[Federal Register Volume 62, Number 142 (Thursday, July 24, 1997)]
[Notices]
[Pages 39809-39824]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 97-19552]


-----------------------------------------------------------------------

DEPARTMENT OF COMMERCE

International Trade Administration
[A-580-812]


Notice of Final Results of Antidumping Duty Administrative Review 
and Determination Not To Revoke Order In Part: Dynamic Random Access 
Memory Semiconductors of One Megabyte or Above From the Republic of 
Korea

AGENCY: Import Administration, International Trade Administration, 
Department of Commerce.

-----------------------------------------------------------------------

SUMMARY: On March 18, 1997, the Department of Commerce (the Department) 
published the preliminary results of its administrative review of the 
antidumping duty order and notice of intent not to revoke, in part, the 
antidumping duty order on dynamic random access memory semiconductors 
(DRAMs) of one megabyte or above from the Republic of Korea (61 FR 
36029). The review covers exports of the subject merchandise to the 
United States by LG Semicon Co., Ltd. (LGS, formerly Goldstar Electron 
Co., Ltd.) and Hyundai Electronics Industries, Inc. (Hyundai). The 
period of review (POR) is May 1, 1995 through April 30, 1996. This is 
the third review period.
    As a result of our analysis of the comments received, the 
antidumping margins have changed from those presented in our 
preliminary results.

EFFECTIVE DATE: July 24, 1997.

FOR FURTHER INFORMATION CONTACT: Thomas F. Futtner, AD/CVD Enforcement, 
Group II, Office 4, Import Administration, International Trade 
Administration, U.S. Department of Commerce, 14th Street and 
Constitution Avenue, NW., Washington, DC 20230, telephone: (202) 482-
3814.

SUPPLEMENTARY INFORMATION:

The Applicable Statute and Regulations

    Unless otherwise indicated, all citations to the statute are 
references to the provisions effective January 1, 1995, the effective 
date of the amendments made to the Tariff Act of 1930 (the Act) by the 
Uruguay Round Agreements Act (URAA). In addition, unless otherwise 
indicated, all citations to the Department's regulations are to 19 CFR 
Part 353 (1997).

Background

    On May 10, 1993, the Department published in the Federal Register 
(58 FR 27250) the antidumping duty order on DRAMs from the Republic of 
Korea. On May 8, 1996, the Department published a notice of 
``Opportunity to Request an Administrative Review'' of this antidumping 
duty order for the period May 1, 1995, through April 30, 1996 (61 FR 
20791). In accordance with 19 CFR 353.22(a)(2), in May 1996, LGS and 
Hyundai (collectively the respondents) requested that the Department 
conduct an administrative review of their shipments of DRAMs to the 
United States during this period. In addition, both respondents 
requested that the Department revoke the antidumping order, in part, 
pursuant to section 353.25(a)(2) of the Department's regulations. We 
also received a request from the petitioner, Micron Technologies Inc., 
that an administrative review of these same two Korean manufacturers of 
DRAMs be conducted. On June 25, 1996, the Department published a notice 
of initiation of administrative review (61 FR 32771). Based upon the 
fact that we disregarded sales found to have been made below the cost 
of production (COP) in the original less-than-fair-value (LTFV) 
investigation, which was the most recent period for which final results 
were available when this review was initiated, on the same date we 
automatically initiated an investigation to determine whether Hyundai 
and LGS made sales of subject merchandise below the COP during the POR.
    On March 18, 1997, the Department published a notice of preliminary 
results of administrative review and intent not to revoke the order on 
DRAMs of one megabyte or above from the Republic of Korea (62 FR 
12794). Case and rebuttal briefs were submitted on April 18, 1997, and 
April 29, 1997, respectively, by the petitioner, both respondents and 
the following interested parties: (1) Compaq Computer Corporation 
(Compaq); (2) Digital Equipment Corporation (Digital), and (3) Dell 
Computer Corporation (Dell). At the request of LGS and Hyundai, a 
public hearing was held on May 5, 1997. The Department has now 
completed its administrative review in accordance with section 751 of 
the Act.

Scope of the Review

    Imports covered by the review are shipments of DRAMs of one 
megabyte and above from the Republic of Korea (Korea). Included in the 
scope are assembled and unassembled DRAMs of one megabyte and above. 
Assembled DRAMs include all package types. Unassembled DRAMs include 
processed wafers, uncut die and cut die. Processed wafers produced in 
Korea, but packaged, or assembled into memory modules in a third 
country, are included in the scope; wafers produced in a third country 
and assembled or packaged in Korea are not included in the scope.
    The scope of this review includes memory modules. A memory module 
is a collection of DRAMs, the sole function of which is memory. Modules 
include single in-line processing modules (SIPs), single in-line memory 
modules (SIMMs), or other collections of DRAMs, whether unmounted or 
mounted on a circuit board. Modules that contain other parts that are 
needed to support the function of memory are covered. Only those 
modules which contain additional items which alter the function of the 
module to something other than memory, such as video graphics adapter 
(VGA) boards and cards, are not included in the scope.
    The scope of this review also includes video random access memory 
semiconductors (VRAMs), as well as any future packaging and assembling 
of DRAMs.
    The scope of this review also includes removable memory modules 
placed on motherboards, with or without a central processing unit 
(CPU), unless the importer of motherboards certifies with the Customs 
Service that neither it, nor a party related to it or under contract to 
it, will remove the modules from the motherboards after importation. 
The scope of this review does not include DRAMs or memory modules that 
are reimported for repair or replacement.
    The DRAMs subject to this review are classifiable under subheadings 
8542.11.0001, 8542.11.0024, 8542.11.0026, and 8542.11.0034 of the 
Harmonized Tariff Schedule of the United States (HTSUS). Also included 
in the scope are those removable Korean DRAMs contained on or within 
products classifiable under subheadings 8471.91.0000 and 8473.30.4000 
of the

[[Page 39810]]

HTSUS. Although the HTSUS subheadings are provided for convenience and 
customs purposes, the written description of the scope of this review 
remains dispositive.

Intent Not To Revoke in Part

    Section 751(d)(1) of the Act provides that the Department ``may 
revoke'' an antidumping order, in whole or in part, after conducting an 
appropriate review. 19 U.S.C. 1675(d)(1) (1995). The Department's 
regulations elaborate upon this standard. Section 353.25(a)(2) provides 
that the Department may revoke an order, in part, if the Secretary 
concludes: (1) ``One or more producers or resellers covered by the 
order have sold the merchandise at not less than foreign market value 
for a period of at least three consecutive years;'' (2) ``it is not 
likely that those persons will in the future sell the merchandise at 
less than foreign market value;'' and (3) * * * ``the producers or 
resellers agree in writing to their immediate reinstatement in the 
order as long as any producer or reseller is subject to the order, if 
the Secretary concludes under section 353.22(f) that the producer or 
reseller, subsequent to the revocation, sold the merchandise at less 
than foreign market value.''
    As noted above, this administrative review is being conducted 
pursuant to the Tariff Act, as amended by the URAA. The URAA revised 
certain terminology in the Act, including substituting the term 
``normal value'' for ``foreign market value'' and ``exporter'' for 
``reseller.'' However, because this review was initiated prior to the 
date the revised regulations became final, the 1996 regulations are 
still applicable. These regulations use the previous terminology. We 
note that the new regulations do not alter the substantive requirements 
for revocation. See Antidumping Duties; Countervailing Duties; Final 
Rule, 62 FR 27296, 27399 (May 19, 1997) (section 351.222(b)(2)).
    In this case, the first and third criteria for revocation have been 
met. The Department found that LGS and Hyundai did not sell at less 
than foreign market value in the first and second reviews under this 
order. Also, in this administrative review, the respondents were found 
not to have made sales at less than normal value. Further, both 
respondents have certified to their immediate reinstatement in the 
order pursuant to the third criterion noted above. Accordingly, the key 
question is whether the Department is satisfied that it is ``not 
likely'' the respondents will sell at prices below normal value in the 
future.
    In evaluating the ``not likely'' issue in numerous cases, Commerce 
has considered three years of no dumping margins, plus a respondent's 
certification that it will not dump in the future, plus its agreeing to 
immediate reinstatement in the order all to be indicative of expected 
future behavior. In such instances, this was the only information 
contained in the record regarding the likelihood issue. See, e.g., 
Fresh Cut Flowers from Mexico, 61 FR 63822, 63825 (December 2, 1996); 
Polyethylene Terephthalate Film from Korea, 61 FR 58374, 58376 
(November 14, 1996); Tapered Roller Bearings and Parts Thereof from 
Japan, 61 FR 57629, 57651 (November 7, 1996).
    In other cases, when additional evidence is on the record 
concerning the likelihood of future dumping, Commerce is, of course, 
obligated to consider that evidence. In this regard, in evaluating such 
record evidence to determine whether future dumping is not likely, the 
Department has a longstanding practice of examining all relevant 
economic factors and other information on the record in a particular 
case. In particular, depending upon the facts of a case, we consider 
such ``factors as conditions and trends in the domestic and home market 
industries, currency movements, and the ability of the foreign entity 
to compete in the U.S. marketplace without [sales at less than normal 
value].'' Brass Sheet and Strip from Germany, 61 FR 49727, 49730 
(September 23, 1996) (Brass Sheet and Strip); accord Frozen 
Concentrated Orange Juice from Brazil, 56 FR 52510, 52511 (October 21, 
1991) (FCOJ); and Titanium Sponge from Japan, 53 FR 26099, 26100 (July 
11, 1988) (Titanium Sponge).
    In summary, the Department engages in an impartial, balanced 
analysis of all of the information on the record. Pursuant to the 
Department's regulations, the Department cannot revoke this order 
unless it concludes that it is not likely that the respondents will 
dump in the future. As we fully explain below, the Department is not 
satisfied, based on the evidence on the record, that the not likely 
standard has been made.
    Prior to issuing the preliminary results in this administrative 
review, the Department, at the request of the parties, established a 
procedure for the submission of factual information regarding 
revocation. The petitioner and both respondents made several 
submissions of information relevant to whether future dumping is not 
likely, including various in-depth economic analyses. Accordingly, at 
the time of its preliminary results, the Department had an extensive 
factual record before it.
    Based on an analysis of that record, the Department preliminarily 
determined that the likelihood criterion for revocation had not been 
met. Therefore, on March 18, 1997, the Department published a notice of 
intent not to revoke the order concerning DRAMs from Korea (62 FR 
12794) with respect to LGS and Hyundai. Thereafter, the Department 
received a number of comments on the Department's preliminary results 
from the petitioner, LGS, Hyundai, Compaq, Digital and Dell in the case 
and rebuttal briefs. The case and/or rebuttal briefs of the petitioner, 
LGS, Hyundai and Compaq contained additional factual information, which 
the Department had previously requested. The data presented in these 
briefs was therefore taken into consideration in the Department's final 
analysis, as well as publicly available data regarding current market 
conditions.
    The DRAM industry is highly cyclical in nature with periods of 
sharp upturn and downturn in market prices. In the past, the DRAM 
industry has been characterized by dumping during periods of 
significant downturn. For instance, various foreign producers were 
found to have dumped during the downturn in the mid-1980s (see Dynamic 
Random Access Memory Devices from Japan, 51 FR 15943 (April 29, 1986)), 
and the Korean respondents in this proceeding were found to have dumped 
in the less than fair value investigation during 1991-1992, the last 
period when there was a significant downturn in the DRAM industry. 
Because DRAMs are a commodity product, DRAM producers/resellers must 
price aggressively during a downturn period in order to stay 
competitive and maintain their customer base. This is especially true 
during the lowest point in the downturn. Therefore, it is reasonable to 
conclude that information regarding the selling activities and pricing 
practices of respondents, as well as other market conditions, during 
periods of significant downturn are relevant to whether dumping is not 
likely to occur in the future. Thus, as discussed further in comment 3, 
below, we found the January through December 1996 time period to be 
particularly relevant to the ``not likely'' issue because it 
corresponded with a significant ``downturn'' in the DRAM industry.
    In its April 18, 1997, case brief, Compaq proposed that the 
respondents participate in a DRAM data collection program. In its 
proposal, Compaq presumed that the antidumping order

[[Page 39811]]

would be revoked, and that under such a program, respondents would 
agree to maintain cost and pricing data which the respondents would 
submit to the Department should an antidumping petition be filed in the 
future. On June 17, 1997, the Government of Korea submitted a similar 
proposal. On the same date, the respondents stated their willingness to 
participate in such a program, and argued that this proposal should be 
taken into consideration in the Department's likelihood determination 
in this proceeding. The petitioner submitted its opposition to any such 
data collection program on June 14, 1997, and July 3, 1997.
    Other than Compaq's April 18, 1997, submission, all submissions 
regarding the proposed data collection program were received late in 
the proceeding, after the deadline for submitting new information. We 
note further that the proposal itself is precatory in nature. No such 
data collection program is currently in place. Therefore, while we have 
considered this proposed data collection program, we find that this 
program has no bearing on the likelihood issue.
    As discussed further in comment 4, below, based on our analysis of 
the DRAM industry generally and, in particular, during the 1996 time 
frame, we find that the likelihood standard has not been met. 
Therefore, we have not revoked the antidumping duty order on DRAMs from 
Korea with respect to LGS and Hyundai.

Analysis of Comments Received

    We invited interested parties to comment on the preliminary results 
of this administrative review. As noted above, we received timely 
comments from the petitioner, LGS, Hyundai, Compaq, Digital and Dell.

I. Revocation Comments

    Comment 1: Whether the Department Erred when it Issued a 
Preliminary Intent Not to Revoke the Order In Part.
    Hyundai and Compaq argue that the Department's failure to publish a 
notice of ``Intent to Revoke Order (In Part)'' with its preliminary 
results is contrary to case precedent. Both parties contend that, 
barring extremely unusual circumstances not present in this proceeding, 
it is the Department's practice to revoke orders whenever a respondent 
has established three consecutive years of no dumping and has furnished 
a written statement agreeing to the immediate reinstatement of the 
order in the event the Secretary concludes that the respondent sells at 
less than normal value in the future. Hyundai and Compaq cite numerous 
cases where the Department has granted revocation, including Steel Wire 
Rope from the Republic of Korea, 62 FR 17171 (April 9, 1997) (Steel 
Wire Rope); Certain Forged Steel Crankshafts from the United Kingdom, 
62 FR 16768, 16771 (April 8, 1997) (Crankshafts); and Fresh Cut Flowers 
from Mexico, 61 FR 63825 (December 2, 1996).
    Hyundai further claims that the Department's failure to issue a 
preliminary intent to revoke the order, in part, despite three 
consecutive years of de minimis margins, is in conflict with the intent 
of Article 11 of the WTO Antidumping Agreement, which states that an 
antidumping duty order ``shall remain in force only as long and to the 
extent necessary to counteract the dumping which is causing injury,'' 
and that an order must be terminated ``immediately'' if the authorities 
determine that the order is no longer warranted.
    Finally, Hyundai argues that the Department's reliance on Brass 
Sheet and Strip as case precedent for its preliminary finding regarding 
the ``not likely'' issue was misplaced. Specifically, Hyundai asserts 
that the facts in Brass Sheet and Strip differ from the facts in this 
proceeding in the following ways: (1) In contrast to Brass Sheet and 
Strip where the respondent's exports had fallen to commercially 
insignificant levels, Hyundai's shipments of DRAMs have increased 
substantially since the order was put in place; (2) unlike the 
respondent in Brass Sheet and Strip, the ability of the Korean 
respondents to sell at fair value in the United States has not been 
impaired by a strengthening currency; (3) in contrast to Brass Sheet 
and Strip where the respondent was planning to use the imported product 
as an input for a plant located in the United States (making increased 
imports of the subject merchandise in the future almost certain), 
Hyundai will not use the subject merchandise as an input product; and 
(4) in contrast to Brass Sheet and Strip where the worldwide demand for 
the product was declining, the worldwide demand for DRAMs is strong and 
is predicted to increase in the future.
    The petitioner argues that the Department's preliminary 
determination not to revoke was correct and in accordance with the law. 
The petitioner claims that section 353.25(a)(2) of the Department's 
regulations specify that before an antidumping duty order can be 
revoked, the Department must be satisfied that future dumping by the 
respondents is not likely. Therefore, the petitioner contends that 
although three consecutive years of de minimis margins and the 
respondents' certification regarding the immediate reinstatement of the 
order if dumping resumes are requirements for revocation, these factors 
alone are not a sufficient basis for revocation. The petitioner claims 
that because the Department's preliminary results found no basis to 
conclude that it is not likely that the Korean respondents will resume 
dumping in the future, the Department had a ``reasonable basis'' to 
believe that the requirements for revocation had not been met. 
Therefore, the petitioner asserts that the order continues to be 
warranted in order to counteract injurious dumping. Accordingly, the 
petitioner contends that the Department's preliminary decision not to 
revoke the order in part was in compliance with the law and the 
international obligations of the United States under Article 11 of the 
WTO Antidumping Agreement.
    The petitioner further argues that although the cases differ with 
regard to certain facts, the Department's reliance on Brass Sheet and 
Strip was not misplaced. The petitioner contends that the factors 
identified by Hyundai do not diminish the relevance of Brass Sheet and 
Strip as important case precedent on the issue of revocation. In 
particular, the petitioner contends that factual similarities between 
this proceeding and Brass Sheet and Strip, such as the relationship 
between global oversupply and declining prices and the relative size of 
the U.S. market, are more probative than the differences cited by 
Hyundai.

DOC Position

    We disagree with respondents' interpretation both of the proper 
revocation standard and the Department's previous determinations. 
Regarding the proper revocation standard, 19 C.F.R. 353.25(a)(2) 
requires not only a showing of three years of no dumping and a 
respondent's certification and agreement to immediate reinstatement in 
the order, but also a determination that future dumping is not likely. 
This ``second requirement for revocation, that the respondent is not 
likely to resume dumping, necessarily involves an exercise of 
discretion and judgment.'' Tatung Co. v. United States, 18 CIT 1137, 
1144 (1994). In certain cases, the record may only contain evidence 
regarding the parties' history of no dumping, which ``[o]rdinarily * * 
* would constitute substantial evidence of expected future behavior.'' 
Id.; see also Antifriction Bearings (Other Than Tapered Roller 
Bearings) and Parts Thereof From Italy, 60 FR 10950, 10967

[[Page 39812]]

(Feb. 28, 1995). In other cases, respondents are able to produce 
additional evidence demonstrating that future dumping is not likely. 
See Steel Wire Rope From Korea, 62 FR at 17174; FCOJ From Brazil, 56 FR 
at 52510.
    In still other cases, the Department has not been satisfied, based 
on the record before it, that future dumping is not likely. Contrary to 
respondents' argument, these cases do not necessarily only involve 
``extremely unusual circumstances.'' The Department reaches its 
revocation determinations on a case-by-case basis, depending upon the 
industry in question, the relevant market conditions and the evidence 
submitted on the record. See, e.g., Brass Sheet and Strip from Germany, 
61 FR at 49730; Certain Circular Welded Carbon Steel Pipes and Tubes 
From Taiwan, 56 FR 8741, 8742 (March 1, 1991). The Court of 
International Trade (``CIT'') has upheld several determinations by the 
Department denying revocation. See Sanyo Elec. Co. v. United States, 15 
CIT 609 (1991); Toshiba Corp. v. United States, 15 CIT 597 (1991). 
While the Court distinguished cases granting revocation based upon the 
absence of evidence regarding the likelihood of future dumping, in 
neither case did the Court indicate that revocation should be the rule 
and denying revocation the exception. See Toshiba at 601. Like the 
Department, the Court properly focused instead upon the facts at issue 
and the ``predictive nature of the revocation proceeding.'' Id. at 603; 
see also Matsushita Elec. Indus. Co. v. United States, 750 F.2d 927, 
933 (Fed. Cir. 1984). In the end, the Court concluded that because 
respondents requested revocation ``it was for [respondents] to come 
forward with `real evidence' to persuade Commerce to revoke the 
order.'' Toshiba at 603 (citation omitted).
    We also disagree with Hyundai's assertion that the Department erred 
by relying on Brass Sheet and Strip as support for its preliminary 
determination not to revoke. The Department did not rely upon Brass 
Sheet and Strip as support for each of the elements addressed in the 
Department's preliminary determination regarding the ``not likely'' 
issue. Rather, the Department relied upon Brass Sheet and Strip 
primarily to confirm the legal standard for the type of factors the 
Department has considered relevant in the past (e.g., conditions and 
trends in the industry, currency movements and the ability of the 
foreign entity to compete in the U.S. without dumping).
    Finally, we disagree with Hyundai's interpretation of the 
revocation standard under the Antidumping Agreement. We note at the 
outset that all parties agree that the revocation standard, as set 
forth in the Department's regulations, does not violate the Antidumping 
Agreement. See e.g., LGS Case Brief at 15 (April 18, 1997). The sole 
issue involves how this standard is applied to the facts and 
circumstances of this case. The Department believes that its likelihood 
determination, given the facts of this case, is entirely consistent 
with Article 11.2 of the Antidumping Agreement, which establishes a 
broad based standard under which revocation is warranted if the 
authorities determine that the order ``is no longer warranted.''
    Comment 2: Whether the Department Applied a Proper and Fair 
Revocation Standard in its Preliminary Results.
    LGS, Hyundai, Compaq and Dell argue that in its preliminary results 
the Department improperly used the phrase ``no likelihood'' in lieu of 
``not likely'' in determining whether the requirements for revocation 
under section 353.25(a)(2) of the Department's regulations had been 
met. These parties contend that the Department's use of a ``no 
likelihood'' standard was unlawful under the Antidumping Agreement 
because it altered the meaning of the regulation and created a 
revocation standard which is virtually impossible for respondents to 
attain. Specifically, LGS, Hyundai, Compaq and Dell contend that the 
phrase ``not likely'' connotes only a lack of probability but the 
phrase ``no likelihood'' creates a much higher standard which implies 
that the respondents must demonstrate that there is almost zero 
probability of dumping in the future. LGS further claims that ``not 
likely'' means a probability of 51 percent or greater while ``no 
likelihood'' means a probability of 99 percent or greater that the 
respondent will not dump in the future.
    Hyundai and LGS further contend that the Department's use of the 
``no likelihood'' standard is particularly insupportable given that the 
Department amended its regulations in 1989 to specifically change the 
phrase ``no likelihood'' to ``not likely.'' Hyundai asserts that this 
change was made to clarify the regulation to avoid imposing an 
impossible burden on respondents seeking revocation. Accordingly, LGS 
and Hyundai argue that in its final results the Department should 
follow the ``not likely'' standard outlined in its current regulations, 
not the ``no likelihood'' standard abolished a decade ago.
    In addition, LGS argues that the Department's preliminary finding 
that LGS ``may have dumped in the post 1996 period'' is irrelevant to 
the ``not likely'' test. LGS asserts that the relevant question is not 
whether LGS ``may'' have dumped but whether the company is ``not 
likely'' to dump. LGS cites Crankshafts to argue that the Department's 
reliance on something that ``may'' happen is tantamount to sheer 
speculation, a standard prohibited by the Department's regulations and 
explicitly rejected by the Department in practice.
    The petitioner counters stating that the Department properly 
applied the long-standing and judicially recognized ``no likelihood'' 
standard. Specifically, the petitioner contends that the Department's 
long-standing administrative practice has been to use the terms ``not 
likely'' and ``no likelihood'' interchangeably. The petitioner cites 
Brass Sheet and Strip, Elemental Sulphur from Canada, 56 FR 5391 
(February 11, 1991) (Sulphur) and FCOJ from Brazil, 56 FR 52510, in 
support of its argument. In addition, the petitioner claims that 
because the Department has used the terms ``no likelihood'' and ``not 
likely'' interchangeably in the past, the regulatory change in 1989 was 
simply to clarify the revocation standard, not change it. In support of 
this contention the petitioner cites the CIT's decision in Toshiba in 
which the Court found that the ``no likelihood test'' does not impose 
an unattainable standard.

DOC Position

    The Department has applied the proper revocation standard, 
consistent with our longstanding practice, throughout the proceeding. 
Despite the potential difference in meaning between the phrases ``not 
likely'' and ``no likelihood'' as used in the revocation provisions of 
the 1988 regulations and the regulations applicable to this proceeding, 
the Department has consistently applied the same likelihood standard 
under both sets of regulations. As our practice shows, and as we 
explain below, the Department has never applied the likelihood standard 
to require the degree of certainty that dumping will not recur that the 
respondents claim the phrase ``no likelihood'' implies.
    Prior to 1989, the applicable regulation expressly conditioned 
revocation upon a finding of ``no likelihood'' of future dumping. See 
19 CFR 353.54(a) (1988). When the Department first proposed the 
amendment to the regulation in 1986, the Department offered no 
explanation for substituting ``not likely'' for ``no likelihood,'' 
stating only that revocation ``is premised on the Secretary's finding 
that it is not likely that the person or

[[Page 39813]]

persons will in the future sell the merchandise at less than foreign 
market value.'' 51 FR 29046, 29052 (1986) (Preamble to Proposed 
Regulations) (emphasis added). The one comment received regarding this 
regulatory provision argued only that the Department should not 
consider the issue of future dumping at all. Id. Antidumping Duties; 
Final Rule, 54 FR 12742, 12758 (March 28, 1989) (Preamble) (emphasis 
added). The Department disagreed, retained the proposed amendment 
without revision, and responded to the comment as follows:

    The statute gives the Secretary broad discretion in deciding 
when to revoke an order. The Secretary has determined that a pre-
condition to revocation under this paragraph is that the Secretary 
be satisfied that there is no likelihood of future sales at less 
than foreign market value.

Hence, even in the preamble to the regulation, which substituted ``not 
likely'' for ``no likelihood,'' the Department continued to describe 
the standard using the phrase ``no likelihood.'' Similarly, the 
Department substituted ``not likely'' for ``no likelihood'' when it 
amended the countervailing duty regulations in 1988. Compare 19 CFR 
355.42(a) (1988) with 19 CFR 355.25(a) (1996). Again, the Department 
gave no explanation.
    Thus, in amending the revocation regulation, the Department used 
the phrases ``not likely'' and ``no likelihood'' interchangeably, and 
consistently failed to draw a legal distinction between the two. The 
Department has also used the two phrases interchangeably in its 
administrative practice. See Silicon Metal From Brazil, 62 FR 1954, 
1957 (Jan. 14, 1997) (Silicon Metal); Fresh Cut Flowers From Colombia, 
61 FR 42833, 42838 (Aug. 19, 1996). In many determinations since 
amending the regulation in 1989, the Department has described the 
future dumping standard in terms of ``no likelihood'' just as it did in 
this proceeding. See, e.g., Brass Sheet and Strip, 61 FR at 49730; 
FCOJ, 56 FR at 52511.
    Moreover, contrary to the assertions of LGS and Hyundai, the 
Department has never interpreted ``no likelihood,'' in practice, to 
mean a zero probability of dumping, either before the regulations were 
amended in 1989 or after. The very fact that the Department has revoked 
numerous orders, in whole or in part, before and after the 1989 
amendments, confirms this conclusion. Never once has the Department 
indicated that it was 100 percent certain there was ``no likelihood'' 
of future dumping in any of these cases. As stated by the CIT in 
Toshiba, ``rarely, if ever, will Commerce be able to predict with 
certainty what will occur upon revocation.'' 15 CIT at 599 (citing 
Matsushita, 750 F. 2d at 933). Hence, it is clear that the standard is 
not an impossibly high one, as the respondents suggest.
    Contrary to the assertions of LGS, evidence indicating that a 
respondent ``may have dumped'' in the period following the third 
administrative review is relevant to the Department's ``not likely'' 
test. As the Department's practice and the decisions of the courts make 
clear, the determination regarding the likelihood issue is ``inherently 
predictive'' in nature. See, e.g., Matsushita, 750 F.2d at 933. The 
Department ordinarily does not have actual sales and cost data to 
examine. Therefore, in assessing the likelihood of future dumping, as 
discussed in more detail in comment 3, below, the Department examines 
all available record evidence.
    Likewise, we are not persuaded by LGS' contention that the ``not 
likely'' standard implies that revocation is appropriate if the 
Department finds at least a 51 percent chance that the respondent will 
not dump in the future. The Department's regulations and administrative 
practice properly do not establish a specific, quantifiable standard 
for determining whether revocation is appropriate. As noted above, in 
most cases, the presence of three years of no dumping margins and a 
respondent's certification and agreement to immediate reinstatement in 
the order are indicative that future dumping is not likely because, in 
most cases, this is the only record evidence regarding likelihood. Here 
the facts of record, reasonably interpreted, lead us to a contrary 
conclusion.
    Based on the foregoing, we therefore find that when the Department 
amended the revocation regulation in 1989 to change the phrase ``no 
likelihood'' to ``not likely,'' the purpose of the regulatory change 
was simply to clarify the revocation standard, not amend it. Therefore, 
the Department has applied the proper revocation standard throughout 
this proceeding.
    Comment 3: What Time Frame Should be Considered When Determining 
Whether Future Dumping is Not Likely.
    LGS and Hyundai argue that the Department improperly focused on the 
period immediately following the third administrative review in 
conducting its preliminary ``not likely'' analysis. LGS and Hyundai 
assert that section 353.25(a)(2)(ii) of the Department's regulations 
instruct the Department to examine whether it is not likely that a 
respondent will in the future sell the merchandise at less than normal 
value. LGS and Hyundai interpret this reference to a period ``in the 
future'' as being a time period after revocation of the order. 
Therefore, LGS and Hyundai assert that in the final results the 
Department should conduct its ``not likely'' analysis for the time 
period beginning the day after the Department issues a revocation 
determination (i.e., beginning in second quarter 1997).
    In addition, LGS and Hyundai argue that because the DRAM industry 
is highly cyclical, the Department must take into account a 
respondent's behavior over the long term (i.e., during both market 
upturns and downturns). In addition, the respondents contend that the 
Department's preliminary conclusion that DRAM producers ``dump during 
periods of significant downturn'' is flawed. If this were true, 
respondents argue, antidumping duty orders could never be revoked in 
cases involving cyclical industries.
    Hyundai further argues that by implying that respondents must prove 
they were not dumping after the end of the third administrative review, 
the petitioner is essentially seeking to restore the old ``gap period'' 
reviews which the Department conducted under the former regulations 
during the 1980's. As Hyundai explains, under the Department's old 
regulations, a respondent could qualify for revocation on the basis of 
two years of zero or de minimis margins if the respondent was also 
found not to have dumped during a period of at least nine months after 
the completion of the second administrative review. Hyundai claims that 
upon amending the regulations in 1988, the Department eliminated the 
need for ``gap period'' reviews, stating instead that revocation would 
become effective the day after the three-year period.
    The petitioner asserts that in conducting its preliminary ``not 
likely'' analysis the Department properly examined the period 
immediately following the end of the third review period. The 
petitioner claims that the period immediately following the close of 
the third review period must be examined because any evidence 
indicating that dumping was likely to have occurred anytime after this 
period demonstrates the continued need for the protection afforded by 
the antidumping duty order. The petitioner cites Silicon Metal and 
Brass Sheet and Strip as recent cases where the Department examined the 
period immediately following the third POR to determine whether the 
requirements for revocation had been met.

[[Page 39814]]

DOC Position

    We disagree with Hyundai and LGS. While 19 CFR 353.25(a)(2)(ii) 
requires the Department to assess whether the evidence supports a 
conclusion that it is not likely the respondents will dump ``in the 
future,'' respondents are incorrect to interpret this provision as 
requiring the Department to consider only a time period beginning after 
the date the Department would issue a revocation determination. Rather, 
this provision requires the Department to examine all of the evidence 
available on the record. There is nothing in the Act, the Department's 
regulations or case precedent that defines the relevant time period in 
considering the likelihood issue. Common sense, however, dictates that 
the Department should, as always, base its determination on all record 
evidence.
    In this revocation proceeding the Department considered all 
publicly available data and information placed on the record by all 
parties (including data regarding the January 1997 through April 1997 
time period, which respondents characterize as a market upturn). We 
agree that a respondent's past conduct is relevant, including a showing 
of three years of de minimis margins. Market trends and forecasts 
beyond the possible revocation date may also be relevant. In this case 
we find the January through December 1996 period to be particularly 
probative because it corresponded with a significant downturn in the 
DRAM industry. The DRAM industry is highly cyclical, market prices for 
DRAMs are generally lower during periods of downturn and there is a 
history of dumping in the DRAM industry during such periods. It is 
therefore reasonable to conclude that an examination of the selling 
activities and pricing practices of respondents during such downturn 
periods will provide the Department with a reasonable indication as to 
whether dumping is not likely to occur in the future. Further, the 1996 
period is not only the most recent downturn, but one which occurred 
since the order has been in place.
    As discussed further in comment 4, below, based on our analysis of 
the DRAM industry during the 1996 downturn and other factors, we find 
that the likelihood standard for revocation set forth in section 
353.25(a)(2) of the regulations has not been met. Although we agree 
with the respondents that market conditions in the DRAM industry have 
recovered somewhat in 1997 (though not to the extent that respondents 
argue), neither this fact nor any other evidence regarding future 
conditions in the DRAM industry contradicts or significantly detracts 
from other record evidence indicating that dumping may have taken place 
during the 1996 downturn. Such evidence suggests that the not likely 
criterion for revocation has not been satisfied in this case.
    For much the same reasons, we disagree with Hyundai that the 
Department's approach effectively reinstates the ``gap period'' reviews 
disavowed when the regulations were amended in 1989. See Preamble to 
1989 Regulations, 54 FR at 12758 (discussing ``gap period'' reviews). 
At that time, the regulations required only two years of no dumping 
before the Department would consider revocation. Pursuant to the so-
called ``gap period'' reviews, however, the Department would not revoke 
the order until after determining that no dumping had occurred during 
the gap period. This required that the Department conduct an additional 
administrative review of the respondent's data, involving at least nine 
months. As discussed above, in evaluating whether future dumping is not 
likely, the Department may find that the market conditions and trends 
during a certain period or periods are probative. In this case we found 
the January through December 1996 time frame to be particularly 
important to our consideration of the ``not likely'' issue because it 
corresponded with a significant downturn in the DRAM industry. We 
consider it merely coincidental that this time frame coincided with the 
end of the third administrative review and the period immediately 
following. Had the most recent downturn occurred during a different 
time frame, it may have been appropriate to take that period into 
account in our analysis.
    Comment 4: Whether Record Evidence Indicates that Future Dumping by 
the Korean Respondents is Not Likely.
    The petitioner argues that in its preliminary results, the 
Department drew upon an extensive record, including submissions on 
market conditions, pricing trends, econometric analyses, newspaper 
articles and market studies and properly concluded, based on the 
totality of data, that there was no basis on which to conclude that 
future dumping by the Korean respondents was not likely.
    LGS and Hyundai argue that the Department's preliminary conclusion 
regarding the ``not likely'' issue was contrary to law and based on 
incorrect and outdated data that do not reflect current market 
conditions. LGS and Hyundai contend that when current market conditions 
are viewed, the record indicates that future dumping is not likely. 
Hyundai submits that in order to make a reasonable prediction of the 
future, the Department's final decision must be based on the most 
recent information available. LGS adds that the Court of Appeals for 
the Federal Circuit has found it be ``reversible error'' for the 
Department, in a revocation proceeding, to fail to obtain and consider 
the most up-to-date information available. See Freeport Minerals Co. v. 
United States, 776 F.2d 1029, 1032 (Fed. Cir. 1985).
    In addition to the general comments concerning the Department's 
preliminary revocation determination noted above, the petitioner and 
respondents make a number of arguments regarding the specific data 
relied upon by the Department in its preliminary ``not likely'' 
analysis. These arguments are summarized according to topic, below.
A. Pricing Trends in the DRAM Industry
    The petitioner argues that during 1996 the DRAM market was in a 
downturn, with steep worldwide price declines. Citing to data obtained 
from publicly available reports, the petitioner claims that these price 
declines are forecasted to continue throughout 1997.
    LGS, Hyundai, Compaq, Digital and Dell argue that the worldwide 
price decline noted in the Department's preliminary results has ended 
and that current market information indicates that DRAM prices have 
rebounded significantly in 1997. LGS, Hyundai and Dell further contend 
that the recent trend towards an equilibrium between supply and demand 
in the DRAM industry indicates that higher prices are likely in the 
future. In support of these arguments, LGS, Hyundai, Compaq, Digital 
and Dell reference actual prices paid in the U.S. market for DRAMs, 
public statements made by the company officials at Micron, average U.S. 
prices reported by Dataquest and the American IC Exchange, studies by 
independent analysts and numerous newspaper and magazine articles. LGS 
further asserts that because costs in the DRAM industry are constantly 
declining, in the event that market prices were stable, rather than 
rising, the likelihood that a respondent would have to sell below cost 
in order to remain competitive in the U.S. market decreases over time.
    The petitioner rebuts the arguments of LGS, Hyundai, Compaq, 
Digital and Dell. The petitioner argues that the DRAM market is still 
volatile and that price declines will continue throughout 1997. The 
petitioner cites recent price

[[Page 39815]]

reports, newspaper and magazine articles and market reports which 
suggest that the temporary rebound in DRAM pricing will soon be over 
and that prices thereafter will continue to decline throughout 1997. 
Finally, the petitioner attempts to demonstrate that the DRAM market is 
still volatile and difficult to predict by pointing out that just 48 
hours after the date the respondents cited recent price increases in 
their case briefs, the worldwide market prices for DRAMs fell more than 
10 percent.
B. Inventory Levels
    The petitioner argues that, despite the 1996 ``glut in the global 
DRAM market,'' publicly available data indicate that Korean producers 
have continued to increase production by bringing new facilities on-
line. The petitioner claims that this additional increase in DRAM 
production will add to the oversupply problem being experienced in the 
marketplace and will keep DRAM prices depressed throughout 1997. In 
support of this argument, the petitioner cites public studies by 
independent analysts and numerous newspaper and magazine articles. In 
addition, the petitioner cites Brass Sheet and Strip as a recent case 
where the Department was unable to conclude that future dumping was not 
likely, based, in part, on competitive conditions in an industry 
characterized by oversupply.
    LGS, Hyundai and Compaq argue that in its preliminary results the 
Department incorrectly concluded that there is no evidence that the 
announced DRAM production cutbacks ``have occurred.'' Specifically, 
LGS, Hyundai and Compaq argue that numerous industry reports confirm 
that the Korean producers have trimmed production and will continue to 
reduce their operations in 1997 in order to bring supply and demand 
into balance. In support of this argument LGS and Hyundai cite publicly 
available reports and newspaper and magazine articles. The respondents 
contend that these documents suggest that recent cutbacks in production 
by Korean DRAM producers have led to market price increases. LGS 
further argues that the Department's conclusion that ``there is a 
significant DRAM oversupply'' and that ``the existing DRAM oversupply 
is likely to cause prices to remain low or fall lower in the future'' 
was based on data which are now outdated. LGS, Hyundai, Compaq and Dell 
claim that the oversupply conditions present in the DRAM industry in 
1996 have disappeared and that the recent cutback in production by the 
Korean producers, in conjunction with an exploding global demand, has 
resulted in a market equilibrium between supply and demand.
    Finally, as noted in comment 1 above, LGS contends that reliance on 
Brass Sheet and Strip as case precedent is misplaced. LGS asserts that 
unlike Brass Sheet and Strip, where the Department found that there had 
been a decrease in demand in the European market and that the U.S. 
market continued to be desirable for exporters, the DRAM demand is 
booming worldwide. In addition, LGS and Hyundai contend that as a 
result of the shrinking global supply of DRAMs many producers, 
including the petitioner, are beginning to return to profitability.
    The petitioner rebuts the arguments of LGS, Hyundai and Compaq. 
According to the petitioner, Korean DRAM producers have not made 
production cutbacks, but instead have shifted production increases to 
64M DRAMs while continuing to produce other DRAM configurations at 
prior levels and withholding them temporarily from the market. The 
petitioner cites brokerage house, press and other recent market reports 
as support for its argument. The petitioner claims that these articles 
suggest that Korean DRAM producers will stockpile DRAMs long enough to 
lift prices, but that the eventual release of this inventory into the 
marketplace will result in continued price declines.
C. The Petitioner's Allegation That LGS and Hyundai Were Dumping in 
1996
    The petitioner argues that the sales and cost data submitted by 
Hyundai and LGS in the third administrative review, when viewed in 
conjunction with publicly available information regarding pricing 
trends since the end of the third review period, demonstrate that LGS 
and Hyundai made sales at less than normal value during the second half 
of 1996 (i.e., the period immediately following the third review 
period). Specifically, the petitioner contends that the home market 
sales and cost data submitted by Hyundai and LGS in the present 
administrative review demonstrate that the two respondents made sales 
at prices which were below COP during the two months immediately 
following the end of the third review period (i.e., May and June 1996).
    In addition, the petitioner asserts that when the reported costs of 
LGS and Hyundai are extrapolated through to the end of the fourth 
quarter 1996 using the same rate of decline actually experienced by the 
producers in 1995, and then compared to publicly available, average 
U.S. DRAM price data (compiled by Dataquest and Lehman Brothers), there 
is evidence that LGS and Hyundai made U.S. sales at prices below COP 
during the third and fourth quarters of 1996 as well. Based on the 
foregoing, the petitioner contends that the Korean respondents were 
dumping during the second half of 1996.
    LGS and Hyundai contend that the Department's preliminary 
conclusion that the respondents made U.S. sales during the second half 
of 1996 at prices that appeared to ``be near or below normal value and 
production costs'' was based on incomplete and inaccurate data 
presented by the petitioner. Specifically, regarding the data relied 
upon in the preliminary results, LGS contends the following: (1) 
Verified data demonstrate that LGS' actual contract prices with its 
U.S. customers during 1996 were significantly higher than the average 
U.S. spot prices provided in the petitioner's analysis; (2) the fact 
that LGS may have made certain home market sales at prices below its 
COP does not definitively demonstrate that dumping occurred; and (3) 
the U.S. price quotes referred to in the petitioner's analysis cannot 
be relied upon because neither the underlying data nor source for the 
data were provided by the petitioner.
    LGS further argues that the petitioner's analysis overstates the 
degree to which DRAM prices declined in 1996 because the analysis was 
based on quarterly prices calculated from prices which were averaged on 
a simple, rather than a weighted-average basis. LGS claims that when 
projections based on ``corrected'' price and cost data are used, the 
data demonstrate that LGS continued to sell at prices above both the 
average U.S. spot price and its COP during the second half of 1996. As 
additional support for its claim that it was not dumping during the 
second half of 1996, LGS provided what it claimed were actual price and 
cost data for the post-April 1996 period.
    Hyundai also asserts that there were distortions and inaccuracies 
in the petitioner's data. First, Hyundai contends that the average U.S. 
price calculated by the petitioner was based on spot prices, rather 
than OEM contract prices. Hyundai asserts that verified data on the 
record in the third administrative review indicate that Hyundai's 
actual U.S. prices during the POR were higher than the average U.S. 
prices for the first quarter 1996 presented by the petitioner. 
Therefore, Hyundai claims that there is no correlation between 
Hyundai's actual prices and the average spot prices provided by the 
petitioner. In addition, Hyundai asserts that based on an econometric 
analysis conducted by Dr.

[[Page 39816]]

Kenneth Flamm, the market price for DRAMs is expected to exceed 
Hyundai's COP by substantial margins during 1997 and 1998. Hyundai 
further attacks the petitioner's analysis stating that it mistakenly 
compared the average spot price for all 16M DRAMs with the COP of only 
the 1X16 configuration. Finally, Hyundai argues that the petitioner's 
data failed to take into account the reductions in cost resulting from 
the depreciation of the Korean won. Hyundai asserts that when 
``corrected'' price and cost data are used, the average U.S. price 
remains above Hyundai's COP during the second half of 1996.
    The petitioner responds that the data LGS claimed in its case brief 
were its actual price and cost data actually confirm that LGS was 
dumping during the second half of 1996. The petitioner contends that 
the costs reported by LGS are understated for the following reasons: 
(1) LGS did not include foreign exchange losses on long-term foreign 
debt in its reported COP; and (2) LGS lengthened its reported 
depreciation schedule for the second half of 1996. The petitioner 
claims that this one-time restatement of depreciation expenses caused 
the sharp decline in costs in July 1996 reported by LGS. The petitioner 
cites numerous publicly available reports and articles which state that 
LGS, as well as other Korean DRAM producers, lengthened their 
depreciation schedules during the second half of 1996 to avoid 
reporting substantial losses for fiscal year 1996. The petitioner 
argues that, had LGS not manipulated its costs for the second half of 
1996, its reported (but unverified) U.S. prices would have been below 
its reported COP.
    The petitioner rebuts Hyundai's arguments as well. The petitioner 
argues that the so-called ``corrected'' prices provided by Hyundai do 
not reflect actual prices but are, instead, merely derived prices. The 
petitioner contends that the actual prices paid were usually below the 
average U.S. DRAM prices provided in the petitioner's analysis. In 
addition, the petitioner asserts that its analysis correctly compared 
cost and price data for the 1X16 configuration, not all DRAM models as 
suggested by Hyundai.
D. Whether Korean DRAM Producers Can Remain Competitive in the U.S. 
Market Without Dumping
    The petitioner argues that due to the market conditions noted in 
points B and C above, LGS and Hyundai cannot remain competitive in the 
U.S. market without selling DRAMs at less than normal value.
    LGS responds that, regardless of market circumstances, LGS is 
likely to continue to sell DRAMs in the United States at fair value 
prices. Specifically, LGS contends that in contrast to the respondents 
in Brass Sheet and Strip and Steel Wire Rope, the U.S. market is not 
LGS' principal export market and LGS is not a major supplier to the 
United States. Therefore, LGS argues, it has no incentive to sell in 
the United States unless it can make a reasonable profit. In addition, 
LGS relies upon an economic study by the Law & Economics Consulting 
Group (LECG study) to contend that LGS has no economic incentive to 
dump in the United States for a number of reasons. In addition to the 
argument that its share of the U.S. market is too small to make 
predatory pricing appealing, LGS contends that, because its prices with 
OEM customers are based on contracts, it is able to command higher 
prices from OEM customers during market downturns. In support, LGS 
asserts that actual, verified prices collected by the Department prove 
that LGS' contract prices were higher than the spot market prices 
during 1996. Moreover, the won is currently depreciating against the 
dollar, negating the possibility of exchange rate dumping. LGS cites 
Steel Wire Rope and Flowers as confirming the Department's view that 
``devaluation of the home market currency makes dumping less likely.''
    In addition, LGS argues that the Department incorrectly found that 
``the history of the DRAM industry is one of dumping in periods of 
significant downturn.'' Specifically, LGS asserts that the behavior of 
Japanese DRAM producers in 1986 has no bearing on the pricing behavior 
of unaffiliated Korean producers in 1996. In addition, LGS claims that 
the fact that the Korean producers were found to be dumping in 1991 and 
1992 is not indicative of future dumping. If this were true, LGS 
asserts, no antidumping duty order could ever be revoked since 
revocation findings can only exist once an antidumping duty order has 
been issued.
    Finally, LGS and Hyundai argue that the fact that neither 
respondent has had dumping margins through a variety of market 
conditions (including downturns) over the past three review periods is 
indicative that future dumping during any market condition is not 
likely. See, e.g., Steel Wire Rope (stating that because past 
appreciation of the Korean won did not cause the respondents to dump, 
the Department had no basis to conclude that a possible currency 
appreciation in the future would cause the respondents to change their 
pricing practices); Tatung 18 CIT at 1144 (finding that with regard to 
the likelihood requirement for revocation ``ordinarily past behavior 
would constitute substantial evidence of expected future behavior'').
    The petitioner counters that LGS has the following compelling 
reasons to dump: (1) OEM customers have leverage over the DRAM 
suppliers; therefore, OEM customers will not pay significantly higher 
prices for commodity products such as DRAMs; (2) because of the sheer 
size of the DRAM market in the United States, LGS' market share 
accounts for substantial revenues; and (3) LGS needs an outlet for the 
additional DRAMs it has already committed to producing in 1997. The 
petitioners contend that the United States is the logical outlet for 
these additional DRAMs because Europe has recently ended a two-year 
suspension of a reference price system on Korean DRAMs and Japan is 
currently flooded with Japanese produced DRAMs.
    The petitioner further argues that, unlike in Steel Wire Rope 
(where the Department concluded that there was no evidence of imported 
production inputs) and Flowers (where there were ``virtually no fixed 
costs''), Korean DRAM producers import raw materials that account for a 
large portion of their costs. Therefore, the petitioner asserts that 
the depreciation of the won increases the COP, making dumping more 
likely in the United States.

DOC Position

    We continue to find that the record supports a conclusion that the 
not likely criterion for revocation has not been satisfied. In reaching 
this decision, we have examined all the information on the record, 
including publicly available data regarding current market conditions. 
Based on this analysis, we found the January through December 1996 time 
frame to be particularly relevant because of the significant downturn 
in the DRAM industry during this period.
A. Pricing Trends in the DRAM Industry
    The DRAM market has suffered periodic set-backs over the past 25 
years. During the most recent downturn, industry revenues significantly 
declined. For instance, according to Electronic Buyers News, total 
worldwide market revenue plunged 38% to $25.13 billion in 1996. Both 
Hyundai and LGS reported dramatic decreases in revenues in their 1996 
publicly available financial statements. Therefore, as discussed above, 
we find this time frame to be particularly relevant to the Department's 
``not

[[Page 39817]]

likely'' analysis. Although we agree with the respondents that DRAM 
prices have recovered somewhat during 1997, this does not detract from 
the fact that prices fell significantly during the 1996 downturn. In 
any case, it appears that pricing in the DRAM market has not yet fully 
recovered. Current prices are still lower than in the years preceding 
the 1996 market downturn, years in which the respondents were found not 
to be dumping. Furthermore, prices have, in fact, decreased recently. 
According to Dataquest (``The Semiconductor DQ MONDAY Report'', Issue 
24, June 23, 1997, and Issue 25, June 30, 1997) the spot market price 
for the 1Mx16 EDO DRAM decreased from the $7.45 to $8.09 range on June 
13 to the $6.30 to $6.85 range on June 27. Similarly, the price for the 
higher-density 64M DRAMs continues to fall. In fact, the average price 
for a 64M DRAM is now in the mid $40 range, down from $55 earlier this 
year. In sum, although the DRAM market has stabilized somewhat, prices 
continue to fluctuate and a large degree of uncertainty about the 
direction of the market remains.
B. Inventory Levels
    In regard to inventory levels and the supply of DRAMs, the record 
demonstrates that supply exceeded demand during 1996 and thus far in 
1997. While there were conflicting reports as to whether respondents 
were actually decreasing their DRAM production levels during the 1996 
downturn period, prices fell dramatically during 1996 and have not yet 
fully stabilized. In addition, although the respondents have made 
public announcements regarding DRAM production cut-backs and it appears 
that the market has reacted with higher prices, it is unclear how much 
of an effect this will have on the overall supply of DRAMs. Similarly, 
it is uncertain how long it will be before production returns to 
previous levels in anticipation of increased demand in the marketplace. 
According to Electronic Buyer's News (January 27, 1997, Issue 1042), an 
upturn in demand in October, 1996, triggered a simultaneous increase in 
production. As a result, the DRAM market was glutted, driving prices 
down in December, 1996 to one of the lowest levels during the downturn. 
A question in the DRAM industry today is whether another temporary 
spike in demand will trigger a new flow of production, resulting in a 
new round of market saturation. According to Dataquest (see ``When Will 
the DRAM Market Turn?'', February 3, 1997), supply is expected to 
moderate throughout 1997, but it may be 1998 before supply will come 
into balance with demand.
C. The Petitioner's Allegation That LGS and Hyundai Were Dumping in 
1996
    Throughout this proceeding the petitioner has made a number of 
submissions, including numerous charts and graphs using the sales and 
cost data submitted by the respondents during the third administrative 
review and publicly available information regarding pricing trends, 
which the petitioner claims demonstrate that LGS and Hyundai made sales 
at less than normal value during the 1996 downturn. The respondents 
claim that the petitioner's analysis is flawed because it made a number 
of erroneous assumptions and was based on incomplete and inaccurate 
data. In addition, the respondents' contend that when current market 
conditions are viewed, the record indicates that future dumping is not 
likely.
    We have reviewed the data submitted by the petitioner as well as 
all arguments and information on the record regarding the veracity of 
the data and the underlying assumptions. As discussed more fully below, 
on the basis of that examination, we find that the not likely criterion 
for revocation has not been satisfied for the following reasons: (1) 
The respondents' own sales and cost data indicate that there were a 
substantial number of home market sales made at prices below COP during 
the two months immediately following the close of the third 
administrative review; (2) the lowest point of the downturn, in terms 
of DRAM pricing and other market conditions, did not occur until after 
mid-1996 (well after the end of the third administrative review 
period); (3) publicly available spot market pricing data, when viewed 
in conjunction with the respondent's cost data, extrapolated to a 
future point in time, indicate that LGS and Hyundai may have made U.S. 
sales at prices below COP during 1996; (4) respondent's own pricing 
data indicate that contract prices generally follow the same pricing 
patterns as spot market prices; and (5) many of the respondents' 
arguments concerning the alleged distortions and inaccuracies in the 
petitioner's analysis lack merit. In addition, we find that the 
respondents made several changes to their costs in the period 
immediately following the third review period, including changes in 
depreciation and foreign exchange loss write-offs. For a complete 
analysis, see the Memorandum to the File from Tom Futtner to Jeffrey P. 
Bialos, dated July 16, 1997, on file in room B-099 of the main Commerce 
building.
    As the petitioner points out, respondents' data indicate that 
products were sold in the home market at prices below the COP during 
May and June of 1996, the two months immediately following the end of 
the third review period. According to the Department's standard 
questionnaire for the third review, the respondents were required to 
report costs and sales for May and June of 1996 to ensure that the 
proper cost test and contemporaneous sales comparisons could be 
performed. These data demonstrate that the sales made below cost for 
both respondents increased in these two months, as the downturn in the 
DRAM market worsened. We note that, according to the Department's cost 
test methodology, these below cost sales were not sufficiently numerous 
for the Department to reject as a basis for determining normal value in 
this third review. We also agree with LGS that whether it made home 
market sales at prices below the COP during the two months immediately 
following the close of the third review period in and of itself does 
not demonstrate that dumping occurred. However, in light of the market 
conditions during the downturn and the fact that the months actually 
examined during the POR did not include the lowest point in the 
downturn, we find that the existence of below-cost sales during May and 
June of 1996 suggests that the number of below-cost sales increased 
following the end of the third review period as the DRAM market 
worsened. As prices in the DRAM market fell, a substantial number of 
sales were made below cost. This pattern is suggestive of deteriorating 
market conditions that often give rise to dumping.
    In order to derive the estimated COP for 4M and 16M DRAMs for the 
third and fourth quarters of 1996, the petitioner took the respondent's 
actual reported costs for the third administrative review and projected 
these costs through the year using the same rate of decline experienced 
in the industry during 1995. Given that costs typically decline over 
time in the DRAM industry, we find the petitioner's approach to 
estimating the respondents' COP to be reasonable.
    We disagree with the respondents' assertion that the average U.S. 
prices presented in the petitioner's analysis bear no relation to their 
actual U.S. prices. We recognize that the petitioner based its analysis 
upon average U.S. spot market prices instead of contract prices. 
However, based upon the average gross unit prices calculated using 
respondent's own data from the POR, it appears that contract prices

[[Page 39818]]

generally follow the same pricing patterns as spot market prices. There 
is even evidence on the record indicating that the actual contract 
prices were sometimes lower than the average spot prices presented in 
the petitioner's analysis. We also disagree with LGS' claim that the 
U.S. price quotes referred to in the petitioner's analysis cannot be 
relied upon because the source documentation was not provided. The 
record is clear that the petitioner used prices compiled by Lehman 
Brothers. These data were similar to other pricing data submitted on 
the record, including the pricing data obtained from the American 
Integrated Chip Exchange (AICE) and Dataquest.
    Regarding Hyundai's claim that the petitioner's data failed to take 
into account reductions in cost resulting from the depreciation of the 
won, we note that Korean DRAM producers import machinery and equipment 
and many raw materials. In fact, both respondents recorded large 
foreign exchange losses for fiscal year 1996. Therefore, the 
depreciation of the won may have actually tended to increase the 
respondent's COP, making dumping more likely in the United States. At 
the very least, we find no basis in the record to conclude that this 
exchange rate depreciation entirely favored the respondents.
    Regarding LGS'' contention that the petitioner's analysis 
overstated the degree of DRAM price decline because it was based on 
monthly prices averaged on a simple, rather than weighted-average 
basis, we note that petitioner's pricing data generally followed the 
same downward trend of other pricing data on the record, including the 
AICE data noted above. In fact, all pricing data on the record followed 
the same downward trend throughout 1996, whether they were based on a 
simple average or not. Finally, we disagree with Hyundai's assertion 
that the preliminary analysis was flawed because it compared the 
average spot price for all 16M DRAMs with the COP of only the 1X16 
configuration. In fact, both the cost and sales data used for this 
comparison were for the 1X16 configuration, not all DRAM models.
    In its case brief, LGS submitted what it claimed were actual price 
and cost data for the second half of 1996. Our review of this 
information, however, indicates that there are serious questions 
whether the reported costs were understated due to significant changes 
in LGS' depreciation schedule and write-offs of foreign exchange 
losses. Publicly available data indicate that, for their 1996 financial 
statements, both LGS and Hyundai changed the useful life of fixed 
assets from three years to five years. However, it is unclear exactly 
to what extent this change reduced the reported costs. Similarly it is 
unclear how the reported costs were affected by the losses on foreign 
exchange. Moreover, the fact that LGS failed to identify these 
adjustments to its costs significantly reduces the reliability of the 
information. We are uncertain whether LGS made other adjustments to its 
reported costs. Additionally, we note that LGS did not provide these 
data until its April 18, 1997, case brief, despite having ample 
opportunity to do so before the Department's March 10, 1997, 
preliminary results. Although the Department accepted these data into 
the record because of the extended deadline for submitting factual 
information during this revocation proceeding, LGS' delay in submitting 
the information greatly limits its usefulness. The Department was 
unable to fully examine the data and perhaps question LGS concerning 
the composition of the data.
    In its case brief Hyundai presented a detailed econometric study 
conducted by Dr. Kenneth Flamm. Senior Fellow, the Brookings 
Institution. The cost projections in this analysis included assumptions 
regarding certain production indices and yields and exchange rates. 
Prices were projected using econometric techniques including various 
scenarios for supply, economic growth, and technological change. The 
study concluded that Hyundai's prices would exceed its cost of 
production ``by a comfortable margin'' in all scenarios considered.
    We find that the cost portion of the Flamm study was based on 
several questionable premises including the assumption of certain 
production yields and rates. The study utilizes a ``best case 
scenario'' in terms of certain of these assumptions. Optimistic 
capacity rates in particular are difficult to accept in a time when 
major producers, Hyundai included, have announced major cutbacks in the 
production of DRAMs. Furthermore, as the Flamm study itself points out, 
the capacity scenario is based on the assumption that DRAM demand will 
continue to strengthen. However, current market conditions do not bear 
the strong demand assumption out. According to the AICE's Bulletin for 
the Day (June 13th), activity in the U.S. market continues to be slow. 
Similarly, according to Dataquest (``The Semiconductor DQ Monday 
Report'', Issue 24, June 23, 1997), there continues to be a ``serious 
oversupply or inventory excess'' in the DRAM market. Also, 
technological shifts in demand are difficult to predict. For instance, 
the study does not mention the rate at which the supply of competing 
64M DRAMs can be expected to expand, and put downward pressure on the 
prices for the 16M generation.
    In addition, wholly apart from the data concerning the 1996 
downturn, as discussed in sections B and C, above, our analysis 
indicates that market conditions in the DRAM industry remain volatile. 
As stated previously, while the plunge in prices began to stabilize 
somewhat in early 1997, recent data indicate that prices are headed 
downward again. For example, according to publicly available data, the 
average U.S. price for a 16M DRAM fell from approximately $18.00 in May 
1996 to approximately $7.00 in December 1996. According to Dataquest, 
the price for the 16M as of June 30, 1997, is approximately $6.50. This 
represents a 64 percent decline in prices between the end of the third 
period of review (April 30, 1996) and June 1997. Since DRAMs are a 
commodity product, it is reasonable to expect that Korean producers 
will match prevailing market prices in the United States.
D. Whether Korean DRAM Producers Can Remain Competitive in the U.S. 
Market Without Dumping
    As noted above, LGS argues that it has no economic incentive to 
dump DRAMs in the U.S. market. LGS' key arguments are that its share of 
the U.S. market is too small for predatory pricing to be successful; 
that the company's U.S. market share is, nevertheless, steady enough to 
discourage ``promotional'' dumping; that dumping did not result from 
exchange movements; and that LGS knows the U.S. antidumping laws well 
enough to have avoided ``accidental'' dumping. LGS concludes its 
analysis by forecasting increasing demand and price levels in 1997.
    The antidumping law is designed to counteract price discrimination 
by foreign producers and exporters which injures a domestic industry. 
This requires only a comparison of U.S. prices and normal value and 
does not allow for the Department to consider the intent of producers 
and exporters who sell here. That being said, in determining whether it 
is not likely parties will sell at less than normal value in the 
future, the issue of whether those parties have an economic incentive 
to dump is relevant to the Department's analysis. See Preliminary 
Results, 62 FR at 12796 (citing Brass Sheet and Strip from Germany, 61 
FR at 49730). However, it may not be an overriding factor, and must be 
considered in conjunction with the remaining record evidence and in 
light

[[Page 39819]]

of the Department's experience in administering the revocation 
provisions. For instance, whether parties can price competitively 
without dumping depends, among other things, upon short-term and long-
term market conditions. In this regard, LGS argues that it has a 
relatively small share of the U.S. market, which decreases its economic 
incentive to dump. However, the United States is part of the world's 
largest regional market for DRAMs, with considerable growth potential. 
Given the importance of the U.S. market, as a general matter, even a 
producer with a relatively small market share would have an incentive 
to ride out industry downturns. The fact that DRAM producers, including 
the Korean respondents, have historically been found to have dumped 
during downturns supports this conclusion.
    LGS states that its OEM contract customers pay higher-than-spot 
market prices in a market downturn, and lower-than-spot market prices 
in a market upturn. In actuality, the record demonstrates that contract 
prices to OEM customers, which are negotiated on a quarterly basis, 
follow the direction of prices on the spot market. Dell and Digital 
both noted such trends based on their own experience. Thus, according 
to our record, changes in prices of OEM customers simply lagged behind 
spot prices. In fact, even into 1997, prices to OEM customers remained 
depressed, and below spot market prices, even as the spot market prices 
began to show some increase.
    Finally, LGS argues that the company did not dump subsequent to the 
third review period because its production costs were also declining. 
Historical data support the premise that both costs and prices of any 
given generation of DRAM will decline over time. What respondents have 
been unable to demonstrate, however, is that the decline in costs kept 
up with the rapid rate of decline in prices during the second half of 
1996.
    In sum, the current condition of the DRAM market and the data on 
the record supports a conclusion that the not likely criterion for 
revocation has not been satisfied.
    Comment 5: Whether the Antidumping Order is Constraining LGS and 
Hyundai from Dumping in the U.S. Market.
    The petitioner argues that during the third review period LGS and 
Hyundai were constrained by the antidumping duty order in that both 
companies took significant steps to minimize the size of their dumping 
margins. Regarding LGS, the petitioner contends that the company's U.S. 
sales volume and number of customers decreased dramatically during 
1996, demonstrating that the antidumping duty order was constraining 
LGS from dumping. In addition, the petitioner claims that LGS' average 
U.S. DRAM price decline during 1996 was not as severe as the general 
price declines experienced in the industry during the same period, 
indicating that LGS was selecting the customers to which it would sell 
DRAMs directly. Regarding Hyundai, the petitioner asserts that the 
dumping order forced Hyundai to take measures to ensure that its home 
market sales were used as the basis for normal value, and that its home 
market sales prices were always higher than its United States sales 
prices.
    LGS argues that the Department's attempt to speculate as to whether 
LGS' prices may have been at less than normal value ``in the absence of 
the order'' is fundamentally flawed. LGS asserts that no amount of 
speculation could produce a reliable conclusion as to what ``might have 
happened'' if the dumping order had not been in effect during a 
historical period when the dumping order did in fact exist. Hyundai 
argues that the Department's findings that the majority of its United 
States sales were at prices well above normal value in the preliminary 
results demonstrates that Hyundai's prices were not constrained by the 
order.
    LGS rebuts the petitioner's arguments by arguing that the facts on 
the record indicate that LGS maintained a consistent U.S. presence 
during 1996. Specifically, LGS contends that publicly available data 
indicate that the company's U.S. market share remained stable during 
1995 and 1996. In addition, LGS asserts that the petitioner's analysis 
was flawed because, first, it compared the volume of sales and customer 
base from the middle of 1995 to the volume of sales and customer base 
at the beginning of 1996. LGS asserts that such a comparison is not 
fair, given the seasonal nature of DRAM prices. When prices and costs 
are compared for the same time period, LGS asserts, verified data show 
that direct sales in the United States actually increased during 1996. 
Second, LGS contends that the petitioner's analysis compared unit 
quantities rather than megabyte quantities. LGS asserts that by only 
examining unit quantity declines, the petitioner failed to capture the 
natural shift to higher DRAM generations with larger memory capability. 
Regarding the petitioner's contention that LGS' price declines were not 
in line with general industry declines, LGS maintains that during 
market downturns, the company's OEM customers pay higher prices than 
they would on the spot market.
    The petitioner contests LGS' assertion that it is illogical to 
attempt to determine what a respondent's pricing behavior ``may'' have 
been if an antidumping duty were not in place. According to the 
petitioner, it is entirely reasonable for the Department to analyze 
what a respondent's pricing practices ``would have been'' in the 
absence of an order.

DOC Position

    We agree with respondents that in the circumstances of this case it 
would be inappropriate for the Department to speculate as to whether or 
to what degree, during the first three review periods, the antidumping 
order on DRAMs from Korea constrained LGS and Hyundai from pricing at 
less than normal value. At the same time, the Department does not have 
to find that the order has had no effect on the parties' pricing 
behavior. The more relevant question is whether the recent significant 
downturn in the industry affects the likelihood that the Korean 
respondents will dump in the future. As discussed in Comment 2, above, 
this is not a question the Department can or needs to answer with 
certainty. Rather, the Department must be satisfied that future dumping 
is not likely in order to revoke an order. In this case, based upon the 
evidence in the record, this standard has not been met and, therefore, 
we conclude that there is a need for the order to remain in place. 
Accordingly, we have determined not to revoke, in part, the antidumping 
duty order on DRAMs from Korea.

II. General Comments

    Comment 6: New Factual Information Allegation.
    The petitioner argues that LGS, Hyundai, and Compaq submitted new 
factual information in their April 18, 1997, case briefs. The 
petitioner asserts that such information is untimely since the 
established deadline for the submission of factual information 
regarding revocation was January 27, 1997.
    LGS, Hyundai and Compaq argue that the information submitted in 
their case briefs was not untimely, but instead was responsive to the 
Department's request in its preliminary results for views on ``current 
and projected market circumstances'' regarding the issue of revocation.
    The petitioner rebuts the respondents' argument stating that the 
common meaning of ``views'' refers to opinions, arguments and 
conclusions concerning

[[Page 39820]]

a given issue, not the submission of new factual information. In 
addition, the petitioner asserts that in the event the Department 
determines it is appropriate to accept the additional market 
information presented in the respondents' case briefs, the data claimed 
to be the actual price and cost information of LGS cannot be used to 
support revocation because it is not accurate as discussed in comment 
5, above, and was not verified.

DOC Position

    We agree with LGS, Hyundai and Compaq. In our preliminary analysis 
of the revocation issue, we cited trends in DRAM prices and costs as 
part of our rationale for publishing a preliminary notice of intent not 
to revoke the order, in part. Our preliminary results also specifically 
invited comments from interested parties regarding ``current and 
projected market circumstances.'' The information submitted by the 
interested parties in their case and rebuttal briefs pertain to current 
and projected market conditions directly relating to the factors 
underlying the Department's preliminary ``not likely'' analysis. 
Therefore, we agree with LGS, Hyundai and Compaq that this information 
was solicited by the Department and may have a direct bearing on the 
factors the Department will consider in making in its final ``not 
likely'' analysis. Therefore, we find that this data was not untimely 
filed.
    Comment 7: Whether the Department Properly Applied the CEP Offset 
in the Preliminary Results.
    The petitioner argues that the Department should not have applied 
the CEP offset in its preliminary results because neither LGS nor 
Hyundai has demonstrated that they were entitled to an adjustment for 
differences in level of trade. Specifically, the petitioner maintains 
that the Department erred in determining that one level of trade 
existed in the home market (direct sales by the parent corporation to 
the domestic customer) and that a different level of trade existed in 
the U.S. market, where the Department used the level of trade of the 
sale to the affiliated importer rather than the resale to the 
unaffiliated customer (i.e., a ``constructed'' level of trade). The 
petitioner asserts that neither the Act nor the SAA permit the 
Department to use a ``constructed'' level of trade for constructed 
export price (CEP) sales when identifying the level of trade. The 
petitioner argues that section 773(a)(7)(A) of the Act, which provides 
for a level of trade adjustment, does not make any distinction between 
export price (EP) sales and CEP sales, and that the distinction between 
EP and CEP sales in subsections 772(a) and 772(b) of the Act also does 
not warrant any different treatment when identifying levels of trade.
    The petitioner argues that, in view of the sections of the Act 
mentioned above, the Department's interpretation of the SAA as 
permitting a constructed level of trade means that the home market 
level of trade will always be at a more advanced stage of distribution 
than the level of trade of the CEP, the data available will never 
provide an adequate basis to quantify a level of trade adjustment, and 
thus, the CEP offset will always be used. The petitioner contends that 
the SAA intended the application of the CEP offset to be an exception, 
rather than the rule. Therefore, the petitioner asserts that the 
Department's acceptance of a constructed level of trade contradicts the 
intent of the SAA and the intent of the statue in section 773(a)(7)(A).
    The petitioner further argues that, even if the Department adheres 
to the distinction between EP and CEP sales in determining the starting 
price for determining the level of trade, neither respondent has 
adequately demonstrated that it is entitled to a level of trade 
adjustment. The petitioner argues that the simple enumeration of 
selling functions in both the home market and U.S. market is not 
sufficient to demonstrate the significance of the differing selling 
functions in both markets.
    LGS and Hyundai argue that the Department correctly applied the CEP 
offset to adjust for differences in the levels of trade in the two 
markets which were not capable of being quantified. Both respondents 
assert that the Department's use of a ``constructed'' level of trade 
when analyzing CEP sales is in accordance with past interpretation of 
the SAA and the Act. In addition, LGS maintains that the Department has 
consistently followed this approach and has explicitly stated in the 
antidumping questionnaire that a constructed level of trade will be 
used for CEP sales.
    LGS and Hyundai also reject the petitioner's argument that 
respondents have not adequately documented differences in selling 
functions in the U.S. and home markets. The respondents claim that in 
its case brief, the petitioner only referenced the brief discussion of 
the selling function differences contained in the notice of preliminary 
results and ignored the detailed analysis presented in the respondents' 
questionnaire responses and in the Department's preliminary analysis 
memorandum. Hyundai and LGS contend that the Department's preliminary 
analysis memorandum shows that the selling functions actually performed 
by the respondents on home market sales are much more significant than 
the selling functions performed for U.S. sales. LGS and Hyundai contend 
that, because their home market sales were at levels of trade more 
advanced than their U.S. sales and it was not possible to quantify the 
price differential caused by these differences, the Department should 
continue to allow a CEP offset to NV or to constructed value (CV) in 
order to adjust for the differences in levels of trade between the two 
markets.

DOC Position

    We agree with LGS and Hyundai. We do not base the level of trade on 
the starting price for both EP and CEP sales. While the petitioner is 
correct in noting that the starting price for calculating the CEP is 
that of the subsequent resale by the affiliated importer to an 
unaffiliated buyer, the Act, as amended by the URAA, and the SAA 
clearly specify that the relevant sale for our level of trade analysis 
is the constructed export price transaction between the exporter and 
the importer.
    While the starting price for CEP is that of a subsequent resale to 
an unaffiliated buyer, the calculation of the CEP results in a price 
that corresponds, as closely as possible, to an export price between 
non-affiliated exporters and importers, as explained in the SAA. See H. 
Doc. No. 316, 103d Con., 2d Ses., Vol. I, at 823 (1994). In other 
words, constructing an export price removes a link from a respondent's 
U.S. distribution chain--the link between the affiliated U.S. importer 
and its customers. Thus, the CEP is a price exclusive of all expenses 
and profit associated with economic activates occurring in the United 
States. The expenses specified in section 772(d) of the Act and the 
profit associated with those expenses represent activities undertaken 
in the United States to support U.S. resales to unaffiliated customer. 
Generally these activities are undertaken by the affiliated importer 
and occur after the transaction between the exporter and the importer. 
Because the expenses and profit deducted under section 772(d) represent 
activities undertaken to support the U.S. resale, the deduction of 
these expenses normally yields a different level of trade for the CEP 
than for the later resale. Movement charges, duties and taxes deducted 
under section 772(c) do not represent activities of the affiliated 
importer, and we do not remove them from starting price to obtain the 
CEP level of trade. See, e.g., Antifriction Bearings (Other than 
Tapered Roller

[[Page 39821]]

Bearing) and Parts Thereof from France, et. al.; Final Results of 
Antidumping Administrative Review, 62 FR 2083, 2105 (January 15, 1997); 
Roller Chain, other than Bicycle from Japan; Preliminary Results of 
Antidumping Duty Administrative Review, 62 FR 25165, 25168 (May 8, 
1997); and Certain Corrosion-Resistant Carbon Steel Flat Products and 
Certain Cut-to-Length Carbon Steel Plate from Canada; Final Results of 
Administrative Review, 62 FR 18448, 18466 (April 15, 1997). In 
accordance with our practice, the instructions in the questionnaire 
issued to respondents in this administrative review properly stated 
that a constructed level of trade would be used for our level of trade 
analysis.
    We also disagree with the petitioner's assertion that LGS and 
Hyundai have not adequately documented their respective differences in 
selling functions in the home and U.S. markets so as to warrant level 
of trade adjustments (or a CEP offset, as was actually calculated). As 
noted by respondents, the petitioner referred primarily to the 
Department's preliminary results of review as published, and 
disregarded the more detailed data and analysis on the record 
concerning the differences in selling functions and other factors 
contained in the Department's preliminary analysis memoranda for both 
respondents.
    In addition to the analysis contained in the preliminary results, 
these memoranda contain more detailed descriptions of the information 
provided by respondents and the differences in selling functions 
between the two markets. Based on this analysis, we concluded that U.S. 
and home market sales made by both respondents were at different points 
in the channel of distribution and that the selling functions performed 
by the respondents for home market sales were sufficiently different 
from those performed by the respondents for U.S. sales. Therefore, the 
Department properly determined that the sales made by Hyundai and LGS 
in the home market were at a different level of trade than the sales 
made in the United States. As explained in the preliminary results of 
review, however, we also determined that it was not possible to 
quantify the price differences resulting from the differing levels of 
trade, thus justifying a CEP offset to normal value for both 
respondents pursuant to section 773(a)(7)(B) of the Act. See 
Preliminary Results, 62 FR at 12798-99.

III. Company Specific Comments

A. Hyundai
    Comment 8: Whether Hyundai's Reported Home Market Sales Constitute 
a Fictitious Market.
    The petitioner argues that Hyundai's reported home market sales 
constitute a fictitious market and cannot be used as a basis for normal 
value. Specifically, the petitioner contends that beginning in February 
1996, Hyundai created a fictitious market by manipulating its home 
market sales prices in the following manner: (1) Hyundai essentially 
quit making sales to OEM customers and instead made sales only to a 
small number of distributors. The petitioner asserts that this allowed 
Hyundai to control its home market prices; (2) Hyundai stopped making 
sales at different times throughout the month, and instead only made 
sales at the end of the month. The petitioner claims that this practice 
allowed Hyundai to determine the necessary price to charge for those 
home market sales that would be matched to the U.S. sales prior to 
making the sale; (3) although the number of home market customers 
decreased, the quantity of DRAMs sold in the home market increased as 
the price collapsed. The petitioner asserts that Hyundai did not 
explain how the Korean market was able to absorb the surge in DRAMs; 
(4) the Department did not conduct a thorough verification of this 
issue; and (5) the average unit prices for home market sales which were 
used as matches to U.S. sales were significantly lower than the average 
unit prices for DRAM sales not matched to U.S. sales. The petitioner 
contends that in most instances, the price difference was not warranted 
because the products which were not used as matches for U.S. sales 
generally had only one characteristic (e.g., speed) different from 
those sales that were matched to U.S. sales. Based on these assertions, 
the petitioner contends that in the final results, the Department 
should find that a fictitious market exists, disregard Hyundai's 
reported home market sales and base normal value on facts available.
    Hyundai argues that the petitioner's arguments hold no merit and 
are based on a distorted analysis of the record. Specifically, Hyundai 
asserts the following: (1) The Department's verification report 
confirms that the sales made to home market distributors were in fact 
real sales made to real customers. In addition, Hyundai contends that 
the Department examined numerous home market sales, including receipts 
and other documents verifying delivery of the merchandise, at 
verification. Therefore, Hyundai asserts that the record indicates that 
Hyundai's home market sales were bona fide sales; (2) Hyundai contends 
that the petitioner's assertion that the company priced its home market 
sales which were matched to U.S. sales at prices that were lower than 
the prices it charged on sales not used for comparison purposes is 
factually incorrect and based on a flawed analysis. In addition, 
Hyundai claims that given that 99.9 percent of its home market sales 
were used as comparison sales, the petitioner's apparent assumption 
that Hyundai made up for the revenues sacrificed on lower-priced 
matched sales with the revenues earned on higher priced non-matched 
sales is mathematically impossible; (3) Hyundai asserts that the 
petitioner's claim that the company began making sales only at the end 
of the month is inaccurate. Hyundai asserts that throughout the POR, 
its home market sales were usually made during the last 10 days of the 
month, although on occasion, Hyundai made sales earlier in the month 
(e.g., in March 1996, Hyundai made sales at various times during the 
beginning, middle and end of the month); (4) Hyundai argues that its 
reported home market sales information demonstrates that most of 
Hyundai's sales throughout the entire POR were to distributors. 
Therefore, Hyundai asserts that there was nothing unusual about its 
sales to distributors, as alleged by the petitioner; (5) Hyundai claims 
that the petitioner's contention that the quantity of DRAMs sold in the 
home market increased fails to demonstrate anything other than that 
price reductions stimulate demand; and (6) the petitioner's 
presentation of pricing patterns in the home market does not satisfy 
the statutory definition of fictitious market in that it only shows 
prices moving in tandem, not ``differences in movements.'' 
Specifically, Hyundai asserts that the petitioner's pricing data do not 
show that prices for non-matched sales increased while prices for 
matched sales decreased. Instead, Hyundai asserts that the petitioner's 
data show that prices for both types of sales declined over time, a 
pricing pattern entirely consistent with the normal pricing patterns 
for the DRAM industry. For all of these reasons, Hyundai argues that 
the Department should reject the petitioner's assertion that Hyundai's 
home market is fictitious.

DOC Position

    The petitioner failed to raise its fictitious market allegation 
until filing its case brief following the preliminary results of 
review. Therefore, the

[[Page 39822]]

petitioner's allegation was untimely filed and not adequate to warrant 
determining that Hyundai's home market sales constitute a fictitious 
market.
    A fictitious market analysis is extraordinary. As the Department 
stated recently in the preamble to its final regulations implementing 
the URAA, the Department typically does not engage in a fictitious 
market analysis under section 773(a)(2) of the Act, or a variety of 
other analyses called for by section 773, ``unless it receives a timely 
and adequately substantiated allegation from a party.'' Antidumping 
Duties; Countervailing Duties; Final Rule, 62 FR 27296, 27357 (May 19, 
1997) (Final Regulations) (citing Tubeless Steel Disc Wheels from 
Brazil, 56 FR 14083 (1991); Porcelain-on-Steel Cooking Ware from 
Mexico, 58 FR 32095 (1993)). The various provisions of section 773, 
including section 773(a)(2), ``call for analyses based on information 
that is quantitatively and/or qualitatively different from the 
information normally gathered by the Department as part of its standard 
antidumping analysis.'' Final Regulations, 62 FR at 27357. The 
Department must determine, as a threshold matter, whether such an 
analysis is warranted based upon the adequacy of the allegation. See 
Porcelain-on-Steel Cooking Ware, 58 FR at 32096; Electrolytic Manganese 
Dioxide From Japan, 56 FR 28551, 28555 (May 14, 1993).
    The untimely nature of petitioner's allegation during this review 
prevented the Department from making this threshold determination at an 
appropriate point in the proceeding. Therefore, we reject petitioner's 
allegation on this basis alone.
    Comment 9: Whether the Normal Value of Further-Manufactured Models 
Should be Based on Constructed Value.
    Hyundai argues that in its preliminary results, the Department 
improperly compared the prices of its further-manufactured sales of 
memory modules to the CV of the imported merchandise. Hyundai asserts 
that this approach is inconsistent with the Department's standard 
practice of comparing the U.S. price of the product as imported, to the 
normal value of the identical product. Hyundai cites Certain Internal-
Combustion, Industrial Fork Lift Trucks from Japan, 53 FR 12552, 12559 
(1988), as case precedent for this practice. Hyundai contends that in 
its final results, the Department should make price-to-price 
comparisons for all further manufactured models using the net price of 
the imported product. Alternatively, in the event the Department 
determines that it is too complicated to determine the net price for 
mixed modules (i.e., modules that include two types of DRAMs), Hyundai 
argues that the Department could use CV for the mixed modules. Hyundai 
notes that sales of mixed modules accounted for less than ten percent 
of its further manufactured sales during the POR.
    The petitioner argues that the Department was correct in comparing 
all of Hyundai's further manufactured U.S. sales to CV. The petitioner 
asserts that in the first administrative review, the Department stated 
that ``there were no comparable home market sales for U.S. sales of 
mixed modules and that the configuration and application of mixed 
memory modules are critical factors in determining the foreign market 
value of these modules.'' Based on these facts, the petitioner claims 
that the Department was compelled to use CV in its preliminary results.

DOC Position

    The Act sets forth a preference for basing normal value on the 
price of the foreign like product and for making price-to-price 
comparisons, whenever possible. See 19 U.S.C. 1677 (b)(1); 19 CFR 
353.46(2)(1996). Therefore, for single memory modules, because there 
were home market sales of merchandise identical to the merchandise 
imported into the United States, we agree with Hyundai that, rather 
than resorting to CV, the Department should have followed its practice 
of comparing the U.S. price of the imported product (i.e., the DRAM) to 
the weighted-average price of the comparison product sold in the home 
market for single memory modules. We have made this correction in the 
final results.
    With regard to mixed memory modules, we agree with the petitioner 
that the Department correctly applied CV. Mixed memory modules are 
modules which contain more than one type of DRAM. In order to determine 
the net imported price for each type of DRAM, it would be necessary to 
allocate the net price of all DRAMs included in the mixed module to the 
individual DRAM types on the basis of relative costs. Due to the small 
quantity of mixed module sales in the United States and the complexity 
of such a calculation, we find that the use of CV is reasonable for 
mixed memory modules.
    Comment 10: Clerical Errors.
    The petitioner argues that the Department made the following 
clerical errors in its preliminary margin calculation for Hyundai: (1) 
The Department calculated CV profit on the basis of all home market 
sales, instead of using only those sales that were found to be above 
cost; and (2) the Department improperly excluded imputed credit and 
inventory carrying costs from the calculation of total U.S. expenses 
for the CEP profit calculation.
    Hyundai agrees that the Department incorrectly calculated CV profit 
using all home market sales, rather than only those sales that were 
found to be above COP. With respect to CEP profit, Hyundai argues that 
the Department properly excluded imputed credit and inventory carrying 
costs from both the calculation of the profit percentage and the 
calculation of total U.S. expenses used in the CEP profit calculation.

DOC Position

    We agree with the petitioner that the Department inadvertently 
included those home market sales which did not pass the COP test in the 
pool of sales used to calculate CV profit. We have corrected this error 
in these final results. In reviewing the margin calculation program it 
was noted that in the calculation of CEP profit duty drawback was 
inadvertently subtracted, rather than added. In addition, we noted that 
imputed credit and inventory carrying costs were inadvertently included 
in the pool of expenses used to calculate the selling expenses for CV. 
We have corrected these errors. Regarding the calculation of CEP 
profit, we agree with the petitioner that imputed credit and inventory 
carrying costs should have been included in the calculation of total 
U.S. expenses used to calculate CEP profit, although this did not 
necessarily constitute a clerical error. Including these expenses is 
consistent with section 772(f)(2)(B) of the Act. This provision defines 
the term ``total United States expenses'' as those expenses described 
under sections 772(d)(1) and (2) of the Act, which in turn include 
these imputed credit and inventory carrying costs. We have corrected 
this error in the final results.
    However, the Department properly excluded imputed credit and 
inventory carrying costs from the pool of selling expenses used to 
calculate the company's actual profit percentage. Because Hyundai's 
actual interest expense (as reported in the CV database) is accounted 
for in the calculation of profit there is no need to include imputed 
interest amounts. ``Although the actual and imputed amounts may differ, 
if we were to account for imputed expenses in the denominator of the 
CEP allocation ratio, we would double count the interest expense 
incurred for credit and inventory carrying costs because these expenses 
are already included in the denominator.'' Certain Cold-Rolled

[[Page 39823]]

and Corrosion-Resistant Carbon Steel Flat Products from Korea, 62 FR 
18404, 18440 (April 15, 1997); accord Preliminary Determination of 
Sales at Less Than Fair Value: Fresh Tomatoes from Mexico, 61 FR 56612 
(November 1, 1996).
B. LGS
    Comment 11: Research and Development Expenses.
    The petitioner argues that the Department erred in its preliminary 
results by accepting LGS' reported DRAM research and development (R&D) 
expenses which allocated DRAM R&D expenses over DRAM cost of sales. The 
petitioner maintains that, in accordance with the first and second 
administrative reviews, the Department should allocate LGS' R&D 
expenses related to all semiconductors over its 1995 total cost of 
sales for all semiconductors.
    LGS responds that the Department did revise LGS' reported R&D 
expenses in the preliminary results. However, LGS takes issue with the 
Department's recalculation. Specifically, LGS contends that the 
Department erroneously included R&D costs for products other than 
subject DRAMs in its calculation. LGS asserts that the same methodology 
was used in the less than fair value investigation and was reversed by 
the CIT, which found that the record evidence did not support a 
departure from the Department's practice of assigning research and 
development as specifically as possible to individual products. LGS 
argues that in the final results the Department should calculate the 
research and development rate by dividing the company's total DRAM 
research and development expenses for 1995 by its total DRAM cost of 
sales.
    In its rebuttal brief the petitioner states that if the Department, 
in fact, re-calculated the research and development expense ratio in 
its preliminary results by allocating the company's 1995 R&D expenses 
for all semiconductors over its 1995 total cost of sales, the 
petitioner fully supports the Department's preliminary calculation.

DOC Position

    In the preliminary results we properly calculated a R&D rate for 
LGS by allocating all semiconductor R&D expenses over the company's 
cost of sales for all semiconductors as reported in its audited 1995 
financial statements. This method of allocation is consistent with our 
practice in the last two administrative reviews, where we determined 
that sufficient evidence of cross-fertilization exists in the 
semiconductor industry to rule out the use of product or DRAM-specific 
research and development expenses. See Dynamic Random Access Memory 
Semiconductors from the Republic of Korea; Final Results of Antidumping 
Duty Administrative Review, 62 FR 965, 967 (January 7, 1997); 61 FR 
20216, 20218 (May 6, 1996). We have included in the record of this 
review a memorandum from a non-partisan expert relied upon in previous 
reviews, which describes the cross-fertilization and includes relevant 
pages from verification exhibits. See Memorandum regarding cross-
fertilization of research and development costs for DRAMs, August 14, 
1995.
    Comment 12: Clerical Errors.
    The petitioner argues that the Department made the following 
clerical errors in its preliminary margin calculation for LGS: (1) The 
Department failed to deduct early payment discounts from the 
calculation of the net price used in the cost test; (2) the 
Department's preliminary margin program used the wrong customer codes 
to identify sales made to home market customers which failed the 
Department's arm's-length test; as a result, the petitioner contends 
that sales to these customers were improperly included in the 
calculation of normal value; (3) although the preliminary margin 
calculation properly recalculated G&A and interest expenses for DRAMs, 
the Department failed to similarly recalculate G&A and interest 
expenses for modules; (4) the Department inadvertently double counted 
home market indirect selling expenses, bank fees and packing expenses 
in its calculation of total costs for the CEP profit calculation; and 
(5) the Department improperly excluded imputed credit expenses from the 
calculation of total U.S. expenses used to calculate CEP profit.
    LGS rebuts the petitioner's first alleged clerical error. LGS 
states that the Department should not deduct early payment discounts 
from the net price used in the cost test because these discounts were 
included in the build-up of the COP to which the net price was 
compared.
    LGS alleged the following clerical errors in the Department's 
preliminary margin calculations: (1) The Department inadvertently 
double counted home market indirect selling expenses in its calculation 
of COP; (2) the Department improperly excluded U.S. imputed credit 
expenses from the calculation of total expenses used to calculate the 
CEP profit percentage; and (3) the Department improperly calculated a 
single, weighted-average home market direct selling expense and 
indirect selling expense for CV based on the quantity of sales. LGS 
asserts that because direct and indirect selling expenses are allocated 
to sales based on value, and products with a relatively higher sales 
value carry a proportionately higher share of selling expenses, the 
Department should calculate weighted-average indirect and direct 
selling expenses based on density, not quantity.
    The petitioner argues that LGS did not explain why basing the 
calculation of the weighted-average selling expenses for CV on sales 
volume is inherently wrong or a clerical error. Therefore, the 
petitioner argues that there is no need for the Department to make the 
proposed change in allocation in its margin calculations. In addition, 
the petitioner asserts that the Department correctly deducted U.S. 
imputed credit expenses from the calculation of total expenses used to 
calculate the actual CEP profit percentage.

DOC Position

    We agree that the Department committed all five clerical errors 
alleged by the petitioner and the first clerical error alleged by LGS. 
These errors have been corrected in the final results. In addition, in 
reviewing the margin calculation program we discovered that U.S. re-
packing expenses had been deducted twice in the calculation of the CEP 
profit rate, that imputed credit and inventory carrying costs were 
inadvertently included in the pool of expenses used to calculate 
selling expenses for CV, and that the weighted-average direct and 
indirect selling expenses for CV had been calculated based on all home 
market sales, rather than just those sales which passed the COP test. 
We have corrected these errors. Finally, in response to LGS' concern, 
we have ensured that the calculation of the net price and COP used in 
the cost test were on the same basis.
    We disagree with LGS that the Department should have calculated the 
weighted-average direct and indirect selling expenses to be included in 
the calculation of CV based on density not quantity. LGS has not 
explained why it would be more accurate to calculate selling expenses 
for DRAMs based on density. In addition, based on information on the 
record it does not appear that selling expenses are incurred by LGS 
based on the density of different products. Finally, it is the 
Department's practice to calculate weighted-average selling expenses 
for CV based on the quantity of sales.
    We disagree with LGS' contention that the Department improperly

[[Page 39824]]

excluded imputed credit expenses from the pool of expenses used to 
calculate the actual CEP profit percentage. Because the actual interest 
expense of LGS was captured in the profit calculation there is no need 
to include an amount for imputed interest. See Comment 10, above.

Final Results of the Review

    As a result of this review, we determine that the following 
weighted-average dumping margins exist for the POR:

------------------------------------------------------------------------
                                                                Percent 
                    Manufacturer/exporter                        Margin 
------------------------------------------------------------------------
Hyundai Electronic Industries, Inc...........................       0.00
LG Semicon Co., Ltd..........................................       0.01
------------------------------------------------------------------------

    The U.S. Customs Service shall assess antidumping duties on all 
appropriate entries. Individual differences between United States price 
and normal value may vary from the percentages stated above. The 
Department will issue appraisement instructions concerning each 
respondent directly to the U.S. Customs Service.
    Furthermore, the following deposit requirements will be effective 
for all shipments of the subject merchandise entered, or withdrawn from 
warehouse, for consumption on or after the publication date of these 
final results of administrative review, as provided for by section 
751(a)(1) of the Act: (1) The cash deposit rate for the reviewed firms 
will be zero percent; (2) for previously reviewed or investigated 
companies not listed above, the cash deposit rate will continue to be 
the company-specific rate published for the most recent period; (3) if 
the exporter is not a firm covered in this review, a prior review, or 
in the original LTFV investigation, but the manufacturer is, the cash 
deposit rate will be the rate established for the most recent period 
for the manufacturer of the merchandise; and (4) if neither the 
exporter nor the manufacturer is a firm covered in this or any previous 
review conducted by the Department, the cash deposit rate will be 3.85 
percent, the all others rate established in the LTFV investigation. 
Samsung Electronics Co., Ltd. (Samsung), formerly a respondent in 
previous administrative reviews, was excluded from the antidumping duty 
order on DRAMs from Korea on February 8, 1996. See Final Court Decision 
and Partial Amended Final Determination: Dynamic Random Access Memory 
Semiconductors of One Megabyte and Above From the Republic of Korea, 61 
FR 4765 (February 8, 1996).
    These deposit requirements shall remain in effect until publication 
of the final results of the next administrative review.
    This notice serves as the final reminder to importers of their 
responsibility under 19 CFR 353.26 to file a certificate regarding the 
reimbursement of antidumping duties prior to liquidation of the 
relevant entries during this review period. Failure to comply with this 
requirement could result in the Secretary's presumption that 
reimbursement of antidumping duties occurred and the subsequent 
assessment of double antidumping duties.
    This notice also serves as a reminder to parties subject to 
administrative protective order (APOs) of their responsibility 
concerning the disposition of proprietary information disclosed under 
APO in accordance with 19 CFR 353.34(d). Timely written notification of 
the return/destruction of APO materials or conversion to judicial 
protective order is hereby requested. Failure to comply with the 
regulations and terms of an APO is a violation which is subject to 
sanction.
    This administrative review and notice are in accordance with 
section 751(a)(1) of the Act (19 U.S.C. 1675(a)(1)) and 19 CFR 353.22.

    Dated: July 16, 1997.
Robert S. LaRussa,
Acting Assistant Secretary for Import Administration.
[FR Doc. 97-19552 Filed 7-23-97; 8:45 am]
BILLING CODE 3510-DS-U