Hynix Semiconductor Inc. v. United States

425 F. Supp. 2d 1287, 30 Ct. Int'l Trade 288, 30 C.I.T. 288, 28 I.T.R.D. (BNA) 1361, 2006 Ct. Intl. Trade LEXIS 40

Opinion

OPINION

GOLDBERG, Senior Judge:

[Commerce’s remand determination sustained. Previously deferred portions of final affirmative countervailing duty determination sustained.]

In Hynix Semiconductor Inc. v. United States, 29 CIT -, 391 F.Supp.2d 1337 (2005) (“Hynix I”), familiarity with which is presumed, the Court sustained in part, remanded in part, and deferred reviewing in part the final affirmative countervailing duty determination made by the United States Department of Commerce (“Commerce”) regarding dynamic random access memory semiconductors (“DRAMS”) from the Republic of Korea (“Korea”). See Dynamic Random Access Memory Semiconductors from the Republic of Korea, 68 Fed. Reg. 37122 (Dep’t Commerce June 23, 2003) (final determination), amended by 68 Fed. Reg. 44290 (Dep’t Commerce July 28, 2003) (amended final determination) (together, the “Final Determination”). Duly complying with the Court’s remand order in Hynix I, Commerce issued draft redetermination results on November 3, 2005 and then, after receiving comments from Plaintiffs Hynix Semiconductor Inc. and Hynix Semiconductor America Inc. (together, “Hynix”) and Defendant-Intervenor Micron Technology, Inc. (“Micron”), Commerce issued final re-determination results. See Final Results of Redetermination Pursuant to Remand, Inv. No. C-580-851 (Nov. 23, 2005), available at http://ia.ita.doc.gov/remands/05106.pdf (the “Remand Results”).

This case is now properly before the Court following remand and the Court has jurisdiction pursuant to 28 U.S.C. § 1581(c). For the reasons that follow, the Court sustains the Remand Results and, proceeding to an analysis of the issues previously deferred by the Court, also sustains the remainder of the Final Determination.

I. BACKGROUND

A. The Court’s Decision in Hynix I

In Hynix I, the Court recognized the novelty of Commerce’s invocation of au thority under 19 U.S.C. § 1677(5)(B)(iii) 1 for purposes of the Final Determination. Hynix I, 29 CIT at-, 391 F.Supp.2d at 1343. This section of the countervailing duty statute permits Commerce to countervail certain benefit-conferring financial contributions made by private parties pursuant to government entrustment or direction. 2 Invoking this section in the Final Determination, Commerce determined that Hynix had received substantial indirect subsidies from the Korean government through a clandestine program of coercing Hynix’s creditors to give preferential loans and debt-to-equity swaps during Hynix’s ten-month restructuring. Id. at -, 391 F.Supp.2d at 1340 (citing Issues and Decision Memorandum for the Final Determination in the Countervailing Duty Investigation of Dynamic Random Access Memory Semiconductors from the Republic of Korea, Inv. No. C-580-851, (Dep’t Commerce June 16, 2003), available at http://ia.ita.doc.gov/frn/surnmary/korea-southA13-15793-l.pdf (“Decision Memo”) at 20-21).

The Court focused its initial review of the Final Determination on Commerce’s interpretation and application of the first part of the three-prong statutory test required to prove the existence of these so-called “entrusted or directed” subsidies: “the making of a financial contribution by a private entity to another private entity pursuant to government entrustment or direction.” Id. at-, 391 F.Supp.2d at 1343 (citing 19 U.S.C. § 1677(5)(B)(iii)). The Court held that Commerce’s decision to interpret the “entrusts or directs” language of this prong to include “a single program of financial contributions involving multiple financial institutions directed by a foreign government” was in accordance with law. Id. Further, the Court upheld Commerce’s methodology for proving such a program of financial contributions, recognizing that the substantial evidence standard “does not require Commerce to produce conclusive evidence of entrustment or direction of each entity involved in each transaction making up an alleged program” under 19 U.S.C. § 1677(5)(B)(iii), so long as “the cumulated evidence and the reasonable inferences drawn therefrom sufficiently connect all the implicated parties and transactions to the alleged program of government entrustment or direction.” Id.

Nonetheless, the Court remanded the Final Determination. Although Commerce provided an extensive explanation of the record evidence which, in the agency’s view, demonstrated that the Korean government had both a “governmental policy to support Hynix” and “a pattern of practices ... to act upon that policy to entrust or direct” Hynix’s creditors, Decision Memo at 49 (emphasis added), the Court found that Commerce had neglected to adequately consider “counterevidence indicating that the transactions making up [the alleged program in this case] were formulated by an independent commercial actor (not a government) and motivated by commercial considerations.” Hynix I at -, 391 F.Supp.2d at 1343. In the Court’s view, the unusual role played by Citibank and its affiliate Solomon Smith Barney (“SSB”) in Hynix’s restructuring, as well as the apparent presence of commercial options and contingencies in the restructuring, required additional explanation before the Court could proceed with its substantial evidence review of Commerce’s financial contribution analysis. Id. at -, 391 F.Supp.2d at 1344.

Because the Court remanded to Commerce for further consideration of its threshold financial contribution analysis, the Court deferred review of Commerce’s interpretation and application of the other two prongs of the statutory test required to prove the existence of “entrusted or directed” subsidies: the exercise of a government subsidy junction 3 in the provision of the investigated financial contribution and the existence of a benefit from that financial contribution to its recipient. Id.

B. Commerce’s Remand Results

In the Remand Results, Commerce affirmed its original determination that the Korean government entrusted or directed Hynix’s creditors to provide financial contributions within the meaning of 19 U.S.C. § 1677(5)(B)(iii). Remand Results at 1.

Considering first whether Hynix’s restructuring was in fact orchestrated by a commercial actor rather than the Korean government, Commerce found that Citibank/SSB’s role was “quite limited!,]” id. at 6, and more akin to that of a “consultant” than orchestrator. Id. at 7. While acknowledging that “Citibank/SSB certainly did much of the technical work behind the mechanics of Hynix’s financial restructuring[,]” Commerce concluded that “it was the actions taken by the [Korean government] ... that effectuated the restructuring and brought about the financial contributions.” Id. at 9. In Commerce’s view, Citibank/SSB provided necessary expertise in arranging the complicated financial transactions which comprised Hynix’s restructuring, but was able to do so only because the Korean government used its authority to coerce the participation of Korean financial institutions in those highly risky transactions. Id. at 7-8. At most, Commerce found that Citibank/SSB’s involvement could be seen as “working to assist the creditors make the best out of a bad situation” and not as orchestrating commercially-motivated lending and investment opportunities for Hynix’s creditors. Id. at 11.

Next considering whether Hynix’s restructuring featured commercially-based contingencies and options which belied an inference of government control, Commerce found that no such contingencies or options existed in Hynix’s restructuring. Id. With regard to the international offering of Hynix’s equity (the “GDS offering”) made in conjunction with Hynix’s May 2001 restructuring, Commerce concluded that the May 2001 restructuring was not “truly contingent upon the GDS offering!.]” Id. (quotation marks omitted). Commerce noted that, before completion of the GDS offering, Hynix’s creditors approved the new loans and debt restructur ing included in that transaction, id., and they also entered into a related underwriting agreement. Id. at 13. Because Hy-nix’s creditors agreed to important details of the May 2001 restructuring even before the GDS offering closed, Commerce found it “unlikely that the [creditors] were truly waiting until the successful conclusion of the GDS to decide whether to proceed with the May restructuring.” Id. To further support this view, Commerce noted that the May 2001 restructuring was used as an important selling point in the GDS Offering Memorandum. Id. Commerce observed that this memorandum characterized the May 2001 loans and debt restructuring as closing “substantially concurrently” with the closing of the GDS offering period, “highlighting the automat-icity of the assistance agreed to in May” by Hynix’s creditors. Id. at 12. Commerce also noted that the GDS Offering Memorandum underscored the Korean government’s support for Hynix. Id. Commerce concluded its analysis of the GDS offering by characterizing it as simply an attempt to share at least some of the financial burden of saving Hynix which had been imposed on Hynix’s creditors by the Korean government. Id. at 14.

With regard to whether the options provided to creditors participating in Hynix’s October 2001 debt restructuring belied an inference of government control, Commerce concluded that the “true nature of the options was to benefit Hynix at the creditors’ expense.” Id. at 15 (quotation marks omitted). Commerce noted that, with regard to this transaction, Hynix’s creditors were required to select from among three options developed by Hynix’s creditors council. Id. These options were: (1) extend new loans to Hynix and convert/renegotiate existing secured and unsecured debt in a manner more advantageous to Hynix; (2) not extend new loans, but convert all secured debt and 28 percent of unsecured debt in a manner more advantageous to Hynix, and forgive the remaining unsecured debt; or (3) exercise appraisal rights for all secured debt and 25 percent of unsecured debt based on Hy-nix’s liquidation value, and forgive the remaining unsecured debt. Id. Commerce observed that the third option did not provide for an immediate refund of liquidated loans, but instead called for these liquidated funds to be converted into five-year, interest-free loans to Hynix. Id. In Commerce’s view, the result to Hynix under any of the options was either complete debt extinguishment or partial debt extin-guishment coupled with sufficient new loans to service the remaining debt load— all at the expense of the creditors’ balance sheets. Id. at 16. Commerce further observed that Hynix’s creditors were unhappy with these options, as reported in several contemporaneous news accounts. Id. Commerce concluded its analysis of the options featured in the October 2001 restructuring by characterizing them as an attempt to provide Hynix’s creditors with some limited flexibility in the manner in which they participated in the government-mandated bailout of the struggling company. Id. at 17. In Commerce’s view, this flexibility was simply intended to better accommodate the varying levels of investment and financial health of Hynix’s beleaguered creditors. Id.

Having thus found that Hynix’s restructuring “was not the product of market forces,” Commerce concluded the Remand Results by reaffirming its determination that, based on the record evidence, 4 Hynix had been the recipient of government-entrusted or directed financial contributions. Id.

C. The Deferred Portions of the Final Determination

Commerce appropriately limited the Remand Results to the questions concerning its financial contribution analysis raised by the Court in Hynix I, relying on its original analysis in the portions of the Final Determination deferred by the Court.

In the Final Determination, once Commerce found that Hynix had received financial contributions entrusted or directed by the Korean government. Commerce proceeded to the next step in the statutory test to prove their countervailability. Considering the portion of 19 U.S.C. § 1677(5)(B)(iii) that specifies that a financial contribution is only eountervailable “if providing the contribution would normally be vested in the government and the practice does not differ in substance from practices normally followed by governments,” Commerce interpreted this requirement to mean that a “governmental subsidy function” must be performed for an investigated financial contribution to be countervail-able. Decision Memo at 47. Applying this statutory interpretation and in light of the evidence before it, Commerce concluded that this requirement had been met. Id. at 61.

Commerce then proceeded to the third and final prong of the statutory test for countervailability under 19 U.S.C. § 1677(5) (B) (iii). To countervail an entrusted or directed financial contribution given pursuant to a government subsidy function, Commerce determined that it was statutorily required to establish that the financial contribution conferred a benefit on its recipient. Id. at 21. To identify the benefit, if any, received by Hynix during its restructuring, Commerce attempted to compare the investigated financial contributions to commercial benchmarks (i.e., similar loans or equity infusions made by independent actors to Hynix under market conditions). Id. at 6-7.

First analyzing the financial contributions received in the form of credit (i.e., preferential loans), Commerce was unable to find any appropriate commercial benchmarks for use in establishing Hynix’s creditworthiness. 5 Id. at 19-25. To reach this conclusion, Commerce eliminated from consideration Citibank’s loans to Hy-nix. Id. at 11. Although concluding that Citibank was independent of government control, id. at 8, Commerce disqualified Citibank’s loans because (1) Citibank’s involvement was relatively small compared to the overall restructuring; (2) Citibank took into consideration the behavior of the government-entrusted or directed financial institutions in order to hedge its lending risk; and (3) Citibank/SSB stood to earn greater fees as Hynix’s financial advisor than what other financial institutions could expect from their return on investment in Hynix, thus skewing Citibank’s risk calculus. Id. at 9-11. Commerce also disre garded the loans made by Hynix’s other creditors, based on their entrustment or direction by the Korean government. Id. at 11. Lacking an actual commercial benchmark, Commerce attempted to determine if Hynix was otherwise creditworthy during its restructuring. Id. Commerce determined that Hynix was not and constructed a benchmark to calculate the benefit conferred to Hynix by the credit-based financial contributions. Id. at 11, 105. Commerce developed this constructed benchmark using Moody’s U.S. average cumulative default rates for corporate bonds, instead of default rates specific to Korea which were supplied to Commerce by Hynix during the course of the investigation. Id. at 5.

Next analyzing the financial contributions received in the form of equity (i.e., investments), Commerce was similarly unable to identify any commercial benchmarks for use in establishing Hynix’s equity worthiness. 6 Id. at 91. To reach this conclusion, Commerce again eliminated from consideration Citibank’s involvement because, when compared to the size of the investment made by the government-entrusted and directed financial institutions during Hynix’s restructuring, Commerce found that Citibank’s equity investment in Hynix was not “significant” as required by the countervailing duty regulations. 7 Id. at 90 (citing 19 C.F.R. § 351.507(a)(2)(iii)). Commerce also disregarded the equity infusions made by Hynix’s other creditors, based on their entrustment or direction by the Korean government. Id. at 91.

Lacking an actual commercial benchmark, Commerce áttempted to determine if Hynix was otherwise equity-worthy during its restructuring. Id. at 91. As part of that analysis. Commerce considered third party studies of Hynix commissioned by its creditors which discussed Hynix’s investment potential at the time of its restructuring. Id. Commerce ultimately disregarded these studies, finding that then-focus on creditor concerns meant that they did not properly discuss Hynix’s future financial prospects or other factors denoting equity-worthiness. Id. Commerce also questioned the credibility of the methodology and analysis used in some of these reports. Id. Further, Commerce found that Hynix’s financial indicators for the years 1997 through 2001 were too weak to support a commercially reasonable investment decision at that time. Id. at 92. To reach this conclusion, Commerce applied an economic theory known as the Expected Utility Model, which posits that a rational investor focuses on future profitability and does not let the value of past investments in a company affect future investment decisions in that same company. Id. Ultimately finding that Hy-nix was unequity-worthy during its restructuring, Commerce calculated the benefit conferred to Hynix by the equity-based financial contributions. Id.

Based on the foregoing findings and analysis, Commerce determined that the three-prong statutory test had been met and made a final affirmative countervailing duty determination. Final Determination, 68 Fed. Reg. 37122, 37122.

II. STANDARD OF REVIEW

The Court must sustain any determination, finding, or conclusion made by Commerce in the Final Determination and the Remand Results unless it is “unsupported by substantial evidence on the record, or otherwise not in accordance with law.” 19 U.S.C. § 1516a(b)(l)(B)(i) (1999). The Court must also defer to an agency’s reasonable construction of an ambiguous statute. Allegheny Ludlum Corp. v. United States, 367 F.3d 1339, 1343 (Fed. Cir.2004) (citing Chevron, U.S.A., Inc. v. NRDC, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)). Further, “[t]he deference granted to the agency’s interpretation of the statutes it administers extends to the methodology it applies to fulfill its statutory mandate.” GMN Georg Muller Numberg AG v. United States, 15 CIT 174, 178, 763 F.Supp. 607, 611 (1991) (citing, inter alia, Chevron, 467 U.S. at 844-45, 104 S.Ct. 2778; Amer. Lamb Co. v. United States, 785 F.2d 994, 1001 (Fed.Cir.1986)). “Likewise, the [C]ourt may defer to an agency’s interpretation of an ambiguous regulation, so long as that interpretation is not plainly erroneous or inconsistent with the regulation, does not fail to reflect the ‘agency’s fair and considered judgment on the matter in question,’ or, if adopted, does not render the regulation unreasonable or otherwise not in accordance with law.” Decca Hospitality Furnishings, LLC v. United States, 29 CIT-, 391 F.Supp.2d 1298, 1304 (2005) (quoting Auer v. Robbins, 519 U.S. 452, 462, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997) (citations omitted)).

III. DISCUSSION

A. Summary of Analysis

This case is before the Court for review of Commerce’s determination that Hynix received a countervailable benefit from the Korean government through a program of indirect subsidies of the type described in 19 U.S.C. § 1677(5)(B)(iii). For the reasons that follow, the Court concludes that Commerce has satisfied the requirements of the applicable three-prong statutory test in reaching this determination.

First, Commerce adduced substantial evidence in support of its finding that the Korean government entrusted or directed certain financial institutions to provide preferential loans and equity infusions to Hynix during its restructuring. Although Commerce did not provide conclusive evidence for each party or each transaction involved in the program, the agency’s circumstantial and direct evidence (and the reasonable inferences drawn therefrom) adequately connected the various financial institutions involved in Hynix’s multi-phase restructuring to the Korean government’s anticompetitive involvement. Counterevi-dence offered by Hynix does not undermine the agency’s substantiated factual finding.

Second, Commerce’s interpretation of the second prong of the statutory test, concerning the performance of a government subsidy function in connection with the entrusted or directed financial contributions, is in accordance with law. Commerce’s interpretation appropriately narrows the reach of the countervailing duty statute to only those government actions which involve the delegation of a subsidy function to a private entity. Applying this interpretation, Commerce adduced substantial evidence demonstrating that the Korean government delegated its subsidy function to Hynix’s creditors.

Finally, Commerce met the third prong of the statutory test by demonstrating that Hynix received a benefit from the credit and equity-based financial contributions provided by its creditors at the behest of the Korean government. In making this assessment, Commerce appropriately considered the suitability of commercial benchmarks provided by Citibank’s loans and equity infusions in Hynix and reasonably concluded that Hynix was neither creditworthy nor equity-worthy at the time of its restructuring. Commerce also acted within its authority when establishing an unereditworthy benchmark for Hynix.

As a result, the Court sustains both the Remand Results and the remainder of the Final Determination. The Court’s conclusions are discussed more fully below.

B. Commerce’s Financial Contribution Analysis Is Supported by Substantial Evidence

Hynix argues that the record evidence in this case does not support Commerce’s conclusion that Hynix received entrusted or directed financial contributions during its restructuring. First, Hynix claims that the various pieces of evidence in support of Commerce’s conclusion were seriously flawed and insufficient to establish a program of entrustment or direction under the substantial evidence standard. Plaintiffs’ Memorandum in Support of Its Rule 56.2 Motion for Judgment on the Agency Record (“Pis.’ Br.”) at 17-25, 29-33. Second, Hynix contends that Commerce’s proffered evidence was in fact rebutted by counterevidence firmly establishing that an independent third party (not the Korean government) orchestrated Hynix’s restructuring and included commercial options and contingencies in that restructuring. Id. at 11-16, 25-29.

For the reasons that follow, the Court upholds Commerce’s conclusion that Hynix received government-entrusted or directed financial contributions as supported by substantial evidence.

1. Record Evidence Supports Commerce’s Conclusion That Hynix Received Entrusted or Directed Financial Contributions

Notwithstanding Hynix’s specific eviden-tiary arguments (discussed below), the Court finds that the record supports Commerce’s conclusion that Hynix received financial contributions from private entities entrusted or directed by the Korean government.

To support its factual finding of government entrustment or direction, Commerce adduced circumstantial and direct evidence of the Korean government’s motive, proclivity, opportunity, and capacity to support Hynix through private entities. For example, Commerce cited persuasive evidence indicating that the Korean government had a policy of supporting Hynix and, therefore, a motive to entrust or direct private entities to participate in Hy-nix’s restructuring. Commerce noted that, in a 2001 statement, a member of the Korean president’s staff stated that Hynix was part of a strategically important domestic industry which “should not be sold off just to follow market principles.” Decision Memo at 49; see also Appendix to Defendant’s Memorandum in Opposition to Plaintiffs’ Motion for Judgment on the Administrative Record (“Def.’s App.”), App. 5 (Ex. C-20 of Petitioner’s Comments to Commerce dated Mar. 14, 2003) at 17. Commerce further noted that, in a 2002 exchange between the Korean president and a member of Korea’s National Assembly, the assembly member criticized Korea’s president for compelling financial institutions to provide Hynix “astronomical sums of special support ... by mobilizing the resources of financial and government-run institutions.” Id. at 50; see also Def.’s App., App. 5 (Ex. C-20 of Petitioner’s Comments to Commerce dated Mar. 14, 2003) at 17. The official presidential response to this statement was: “[w]e are doing what is deemed necessary to save companies leading the countries [sic] strategic industries.” Decision Memo at 50. Even if this exchange was political banter as asserted by Hynix, see Pis.’ Br. at 17-18, Commerce reasonably found it telling that the presidential response did not deny the allegation of an official policy of supporting Hynix. Cf. United States v. Hale, 422 U.S. 171, 176, 95 S.Ct. 2133, 45 L.Ed.2d 99 (1975) (in criminal context, “[s]ilence gains more probative weight where it persists in the face of accusation, since it is assumed in such circumstances that the accused would be more likely than not to dispute an untrue accusation”). Here, “it would have been natural under the circumstances” for the Korean executive branch to object to an unfounded public accusation of large-scale government waste. Hale, 422 U.S. at 176, 95 S.Ct. 2133. This exchange, particularly when read together with the 2001 presidential statement, 8 gave rise to a reasonable inference by Commerce that the Korean government maintained a policy to financially support Hynix. 9

Commerce’s evidence also demonstrated a strong proclivity on the part of the Korean government to support Hynix through private entities. During the early stages of Hynix’s restructuring, record evidence showed that the Korean government’s Economic Ministers met to discuss possible measures to alleviate Hynix’s liquidity problems. Decision Memo at 50. The execution of the Ministers’ decisions was delegated to government agencies, including the Korea Export Insurance Corporation (“KEIC”), which was advised by the Economic Ministers that the decisions should be “carried out perfectly.” Id. at 51; see also Def.’s App., App. 5 (Ex. C-20 of Petitioner’s Comments to Commerce dated Mar. 14, 2003) at 17. The agencies then took two measures: (1) they waived certain regulatory requirements to enable Hynix’s creditors to increase the credit extended to Hynix and (2) they resumed providing insurance for certain financing transactions undertaken by Hynix and its creditors. Decision Memo at 51-52. Commerce reasonably found that these measures enabled Hynix’s creditors to participate in the company’s restructuring. Id. at 50-51. In other words, early in Hy-nix’s restructuring, the Korean government demonstrated an inclination for using private entities to achieve its policy of supporting Hynix. This early demonstration was followed by the creation of a gov ernment-ran bond placement program used by Hynix’s creditors to extend/refinance credit at a time in which the maturation of existing bonds threatened Hy-nix’s default. Id. at 52. Although this so-called KDB Fast Track program was non-eompulsory and open to firms other than Hynix, see Pis.’ Br. at 21, it was used predominantly by Hynix’s creditors. Decision Memo at 52. This “more than coincidental” participation reasonably led Commerce to characterize the program as a further demonstration of the Korean government’s encouragement of private entity involvement in Hynix’s restructuring. 10 Id.

Further, evidence concerning Hynix’s creditors council demonstrated that the Korean government had ample opportunity to entrust or direct private entities during the later phases of Hynix’s restructuring. 11 Hynix’s creditors council was comprised of the same financial institutions which participated in each phase of Hynix’s restructuring. Decision Memo at 53-54. A majority of Hynix’s outstanding debt was held by financial institutions with varying degrees of government ownership. Id. These debt levels translated to voting interests on the creditors council in an amount sufficient to influence the plans approved by the council and to veto any undesirable proposals. Id. at 54-55. Commerce found that, by virtue of its ownership interests in voting members of the creditors council, the Korean government could have had a unique vantage point from which to orchestrate Hynix’s restructuring using the private entities on Hynix’s creditors council. Id. This inference is reasonable. If the Korean government’s ownership interests in certain Hynix creditors gave the government the ability to influence or direct decisions taken by those financial institutions, 12 then the dominant presence of financial institutions with government ownership could have given the Korean government the opportunity to have a pervasive influence on the decision-making of Hynix’s creditors on the creditors council.

Commerce also adduced evidence indicating that the Korean government recognized the opportunity to exert influence or control over private entities which was presented by the creditors council. Commerce learned during verification that a government official attended a March 2001 creditors council meeting “to urge creditor banks to execute the resolutions made by creditors.” Decision Memo at 59; see also Def.’s App., App. 30 (Korean Government Verification Report dated May 15, 2003) at 19. In addition, the Korean government later enacted a new law requiring all creditor financial institutions to attend creditors council meetings for any major corporate restructuring, such as Hynix’s. Id. A government official quoted in a July 2001 Korea Times article cited by Commerce explained that the purpose of the new law was “to prevent some of [the creditors] from refusing to attend [meetings] and pursuing their own interests by taking advantage of bailout programs[.]” Id. at 59. From this evidence, Commerce could reasonably find that the Korean government recognized that the creditors council was a possible forum to both communicate and effectuate its Hynix support policy through private entities.

Moreover, Commerce’s evidence demonstrated that the Korean government had the capacity to act on the opportunity for entrustment or direction of private entities which was presented by its ownership interests in financial institutions on Hynix’s creditors council. For example, Commerce cited various contemporaneous Korean newspapers and international financial publications which reported that the Korean government influenced at least three of Hynix’s creditors with substantial government ownership. Decision Memo at 56. Specifically, Commerce cited a January 2002 Business Week article which reported that the Korean government forced Woori Bank, the Korean Exchange Bank (“KEB”), and ChoHung Bank to provide significant funding to Hynix. Id. An October 2001 Korea Times article reported that a KEB official had confirmed that the Korean government was “working out a series of powerful measures to ensure the survival of [Hynix.]” Id. Commerce also cited a September 2001 Asiamoney article which discussed general suspicions that banks with substantial government shareholdings were being pressured by the Korean government to support Hynix. Id.

Additional reports cited by Commerce indicated that the Korean government had the capacity to influence even those members of Hynix’s creditors council without significant government ownership. For example, Commerce cited to a Doiv Jones International article which reported that KorAm Bank reversed its decision not to participate in a portion of Hynix’s May 2001 restructuring after the Korean government’s Financial Supervisory Service (the “FSS”) warned of possible sanctions if it did not participate. Decision Memo at 59. Commerce also cited to a Korea Herald article which reported that the FSS had threatened to fine Hana Bank if it did not provide emergency liquidity to HPC, a Hynix affiliate. Id. at 60. Taken together, Commerce reasonably viewed these news reports as circumstantial evidence suggesting that the Korean government was able to influence, or coerce private entities— with and without government ownership— to support Hynix’s restructuring.

Commerce was able to further support the inference of Korean government capacity to influence private entities with additional evidence drawn from the opinions of the independent Korean financial experts interviewed during verification. Commerce noted that “the clear consensus that emerged from the independent financial sector experts ... was that the [Korean government] can and does influence” financial institutions owned whole or in part by the government. Decision Memo at 53-54 n.20. With regard to financial institutions free of government ownership, Commerce also observed that while “many experts interviewed suggested] that the [Korean government] no longer had control over the private banks the way it had in the past[,] ... at least one expert did comment that government influence over the private banks has continued.” Id. at 57. Upon a review of the entire summary of the financial expert interviews as urged by Hynix, see Pis.’ Br. at 25, the Court finds that this evidence supports Commerce’s inference that the Korean government could have exercised a degree of influence over the financial institutions involved in Hynix’s restructuring. See Appendix to Plaintiffs’ Motion for Judgment on the Administrative Record (“Pis.’ App.”), App. 6 (Private Financial Experts Verification Report dated May 15, 2003). Although the opinions of the independent Korean financial experts were far from unanimous or conclusive on the question of the Korean government’s ability to effectuate its Hynix support policy through private financial institutions, see id. at 3, 12; Pis.’ Br. at 22, this evidence lent some additional support for Commerce’s inference that the Korean government had the capacity to entrust or direct the private financial institutions that participated in Hynix’s restructuring.

Commerce built on its evidence of the Korean government’s capacity to influence financial institutions with government ownership by specifically examining actions taken with respect to the KEB. Formerly a fully government-owned bank, the KEB was Hynix’s principal creditor. Decision Memo at 56. Because the Korean government remained the KEB’s largest shareholder with about 43% of the bank’s shares, certain of the financial experts interviewed by Commerce contended that the KEB was still subject to government influence over lending decisions. Id. at 55-56. , Indeed, official correspondence sent to the KEB from the Korean government’s Economic Ministers advised the bank to “carr[y] out perfectly” their decisions to support Hynix. Id. at 50. Commerce could reasonably find it telling that, as discussed above, these were the same •instructions sent by the Economic Ministers to a Korean government agency. Further, confidential internal loan documentation obtained by Commerce at verification also indicated that the KEB took into account non-commercial, economic and social policy considerations when it chose to participate in various stages of Hynix’s restructuring. Id. at 55-56; Pis.’ App., App. 7 (Hynix Verification Report dated May 15, 2003) at 15, 17. This evidence also indicated that the KEB shared these considerations with Hynix’s creditors council. Id. In the Court’s view, Commerce reasonably found this evidence to be a demonstration of the Korean government’s influence on the KEB’s decision to provide credit and equity to Hynix. It is indeed suspect for an allegedly independent financial institution to consider the ramifications of isolated lending and investment decisions on the economic and social health of a country, rather than that institution’s bottom line. Commerce reasonably found that this was not normal behavior for a profit-maximizing market actor. Cf. In re Flat Glass Antitrust Litigation, 385 F.3d 350, 360-61 (3d Cir.2004) (in antitrust context, noting that “[ejvidence that the defendant acted contrary to its interests means evidence of conduct that would be irrational assuming that the defendant operated in a competitive market. Put differently ... a court looks to evidence that the market behaved in a noncompetitive manner.”) (quotation marks omitted). Based on this evidence, Commerce was justified in finding that “the very commercial nature which Hynix states motivated the KEB is fundamentally called into question.” Decision Memo at 57.

Commerce also found evidence of Korean government entrustment or direction with respect to Kookmin Bank (“Kook-min”), a Korean commercial financial institution without substantial government ownership. Commerce initially cited a September 2001 certified filing made by Kookmin to the U.S. Securities and Exchange Commission (“SEC”). Id. at 57. In that filing, Kookmin warned its investors that:

The [Korean government] has promoted, and, as a matter of policy may continue to attempt to promote certain lending to certain types of borrowers. It generally has done this by requesting banks to participate in remedial programs for troubled corporate borrowers.... The government has in this manner promoted low-mortgage lending and lending to technology companies. We expect that all loans made pursuant to government policies will be reviewed in accordance with [Kookmin’s] credit review policies. However, we cannot assure you that government policy will not influence [Kookmin] to lend to certain sectors or in a manner in which [Kookmin] otherwise would not in the absence of government policy.

Id. at 57-58 (emphasis added); see also Def.’s App., App. 12 (Attach. 1 of Petitioner’s Comments to Commerce dated Mar. 28, 2003) at 22. 13 In the Court’s view, Commerce reasonably found that this filing served as an admission by a Hynix creditor of the tendency of the Korean government to direct private banks to provide financial contributions to technology companies, such as Hynix. 14 To tie this tendency specifically to Hynix’s restructuring, Commerce then cited confidential internal loan documentation obtained from Kookmin at verification which indicated that, as warned in its SEC filing, Kookmin took into account Korean government poli cy goals when weighing its participation in Hynix’s December 2000 syndicated loan. Decision Memo at 59; see also Def.’s App., App. 11 (Ex. 11 of Hynix Verification Report dated May 15, 2003) at 12. Taken together, Commerce reasonably found that this evidence demonstrated that the Korean government was successful in enlisting Kookmin, a financial institution without substantial government ownership, to support Hynix during its restructuring.

In sum, the Court concludes that record evidence supports Commerce’s determination that Hynix received financial contributions from private entities entrusted or directed by the Korean government. As the foregoing discussion demonstrates, Commerce did not rely on “past findings” from earlier countervailing duty investigations involving the Korean government to support its finding of government entrustment or direction. Pis.’ Br. at 10. Rather, Commerce appropriately “point[ed] to evidence from which it [was] reasonable to infer that the government’s control continued into the period of investigation.” AK Steel Corp. v. United States, 192 F.3d 1367, 1376 (Fed.Cir.1999).

2. Counterevidence Adduced by Hy-nix Does Not Undermine Commerce’s Conclusion That Hynix Received Entrusted or Directed Financial Contributions

Of course, the Court’s substantial evidence review does not end with an examination of the evidence supporting Commerce’s finding. The Court must also consider whatever “fairly detracts from the substantiality of [that] evidence.” Huaiyin Foreign Trade Corp. v. United States, 322 F.3d 1369, 1374 (Fed.Cir.2003) (quotation marks omitted). Hynix argues that counterevidence on the record soundly refutes Commerce’s finding of Korean government entrustment or direction of the financial institutions involved in Hynix’s restructuring.

First, Hynix argues that Citibank/SSB, not the Korean government, was responsible for orchestrating Hynix’s restructuring. Pis.’ Br. at 11-16, 25-29. Hynix contends that Citibank/SSB initiated and was at the center of Hynix’s multifaceted restructuring, both as an advisor and participant. Plaintiffs’ Comments on the Final Results of Redetermination (“Pis.’ Remand Comments”) at 7. In these roles, Hynix notes that Citibank/SSB was found not to be under Korean government control. Id. at 9 (citing Decision Memo at 5). Further, Hynix observes that SSB’s engagement letter and restructuring proposals recognized that the Korean government might not provide the regulatory flexibility needed to make Hynix’s restructuring successful. Id. at 10-11. Hynix additionally notes that Citibank committed its own funds to Hy-nix’s restructuring. Id. at 12-13. Taken together, Hynix argues that the evidence concerning Citibank/SSB’s commercial involvement demonstrated the independence of Hynix’s restructuring from the Korean government. Id. at 12-13.

The Court finds that the evidence of Citibank/SSB’s involvement in Hynix’s restructuring is insufficient to undermine Commerce’s finding of government en-trustment or direction. The parties agree that, while important to the restructuring from a technical perspective, SSB did not have the ability to ensure the participation of Hynix’s creditors in the various phases of Hynix’s restructuring. See Remand Results at 7; Pls.’ Remand Comments at 8. In other words, for SSB’s restructuring blueprint to work, Hynix’s creditors had to participate — either voluntarily or through government coercion. Hynix places great emphasis on the fact that Citibank/SSB, as demonstrated by its proposals to Hynix’s creditors and affidavits to Commerce, believed that commercial persuasion (not government coercion) was the motivating force behind creditor participation. Indeed, this rightfully is circumstantial evidence, that weighs against Commerce’s determination. However, even the documents cited by Hynix acknowledge that the restructuring devised by SSB was subject to some form of Korean government approval. See Remand, Results at 8. Regardless, Citibank/SSB’s view of the nature of the Korean government’s involvement in Hynix’s restructuring or the reasons for creditor involvement is but one of the opinions collected by Commerce during its investigation. Considering the totality of the evidence before the agency, which included reports of behind-the-scenes Korean government coercion by numerous independent sources, 15 Commerce reasonably chose to disbelieve the minority view of Citibank/SSB.

Further, this choice by the agency was not significantly undercut by evidence of Citibank’s own financial participation in Hynix’s restructuring. As discussed in greater detail infra at Part III.D.l.a, Commerce found that Citibank purposefully waited until the involvement of the other creditors was assured before committing resources to Hynix’s restructuring. The important point for Citibank was the participation of the other creditors — not their rationale (or provocation) for doing so. Citibank’s “symbolic gesture” of support for Hynix (and, by extension, the potentially marketable Korean corporate restructuring blueprint represented by Hynix), Decision Memo at 9-10, was therefore minimally probative on the question of the true nature of the Korean government’s role in the restructuring. As such, it was reasonable for Commerce to find that Citibank/SSB’s involvement did not negate the existence of government entrustment or direction in Hynix’s restructuring.

Second, Hynix contends that Hynix’s restructuring included commercial options and contingencies which belie a finding of government entrustment or direction. Pis.’ Remand Comments at 13-20. With regard to the May 2001 phase of Hynix’s restructuring, which featured an international GDS offering, Hynix argues that the provisions of this offering demonstrate that the success of Hynix’s restructuring depended on the support of international investors — not the Korean government. Id. at 13. Hynix contends that, because the record evidence demonstrates that the May 2001 restructuring was contingent on commercial action, Hynix’s restructuring had to have been independent from the Korean government. Id. at 16.

The Court finds that the inclusion of the GDS offering in the May 2001 restructuring is also insufficient to undermine Commerce’s finding of government entrustment or direction. Record evidence shows that Hynix’s creditors voted to provide the new loan and debt restructuring package featured in the May 2001 restructuring before the GDS offering even began. Remand Results at 12. While the GDS offering was underway, an offering memorandum was circulated to potential investors, characterizing the May 2001 package as one of the “Concurrent Financing Transactions” central to Hynix’s overall restructuring. Id.; see also Pis.’ App., App. 1C (Ex. 5 of Hynix Questionnaire Resp. dated Jan. 27, 2004). Then, before the GDS offering closed, Hynix’s creditors met again to work out important details of the restructuring package. Remand Results at 13. Based on the timing of the creditors’ agreements and the characterization of the restructuring package in the offering memorandum, Commerce found it “unlikely that the banks were truly waiting until the successful conclusion of the GDS to decide whether to proceed with the May restructuring.” Id. Hynix looks to the same evidence and finds that it supports the opposite inference — that the May 2001 restructuring package was contingent on approval by international investors and not the Korean government. Pis.’ Remand Comments at 13-16. Upon a careful review of the record evidence, the Court is forced to conclude that both interpretations of the documents related to the May 2001 restructuring are equally plausible. Faced with this equipoise, the Court must defer to the interpretation made by Commerce as the agency expert. See Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 167, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962) (noting that “[ejxpert discretion is the lifeblood of the administrative process”). As such, it was reasonable for Commerce to find that the GDS offering featured in Hynix’s May 2001 restructuring did not negate the existence of government entrustment or direction.

Hynix next argues that, with regard to the October 2001 phase of Hynix’s restructuring, the existence of multiple options available to creditors (including debt liquidation) negates government control. Pis.’ Remand Comments at 17. Hynix notes that these options were adopted by a vote of Hynix’s creditors council — a vote which required the support of the holders of at least seventy-five percent of Hynix’s outstanding debt. Id. Hynix contends that Commerce failed to demonstrate that the Korean government had control over creditors holding this amount of Hynix’s debt. Id. According to Hynix, at best Commerce showed that the Korean government had ownership interests in certain Hynix creditors, but that even these creditors did not hold the requisite seventy-five percent of Hynix’s outstanding debt. Id. at 18. Hy-nix contends that, as a result, the Korean government was simply not in a position to force Hynix’s creditors to follow any course of action — as reflected in the availability of multiple options during the October 2001 restructuring. Id. at 19. 16

The Court finds that the inclusion of multiple options in the October 2001 re structuring is insufficient to undermine Commerce’s finding of government en-trustment or direction. In the Remand Results, Commerce provided a more detailed explanation of the options made available to creditors during the last stage of Hynix’s restructuring. Although these options offered varying degrees of continued involvement in Hynix, none of the options provided an immediate refund of liquidated loans to creditors. Remand Results at 15. Instead, even under the option most favorable to a creditor seeking to extricate itself from the restructuring, funds from the liquidated loans were converted back into five-year, interest free loans to Hynix. Id. Under any scenario, Hynix stood to benefit from either complete debt extinguishment or partial debt extinguishment coupled with sufficient new loans to service the remaining debt. Id. at 16. These were the options presented to Hynix’s creditors, notwithstanding the fact that the October 2001 restructuring was an “unforeseen event, made necessary by an unexpected slump in the DRAM market” — in other words, made necessary by Hynix’s still worsening financial position. Pis.’ Br. at 12. While it is hard to imagine what a good set of options might have been for Hynix’s creditors in this situation, see Pis.’ Remand Comments at 19, it is suspect that “under any scenario, Hynix would be saved to the detriment of its creditors.” Remand Results at 16. Viewed in this light, Commerce reasonably found that the October 2001 restructuring options were not commercial in nature and, therefore, did not contradict a finding of government entrustment or direction.

In reaching this conclusion, the Court rejects Hynix’s chief rejoinder — that Commerce failed to demonstrate that the Korean government had control over a sufficient percentage of the creditors council in order to vote into place any sort of noncommercial options. See Pis.’ Remand Comments at 17. As the Court found above in Part III.B.l, Commerce did in fact adduce evidence supporting its conclusion that the Korean government was able to influence or coerce multiple members of Hynix’s creditors council, both with and without government ownership. Further, the very existence of the highly suspect options featured in the last restructuring phase actually reinforces Commerce’s determination that government entrustment or direction persisted for the duration of the alleged ten-month program.

The Court recognizes that this last piece of circumstantial evidence- — like each set of evidence related to the Korean government’s motive, proclivity, opportunity, and capacity to support Hynix in a manner at odds with the countervailing duty statute — would fall short of meeting the substantial evidence standard if viewed in isolation. Hynix has ably demonstrated as much throughout its briefing. Unfortunately for Hynix, this observation is of no moment. Commerce need not exclusively rely on any one piece or set of evidence to prove entrustment or direction. Rather, Commerce must show through the totality of its evidence that entrustment or direction has taken place. See Hynix I at -, 391 F.Supp.2d at 1349. Commerce has done so here. Through its substantial direct and circumstantial evidence, Commerce has “connect[ed] ostensibly disparate parties and transactions to a single, interrelated program of government en-trustment or direction.” Id. at-, 391 F.Supp.2d at 1350.

Admittedly, Commerce’s finding of government entrustment or direction here is not without some doubt. This is a close case. In such circumstances, however, “the Court may not substitute its judgment for that of the [agency] when the choice is between two fairly conflicting views[.]” San Francisco Candle Co. v. United States, 27 CIT-, 265 F.Supp.2d 1374, 1381 (2003) (quotation marks omitted). Accordingly, the Court upholds Commerce’s conclusion that Hynix’s creditors were entrusted or directed by the Korean government to provide financial contributions to Hynix as supported by substantial evidence.

C. Commerce’s Government Subsidy Function Analysis Is In Accordance with Law and Supported by Substantial Evidence

Hynix next argues that- Commerce wrongly concluded that the prong of 19 U.S.C. § 1677(5)(B)(iii) pertaining to the performance of a government subsidy function was satisfied in connection with the investigated financial contributions. Pls.’ Br. at 33 (citing Decision Memo at 47). Hynix asserts that, as a matter of law, actions taken by a wholly independent actor cannot possibly be actions or practices normally vested in or followed by governments — the standard established by the relevant portion of 19 U.S.C. § 1677(5)(B)(iii). Id. at 33-34. Hynix contends that Commerce erred by ignoring the close parallels between the actions of Citibank, a concededly independent commercial actor, and other creditors deemed to be under government control. Id. Since Hynix’s creditors generally acted like Citibank during the restructuring, Hynix argues that a government subsidy function simply could not have been performed. Id. Commerce’s conclusion to the contrary was, in Hynix’s view, not in accordance with law and unsupported by substantial evidence. Id.

The Court upholds both Commerce’s interpretation and application of the portion of 19 U.S.C. § 1677(5)(B)(iii) pertaining to the performance of a government subsidy function. First, under two-step Chevron analysis, Commerce’s interpretation of the relevant statutory language is in accordance with law. 19 U.S.C. § 1677(5)(B)(iii) provides that, to be actionable, the provision of a financial contribution must be done by a function or practice normally vested in or followed by government; however, the statute does not define or provide examples of such functions or practices. The relevant legislative history is likewise silent. In light of this statutory ambiguity, Commerce is given deference under Chevron step one to make a reasonable interpretation. See Floral Trade Council v. United States, 23 CIT 20, 24, 41 F.Supp.2d 319, 324 (1999) (noting that courts will defer to Commerce’s reasonable interpretation under Chevron where Congress’s intended definition of a term is not ascertainable through statutory construction).

Turning to Chevron step two, Commerce determined that this portion of 19 U.S.C. § 1677(5)(B)(iii) was best understood as making actionable under the countervailing duty law only those financial contributions which could be characterized as fulfilling a “governmental subsidy function[.]” Decision Memo at 47. In the Court’s view, an example best demonstrates the soundness of this interpretation: in the context of an ordinary civil trial, a government, through its courts, could order a losing party to pay the prevailing party punitive damages. Such a court order would direct a private entity to transfer funds to another private entity without any consideration, resulting in a windfall to the second party. This order would contain the familiar elements of government direction, financial contribution, and benefit — but should such an order reasonably be considered a eountervailable subsidy? 19 U.S.C. § 1677(5)(B)(iii), as interpreted by Commerce, clearly provides the answer: no, because under normal circumstances court-ordered punitive damages do not fulfill a government subsidy function. 17 Commerce’s interpretation of 19 U.S.C. § 1677(5)(B)(iii) avoids the nonsensical result of bringing many more government actions within the ambit of the countervailing duty law than could have been plausibly intended by Congress.

Further, the Court is not persuaded by Hynix’s criticism of Commerce’s interpretation. In essence, Hynix argues for a more limited reading of the government subsidy function requirement of 19 U.S.C. § 1677(5) (B) (Hi) — namely that a government subsidy function cannot be performed if the practice in question is commercially rational. What Hynix fails to recognize is that the countervailing duty statute already requires Commerce to consider the relative commercially of a financial contribution — to determine if a benefit has been conferred. See 19 U.S.C. § 1677(5)(B)(iii) (1999); id. § 1677(5)(E). Hynix’s interpretation seeks to unnecessarily conflate two statutorily distinct inquiries and is therefore unpersuasive. As such, the Court finds Commerce’s reasonable statutory interpretation of the portion of 19 U.S.C. § 1677(5)(B)(ni) pertaining to the government subsidy function requirement to be in accordance with law.

In addition, the Court finds substantial evidence in support of Commerce’s conclusion that the government subsidy function requirement was satisfied by the investigated financial contributions. 18 As discussed above at Part III.B, Hynix’s creditors transferred funds, in the form of preferential loans and equity infusions, pursuant to the entrustment or direction of the Korean government. If the Korean government had undertaken these transfers directly, there can be no question that it would have thereby provided eoun-tervailable subsidies to Hynix. See 19 U.S.C. § 1677(5)(B)(i) (1999) (describing eountervailable subsidy to include benefit-conferring financial contribution provided directly by a government). In effect, the Korean government delegated its subsidy function to Hynix’s creditors, which then performed officially sanctioned “financial support activities])]” Decision Memo at 61. As such, substantial evidence supports the conclusion that the government subsidy function requirement of 19 U.S.C. § 1677(5)(B)(iii) was met.

Accordingly, the Court upholds Commerce’s conclusion that Hynix’s creditors performed a government subsidy function for purposes of 19 U.S.C. § 1677(5)(B)(iii) as in accordance with law and supported by substantial evidence.

D. Commerce’s Benefit Analysis Is In Accordance with Law and Supported by Substantial Evidence

Turning to the final prong of 19 U.S.C. § 1677(5)(B)(iii), Hynix alleges error with the two principal analyses underlying Commerce’s conclusion that a eountervaila-ble benefit was conferred to Hynix during its restructuring. First, Hynix claims that Commerce’s creditworthiness analysis, which determined that Hynix would not have been able to attract loans from commercial sources during its restructuring, was flawed. Pis.’ Br. at 36-40, 42-45, 49-50. Second, Hynix argues that Commerce’s equity worthiness analysis, which determined that Hynix would not have been able to attract equity from commercial sources during its restructuring, was also flawed. Id. at 40-42, 46-49. Hynix’s specific arguments concerning Commerce’s creditworthiness analysis and equity worthiness analysis are addressed separately below. 19

For the reasons that follow, the Court upholds Commerce’s conclusion that a benefit was conferred to Hynix by its receipt of government-entrusted or directed financial contributions as in accordance with law and supported by substantial evidence.

1. Creditworthiness Analysis

a. Commerce’s Rejection of Loans Made by Citibank as Commercial Benchmarks in Hynix’s Creditworthiness Analysis Is Reasonable

Hynix contends that Commerce erroneously rejected as commercial benchmarks the loans made by Citibank to Hynix during its restructuring, ultimately leading to an inaccurate creditworthiness analysis. Pis.’ Br. at 36. Hynix argues that Commerce should have considered these loans, made by a concededly independent commercial actor, as evidence that Hynix was creditworthy. Id. at 37. In support of this position, Hynix points to the countervailing duty regulations, which state that “the receipt [by an investigated company] of comparable long-term commercial loans, unaccompanied by a government-guarantee, will normally constitute dispositive evidence that [the investigated company] is not uncreditworthy.” Id. at 35 (quoting 19 C.F.R. § 351.505(a)(4)(ii) (2005)). Hynix also contends that Commerce ignored affidavits by Citibank officials indicating that the bank’s involvement in Hynix’s restructuring stemmed from purely commercial motivations, rather than the influence of the Korean government. Id. at 42-44. Finally, Hynix argues that Citibank’s dual role as lender and financial advisor to Hy-nix should not have led Commerce to the conclusion that Citibank was different from the average lender. Id. at 44-45.

The Court finds that Commerce reasonably determined that the loans made by Citibank were not suitable commercial benchmarks for use in Hynix’s creditworthiness analysis. First, Commerce acted in accordance with law when it considered the influence of governmental actions on a private entity whose loans were proffered as commercial benchmarks of creditworthiness for an investigated company. The preamble to the countervailing duty regulations explains that Commerce will carefully examine any loan made by a private entity which is part of a package including government loans to determine if the loan is truly “commercial” in nature. Countervailing Duties, 63 Fed. Reg. 65348, 65364 (Dep’t Commerce Nov. 25, 1998) (final rule). This examination is necessary because, as Commerce has noted, “special features” in such a loan package may influence an otherwise independent, commercial lender to “offer lower, more favorable terms than would be offered absent the government/eommereial bank package.” Id. For purposes of this examination, Commerce need not find that a private entity has been entrusted or directed by a government for that entity to nonetheless be influenced by the government’s actions when making investment decisions. For example, it would be fully rational for an independent private entity seeking to make sound business decisions based on market factors to take into consideration a government’s pervasive involvement in the restructuring of a company. Although rightly a factor in the commercial decision-making process, such government influence would render that entity’s loans inappropriate for use as commercial benchmarks in creditworthiness analysis. 20

Further, Commerce’s finding that Citibank’s lending decisions were influenced by the Korean government’s involvement in Hynix’s restructuring is supported by substantial evidence. Commerce appropriately took great care in examining the nature of Citibank’s lending to Hynix during a restructuring which involved significant participation by government-entrusted and directed financial institutions. Commerce found that Citibank waited until the involvement of these financial institutions was assured before it committed resources to Hynix’s restructuring. Decision Memo at 9-10; see also Def.’s App., App. 8 (Hynix Verification Report dated May 15, 2003) at 20 (“Citibank decided to participate in the [bond] issuance that was part of the May restructuring to provide a ‘symbolic gesture of support’ to show that Citibank willing [sic] to stand behind Hy-nix.”); id. at 21 (“Citibank felt that it was best to provide a small additional amount of funding and ‘ride’ with the [Korean] banks to see if Hynix could make it as an ongoing concern. The officials explained that Citibank was making a bet that the [Korean] banks would protect their expo sure.”). Only then did Citibank seek internal credit approval for its portion of the first loan to Hynix. Def.’s App., App. 8 (Hynix Verification Report, dated May 15, 2003) at 20 (“According to Citibank officials, it did not seek internal credit approval for its portion of the syndicated bank loan until after the [Korean] banks committed to the syndicated bank loan.”). In addition, the level of Citibank’s lending to Hynix — only 12.5 percent of the December 2000 syndicated loan and a small percentage of the May 2001 restructuring package — further supports Commerce’s conclusion that Citibank was able to alter its lending risk calculus by relying on the dominant participation of government-entrusted or directed financial institutions. 21 See Def.’s App., App. 7 (Commerce Mem. on Bus. Proprietary Info, for Final Determination dated June 16, 2003) at Attach. 1 (detailing Citibank’s share of new debt extended to Hynix); id., App. 24 (Hynix Supplemental Resp. dated Mar. 4, 2003) at Ex. 8 (detailing Hynix’s various loans). Based on this evidence, it was reasonable for Commerce to conclude that the involvement of government-entrusted or directed financial institutions affected the terms by which Citibank agreed to lend to Hynix. Hynix counters that Commerce failed to take into consideration two affidavits by Citibank officials which indicated that Citibank acted in a purely commercial manner independent of government influence. Pis.’ Br. at 42-44; see Pis.’ App., App. 17 (Citibank Aff. dated Mar. 20, 2003); id., App. 18 (Citibank Aff. dated May 22, 2003). However, Commerce specifically addressed these affidavits in its Decision Memo and in fact altered certain aspects of its preliminary analysis as a result of this evidence. See Decision Memo at 8 (“Since our preliminary analysis, relevant evidence has been added to the record which warrants a reconsideration.... This includes information provided by Citibank officials in an interview at verification, which Citibank further clarified in a second affidavit....”). Commerce nonetheless concluded that these affidavits supported its finding that Citibank considered factors, like the participation of government-entrusted or directed financial institutions, 22 when lending to Hynix. Upon a careful review of these largely confidential affidavits as urged by Hynix, the Court cannot disagree with this conclusion.

As such, the Court concludes that Commerce’s rejection of Citibank’s loans as commercial benchmarks in Hynix’s creditworthiness analysis is reasonable.

b. Commerce’s Rejection of the Korean Default Rates Supplied by Hynix Is Reasonable

Hynix next contends that, once Commerce erroneously determined that Hynix was unereditworthy, Commerce further erred by using Moody’s U.S. average cumulative default rates to construct an un-creditworthy benchmark for use in calculating the benefit received by Hynix from government-entrusted or directed loans. Pis.’ Br. at 49. Hynix argues that Commerce should have instead used Korean default rates for corporate bonds published by Korean bond rating agencies and provided to Commerce by Hynix. Id. at 49. Hynix argues that, while Commerce’s regulations generally require the agency to use Moody’s U.S. data, they also allow Commerce to consider country-specific data when available. Id. (citing Countervailing Duties, 63 Fed. Reg. 65348, 65365). Hynix contends that Commerce should have recognized that the available Korean default data was much more accurate than Moody’s U.S. default data for this investigation. Id. Further, Hynix argues that Commerce inappropriately rejected the Korean default data simply because it was “not sufficiently clear” that the data was comparable to Moody’s U.S. data. Id. (quoting Decision Memo at 105). Hynix argues that Commerce was legally required to seek the needed clarification from Hynix before drawing such an unjustified, adverse inference about the data. Pls.’ Br. at 49 (citing Helmerich & Payne, Inc. v. United States, 22 CIT 928, 24 F.Supp.2d 304 (1998) (holding that Commerce must provide opportunity for clarification/correction before drawing adverse inferences)).

The Court finds that Commerce reasonably rejected the Korean default data provided by Hynix for calculating the uncre-ditworthy benchmark. First, Commerce appropriately found the data provided by Hynix to be unclear and incomplete. The countervailing duty regulations state that Commerce normally uses the average “cumulative” default rates developed by Moody’s to construct unereditworthy benchmarks. 19 C.F.R. § 351.505(a)(3)(iii) (2005). The preamble to Commerce’s countervailing duty regulations reiterates this requirement and makes clear that Commerce will only consider using non-Moody’s data that is “detailed and comprehensive[.]” Countervailing Duties, 63 Fed. Reg. 65348, 65365. Notwithstanding this, Hynix provided default data without any indication that it was cumulative. Decision Memo at 105. Further, the minimal data offered by Hynix provided none of the detail or discussion of methodology which would have allowed Commerce to compare the quality of that data to Moody’s U.S. data. Id. Instead, a portion of the Korean default data was anomalous on its face. See Pis.’ App., App. 10 (Supplemental Questionnaire Resp. of Hynix dated Mar. 4, 2003) at Ex. 18 (report of one Korean bond rating agency indicating a default rate of 2.47 percent for “CCC and below” bonds and a default rate of 4.98 percent for “A” bonds). Hynix did not provide any explanation for these aberrational default rates with its submission. Given these omissions, Commerce had reasonable grounds to question the reliability of the Korean default data provided by Hynix and ultimately disregard it.

Second, Commerce did not err by declining to request clarification of the Korean default data from Hynix. Commerce was not required by the countervailing duty statute or regulations to offer Hynix an opportunity to better explain or correct its proffered data. Rather, the countervailing duty regulations place the onus on the parties to an investigation to convince Commerce that more accurate, country-specific default information is available. 23 To that end, Plaintiffs’ reliance on Helme-rich is misplaced. Helmerich stands for the principle that Commerce must “fairly request” information from a party before drawing an “adverse inference” that the party has failed to cooperate. Helmerich, 22 CIT at 931, 24 F.Supp.2d at 308. This principle is not applicable to Commerce’s rejection of the Korean default data because Commerce did not apply adverse inferences. Rather, Commerce’s standard methodology is to use Moody’s U.S. default data. This court has held that Commerce does not make an adverse inference by “simply following its standard prac-tiee[.]” Rhodia, Inc. v. United States, 26 CIT 1107, 1111, 240 F.Supp.2d 1247, 1251 (2002).

As such, the Court concludes that Commerce’s rejection of the Korean default rates for purposes of Hynix’s creditworthiness analysis is reasonable.

2. Equityworthiness Analysis

a. Commerce’s Rejection of Equity Investments Made by Citibank as Commercial Benchmarks in Hynix’s Equityworthiness Analysis Is Reasonable

Hynix next argues that Commerce erred in finding Citibank’s equity investment in Hynix to be too small for use as a commercial benchmark of the price of Hynix’s equity, ultimately resulting in a flawed eq-uityworthiness analysis. Pis.’ Br. at 40. Hynix notes that, following the October 2001 debt-to-equity conversion, Citibank became Hynix’s fifth largest shareholder. 24 Id. Hynix contends that Citibank’s investment, representing only a small percentage of the total debt-to-equity conversion but valued at tens of millions of dollars, should have been considered “significant” and thus qualified for use as a commercial benchmark. Id. at 41 (citing 19 C.F.R. § 351.507(a)(2)(iii) (2005)). Hynix argues that Commerce should have reached this conclusion in order to be consistent with its past determinations. Id. (citing Small Diameter Circular Seamless Carbon and Alloy Steel Standard, Line and Pressure Pipe from Italy, 60 Fed. Reg. 31992 (Dep’t Commerce June 19,1995) (final determination) (finding investment valued at approx imately $15 million and resulting in 18.3 percent ownership interest in an investigated company to be significant) (“Seamless Pipe from Italy”)). Hynix argues that an appropriate comparison of Citibank’s investment in Hynix and that in Seamless Pipe from Italy should have compared the amount of the investment and not the percent ownership interest involved. Pis.’ Br. at 41. According to Hynix’s calculations, had Commerce compared the amounts invested in the two investigated companies, Commerce would have found that the value of Citibank’s investment was far greater than the investment at issue in Seamless Pipe from Italy, which Commerce had determined to be significant under 19 C.F.R. § 351.507(a)(2)(iii). Id.

The Court concludes that Commerce appropriately found the equity investment made by Citibank was not a suitable commercial benchmark for use in Hynix’s equi-tyworthiness analysis. First, Commerce’s interpretation of the significant investment requirement of 19 C.F.R. § 351.507(a)(2)(iii) is reasonable. Commerce included the significant investment standard in its regulations based on the observation that “the volume of a firm’s traded shares [may] be so low as to preclude the use of those shares as a benchmark.” Countervailing Duties, 62 Fed. Reg. 8818, 8832 (Dep’t Commerce Feb. 26, 1997) (notice of proposed rulemaking and request for public comments). However, neither this language nor the text of the regulation makes clear how the significance of an investment should be evaluated by Commerce. Commerce argues that it should, and here as in Seamless Pipe from Italy properly did, focus its inquiry on the percent involvement by private investors— rather than the dollar value of private investment. The Court agrees. In this case as well as Seamless Pipe from Italy, Commerce largely rested its significance determination on the percent interest held by private investors, even though the dollar amount of private investment could be roughly deduced from the facts presented. See Decision Memo at 90; Seamless Pipe from Italy, 60 Fed. Reg. at 31994. While this construction of the term “significant” is not necessarily compelled by the language of Commerce’s regulation, it is far from being at odds with the regulation. Analysis, of percent interest appears to provide a controlled and predictable way for Commerce to evaluate the significance of equity investments across different industries and investigations. Comparisons of dollar amounts across investigations, as suggested by Hynix, would require Commerce to control for the effects of inflation, exchange rate fluctuations, and, most challenging, the variability in the intrinsic value of companies in order to apply the regulation in an evenhanded manner. The difficulties inherent in Hynix’s approach demonstrate the reasonableness of Commerce’s construction of its own regulation, already entitled to significant deference under this Court’s standard of review.

Second, applying Commerce’s construction of 19 C.F.R. § 351.507(a)(2)(iii) to this case, substantial evidence supports the finding that Citibank’s equity stake in Hy-nix was not significant. Citibank’s equity purchase during Hynix’s restructuring was well below the 18.3 percent equity stake found to be significant in Seamless Pipe from Italy. Although the exact “significant” threshold is not clear from Commerce’s construction, it was reasonable for Commerce to conclude that an investment the size of Citibank’s was not significant.

As such, Commerce’s rejection of Citibank as a commercial benchmark in Hy-nix’s equityworthiness analysis is reasonable.

b. Commerce’s Use of the Expected Utility Model to Reject the October 2001 Debt-to-Equity Conversion as Evidence of Hynix’s Equityworthiness Is Reasonable

Hynix next contends that, despite the absence of commercial benchmarks, Commerce erroneously disregarded the October 2001 debt-to-equity conversion as evidence that Hynix was otherwise equityworthy. Pis.’ Br. at 48. In rejecting this transaction, Hynix argues that Commerce impermissibly grafted an economic theory — the Expected Utility Model — into the countervailing duty statute and regulations. Id. Contrary to this model, which states that the existence and status of previous investments in a company are extraneous considerations when weighing new investment in the same company, Hynix argues that it is natural for an existing creditor to consider the likely effect of a new investment on an existing investment in the same company. Id. According to Hynix, an investor already deeply committed to a company might make an additional capital infusion in the hopes that more resources will help the company to improve. Id. If Commerce had instead applied this principle, Hynix argues that it would have found the October 2001 debt-to-equity conversion to be consistent with the “usual investment practice” of at least some “private investors” — the standard by which Commerce must evaluate the benefit conferred by an equity infusion. Id. (citing 19 U.S.C. § 1677(5)(E)(i) (1999)).

The Court finds that Commerce’s rejection of the October 2001 debt-to-equity conversion as evidence of Hynix’s equity-worthiness is reasonable. First, Commerce’s use of the Expected Utility Model is in accordance with law. Commerce has repeatedly used the Expected Utility Model as a methodological tool to help analyze the equityworthiness of investigated companies. See, e.g., Certain Steel Products from Austria, 58 Fed. Reg. 37217, 37249-50 (Dep’t Commerce July 9, 1993) (final determination); Certain Hot Rolled Lead and Bismuth Carbon Steel Products from the United Kingdom, 58 Fed. Reg. 6237, 6245 (Dep’t Commerce Jan. 27, 1993) (final determination); Steel Wheels from Brazil, 54 Fed. Reg. 15523, 15530 (Dep’t Commerce Apr. 18, 1989) (final determination). Although not identifying the model by name, this court has upheld the logical underpinnings of this methodology in past reviews of Commerce’s countervailing duty determinations. See British Steel Corp. v. United States, 10 CIT 224, 231, 632 F.Supp. 59, 65 (1986); Companhia Siderurgica Paulista, S.A. v. United States, 12 CIT 1098, 1101-03, 700 F.Supp. 38, 42-43 (1988). In British Steel, an analogous case involving a previous version of the countervailing duty statute, the court reviewed Commerce’s finding that the British government’s equity investments in the investigated company were inconsistent with commercial considerations. British Steel, 10 CIT at 224-25, 632 F.Supp. at 60. British Steel argued that such equity investments should be considered commercially reasonable where the funds are used to help cover the operating losses of an investigated company which is able to cover its variable costs; the additional equity would be used to pay down fixed costs. Id. at 228-29, 632 F.Supp. at 62-63. British Steel maintained that it would be economically rational for an investor to support continued operations by such a company so that the company’s overall loss would be minimized by the additional investment funds. Id. This court upheld Commerce’s rejection of British Steel’s arguments, stating:

[I]t would be unrealistic to expect a private sector investor to supply operating funds to a loss-incurring firm merely to permit the firm to continue operations to minimize its losses. Thus, while it may be perfectly rational for an owner to sustain loss-minimizing operations, it would not be commercially reasonable for an investor to provide funds for that purpose without adequate assurance of the future profitability of the enterprise and a return on ... investment within a reasonable time.

Id. at 231, 632 F.Supp. at 65. The poor financial prospects of the investigated company, based on a trend of consistently bad returns, were critical to this court’s implicit support of the Expected Utility Model used by Commerce in British Steel. See also Companhia Siderurgica Paulista, 12 CIT at 1103, 700 F.Supp. at 43 (approving of Commerce’s “comprehensive analysis” which focused on company’s current health, past performance, independent studies, and industry forecasts).

Like British Steel, substantial evidence in this case demonstrates that Hynix was a company whose consistently bleak financial results could provide a reasonable investor with little assurance of future profitability. See Decision Memo at 92; Def.’s App., App. 4 (Creditworthiness Analysis of Hy-nix Semiconductor, Inc. dated Mar. 31, 2003) at 3-5 (analyzing Hynix’s financial records from 1997 to 2002). Hynix’s arguments against application of the Expected Utility Model would perhaps warrant greater consideration in a case where investors in a generally sound company faced only short-term financial troubles. See British Steel, 10 CIT at 231, 632 F.Supp. at 65 (noting that “equity infusions in loss-incurring companies do not per se confer a subsidy”). Such is not the case here. Under these circumstances, application of the Expected Utility Model as a methodological tool for assessing equi-tyworthiness is reasonable. Commerce therefore properly found that the purchase of an additional equity stake in Hynix in October 2001 was inconsistent with the usual investment practice of private investors.

As such, Commerce’s rejection of the October 2001 debt-to-equity conversion as a commercial benchmark for use in Hy-nix’s equityworthy analysis is reasonable.

c. Commerce’s Rejection of Third Party . Studies Commissioned by Hynix’s Creditors During the Restructuring as Evidence of Hynix’s Equitywor-thiness Is Reasonable

Finally, Hynix contends that Commerce improperly rejected studies by third parties which discussed the financial merits of investment in Hynix and proved that Hy-nix was otherwise equityworthy during its restructuring. Pis.’ Br. at 47. Hynix argues that these studies, relied upon by Hynix’s creditors in making their investment decisions, provided important insight into the perceived commercial rationality of transactions like the October 2001 debt-to-equity conversion. Id. at 46-47. Hynix argues that Commerce should not have disregarded these studies simply because they focused on investment in Hynix from a creditor (rather than new investor) perspective. Id. at 47 (citing Decision Memo at 91 (noting that studies focused on “financial mechanisms available to save Hy-nix from collapse”)).

The Court upholds Commerce’s rejection of the third party studies commissioned by Hynix’s creditors as evidence of Hynix’s equityworthiness. Hynix advances these studies as its principal evidence of why an already committed investor would continue to finance the struggling Hynix in order to minimize losses. To the extent that this argument largely relies on acceptance of Hynix’s criticism of the Expected Utility Model, it is rejected by the Court. See supra Part III.D.2.b. In addition, the Court notes that Commerce questioned the credibility and/or reliability of several of these studies for reasons related to soundness of methodology and independence of analysis. See Decision Memo at 91-92. This finding, uncontested here by Hynix, provides an alternative ground for rejecting these studies because they were not “[ojbjective analyses” as required by the countervailing duty regulations. 19 C.F.R. § 351.507(a)(4)(i)(A) (2005).

As such, Commerce’s rejection of the third party studies for use in Hynix’s equityworthy analysis is reasonable and, accordingly, the Court * upholds as in accordance with law and supported by substantial evidence Commerce’s conclusion that the government-entrusted or directed financial contributions received by Hynix during its restructuring conferred a coun-tervailable benefit.

IV. CONCLUSION

For the foregoing reasons, the Court sustains the Remand Results and the remainder of the previously deferred Final Determination. Judgment shall be entered accordingly.

1 . This section provides, in pertinent part:

A subsidy is described in this paragraph in the case in which an authority ...

(iii) makes a payment to a funding mechanism to provide a financial contribution, or entrusts or directs a private entity to make a financial contribution, if providing the contribution would normally be vested in the government and the practice does not differ in substance from practices nor-malty followed by governments, to a person and a benefit is thereby conferred.

19 U.S.C. § 1677(5)(B) (1999) (emphasis added).

2 . References to the countervailing duty statute are to the Tariff Act of 1930, as amended by, inter alia, the Uruguay Round Agreements Act, 19 U.S.C. §§ 1671 ef seq.

3 . The Court has adopted this term as a matter of convenience. It is intended to refer to the portion of the statute which states: "if providing the contribution would normally be vested in the government and the practice does not differ in substance from practices normally followed by governments!.]” 19 U.S.C. § 1677(5)(B)(iii).

4 . The record evidence adduced by Commerce in support of its finding of government en-trustment or direction is discussed in detail infra, at Part III.B.l.

5 . Creditworthiness is a term of art which refers to the "attempt to determine if the company in question could obtain long-term financing from conventional commercial sources" at the time of the government-entrusted or directed loan. Decision Memo at 6; see also 19 C.F.R. § 351.505(a)(4) (2005).

6 . Equityworthiness is a term of art which refers to the attempt to determine if the company in question could, "from the perspective of a reasonable private investor” at the time of the government-entrusted or directed equity infusion, show "an ability to generate a reasonable rate of return within a reasonable time.” Decision Memo et 6; see also 19 C.F.R.'§ 351.507(a)(4) (2005).

7 . References to the countervailing duty regulations are to 19 C.F.R. § 351.101 et seq.

8 . Hynix contends that the 2001 presidential statement is suggestive only of a possible motive for the Korean government to intervene and does not indicate the formulation of an affirmative government support policy toward Hynix. Pis.' Br. at 17. This is one possible reading of that statement. However, "the evidence on which the agency relies does not exist in a vacuum.” Former Employees of Int’l Bus. Mach. v. United States Sec’y of Labor, 29 CIT -, 403 F.Supp.2d 1311, 1324 (2005). Commerce was entitled to consider the statement in conjunction with other evidence and draw reasonable inferences therefrom.

9 . The Court shares Hynix's concerns about Commerce's third piece of governmental policy evidence, concerning the existence of a more general Korean government policy to support the restructuring process of major Korean companies. See Pis.’ Br. at 18. In the Decision Memo, Commerce failed to cite to any record evidence to support this specific contention. See Decision Memo at 50. Without record support, this observation smacks of bootstrapping by the agency. Nonetheless, the other evidence cited by Commerce supports the inference of the existence of a governmental policy to support Hynix.

10 . Hynix correctly notes that evidence concerning regulatory waivers, a government-backed insurance program, and a government-backed bond conversion program did not demonstrate an inclination by the Klorean government to involve private entities in Hy-nix’s restructuring in a manner at odds with the countervailing duty law. See Pis.’ Br. at 19-20. Similarly, this evidence did not establish that the Korean government affirmatively caused any of Hynix’s creditors to participate in the multiple facets of Hynix’s restructuring. See id. What Hynix fails to recognize is that this evidence did demonstrate the willingness of the Korean government to take action to involve private entities in Hynix's restructuring. Commerce could reasonably consider this evidence for that purpose. See Decision Memo at 52 (noting that the Korean government took measures "that would facilitate the new loans from the company’s key creditors”).

11 . The formation of the creditors council was predated by the first major phase in Hynix’s restructuring, a December 2000 syndicated loan. Pis.' Br. at 32. Hynix correctly notes that, as a result, any opportunity presented by the creditors council could not have applied to this early stage of Hynix’s restructuring. Id. However, as Hynix also notes, ”[t]he October 2001 restructuring ... alone accounted] for about two-thirds of the total alleged subsidy[J" Id. at 13.- In other words, it was reasonable for Commerce to find the evidence related to the creditors council highly probative even if the temporal reach of this evidence was somewhat limited.

12 .It is noteworthy that evidence concerning government ownership interests in certain of Hynix’s creditors cannot be considered conclusive proof of Korean government entrustment or direction of these entities. Rather, as argued by Hynix, these financial institutions are subject to the same inquiry as all other private entities under investigation. See Pis.' Br. at 24. Contrary to Hynix’s contention, Commerce recognized this fact in its Decision Memo, noting that financial institutions were not presumed to be under government en-trustment or direction simply by virtue of government ownership interests. Decision Memo at 17. The evidence which led Commerce to find that these entities were in fact subject to government entrustment or di rection is discussed by the Court later in this section.

13 . Kookmin also filed a similar prospectus in June 2002. Decision Memo at 58. Because Kookmin was the sole Korean bank listed on a U.S. stock exchange during the period of investigation, no other such SEC filings were made by Hynix's creditors. Id.

14 . Hynix argues that Commerce failed to consider "a detailed statement by the specific lawyers who drafted the prospectus, which made clear that the language was in no way meant to imply [Korean government] control over Kookmin lending decisions.” Pis.’ Br. at 32 (citing Pis.' App., App. 16 (Hynix’s Supporting Documentation dated Apr. 14, 2003)). However, "absent a showing to the contrary, [the agency] is presumed to have considered all of the evidence in the record.” Nat’l Ass’n of Minor Mfrs. v. United States, 12 CIT 771, 779, 696 F.Supp. 642, 648 (1988). Hynix has failed to rebut this presumption here; Commerce is not required to expressly distinguish every post hoc, self-serving declaration offered by a party which is facially at odds with the plain meaning of non-technical record evidence.

15 . As noted by Commerce, when investigating an alleged clandestine program of subsidization, "secondary sources can be particularly credible as these observers are independent and without a vested interest in the outcome.” Decision Memo at 50 n.13. Of course, secondary information is not necessarily reliable in all circumstances, which is why the countervailing duty statute requires Commerce to corroborate such information to the extent practicable. See 19 U.S.C. § 1677e(c) (1999). Commerce duly carried out its duty to corroborate during the underlying investigation; however, even more telling, Hynix itself has urged both Commerce and the Court to look to "reliable outside commentary” when analyzing the role played by the Korean government in Hynix's restructuring. Pis.' Br. at 31. While the commentary collected by Commerce during its investigation was hardly unanimous, see id., much (if not the majority) lends support to Commerce's finding of government entrustment or direction.

16 . Hynix also argues that Commerce ignored the fact that Hynix's creditors had a statutory right to seek outside mediation concerning the terms for Hynix's October 2001 restructuring set by the creditors council. Pis.’ Remand Comments at 17. In Hynix's view, recourse to mediation contradicts a finding of any sort of Korean government control over these financial institutions. Id. This is an interesting argument; unfortunately, it does not appear that it was made by Hynix during the administrative proceedings below. See Micron’s Rebuttal Comments on the Final Results of Redetermination at 12-13. It is well established that "[a] reviewing court usurps the agency’s function when it sets aside the administrative determination upon a ground not theretofore presented....” Unemployment Comp. Comm’n of Alaska v. Aragon, 329 U.S. 143, 155, 67 S.Ct. 245, 91 L.Ed. 136 (1946); see also 28 U.S.C. § 2637(d) (1999) (requiring exhaustion of administrative remedies where appropriate).

17 . Rather, court-ordered punitive damages are generally considered to implicate a government's police power. See United States v. Morrison, 529 U.S. 598, 120 S.Ct. 1740, 146 L.Ed.2d 658 (2000) (characterizing use of punitive damages to suppress crime as example of state police power).

18 . Hynix asserts that Commerce "completely ignored the second independent prong of 19 U.S.C. § 1677(5)(B)(iii)” (i.e., the government subsidy function requirement) in reaching its determination. Pis.' Br. at 33. The Court disagrees with this characterization. While Commerce's analysis certainly could have been more rigorously demarcated, the Court must “uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned.” Bowman Transp., Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974).

19 . Concerning both of these analyses, Hynix argues that Commerce erred by refusing to use as commercial benchmarks the loans and equity infusions made by Hynix's creditors (other than Citibank) during Hynix’s restructuring. Pis.’ Br. at 45-46. Because the Court concludes that Commerce reasonably found these loans and equity infusions to have been made pursuant to government-entrustment or direction, see supra Part III.B, the Court also upholds Commerce’s decision to disqualify them as commercial benchmarks.

20 . Consideration of the distortive, if non-countervailable, role a government may play in the marketplace is not limited to this section of the countervailing duty statute. For example, with regard to the privatization of government-owned companies, the presumption of subsidy extinguishment which accompanies the sale of such a company for fair market value “may be rebutted upon a showing that the sale process was distorted through government intervention” in the broader market. Allegheny Ludlum Corp. v. United States, 29 CIT -, 358 F.Supp.2d 1334, 1346 (2005) {citing Notice of Final Modification of Agency Practice Under Section 123 of the Uruguay Round Agreements Act, 68 Fed. Reg. 37125, 37127 (Dep’t Commerce June 23, 2003) (notice of modification of agency practice regarding privatizations)).

21 . Indeed, as indicated in the preamble to the countervailing duty regulations, the “relatively small amount” of a long-term commercial loan may rebut the presumption of creditworthiness which accompanies its receipt by a company. Countervailing Duties, 63 Fed. Reg. 65348, 65367. Accordingly, Commerce’s substantiated finding that Citibank’s lending was "relatively small in absolute and percentage terms compared to the involvement” of the government-entrusted or directed financial institutions during Hynix’s restructuring, Decision Memo at 9, provides independent justification for Commerce's rejection of Citibank’s loans as commercial benchmarks. Hynix's attempt to undermine this finding by comparing Citibank's lending with that of individual government-entrusted or directed financial institutions (rather than the group as a whole), see Pis.’ Br. at 38-40, is unavailing.

22 . In addition, Commerce reasonably found that Citibank likely took into consideration its dual role (through SSB) as Hynix’s financial advisor when making lending decisions. As Commerce noted in the preamble to the countervailing duty regulations, "many characteristics could factor into a decision of whether a loan should be considered comparable to the government-provided loan.” Countervailing Duties, 63 Fed. Reg. 65348, 65363. One such characteristic could be the existence of an alternative revenue stream which directly affects the relative risk of entering into a commercial relationship. As a result, even if Commerce had found Citibank's loans to be otherwise suitable commercial benchmarks, the agency still would have been justified in taking this aspect of Citibank’s loans into consideration.

23 . Compare Countervailing Duties, 63 Fed. Reg. 65348, 65365 (”[I]f [detailed and comprehensive country-specific default data] do exist and are brought to our attention in the course of an investigation ... we would consider using the default rate from the country under investigation.”) (emphasis added) with 19 U.S.C. § 1677m(d) (1999) (if "a response to a request for information under this title does not comply with the request, [Commerce] shall promptly inform the person submitting the response of the nature of the deficiency and shall, to the extent practicable, provide that person with an opportunity to remedy or explain the deficiency”) (emphasis added). Absent an affirmative due process, statutory, or regulatory obligation on the part of Commerce to request clarification of unsolicited default data, the Court will not here impose one. However, the Court notes that Commerce's selection of default data must nonetheless be adequately explained and supported by substantial evidence, thereby potentially limiting Commerce’s discretion in future investigations involving country-specific default data which does not feature the serious infirmities present here.

24 . The actual portion of Citibank’s equity purchase which occurred during Hynix’s restructuring is confidential.