes of industry analysts such as Bear Stearns, Morgan Stanley, and Merrill Lynch. Micron also contends that Hynix was lagging behind its competitors in moving to newer technologies, and that the cash from the non-memory asset sale was crucial for Hynix to regain its competitive advantage.
Micron further contends that the debt forgiveness that occurred in connection with the discounting of loans transferred to Magnachip and the CBOs runs contrary to profit-maximizing behavior. Thus, in Micron's view, this act in itself is evidence that debt forgiveness was sought by the GOK as a way to ensure that Hynix's financial condition would not suffer setbacks.
Micron asserts that even if Hynix's financial condition did improve between 2003 and 2004, the evidence shows that the GOK actively supported the asset sale first as a way to keep Hynix going, and later when the DRAM market rebounded and the company's numbers improved, as a means to provide Hynix with necessary capital to continue its recovery. Micron points to several reports by the Financial Supervisory Commission (“FSC”) and Financial Supervisory Service (“FSS”) in 2001 and 2003 which, it argues, indicate GOK interest in the restructuring of large, insolvent corporations. Micron claims that the Department wrongly dismissed a July 2004 KDB report to the Korean National Assembly which stated that the KDB would “continue to push for corporate restructuring” and identified Hynix as among the “affected companies,” as evidence of continued entrustment or direction of Hynix's creditors. In Micron's view, there can be no reason why the GOK would actively seek the sale of Hynix's business divisions if the GOK was not concerned about the company's long-term financial prospects and its continued improvement. Moreover, according to Micron, as the KDB is a GOK entity, its objectives are attributable to the GOK, and the KDB was well-positioned to influence the actions of other members of the Creditors Council.
In the absence of evidence that the GOK terminated its Hynix policy, the Department should find that the policy continued into 2004. According to Hynix, the Department's practice in this proceeding has been that where the record supports a prima facie finding that a policy or practice existed, the burden is on the respondents to submit evidence to the contrary. Neither Hynix nor the GOK did so and nothing on the record, in Micron's view, indicates that the GOK abandoned its policy of supporting the company.
The Department misconstrued evidence of GOK practice. Micron presents several instances in which, it claims, the Department misconstrued evidence of the continued GOK practice of supporting Hynix. First, Micron submitted an article about the 2004 firing of Kookmin Bank's president to demonstrate the GOK's continuing use of pressure and intimidation against Hynix's creditors in that year. The Department incorrectly dismissed the article on the grounds that it referred to the president's 2001 refusal to aid Hynix. Second, the Department wrongly dismissed a 2005 article regarding continued government adherence to the too big to fail' doctrine because it did not talk about GOK policies vis a vis Hynix in 2004. Micron states that it submitted the article to demonstrate continued GOK interference in the market to assist broken companies. Third, the Department dismissed a report of a January 2003 meeting in which the GOK addressed insolvent companies and specifically mentioned the Hynix debt restructuring and asset sales because the meeting did not occur in 2004. Micron states that it submitted the information because it linked the non-memory asset sale to the 2002 restructuring. Finally, Micron contends, even if 2004 is the proper year to focus on, it is logical to assume that the GOK adhered to its 2003 objectives at least through the sale of the non-memory assets in 2004.
The Department overlooked important evidence of GOK lending pressure. Micron argues that the Department overlooked significant evidence of GOK lending pressure before, during, and after the period of review. First, Micron points to statements from Shinhan Bank, Koomin Bank, Woori Bank, and KEB in their SEC filings or annual reports which are similar to statements which the Department found in the first administrative review to be “highly indicative of the general susceptibility of both GOK-owned or controlled banks and private banks to GOK influence.” These statements, in connection with the GOK's firing of the Kookmin president in 2004, show that the GOK continued to exert pressure, coercion, and threats against Hynix's creditors, and contradict the Department's assertion in the Preliminary Results that KEB was no longer pressured or threatened by the GOK.
Micron also claims that there is no indication that, by 2004, the GOK abandoned its vow to maintain silence on matters pertaining to Hynix's restructuring, which the Department relied upon in the first administrative review. Micron further asserts that the Department overlooked the fact that the GOK continued making capital injections into Hynix's creditors, including Shinhan Bank, Kookmin Bank, and Housing and Commercial Bank, and entered into MOUs with these creditors providing the GOK with additional leverage over bank management. According to Micron, the Department overlooked additional evidence of interventionist tactics in 2004, such as the bailout of LG Card, the Ministry of Commerce, Industry, and Energy's designation of the semiconductor industry as a backbone industry, and a May 24, 2004 meeting in which the GOK met with Hynix's creditors to urge Hynix's entry into China.
Continuing clout of GOK entities and GOK-owned or controlled banks in 2004. As noted above, Micron argues that the discounting of loans transferred to Magnachip and CBOs were part of the non-memory asset sale which was approved as part of the 2002 restructuring of Hynix. Consequently, Micron contends that the focus should be on the Creditors Council as it existed at that time and not at the time of the 2004 transactions. Assuming, however, that 2004 is the correct period to examine, Micron argues that the Department failed to find that the Creditors Council was still dominated by either GOK public entities or GOK-owned or-controlled banks.
First, KDB, which the Department has recognized as a specialized bank funded and controlled by the GOK, took a particularly important role in the discounting of loans transferred to Magnachip and the CBOs, and the larger non-memory asset sale. As the Department has recognized, the participation of KDB sends a clear signal of government support. Second, KEB was similarly important. The Department has acknowledged KEB's role as lead bank and, in the first administrative review, concluded that KEB acted at the government's behest in aiding Hynix due to the GOK's significant ownership of the bank. Micron cites press reports which, in its view, show that KEB “pressed” creditors into accepting the Tranche A acquisition financing and the CBOs. KDB and KEB, along with Chohung Bank and Woori Bank, virtually secured the 75 percent bloc needed to control the Creditors Council.
Micron disputes the Department's suggestion that the change in ownership at KEB in 2004 lessened the GOK's control of the bank. First, Micron asserts that the Department has always maintained that government ownership is not controlling for purposes of entrustment or direction and, thus, the Department cannot rely on this change in ownership as dispositive of the issue. Second, Micron claims that the GOK continued to hold 20 percent of the shares in KEB even after the acquisition of shares by Lone Star at the end of 2003, making the GOK the bank's second largest shareholder. Third, the Department incorrectly downplayed the statement by Lonestar at the time it acquired KEB's shares that it would not take part in management matters directly, explicitly leaving Hynix matters under the GOK's control. Micron challenges the Department's weighting of an analyst's statement that it would not be easy for Lone Star to separate itself from management decisions, contending that the Department should give more weight to Lone Star's announcement. In Micron's view, the article makes it clear that Lone Star sought to distance itself from KEBs's management decision because of Newbridge Capital's earlier experience with the KFB. Micron also argues that there is no information on the record that Lone Star actually participated in KEB's decisions to assist Hynix in 2003 and 2004.
Micron continues that KEB, acting as the head of Hynix's Creditors' Council and its Restructuring Committee, approved the sale of Hynix's non-memory assets in December 2002; KEB controlled Hynix's business activities through a business normalization plan; and continued as head of Hynix's Creditors' Council and Restructuring Committee through the 2004 asset sale. Finally, KEB continues to be subject to auditing by the Korean Board of Audit and Inspection, and a recent audit showed that KEB undervalued the bank's worth leading up to its sale to Lone Star.
With regard to new, foreign-owned creditors on the Creditors' Council in 2004, Micron maintains that they are irrelevant for purposes of examining the discounting of loans transferred to Magnachip and the CBOs. This is because these creditors -having purchased Hynix's debt on the secondary market had different motivations than the original participants in the restructuring. Moreover, according to Micron, their ability to influence any decision was limited.
Micron concludes by arguing that upon finding Hynix's creditors to be entrusted or directed by the GOK to provide a financial contribution to Hynix, the Department should find that Hynix received a financial contribution in the form of debt forgiveness and a benefit in the amount of the debt forgiven.
Hynix's Rebuttal:
Hynix contends that the inferences that may have supported the Department's findings of entrustment or direction through 2003 ended in 2003. In Hynix's view, the record evidence in 2004 must, on its own, support an independent basis for finding entrustment or direction. Hynix cites to the First Administrative Review, in which the Department acknowledged that, in considering the possibility of a “single program” of GOK entrustment or direction to save Hynix, “it must be mindful of the totality of the facts.” In Hynix's view, this means that inconsistencies and contradictions found in the individual pieces of evidence relied upon by the Department in considering the existence of a “single program” may weigh heavily on what interferences of GOK entrustment or direction may reasonably be drawn from their totality. Hynix also cites to decisions set forth by the Court of International Trade that state that “Commerce may lawfully support a finding of entrustment or direction with direct and circumstantial evidence drawn from across the alleged program” only when “the cumulated evidence and the reasonable inferences drawn there from sufficiently connect all the implicated parties and transactions to the alleged program of government entrustment or direction.” See Hynix Semiconductor, Inc. v. United States, 91 F. Supp. 2d1337, 1343 (Ct. Int'l Trade 2005). In cumulating the evidence, however, “Commerce must consider counterevidence indicating that the transactions making up that alleged program were formulated by an independent commercial actor (not a government) and motivated by commercial considerations.” Id.
With regard to commercial considerations, Hynix argues that even if there were GOK plans for asset sales in place in 2002, what is ultimately important is the actual terms and conditions of the sales. Hynix argues that Micron makes little to no attempt to address the fact that foreign, unrelated entities (i.e., Citigroup Venture Capital (“CVC”) and Francisco Partners (“FP”)) were involved in Hynix's non-memory asset sale. Hynix argues that, while Micron focuses solely on the conduct of Hynix's creditors in the non-memory assets sale, the conduct of the purchasers is compelling and refutes the alleged entrustment or direction by the GOK. Specifically,
(1) CVC and FP, which are prominent U.S. venture capital firms, agreed to
purchase the assets. These firms committed over $400 million of their
own capital, in addition to assuming the asset.
(2) These creditors were able to flip the entire value of the asset sale
(a sum far in excess of the acquisition financing) in international
bond markets within a matter of weeks of the sale, and
(3) The flip occurred on terms that were better than the asset sale
financing.
Hynix argues that Micron's support for claiming that Hynix remained in a precarious situation is weak because Micron cites to old analyst reports from September 2002 and February 2003. Hynix cites to numerous contemporaneous reports that refute claims of Hynix's financial weakness and constitute there was no need for governmental assistance in 2004. See Hynix Rebuttal at 16 and 17.
Finally, Hynix states that the Department's creditworthiness regulation at 19 CFR 351.505(a)(4)(D)(iii) directs the Department to ignore current and prior subsidies received by the firm in assessing performance in this review. In Hynix's view, the principle set forth in the regulation is applicable to the more general question of Hynix's financial health in relation to the “single program” previously found by the Department.
Department's Position:
We continue to find no entrustment or direction in the discounting of loans transferred to Magnachip or the CBOs. First, while Micron is correct that the sale of non-core assets was considered as a part of the bailout strategy for Hynix in earlier years, we are not investigating the sale of the non-memory assets. Instead, as evidenced by the New Subsidy Allegations Memo, we are investigating the discounting of debt transferred to Magnachip as part of the Tranche A acquisition loans and the CBOs.
There is no evidence that any earlier consideration of asset sales included any contemplation of discounting and early loan repayment. Thus, we find no basis to expand findings of entrustment or direction in earlier periods to the events being examined in this 2004 review. Even if the discounting of loans transferred to Magnachip and the CBOs could be considered part of the non-memory asset sale, the earlier considerations of asset sales set out a general desire or plan to make such sales. They did not set out the terms of the sales. Those details were not decided until Magnachip negotiated the sale and it was approved by the Creditors' Council (see Hynix SQNR at Exhibit 9). Thus, even if the discount and early repayment are considered part of the asset sale, we would treat 2004 as the relevant time period for examining whether the GOK entrusted or directed Hynix's creditors to provide a financial contribution.
Second, we disagree with Micron's contention that in the absence of evidence that the GOK terminated its policy of supporting Hynix, we should find that the policy continued. Micron cites to the Department's position in the investigation where we relied on earlier findings of directed credit to find entrustment or direction in the period 1992 -2000. However, in the investigation we applied findings made in final determinations for the years between 1992 and 2000 to those same years. We did not extend findings for earlier years to later years on the assumption that once begun, the policy continued until it was demonstrably terminated. See Investigation at 12-19. Thus, in this review, as in the first administrative review, we have investigated whether there was entrustment or direction in the POR based on information relevant to the POR.
Beyond these objections to our approach to analyzing entrustment or direction in connection with the discounting of loans transferred to Magnachip and the CBOs, Micron objects to numerous findings we made in the Preliminary Results. Regarding Hynix's financial health in 2004, Micron points to the company's outstanding debt in October 2004 and interest due, a loss in 2003 by the non-memory division, creditor bank ownership, and analyst reports. We find that other evidence shows Hynix's situation showed vast improvement in comparison to 2001 through 2003. As stated in the Preliminary Results, Merrill Lynch reported in October 2004 that, “we do not see any financial distress from Hynix.” See Hynix's January 27, 2006 Rebuttal Factual Information (“Hynix Rebuttal Information”) at Exhibit 22. Other contemporaneous analyses support this position. Id. at 19, 21, 26, 27, 33 and 35. The market analyses referred to by Micron generally date to 2002 and 2003. However, the company as a whole was profitable throughout 2004. See Hynix's June 30, 2006 supplemental questionnaire response at 4, 8-9, and Exhibit NA-3. Thus, we disagree with Micron's assessment of Hynix in 2004. Micron further argues that the GOK actively supported the sale of non-memory assets to provide for Hynix's continued recovery and to help it regain its competitive advantages. While proceeds from the sale put Hynix in a better position, the company's operating cash flows in 2004-2005 covered its capital spending (which exceeded $3.5 billion), according to Merrill Lynch (See Hynix's Rebuttal Information at tab 27, page 5). Micron further disagrees with our position in the Preliminary Results regarding the KDB's desire to promote corporate restructuring, including, inter alia, the sale of Hynix's business divisions. KDB's statement reflects its support for the sale as part of the restructuring, but does not talk about discounting of the debt transferred to Magnachip or the CBOs, nor does it demonstrate continued GOK entrustment or direction.
Regarding Micron's claim that we “misconstrued” certain evidence, we disagree. The article regarding the 2004 firing of the Kookmin Bank president stated this may have occurred because of his opposition to measures on behalf of Hynix in 2001, SK Global in 2003 and LG Card in 2004. We acknowledge that the GOK may have had the power in 2004 to influence banking decisions regarding “recovery measures.” (See NSA at tab 64) However, this article does not provide a basis for finding that the government entrusted or directed banks to make financial contributions to Hynix in 2004. Similarly, the article regarding the LG Card bailout describes ongoing Korean government intervention in support of companies “too big to fail.” (Id. at tab 66) Although this policy is ongoing, according to the author, the article describes the intervention in support of Hynix as having occurred between 2000 -2002. Thus, while we acknowledge that the Korean government may have intervened in banks' decisions on behalf of distressed companies in 2004, this article provides no basis to find that the GOK intervened on behalf of Hynix in 2004. The statements in the SEC filings (or financial statements) filed by Shinhan, Woori, Kookmin and KEB suffer from the same flaw. As we found in the first administrative review, they are “indicative of the general susceptibility of both GOK-owned and controlled banks and private banks to GOK influence.” See First Administrative Review at 37. In 2004, it appears that GOK influence may have been directed at LG Card. For example, Kookmin's SEC filing states that “in light of the financial market instability in Korea resulting from the liquidity problems faced by credit card companies during the first quarter of 2003, the Korean government announced temporary measures in April 2003 intended to provide liquidity support to credit card companies.” See NSA at tab 58.
Micron has argued that there is no indication that the GOK discontinued its policy of silence in directing the banks to assist in Hynix's restructuring. It is not clear what form such a renunciation would take, but as the record shows, the government was not reticent in directing the Seoul Guaranty Insurance Company to save LG Card. (Referred to at p. 40 of Micron brief). Moreover, as Micron has noted, the GOK was not silent about its discussions with Hynix's creditors in regard to building a plant in China (Id. at tab 60).
Finally, Micron contends that the Department failed to recognize the continuing clout of the GOK-owned or -controlled banks in the Creditors' Council. While we acknowledge KDB's continuing role as a Korean government agency, our analysis focused chiefly on the ownership of KEB and the fact that the GOK no longer owned a controlling majority of the voting on Hynix's Creditors' Council. For example, in the first administrative review, we found that banks in which the GOK held at least a 25 percent ownership stake controlled over 75 percent of the votes in the Creditors' Council. See First Administrative Review at 41. In 2004, by contrast, banks in which the GOK held at least a 25 percent ownership stake held 19.57 percent of the voting rights. See Memorandum to File from Team, Regarding Release of Bracketed Information, and (Business Proprietary) Preliminary Calc Memo at 3 and Attachment 3.
Regarding KEB, we agree with Micron that majority government ownership of a bank has never been a prerequisite for finding that a bank is entrusted or directed by the GOK. For example, in the first administrative review, we found only that the GOK was the predominant owner of KEB. However, that high level of GOK ownership, combined with statements by KEB to the effect that it was acting on the basis of the government's social and economic concerns rather than commercial concerns, led us to find that KEB was entrusted or directed by the government to provide financial contributions to Hynix. With Lone Star's purchase of 51 percent of KEB's shares, KEB's situation is more like that of KFB in the investigation and first administrative review. Despite Newbridge Capital's ownership of 51 percent of KFB's shares, we still found GOK entrustment or direction of KFB because the GOK threatened KFB to ensure its participation in the 2001 restructuring. However, as explained in the Preliminary Results, there is no evidence of threats to Lone Star during the 2004 POR.
Micron also makes much of the statement that Lone Star intended to separate the ownership of the bank from its management. A closer reading of the article shows that Lone Star preferred to take a general supervisory role through its outside directors (five of the newly appointed outside directors were recommended by Lone Star). The reason for this, according to the article, was that local managers would be familiar with the characteristics of Korean financial firms, such as strong labor unions. There is no indication as Micron's comment suggests that Hynix matters were left under the GOK's control. Moreover, as noted in the Preliminary Results, at least one analyst expressed skepticism that ownership and management could be separated.
Micron continues that KEB “pressed” other banks to accept the discounting of loans transferred to Magnachip and the 2004 CBOs. Rather, the press reports of statements made by KEB personnel simply state that the parties disagreed and would meet again to discuss the issue (See NSA at tab 38) or that the parties were able to reach agreement after compromising on the terms of the CBOs (Id. at tab 42). There is nothing in these reports to suggest that these events indicate anything other than the normal give and take among negotiating parties.
In sum, the lack of evidence demonstrating a GOK policy or pattern of practices to entrust or direct Hynix's creditors to provide financial assistance to Hynix in 2004, the GOK's lack of sufficient voting rights to dominate the Creditor's Council in 2004, and Hynix's improved financial situation in 2004 lead us to conclude that the discounting of loans transferred to Magnachip and the CBOs were not the result of GOK entrustment or direction. Consequently, we need not address the issues of financial contribution and benefit.
Comment 4: Whether the Department Should Have Investigated Hynix's LCD and
Non-Memory Assets
As explained above, on January 12, 2006, Micron alleged “new” subsidies to Hynix arising from Hynix's sales of its TFT-LCD and non-memory Assets. The Department determined, on June 8, 2006, not to investigate these asset sales because the information submitted by Micron did not support its allegations of countervailable subsidies. See New Subsidy Allegations Memorandum.
Micron's Argument:
Micron objects to the Department's decision not to investigate these alleged subsidies, arguing that evidence shows the assets were worth less than the sales price and, hence, Hynix received subsidies. In support of its claim, Micron points to evidence it presented regarding the allegations and restates several arguments made in its various new subsidies filings regarding the value of the TFT-LCD and non-memory assets.
Moreover, Micron contends that the Department relied upon the incorrect statutory provision in rejecting the new subsidy allegations. Specifically, Micron states that the Department should not have employed the standard in section 702 of the Act, which requires petitioners to support their allegations with “information reasonably available” to them. Instead, Micron claims, the Department should have followed section 775 of the Act (relating to subsidies discovered in the course of a proceeding) which directs the Department to investigate subsidies “that appear” to be countervailable. In relying upon section 702 to reject Micron's allegations, the Department imposed a legal standard that is contrary to the statute, according to Micron.
Micron concludes by arguing that the Department erred in making a factual finding on the basis of an incomplete and contradictory record. Instead, the Department should have investigated the asset sales and developed a full evidentiary record before making its factual findings. Micron points to Bethlehem Steel Corp. v. United States, 140 F. Supp. 2d 1354 CIT 2001 (“Bethlehem I”) where, Micron claims, the Department also refused to investigate an alleged subsidy and was overturned by the CIT.
Hynix's Rebuttal:
Hynix disagrees with Micron's contention that section 775 sets the standard for investigating these subsidy allegations, and argues that the Department was correct in employing the section 702 standard. Hynix further contends that there is little practical distinction between the two standards in terms of the evidence required to initiate.
Hynix claims that Micron seeks a lower initiation threshold when it argues that once a proceeding is underway the petitioner only need raise the prospect of a subsidy without any supporting information. This would lead to subsidy investigations without end, according to Hynix. Instead, if the Department discovers a practice that appears to be a subsidy based on tangible, credible evidence, the Department is required to investigate that subsidy by section 775. Thus, Hynix claims, because the Department did not make such a “discovery” it was appropriate for the Department to demand from Micron reasonably available information supporting its allegations.
Hynix disputes Micron's reliance on Bethlehem I. According to Hynix, the court in Bethlehem I disagreed with the Department's making a factual finding based on assumptions. In this review, to the contrary, the Department's decision not to investigate the alleged subsidies was based on record facts, Hynix claims.
Department's Position:
As evidenced by the New Subsidy Allegation Memorandum, the Department carefully considered Micron's allegations and the information submitted by Micron to support the allegations. For both asset sales, we disagreed with Micron that the sale represented a direct transfer of funds to Hynix, in the form of a grant or its functional equivalent, or that the purchasers functioned as funding mechanisms that funneled funds from the creditors to Hynix. We further concluded that the information submitted by Micron did not support its claims that the buyers merely conveyed funding as an intermediary between two other parties or that the purchasers paid more than adequate remuneration.
In its brief, Micron contests the last finding, i.e., the lack of information supporting its claim that the purchasers paid more than adequate remuneration in these sales. In arguing this point, Micron restates certain claims that were made at the time of its allegations, but were not specifically addressed in the New Subsidy Allegations Memorandum. We address these claims below:
* Micron claims that the failed sale of Hynix's TFT-LCD assets in 2001 and a
predicted worsening of the market in 2003 provide evidence that the TFT-
LCD assets were not worth the $410 million price subsequently paid for
them by BOE.
We disagree. There is no evidence that the intended 2001 purchasers failed to receive financing because they had offered too high a price. Instead, the evidence shows that at the time of the initial offer (June 2001), the intended purchasers were BOE and Cando Corp. of Taiwan. See New Subsidy Allegations at tab 9. According to BOE, this offer did not succeed because of disagreements between itself and Cando. Id. at tab 14, p. 2. A subsequent offer of $400M for the TFT-LCD assets by another Cando-led consortium did fail because the financing fell through, but there is no evidence that the lenders in question (the Korean agencies and entrusted or directed banks, KEB, Chohung, KDB and Hanvit) withdrew financing because the price was too high. If anything, they may have felt that they could wait and receive a better price. One creditor claimed, “as profits are being generated on the TFT-LCD side, there will be no liquidity issues for Hynix in the short run even without the sale of Hydis...but there is no change (to the position) that TFT-LCD must be sold in the long term.” Id. at tab 10. Also, an article regarding the $400M offer is entitled “(Hasty Sale of Hynix LCD) Top of the Line Business Sold at Scrap Value,” which indicates that a least one commentator saw the TFT-LCD business as worth more than was being offered. Id. at tab 11.
Micron also points to a predicted decline in the market for 2003 to argue that the value of the TFTLCD assets would also decline. However, BOE did not share the view. As stated in the article describing BOE's acquisition of the assets:
Up to January of 2003, there were great fluctuations in the TFT-LCD
market price, a 17” LCD screen could only be sold at US$210. (BOE)
observed this opportunity and resolutely fulfilled the acquisition with
US$380,000,000. At that time, some people in the same industry thought
that the market prices for TFT-LCD could still fall, that the price for a
17” LCD monitor could fall to the bottom price line of US$180, so they
believed they should continue to wait. However, based on various market
signals, (BOE) arrived at the conclusion that the IT industry as a whole
would resume to normal and the demand for LCD would rise and the price was
to bounce back.-Id. at tab 14, pg. 3.
* Micron contends that BOE could not secure independent outside financing
for the deal, that some of Hynix's own creditors refused or were reluctant
to participate, and that the terms of the acquisition financing were
inconsistent with market terms.
None of the statements by the concerned banks indicate that their willingness to lend (or not) was a result of the BOE-offered price. Of the $410M paid, $210M came from the acquisition financing, $150M (cash) from BOE, and the remainder from promissory notes issued by BOE. The statements cited by Micron show uncertainty about the TFT-LCD business cycle (Id. at tab 68), BOE's post-sale investment plans, and repayment guarantees (Id. at tab 25). Certain of these concerns were resolved, as Woori Bank, for example, joined in the financing (Id. at tab 16). We note that one report indicates that one of Hynix's creditors, Chohung Bank, did not participate because it viewed the purchase as new financial support for Hynix. Id. at tabs 16 and 69. However, other reports indicate that Chohung did not participate because of “complex situations involving its own sale” (Id. at tab 5) and because the bank was undercapitalized (Id. at tab 17). It is not surprising that BOE turned to Hynix's creditors for financing because those creditors had indicated a willingness to provide financing in the earlier offer (from Cando) (Id. at tab 10) and BOE thought that Hynix's creditors would be anxious to sell (Id. at tab 14, pg. 3). In short, the record indicates that certain lenders did have concerns about various aspects of the financing. However, that they had such concerns does not provide a basis for Micron's claim that BOE paid too much for Hynix's TFT-LCD assets. Finally, while claiming that the financing terms were inconsistent with market terms, Micron provided no evidence about commercial terms for acquisition loans.
* Micron points to Hynix's accountants' valuation of the non-memory assets
to argue that the purchase price exceeded their value.
Other information submitted by Micron indicates that an independent valuation contemporaneous with the sales transaction yielded a different conclusion. Specifically, MagnaChip's 2005 Annual Report, which it filed with the U.S. Securities and Exchange Commission, states:
At the date of the Original Acquisition, the Company obtained an
independent valuation of the acquired property, plant and equipment and
the acquired intangible assets, to ascertain the fair value of the
assets... The Company also re-assessed the remaining useful lives of the
acquired property, plant and equipment, and commenced depreciation on a
straight-line basis... The aggregate fair values obtained exceeded total
purchase consideration of $763,378 thousand, and this difference was
applied to the Company's intangible and non-current assets on a pro-rata
basis... (emphasis added).
See Micron's April 26, 2006 submission (“New Subsidy Supplemental Allegations”) at Exhibit 3, pg. 47.
Regarding the non-memory assets, we note that this sale occurred in 2004, the period covered by this administrative review. To the extent that Micron's allegation relies upon entrustment and direction of Hynix's creditors, as set forth above under the “Analysis of Programs” section, we have determined that such entrustment or direction did not exist in 2004.
Beyond the specifics of its new subsidy allegations, Micron has argued that the Department applied the incorrect initiation standard in its decision. The Department disagrees with Micron that the correct standard is the “appears to be” standard in section 775 of the Act rather than the standard in section 702 of the Act. We interpret the “appears to be” standard as applying when there is an untimely allegation, or no allegation at all (that is, when the Department on its own discovers a potential subsidy), in an investigation or review. This is apparent from the language of section 775 of the Act itself, which refers to situations in which the Department “discovers a practice which appears to be a countervailable subsidy, but was not included in the matters alleged in a countervailing duty petition....” Our regulation implementing section 775 of the Act supports this interpretation. It states: “If during... a countervailing duty administrative review the Secretary discovers a practice that appears to provide a countervailable subsidy with respect to the subject merchandise and the practice was not alleged or examined in the proceeding... the Secretary will examine the practice, subsidy, or subsidy program if the Secretary concludes that sufficient time remains before the scheduled date for the ...final results of review.” 19 CFR 351.311(b) (emphasis added). Timely allegations, whether made in the investigation or in a subsequent administrative review, are governed by section 702 of the Act.
Prior cases involving section 775 of the Act support this interpretation. In the CVD investigation of Stainless Steel Plate in Coils (“SSPC”) from Belgium, the petitioner “alleged” a new subsidy after the regulatory deadline in 19 CFR 351.301(d)(4)(i)(A), and the Department declined to investigation this allegation. See Allegheny Ludlum Corp. v. United States, 112 F. Supp. 2d 1141 (CIT 2000) (“Allegheny Ludlum I”). The CIT remanded to the Department for it to consider whether it had an obligation under section 775 of the Act to investigate the untimely alleged subsidy. Id. at 1149-51. This case does not stand for the proposition that a timely new subsidy allegation in an administrative review is governed by section 775 of the Act.
The same is true for Bethlehem I. In that case, involving the CVD investigation of Certain Cut-to-Length Carbon-Quality Steel Plate from Korea, the petitioner made an untimely new subsidy allegation. Referring back to Allegheny Ludlum I, the CIT explained that “even in cases where an allegation is untimely... the court held Commerce bound to investigate allegations that reasonably appear to be countervailable and are discovered within a reasonable time prior to the completion of its investigation.” Id. at 1361. Accordingly, the CIT remanded to the Department. Again, this case does not speak to the proper standard to apply when there is a timely allegation in an administrative review.
In any event, the “appears to be” standard in section 775 of the Act is functionally equivalent to the initiation standard in section 702 of the Act. In the Department's remand redetermination in SSPC from Belgium, we explained that the agency “ pursues those practices where the basic initiation threshold is met, i.e., there must be evidence on the record indicating that the elements necessary for imposition of countervailing duties are present.'“ Allegheny Ludlum Corp. v. United States, 25 CIT 816, 825 (2001) (“Allegheny Ludlum II”). The Department declined to investigate under section 775 of the Act, and the CIT affirmed. Id. at 816. The CIT rejected the petitioner's interpretation of the “appears to be” standard, which “would force Commerce to engage in a complete countervailing duty investigation in any instance where there is even the slightest suggestion that a certain business practice is a countervailable subsidy.” Id. at 823. Rather, the Department reviews the record, including “both negative evidence and positive evidence,” to determine whether there appears to be a countervailable subsidy. Id. at 824.
Therefore, the “appears to be” standard is in essence no lower than the initiation standard in section 702 of the Act. Congress could not have intended that it be more difficult for the Department to initiate an investigation on the basis of a petition than on the basis of a potential subsidy discovered in an investigation or review. As such, Micron's allegation fails under the benefit element of a countervailable subsidy. That is, Micron did not support its allegation that the assets were valueless or worth less than what the buyers paid and it does not “appear” that the assets were valueless or worth less than what the buyers paid.
Comment 5: Uncreditworthy Benchmark Interest/Discount Rate
The Department uses a benchmark interest rate to measure the benefit conferred by government loans, and that benchmark interest rate is also used as a discount rate to allocate subsidies over time. See 19 CFR 351.505 and 351.524(d)(3)(i). In situations where a company is found uncreditworthy, the Department constructs the benchmark/discount rate. See 19 CFR 351.505(a)(3)(iii) and 351.524(d)(3)(ii). Among the variables included in the uncreditworthy benchmark/discount rate, are a risk measure for creditworthy companies (the probability of default by a creditworthy company within n years) and the interest rate paid by creditworthy companies. See 19 CFR 351.505(a)(3)(iii). In the first administrative review and the Preliminary Results, the Department used the AA -corporate bond rate published by the BOK as the interest rate paid by creditworthy companies.
Micron's Argument:
Micron objects to the Department's use of the AA -corporate bond rate as the average benchmark interest rate for creditworthy companies. Instead, Micron contends that the Department should use an average of the BOK's AA-and BBB-corporate bond rates because this would reflect an average of the interest rates available to all creditworthy companies. Moreover, Micron argues, the Department's formula must be based on an apples-to-apples comparison, i.e., creditworthy interest rates used in the formula should match as closely as possible the data used to calculate the probability of default for creditworthy borrowers. Micron asserts that, because the Department uses data on the probability of default for all A and B rated companies, it is incorrect not to rely on the interest rates for both A and B rated companies.
Micron argues that, in the first administrative review, the Department defended its use of the AA-rate based on language in the Preamble to the proposed regulation (“Proposed Preamble”) establishing the formula for calculating the uncreditworthy benchmark. Micron contends that the Department inappropriately relied on the Proposed Preamble language regarding “a healthy, low-risk company,” although this language changed to “an average, creditworthy company,” in the Preamble. n6 Moreover, Micron argues, elsewhere in the Proposed Preamble, the Department did not limit the appropriate benchmark so as to apply only to healthy, low-risk companies such as those with AA-credit ratings. Rather, the Proposed Preamble refers to “safer, creditworthy companies” and “creditworthy” companies. Thus, according to Micron, the Preamble language actually directs the use of an “average” rate.
n6 See Countervailing Duties; Notice of proposed rulemaking and request for public comments, 62 FR 8818, 8830 (February 26, 1997) (“Proposed Preamble”); see Preamble, 63 FR at 65365.
Micron concludes by urging the Department to use an average of the rates available to creditworthy companies because this is consistent with the Department's practice of using a national average benchmark rate when no company-specific rate is available and is also consistent with the Department's past practice in calculating the uncreditworthy benchmark/discount rate (citing Final Affirmative Countervailing Determination: Sulfanilic Acid from Hungary, 67 FR 60223 (September 25, 2002) and accompanying Issues and Decision Memorandum at 7 (“Sulfanilic Acid from Hungary”) and Final Affirmative Countervailing Determination: Carbon and Certain Alloy Steel Wire Rod from Canada, 67 FR 55813 (August 30, 2002) and accompanying Issues and Decision Memorandum at 8 (“Wire Rod from Canada”)).
Hynix's Rebuttal:
Hynix argues that the Department applied the appropriate national average interest rate in calculating the uncreditworthy benchmark and discount rates, consistent with its decision in the first administrative review. Hynix claims that the Department's regulations do not dictate use of a specific bond rating over another or require an average bond rating. Moreover, according to Hynix, the Preamble indicates the Department's intention to use an interest rate that an “average, creditworthy company” would pay. Absent a finding by the Department on what constitutes an “average, creditworthy” company in Korea, Hynix contends that the Department should not rely in whole, or in part, on a lesser investment grade rate in calculating the uncreditworthy benchmark and discount rates.
Hynix concurs in the Department's finding from the first administrative review, that “there is no evidence to indicate that the BBB -rate is the rate that would be paid by a creditworthy borrower” in Korea. Id. at 79. Consequently, Hynix states that the Department should use the discretion afforded under 19 CFR 351.524(d)(3)(iii) to use the rate it considers most appropriate, the AA -rate. Department's Position:
We have continued to use the AA - rate as the measure of the interest rate that would be paid by creditworthy companies in Korea. We believe that Micron is correct that the interest rate used should correspond as closely as possible to the risk measure in the uncreditworthy benchmark/discount rate, and this is accomplished in this case by using the AA -rate. This is because the risk measure is the default rate for “investment grade” companies, and “investment grade” companies include companies with ratings from Aaa to Baa. If we had interest rates for Aaa and Baa companies in Korea, an average of those rates would be appropriate because it would provide the proverbial apples-to-apples comparison with the risk measure. Lacking such information, however, we have selected the AA -rate as it likely falls near the middle of this range. An average of the AA -and BBB-rates, as suggested by Micron, would be skewed toward the high end of the range and, hence, would not produce an “applesto-apples” comparison.
As evidenced by this review, the interest rate for creditworthy companies that is used in calculating the uncreditworthy benchmark/discount rate will depend on the data available in a particular case. The same is true regarding the selection of the national average benchmark rates. Thus, while Micron can cite to cases where particular data was used, those examples do not provide guidance for all cases. For example, Micron cites Sulfanilic Acid from Hungary and Wire Rod from Canada as instances where the Department used averages to represent the interest rate that would be paid by creditworthy companies in the calculation of the uncreditworthy interest/discount rate. However, there is no discussion in those cases of what was included in the averages or other alternatives that were available. As explained above, the Department is not rejecting use of an average in the instant proceeding. Rather, we simply disagree with the average Micron has proposed.
We acknowledge that we incorrectly relied on language from the Proposed Preamble in explaining our position in the first administrative review. However, despite Micron's arguments, we do not agree that the change in language between the Proposed Preamble and the Preamble language reflects a change in the Department's position on this matter. A close reading of the Preamble shows that no commenters addressed this point and there is no indication that the Department sought to change or correct its earlier characterization. In any case, as we stated in the first administrative review, there is no information on the record of this proceeding to indicate that creditworthy companies in Korea pay the BBB-rate.
Recommendation
Based on our analysis of the comments received, we recommend adopting all of the above positions. If these recommendations are accepted, we will publish the final determination in the Federal Register.
AGREE ---- DISAGREE ----
David M. Spooner
Assistant Secretary for Import Administration
(Date)