E6-13167. Dynamic Random Access Memory Semiconductors from the Republic of Korea: Preliminary Results of Countervailing Duty Administrative Review

AGENCY:

Import Administration, International Trade Administration, Department of Commerce.

SUMMARY:

The Department of Commerce is conducting an administrative review of the countervailing duty order on dynamic random access memory semiconductors from the Republic of Korea for the period January 1, 2004, through December 31, 2004. We preliminarily find that Hynix Semiconductor, Inc. received countervailable subsidies during the period of review. If the final results remain the same as these preliminary results, we will instruct U.S. Customs and Border Protection (“CBP”) to assess countervailing duties as detailed in the “Preliminary Results of Review” section of this notice.

Interested parties are invited to comment on these preliminary results (see the “Public Comment” section of this notice, below).

EFFECTIVE DATE:

August 11, 2006.

FOR FURTHER INFORMATION CONTACT:

Steve Williams and Andrew McAllister , Office of Antidumping/Countervailing Duty Operations, Office 1, Import Administration, International Trade Administration, U.S. Department of Commerce, Room 3069, 14th Street and Constitution Avenue, N.W., Washington, D.C. 20230; telephone: [phone] or [phone], respectively.

SUPPLEMENTARY INFORMATION:

Case History

On August 11, 2003, the Department of Commerce (“the Department”) published a countervailing duty order on dynamic random access memory semiconductors (“DRAMS”) from the Republic of Korea (“ROK”). See Notice of Countervailing Duty Order: Dynamic Random Access Memory Semiconductors from the Republic of Korea, 68 FR 47546 (August 11, 2003) (“CVD Order”). On August 1, 2005, the Department published a notice of “Opportunity to Request Administrative Review” for this countervailing duty order. On August 30, 2005, we received a request for review from the petitioner, Micron Technology, Inc. (“Micron”). On August 31, 2005, we received a request from Hynix Semiconductor, Inc. (“Hynix”). In accordance with 19 CFR 351.221(c)(1)(i) (2004), we published a notice of initiation of the review on September 28, 2005. See Initiation of Antidumping and Countervailing Duty Administrative Reviews and Request for Revocation in Part, 70 FR 56631 (September 28, 2005) (“Initiation Notice”).

On November 2, 2005, we issued countervailing duty questionnaires to the Government of the Republic of Korea (“GOK”) and Hynix. We received responses to these questionnaires in December 2005. Micron submitted comments on Hynix's questionnaire responses in January 2006. In March 2006, we issued supplemental questionnaires to the GOK and Hynix, and we received responses to these supplemental questionnaires in April 2006.

On January 12, 2006, we received a new subsidies allegation from Micron. On April 26, 2006, Micron submitted a supplement to its January 12, 2006, new subsidies allegation. On June 8, 2006, we initiated an investigation of two of the five new subsidies that Micron alleged in this administrative review. See New Subsidy Allegations Memorandum, dated June 8, 2006, available in the Central Records Unit (“CRU”), Room B-099 of the main Department building.

On April 25, 2006, we published a postponement of the preliminary results in this review until August 7, 2006. See Dynamic Random Access Memory Semiconductors from the Republic of Korea: Extension of Time Limit for Preliminary Results of Countervailing Duty Review, 71 FR 23898 (April 25, 2006).

In June 2006, we issued supplemental questionnaires to the GOK and Hynix regarding the new subsidies alleged by Micron. We received responses to the supplemental questionnaires on June 30, 2006. On July 13, 2006, Micron submitted pre-preliminary comments and a separate compilation of rebuttal factual information. On July 18, 2006, Hynix responded to Micron's July 13, 2006 submissions. On July 21, 2006, Micron submitted comments on the GOK and Hynix's supplemental questionnaire responses. On July 26, 2006, we issued another supplemental questionnaire to Hynix, and we received Hynix's response on August 2, 2006.

Scope of the Order

The products covered by this order are DRAMS from the Republic of Korea, whether assembled or unassembled. Assembled DRAMS include all package types. Unassembled DRAMS include processed wafers, uncut die, and cut die. Processed wafers fabricated in the ROK, but assembled into finished semiconductors outside the ROK are also included in the scope. Processed wafers fabricated outside the ROK and assembled into finished semiconductors in the ROK are not included in the scope.

The scope of this order additionally includes memory modules containing DRAMS from the ROK. A memory module is a collection of DRAMS, the sole function of which is memory. Memory modules include single in-line processing modules, single in-line memory modules, dual in-line memory modules, small outline dual in-line memory modules, Rambus in-line memory modules, and memory cards or other collections of DRAMS, whether unmounted or mounted on a circuit board. Modules that contain other parts that are needed to support the function of memory are covered. Only those modules that contain additional items which alter the function of the module to something other than memory, such as video graphics adapter boards and cards, are not included in the scope. This order also covers future DRAMS module types.

The scope of this order additionally includes, but is not limited to, video random access memory and synchronous graphics random access memory, as well as various types of DRAMS, including fast page-mode, extended data-out, burst extended data-out, synchronous dynamic RAM, Rambus DRAM, and Double Data Rate DRAM. The scope also includes any future density, packaging, or assembling of DRAMS. Also included in the scope of this order are removable memory modules placed on motherboards, with or without a central processing unit, unless the importer of the motherboards certifies with CBP that neither it, nor a party related to it or under contract to it, will remove the modules from the motherboards after importation. The scope of this order does not include DRAMS or memory modules that are re-imported for repair or replacement.

The DRAMS subject to this order are currently classifiable under subheadings 8542.21.8005 and 8542.21.8020 through 8542.21.8030 of the Harmonized Tariff Schedule of the United States (“HTSUS”). The memory modules containing DRAMS from the ROK, described above, are currently classifiable under subheadings 8473.30.10.40 or 8473.30.10.80 of the HTSUS. Removable memory modules placed on motherboards are classifiable under subheadings 8471.50.0085, 8517.30.5000, 8517.50.1000, 8517.50.5000, 8517.50.9000, 8517.90.3400, 8517.90.3600, 8517.90.3800, 8517.90.4400, and 8543.89.9600 of the HTSUS.

Scope Rulings

On December 29, 2004, the Department received a request from Cisco Systems, Inc. (“Cisco”), to determine whether removable memory modules placed on motherboards that are imported for repair or refurbishment are within the scope of the CVD Order. The Department initiated a scope inquiry pursuant to 19 CFR 351.225(e) on February 4, 2005. On January 12, 2006, the Department issued a final scope ruling, finding that removable memory modules placed on motherboards that are imported for repair or refurbishment are not within the scope of the CVD Order provided that the importer certifies that it will destroy any memory modules that are removed for repair or refurbishment. See Final Scope Ruling Memorandum from Stephen J. Claeys to David M. Spooner, dated January 12, 2006

Period of Review

The period for which we are measuring subsidies, i.e., the period of review (“POR”), is January 1, 2004, through December 31, 2004.

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”).

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”). For 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 non-recurring 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 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 either did not obtain bonds or obtained bonds through countervailable debt restructurings during 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 won-denominated corporate bonds as reported by the BOK and the U.S. dollar lending rates published by the IMF for each year.

Short-Term Loans

Consistent with the methodology used in the first administrative review, we use the money market rates as reported in the IMF's International Financial Statistics Yearbook for short-term interest rates. For countries (or currencies) for which a money market rate was not reported, we are utilizing the lending rate from the same source.

Creditworthiness

We have not analyzed Hynix's creditworthiness for 2004.

Analysis of Programs

I. Programs Previously Determined to Confer Subsidies

We examined the following programs determined to confer subsidies in the investigation and first administrative review, and preliminarily 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, in order to assist Hynix through its financial difficulties. 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 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 preliminarily 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.

B. Operation G-7/HAN Program

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 Decision Memorandum 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 Operation 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 preliminarily 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. 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 in the POR to calculate the countervailable subsidy rate. On this basis, we calculated a preliminarily subsidy rate of less than 0.005 percent ad valorem for this program and, therefore, we did not include this program in our preliminary net countervailing duty rate, which is consistent with our past practice. See e.g., Notice of Preliminary Results of Countervailing Duty Review: Certain Softwood Lumber Products from Canada, 70 FR 33088, 33091 (June 7, 2005).

II. Programs Preliminarily 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 Issues and Decision Memorandum for the Final Determination in the Countervailing Duty Investigation of Dynamic Random Access Memory Semiconductors from the Republic of Korea, June 16, 2003 (“Investigation Decision Memorandum”) at 47-61 and Issues and Decision Memorandum for the Final Results in the First Administrative Review of the Countervailing Duty Order on Dynamic Random Access Memory Semiconductors from the Republic of Korea, March 14, 2006 (“AR1 Decision Memorandum”) 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 Decision Memorandum 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, 2005 submission 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 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 System IC sale. See New Subsidy Allegations Memorandum, dated June 8, 2006. 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 System IC unit 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 in 2004, 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 two-part test that we used in the investigation and first administrative review to determine whether the GOK entrusted or directed creditors to reduce Hynix's debt in 2004. 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).

In the final results of the first administrative review, the Department found that the nexus of Hynix's poor financial condition in 2002, the GOK's involvement in various solutions to Hynix's financial woes (including the possible sale of Hynix to Micron), the GOK's dominance of the Creditors' Council (through its ownership and control of various member-creditors), GOK threats towards Hynix's creditors, and various statements made by high-ranking GOK officials with respect to dealing with Hynix's troubles, among other things, demonstrated that the GOK entrusted or directed Hynix's creditors to participate in the December 2002 financial restructuring. See AR1 Decision Memorandum at 5-10 and Comment 1. Most of the evidence supporting the Department's finding was contemporaneous with Hynix's financial restructurings in 2002. The record evidence in this review, however, either fails to demonstrate that the GOK entrusted or directed Hynix's creditors in 2004 or relates to GOK actions that occurred prior to 2004.

First, 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 was drastically improved in comparison to 2001 through 2003. For instance, Hynix consistently generated significant revenue, profit, and return on equity throughout 2004. See Hynix's June 30, 2006 supplemental questionnaire response at 4, 8-9, and Exhibit NA-3. In fact, Hynix reported a record net profit of 26 percent in 2004, in contrast to the double-digit negative profit margins that Hynix generated during 2001 through 2003. Similarly, Hynix reported a strong return on equity during 2004, as opposed to significant negative returns on equity during 2001 to 2003. Id. at 11 and Exhibit NA-3. As a result, the key financial measures that creditors turn to in their evaluations of credit risk were quite positive in 2004. Id. at 6-7 and Exhibit NA-1. See also Hynix's January 27, 2006 rebuttal factual information submission at Exhibits 28-30.

In addition, industry analysts held favorable views of Hynix throughout the POR. For example, Merrill Lynch reported in October 2004 that “{w}e do not see any financial distress from Hynix.” See Hynix's January 27, 2006 Rebuttal Factual Information at Exhibit 22. Additional evidence of Hynix's financial health in 2004 are in Hynix's January 27, 2006 Rebuttal Factual Information at Exhibits 3, 10, 19, 21, 26, 27, 33, and 35.

Thus, Hynix was no longer at risk of failure during the POR, as it was in prior years, which eliminated the principal motivation and basis for the GOK's past policy regarding Hynix.

Nevertheless, Micron has submitted various information as evidence that the GOK continued to entrust or direct Hynix's creditors to provide support for Hynix during the POR. For example, Micron cites to a July 2004 report from the Korea Development Bank (“KDB”) to the Korean National Assembly's Committee on Finance and Economy as evidence that the GOK's policy to support Hynix continued in 2004. See Micron's January 12, 2006, New Subsidies Allegation (“Initial Allegation”) at Exhibit 31. This report describes various activities of the KDB, which include “{w}ork toward 2004 key objectives of supporting government goals, such as balanced national development and building a Northeast Asian economic hub...,” as well as, “{c}ontinue to push for corporate restructuring,” and, “{a}s of June 2004, pushing for restructuring of 36 corporations through court receivership, joint management by creditor groups, etc.” Id. at 11 and 16. The report identifies Hynix among the “affected companies” and “sale of business divisions” as the “restructuring method.” Id. at 11 and 16. Although this document shows that the KDB supported the sale of Hynix's business divisions as part of the company's restructuring, we do not find that this document demonstrates that the GOK continued a policy to prevent Hynix's failure in 2004, or took actions to entrust or direct Hynix's other creditors to forgive debt in 2004.

Micron also points to a September 15, 2004 newspaper article entitled, “Revival of Government-Directed Banking,” to show that the GOK continued to interfere in the lending decisions of Korean banks, and in the lending decisions of Hynix's creditors in particular. See Initial Allegations at Exhibit 64. According to this article,

Government-directed banking has now been transformed from explicit to something implicit. Despite the very questionable legitimacy of government control, this transition is taking place under the banner touting 'soundness and transparency'...Interfering with the management of financial institutions through the willful enforcement of vague regulations and accounting standards is the newest form of government-directed banking, and it must be abolished...Jeong-tae Kim has...strongly objected to the recovery measures offered by the government on behalf of Hynix Semiconductor in 2001, SK Global in 2003, and LG Card earlier this year. Id.

While this article may serve as evidence of the GOK's well-documented actions to entrust or direct Korean banks to assist Korean companies in financial crisis, including Hynix in 2001, we do not consider this evidence of GOK entrustment or direction of Hynix's creditors in 2004. Moreover, we note that this article specifically identifies the GOK's involvement in Hynix's 2001 financial restructuring, but makes no mention of GOK entrustment or direction of Hynix's creditors in 2004.

Similarly, an April 5, 2005 Korea Times article, entitled “Too-Big-To-Fail Myth Dies Hard,” reaffirms the Department's past findings regarding GOK entrustment or direction of Hynix's creditors, yet makes no mention of the GOK's policies or actions in 2004, with regard to Hynix:

The government led the bailout of LG Card and Hynix Semiconductor to prevent them from triggering systemic risks over the past several years...Hynix is another sign of the government's intervention policy...The government's moves to direct banks to provide massive loans to Hynix from late 2000 to early 2002 are frankly not seen as credible by non-interested parties outside Korea. Initial Allegations at Exhibit 66.

Again, although we find that this article supports the Department's prior findings with respect to GOK entrustment or direction in 2001-2003, it fails to establish that the GOK entrusted or directed Hynix's creditors in 2004.

Other record evidence in this review relates to periods well before the POR and, therefore, does not pertain to the question of whether the GOK entrusted or directed Hynix's creditors to forgive debt in 2004. For example, Micron points to the January 8, 2003, “Meeting Agenda for the Ministers in the Economic Sector, Direction of Steering the Economy for Year 2003.” This document indicates the GOK's plans to

...complete processing of pending cases of insolvent corporations at expeditious stage. To implement restructuring of insolvent corporations that have become the main issue of our economy with creditor group at the forefront. As for Hynix, business restructuring such as debt restructuring and sales shall be implemented more aggressively following the restructuring method that is confirmed through discussion of the creditor group. Initial Allegation at Exhibit 43.

Micron also cites to a January 9, 2003 newspaper article, which states, “{t}he Government will try to conclude dealing with insolvent companies including Hanbo Steel and Hynix Semiconductor as soon as possible, and improve the system to help create an environment for on-going corporate restructuring.” See Initial Allegation at Exhibit 48. Although these documents clearly relate to the GOK's activities in 2003, there is no indication that they relate to the GOK's actions or policies towards Hynix in 2004. Additional examples of record evidence that do not relate to the GOK's actions or policies in 2004 are exhibits 47, 49, 50, and 51 of Micron's Initial Allegation.

In the first administrative 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 AR1 Decision Memorandum at 10 and Section B and C of Comment 1. However, the record evidence in this review suggests 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.

In September 2003, Lone Star, a Texas-based private equity firm, purchased a 51 percent ownership stake in KEB, and thus became the largest single shareholder in the bank. The GOK maintained a 20 percent ownership stake in KEB in 2003 and 2004. See Initial Allegation at Exhibit 56 and the August 7, 2006 Preliminary Calculations Memorandum at Attachment 3. Throughout 2003 and 2004, KEB's other foreign-owned shareholder, Commerzbank, maintained its ownership stake of just under 15 percent. Combined with Lone Star's ownership, KEB's total foreign ownership was approximately 65 percent in 2004. Id. By comparison, in 2002, the GOK was KEB's single largest shareholder (36 percent) and Commerzbank was the only foreign shareholder. The Department found, “that through its ownership of KEB, the GOK was indeed able to, and did, influence KEB's credit decisions with respect to Hynix's financial restructurings in 2002.” See AR1 Decision Memorandum at 34-35.

In prior segments of this proceeding, we found that the GOK was able to influence the lending decisions of Korea First Bank (“KFB”), despite the fact that a U.S. firm, Newbridge Capital, owned 51 percent of KFB. We based this finding, in part, on the GOK's 49 percent ownership stake in KFB. However, record evidence also demonstrated that the GOK threatened KFB to ensure that it participated in Hynix's 2001 financial restructuring. 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 AR1 Decision Memorandum at 34. The record evidence in this review, however, does not indicate that the GOK threatened, or otherwise entrusted or directed KEB to forgive Hynix's debt in 2004.

Micron cites to a newspaper article which states that “{Lone Star} has expressed its intention to separate the state-funded bank's {(i.e. KEB's)} ownership from management.” See Initial Allegation at Exhibit 56. However, that same article quoted a market analyst's opinion that “the professional management may not easily pursue its own strategy and exclude the bank's largest shareholder,” despite Lone Star's reported desire to separate ownership from management. Id. According to this article, “KEB appointed seven new outside directors, including five recommended by Lone Star following the acquisition,” and that Lone Star was waiting to “announce its official position on management strategy after paying out its takeover money.” Id.

As we stated in the AR1 Decision Memorandum, we considered creditors in which the GOK was the majority or single largest shareholder as GOK-owned or -controlled. See AR1 Decision Memorandum at Comment 1-C. Thus, given Lone Star's majority ownership of KEB and significant presence on KEB's board of directors, coupled with Commerzbank's continuing minority stake in KEB, we find that in 2004 the 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 the Department's August 7, 2006 Preliminary Calculations Memorandum 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 save Hynix in 2004 (and, as we indicated above, the record evidence does not show that they 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 preliminarily find that debt reductions or debt forgiveness 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 AR1 Decision Memorandum 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 preliminarily 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 preliminarily 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 the System IC Unit to MagnaChip

Record information indicates that in July 2004, Hynix's Creditors' Council agreed to provide acquisition financing for MagnaChip's purchase of the System IC unit from Hynix. Concurrently, the Creditors' Council agreed to the terms for the October CBO. See Hynix's March 30, 2006 submission at Exhibit 9. Tranche A of the System IC 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 System IC. 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. According to Micron, this debt reduction constitutes a direct transfer of funds in the form of debt forgiveness, within the meaning of section 771(5)(D)(ii) 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-owned foreign 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. See Hynix's March 30, 2006 supplemental questionnaire response at 6. Consequently, we focused our analysis on the October and December CBOs, in which the Korean government authorities did participate.

The October and December CBOs

According to Hynix, the expected cash proceeds from the System IC sale and income from its normal business operations enabled Hynix to repay numerous outstanding loans in 2004, prior to their maturity.¹ These repayments were made under the October CBO, which occurred concurrently with the System IC sale and Tranche A acquisition financing. Hynix also repaid debt early and at a discount under the December CBO, which occurred after the System IC sale. See Hynix's March 30, 2006 submission at 5-8 and Exhibit 9. See also Hynix's June 30, 2006 submission at Exhibit NA-9.

The terms of the October CBO included a maximum cash buyout rate of 70% for unsecured loans and a fixed cash buyout rate of 96% 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 debt that would be repaid under the CBO, and a hierarchy of loans that were eligible for the CBO. See Hynix's March 30, 2006 submission at 5-8 and Exhibit 9. See also Hynix's June 30, 2006 submission at Exhibit NA-9.

In addition, 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's March 30, 2006 submission at 5-8 and Exhibit 9. See also Hynix's June 30, 2006 submission 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's June 30, 2006 submission at Exhibits NA-11 and NA-12.

We preliminarily 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. We further preliminarily determine that the discounts taken by the participating creditors do not constitute debt 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. Thus, the issue we need to address is whether the terms of repayment of these loans conferred a benefit on Hynix.

According to section 771(5)(E)(ii) of the Act, 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.

For the reasons explained below, we preliminarily determine that there was significant participation by commercial creditors in the CBOs, that the Korean government authorities participated on the same terms as the commercial creditors and, consequently, that Hynix received no benefit from early repayment of its debt at a discount.

In the investigation and first administrative review, we found that wholly-owned foreign 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-owned foreign 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 AR1 Decision Memorandum at Comments 5 and 6. In the instant review, wholly-owned foreign 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-owned foreign creditors accounted for over 40 percent of the debt discounted under the two CBOs. See the August 7, 2006 Preliminary Calculations Memorandum at Attachment 3. Therefore, we find that the wholly-owned foreign creditors held a significant portion of the debt discounted in the October and December CBOs.

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 AR1 Decision Memorandum 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.

We further determine that the government authorities and the wholly-owned foreign 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.

Therefore, we preliminarily 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-owned foreign banks on identical terms means the 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 Decision Memorandum 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 AR1 Decision Memorandum at 10-11.

Nothing on the record of this review would lead us to reconsider these prior specificity findings.

III. Programs Previously Found Not to Have Been Used or Provided No Benefits

We preliminarily determine 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. 1.Tax Programs Under the TERCL and/or the RSTA

2. Tax Credit for Investment in Facilities for Productivity Enhancement (Article 25 of RSTA/Article 25 of TERCL)

3. Tax Credit for Investment in Facilities for Special Purposes (Article 25 of RSTA)

4. Reserve for Overseas Market Development (formerly, Article 17 of TERCL)

5. Reserve for Export Loss (formerly, Article 16 of TERCL)

6. Tax Exemption for Foreign Technicians (Article 18 of RSTA)

7. Reduction of Tax Regarding the Movement of a Factory That Has Been Operated for More Than Five Years (Article 71 of RSTA)

C. Tax Reductions or Exemption on Foreign Investments under Article 9 of the Foreign Investment Promotion Act (“FIPA”)/ FIPA (Formerly Foreign Capital Inducement Law)

D. Duty Drawback on Non-Physically Incorporated Items and Excessive Loss Rates

E. Export Insurance

F. Electricity Discounts Under the RLA Program

G. System IC 2010 Project

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 AR1 Decision Memorandum at 15. No new information has been provided with respect to this program. Therefore, we preliminarily find that Hynix did not receive any countervailing benefits from the System IC 2010 Project during the POR.

Preliminary Results of Review

In accordance with 19 CFR 351.221(b)(4)(i), we calculated an individual subsidy rate for Hynix Semiconductor, Inc., the producer/exporter covered by this administrative review. We preliminarily determine that the total estimated net countervailable subsidy rate for Hynix for calendar year 2004 is 31.86 percent ad valorem.

If the final results of this review remain the same as these preliminary results, the Department intends to instruct CBP, within 15 days of publication of the final results of this review, to liquidate shipments of DRAMS by Hynix entered or withdrawn from warehouse, for consumption from January 1, 2004, through December 31, 2004, at 31.86 percent ad valorem of the F.O.B. invoice price.

The Department also intends to instruct CBP to collect cash deposits of estimated countervailing duties at 31.86 percent ad valorem of the F.O.B. invoice price on all shipments of the subject merchandise from Hynix, entered, or withdrawn from warehouse, for consumption on or after the date of publication of the final results of this administrative review.

We will instruct CBP to continue to collect cash deposits for non-reviewed companies covered by this order at the most recent company-specific rate applicable to the company. Accordingly, the cash deposit rate that will be applied to non-reviewed companies covered by this order will be the rate for that company established in the investigation. See Notice of Amended Final Affirmative Countervailing Duty Determination: Dynamic Random Access Memory Semiconductors from the Republic of Korea, 68 FR 44290 (July 28, 2003). The “all others” rate shall apply to all non-reviewed companies until a review of a company assigned this rate is requested. The Department has previously excluded Samsung Electronics Co., Ltd. from this order. Id.

Public Comment

Interested parties may submit written arguments in case briefs within 30 days of the date of publication of this Notice. Rebuttal briefs, limited to issues raised in case briefs, may be filed not later than five days after the date of filing the case briefs. Parties who submit briefs in this proceeding should provide a summary of the arguments not to exceed five pages and a table of statutes, regulations, and cases cited. Copies of case briefs and rebuttal briefs must be served on interested parties in accordance with 19 CFR 351.303(f).

Interested parties may request a hearing within 30 days after the date of publication of this notice. Unless otherwise specified, the hearing, if requested, will be held two days after the scheduled date for submission of rebuttal briefs.

The Department will publish a notice of the final results of this administrative review within 120 days from the publication of these preliminary results.

We are issuing and publishing these results in accordance with sections 751(a)(1) and 777(i)(1) of the Act.

Dated: August 7, 2006.

Joseph A. Spetrini,

Acting Assistant Secretary for Import Administration.

Footnotes

1.  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.