66 FR 50611, October 4, 2001
                                                            A-357-812
                                                        Investigation
                                                     IA/Group III: MB
                                                      Public Document

MEMORANDUM TO: Faryar Shirzad
               Assistant Secretary 
                 for Import Administration

FROM:          Joseph A. Spetrini
               Deputy Assistant Secretary
               AD/CVD Enforcement Group III


SUBJECT: Issues and Decision Memorandum for the Antidumping Investigation
of Honey from Argentina; Notice of Final Determination of Sales at Less
Than Fair Value (A-357-812)


Summary

We have analyzed the comments and rebuttals of interested parties in the
antidumping duty investigation of honey from Argentina (A-357-812). As a
result of our analysis, we have made changes, including corrections of
certain inadvertent programming and ministerial errors, in the margin
calculations. We recommend that you approve the positions we have
developed in the Discussion of the Issues section of this memorandum.
Below is the complete list of the issues in this investigation for which
we received comments and rebuttals by parties:

General 

1. Adverse Facts Available 
2. Cost of Production/Constructed Value Profit
3. Middlemen Reseller Expenses

Radix 

4. Facts Available 
5. Testing Expenses
6. General and Administrative and Indirect Selling Expenses
7. Sales Reconciliation
8. Reembolso Reimbursements
9. Interest Expense
10. Inventory Carrying Costs

ACA

11. Indirect Selling Expenses
12. General and Administrative Expenses
13. Interest Expenses
14. Testing Expenses
15. German Warranty Expenses
16. International Freight Expenses
17. Differences in Physical Characteristics in Merchandise
18. Level of Trade 


Background

We published in the Federal Register the preliminary determination in
this investigation on May 11, 2001. See Notice of Preliminary
Determination of Sales at Less Than Fair Value: Honey from Argentina, 66
FR 24108 (May 11, 2001) (Preliminary Determination). 

The period of investigation (POI) is July 1, 1999, through June 30, 2000.
The investigation covers honey sales exported by three exporters: Radix
S.R.L. (Radix), Asociacion Cooperativas Argentinas (ACA), and ConAgra
Argentina. We invited parties to comment on our preliminary determination
and the preliminary analysis of Radix's margin. We received case briefs
from ACA, Radix, and petitioners, (1) on August 6, 2001. We received
rebuttal briefs from the same parties on August 13, 2001. We received
petitioners' and Radix's comments on the preliminary margin analysis for
Radix on August 22, 2001. We received rebuttal comments from the same
parties on August 27, 2001. At the request of petitioners and ACA, we held
a public hearing on August 28, 2001.


Scope of Investigation

For purposes of these investigations, the products covered are natural
honey, artificial honey containing more than 50 percent natural honey by
weight, preparations of natural honey containing more than 50 percent
natural honey by weight, and flavored honey. The subject merchandise
includes all grades and colors of honey whether in liquid, creamed, comb,
cut comb, or chunk form, and whether packaged for retail or in bulk form.
For purposes of these investigations, the products covered are natural
honey, artificial honey containing more than 50 percent natural honey by
weight, preparations of natural honey containing more than 50 percent
natural honey by weight, and flavored honey. The subject merchandise
includes all grades and colors of honey whether in liquid, creamed, comb,
cut comb, or chunk form, and whether packaged for retail or in bulk form. 


Discussion of the Issues

Comment 1: Adverse Facts Available

Petitioners in this investigation assert that the Department should draw
an adverse inference against respondents with respect to the cost of
production (COP) data used in the calculation of the dumping margin.
Petitioners claim that respondents willfully withheld cost information
requested by the Department. Petitioners argue that, according to
submissions by counsel for the cost respondents (i.e., "the beekeepers"),
the beekeepers were committed to responding to the Department's cost
questionnaire by providing complete, accurate, and verifiable data. The
beekeepers retained counsel and two economic consulting firms specializing
in the preparation of antidumping questionnaire responses. According to
petitioners, the Department extended the deadline for submitting the cost
information twice because the beekeepers claimed that they needed more
time to collect the required information. Petitioners argue that, instead
of providing the required information, counsel for one of the exporters
under investigation merely submitted a letter to the Department stating
that no cost information would be submitted. Petitioners contend that both
the beekeepers and the exporters under investigation should have submitted
any cost information that was collected at the beekeeper level. In
petitioners' view, the non-responsiveness of the beekeepers and the
exporters demonstrated their preference for the Department's use of the
five cost studies contained in the petition. Petitioners conclude this
information must be less detrimental to their interests than actual COP
data. Petitioners also urge the Department to reject the COP information
submitted by ACA and Radix after the preliminary determination.
Petitioners state the following rationale for this rejection: (1) the
Department did not request the information submitted by respondents; (2)
the information submitted by Radix was untimely filed; (3) the information
submitted by respondents would be self-serving given that the beekeepers
supposedly withheld cost information; (4) the information is irrelevant to
the costs incurred by the selected beekeepers; and (5) the information
submitted by Radix is unreliable in determining accurate beekeeper COP
because it was prepared at the request of the Argentine Secretary of
Agriculture for use in the companion countervailing duty investigation on
honey from Argentina. 

Petitioners next assert that the Department should use adverse facts
available because respondents deliberately withheld actual cost data in an
attempt to improve their results. Petitioners cite to Certain Circular
Welded Carbon Steel Pipes and Tubes from Taiwan: Final Results of
Antidumping Duty Administrative Review, 64 FR 69488 (Dec. 13, 1999) (Pipes
and Tubes from Taiwan), asserting that, as in this case, the respondent
withheld requested cost information which resulted in the use of adverse
facts available in the final determination. Petitioners also cite to
Pistachio Group of the Association of Food Industries, Inc. v. United
States, 11 C.I.T. 668, 679, 671 F. Supp. 31, 40 (1987) (Pistachio Group),
asserting that the Department is justified in using adverse inferences to
prevent a respondent from manipulating the results of an investigation by
providing selective information or withholding requested information.
Further, petitioners note in Atlantic Sugar, Ltd. v. United States, 2 Fed.
Cir. (T) 130, 133-34, 744 F.2d 1556, 1560 (Fed. Cir. 1984) (Atlantic
Sugar) the court found that non-cooperative parties may be penalized by
using whatever other best information is available. Petitioners claim that
it is the Department's normal practice to use information more detrimental
to a respondent when it is apparent from the administrative record that
information has been intentionally withheld. (2) According to petitioners,
respondents would routinely ignore the Department's requests for
information without the Department's use of adverse facts available.

Petitioners also contend that section 776(a)(2) of the Tariff Act of
1930, as amended (the Act), clearly mandates the use of facts otherwise
available when parties fail to cooperate. According to petitioners'
interpretation of section 776(b) of the Act, adverse facts available
should be applied when parties do not act to the best of their ability
when complying with the Department's request for information, and the
Department may use information from the petition or any other information
on the record of the investigation.

Petitioners contend that the Department should have applied an adverse
inference against respondents in the preliminary determination with
respect to the cost data and urge the Department to do so now for purposes
of the final determination. Petitioners claim that the COP information
used in the preliminary determination was inadequate as adverse facts
available and, instead, was beneficial to respondents and rewarded their
decision to withhold cost information. Averaging the five cost studies
contained in the petition, petitioners assert, results in a significant
understatement of the actual average COP, because two of the three cost
studies report yields that greatly exceed the national average.
Petitioners suggest that the Department should use only the cost studies
which show yields comparable to the reported national average of 35-40
kilograms of honey per hive per year to calculate COP for the final
determination. To apply an adverse inference, petitioners contend the
Department should use only the March 1999 cost study, as that study
reports a yield rate comparable to the national average and contains the
highest per-unit costs of all five cost studies contained in the petition. 

ACA insists that it has cooperated fully in the course of this
investigation, by responding within the applicable deadlines to the
Department's numerous requests for information, by submitting supporting
documentation per the Department's requests, and by fully facilitating the
verification process in Buenos Aires. ACA asserts that it assisted the
Department in its attempts to acquire COP information from the selected
beekeepers. ACA observes that neither ACA's counsel, nor the beekeepers'
counsel was required to submit unreliable and internally inconsistent
data. ACA adds that it had no control over the beekeepers and could not
oblige them to keep verifiable records or respond fully to the
Department's request for cost information. With respect to petitioners'
arguments regarding the need for the Department to apply adverse facts
available in light of ACA's deliberate decision to withhold information,
ACA argues it did not deliberately withhold data. First, ACA asserts that
there is no information on the record to indicate that ACA did not act to
the best of its ability. Second, according to ACA, the application of
adverse facts available due to the inability of unaffiliated beekeepers to
provide reliable and verifiable data would be considered "perverse" and
raises the standard of cooperation so high that no respondent could
achieve it. Third, ACA states that, given that adverse facts available
cannot be used in the "all others" rate, ACA would receive a higher
dumping margin than its competitors who were not selected by the
Department to participate in this investigation. Finally, ACA contends
that punishing a cooperative respondent is not the intent of the dumping
statute. (3)

Radix states that the Department's regulations require that counsel
provide certification that information submitted on the record is not
misrepresented or that there are no omissions of facts. See 19 CFR
353.303(g)(2) [sic]. Radix insists it had to make a decision as to whether
the information collected from the beekeepers was reliable and usable.
Given that the information collected from beekeepers was incomplete,
contained clear errors, and contained internal inconsistencies, it was not
presented to the Department. Radix insists that it was not the exporters
in this investigation, ACA and Radix, who withheld the data. Given that
these two companies were not affiliated with the beekeepers, they were
unable to compel them to respond fully to the Department's requests for
information. 

Department's Position: We disagree with petitioners that an adverse
inference is warranted with respect to ACA and Radix in selecting facts
otherwise available for cost purposes. 

In the instance case, as we stated in the Preliminary Determination, the
beekeepers failed to provide cost data. Therefore, pursuant to 776(a)(1)
of the Act, we will use facts available to calculate COP because the
necessary information is not on the record. When the Department determines
that use of facts otherwise available is warranted, pursuant to section
776(a) of the Act, it is then permitted to apply an adverse inference when
selecting facts available, but only under limited circumstances. When
making a determination of whether such inference is warranted, section
776(b) of the Act requires the Department to make a finding that "an
interested party has failed to cooperate by not acting to the best of its
ability to comply with a request for information." In complying with this
standard, the Court of International made it clear the Department must
articulate its reasons for concluding that such lack of cooperation in
providing necessary information indeed took place, and explain why the
absence of this information is of significance to the investigation.
Mannesmannrohren-Werke AG et al. v. United States, 77 F. Supp. 2d 1302,
1313-14 (CIT 1999) (Mannesmannrohren-Werke). In doing so, the Department
must also explicitly state why it finds that the respondent's failure to
respond does not reflect an inability to respond. See Ferro Union v.
United States, 44 F. Supp. 2d 1310, 1331 (CIT 1999). As the Court further
clarified in a recent case, "[t]he deficient response must be analyzed in
light of the respondent's overall conduct, the importance of the
information, the particular time pressures of the investigation, and any
other information that bears on the issue of whether the deficiency was an
excusable inadvertence or a demonstration of a lack of regard for its
responsibilities in the investigation." Tung Mung Development Co., Ltd. v.
United States, Slip Op. 2001-83, Court No. 99-07-00457 (CIT 2001), citing
Nippon Steel Corp. v. United States, 118 F. Supp. 2d 1366, 1379 (CIT 2000).

Taking into consideration the totality of the facts involved in this
case, we do not believe that we are obliged to use an adverse inference
when choosing the facts available. In this case, the Department selected
exporters ACA and Radix (along with the non-responding ConAgra) and
certain producers to participate in this investigation. Rather than
request cost information from the respondents' own suppliers, the
Department selected 12 independent, unaffiliated honey producers (i.e.,
beekeepers) in Argentina to provide the necessary cost information. The
Department selected the 12 beekeepers out of a potential field of
approximately 25000 honey producers in order to calculate a country-wide
cost. The beekeepers selected were of varying sizes, levels of operation
and from regions in Argentina. According to Radix in its January 11, 2001
submission, these independent beekeepers are members of a highly
fragmented industry. Petitioners did not provide information to the
contrary. No evidence on the record indicates that any of the twelve
selected beekeepers were suppliers of ACA or Radix during the POI.
Although the beekeepers failed to produce the requested information, the
record demonstrates that both ACA and Radix made various efforts to assist
the beekeepers in gathering necessary data and presenting it in a format
required by the Department (see ACA's and Radix's submissions to the
Department of February 23, 2001, March 29, 2001, and April 9, 2001).
Further, the record also discloses that ACA and Radix participated fully
in the investigation.

Given these factual circumstances, we find that using an adverse
inference in selecting facts available, pursuant to section 776(b), is not
warranted. See Ferro Union 44 F. Supp. at 1331. Although by not responding
to our questionnaire the selected suppliers failed to cooperate by not
acting to the best of their ability to comply with our request for
information, we do not find it appropriate to use an adverse inference
under the circumstances here, particularly in light of the fact that there
is no evidence on the record that indicates that any of the selected
beekeepers were the suppliers of ACA or Radix. We also find that the
beekeepers' subsequent failure to respond to the Department's request for
cost information does not lead to a conclusion that Radix and ACA were
"unwilling" to cooperate to the best of their ability within the meaning
of section 776(b). The record does not contain any evidence that would
lead us to conclude that either ACA or Radix were able to exert control
over the producers, which would in turn raise questions as to these
companies' ability to control the information. See, e.g., Notice of Final
Determination of Sales At Less Than Fair Value: Certain Cold-rolled Carbon
Quality Steel Products from Brazil, 65 FR 5554, 5580 (February 4, 2000)
(adverse inference not warranted in calculating the major input rule where
respondents were unable to compel suppliers to provide cost information);
Elemental Sulphur from Canada; Final Results of Antidumping Finding
Administrative Review, 61 FR 8239, 8250 (March 4, 1996) (total cooperative
best information available applied to respondent whose supplier failed to
provide COP information); cf. Frozen Concentrated Orange Juice from
Brazil; Final Results and Partial Rescission of Antidumping Duty
Administrative Review, 64 FR 43650 (August 11, 1999) (use of adverse
inference warranted where there is no evidence that it was "outside of
[respondent's] ability to maintain control over the data necessary to
respond to the cost questionnaire"). 

We disagree with the petitioners' interpretation of the Department's
determination in Pipes and Tubes from Taiwan. In that case, unlike here,
we used an adverse inference in selecting facts available because we made
a finding that both the exporter and the producer withheld cost
information that was necessary in the Department's margin calculations.
See 64 FR at 69489 (Dec. 13, 1999). In the instant case, as discussed
above, the record evidence does not demonstrate that either ACA or Radix
attempted to or was in a position to withhold the beekeepers' information.
Nor does the record indicate that ACA or Radix could compel the beekeepers
to release this cost information to the Department. Irrespective of the
participation of the exporters in assisting in the collection of the cost
data requested by the Department, the ultimate responsibility to respond
to the Department's cost questionnaire and submit cost data lies with the
beekeepers because they are the owners of their own cost information.
Moreover, the petitioners incorrectly rely on the judicial opinion in
Pistachio Group, in which the Court upheld the Department's use of facts
available (formerly best information available). In that case, the Court
stated that the absence of diplomatic relations between Iran and the
United States made it "highly reasonable" for the Department to rely on
one party to provide the necessary information. The Court then agreed with
the Department that use of best information available was warranted, given
that "[t]he record reflects a reluctance to comply with the investigation
on the part of [respondent]." 671 F. Supp. at 31. Moreover, we note that
the requirement to find that a respondent had not cooperated by acting to
the best of its ability prior to drawing an adverse inference was added to
the statute after the Pistachio Group case was decided.

We agree with Radix that all parties in an investigation must provide
certification that all information submitted on the record is complete and
accurate pursuant to 19 CFR 351.303(g). However, we do not agree with the
implication that 351.303(g) alleviates the Department's responsibility
independently to assess information put before it. Nevertheless, there is
no evidence on the record which indicates that ACA and Radix did have
complete factual beekeeper cost of production information.

With respect to the alternative cost information provided by Radix on
June 18, 2001, we find that it would be inappropriate to determine COP of
the respondents using a study that was prepared at the behest of the
Argentine Secretary of Agriculture specifically for use in the
Department's companion countervailing duty investigation. This is
particularly so given that this report was prepared subsequent to and as a
consequence of the initiation of the investigation. For the final
determination we will rely upon, instead, the 1999 cost studies to
calculate COP because these studies were prepared: 1) by an independent
author; and 2) not in anticipation or in response to the antidumping
investigation. We find these studies to be contemporaneous information and
to provide more compelling evidence for use as facts available. 

For these reasons, consistent with our preliminary determination, as
facts available we calculated COP based on the cost studies submitted in
the petition. The COP of the producer is based on the simple average of
the five cost studies provided in the petition. In addition to these
costs, we included selling expenses and general and administrative
expenses incurred by ACA and Radix in connection with its sale of the
foreign like product to arrive at total COP through the exporter level.

Comment 2: Cost of Production/Constructed Value Profit

ACA argues that the cost studies used by the Department in the
preliminary determination do not conform to the Department's normal
practice regarding the treatment of revenue of joint products. According
to ACA, the Department's normal practice is to recognize a joint product
that is produced in the production process of the primary product as
either a co-product or a by-product (see Elemental Sulphur From Canada;
Final Results of Antidumping Finding Administrative Review, 61 FR 8239,
8241-8242 (March 4, 1996)). ACA asserts that although the cost studies
submitted by petitioners each report production for nuclei and attribute a
significant portion of beekeeper revenue to nuclei, the studies do not
account for nuclei as a joint product in calculating honey COP. 

ACA asserts the accounting methodology applied to nuclei revenue in the
cost studies is unknown. ACA states that when the Gestion Apicola began
publishing beekeeper cost studies in 1997, it applied a by-product cost
allocation for nuclei production. According to ACA, revenue from nuclei
was subtracted from total cost and honey COP was then calculated by
dividing the remaining costs by the kilograms of honey produced. ACA
asserts that without explanation, a new methodology was adopted for the
petitioners' studies, which appears to disregard the nuclei entirely in
the calculation of COP. 

ACA asserts the petitioners' study includes other accounting flaws. ACA
states the cost of both producing and purchasing the same nuclei are
included in the cost calculation, and notes that the March 1999 study
reports costs for feeding nuclei and health treatments for nuclei.
According to ACA, presumably the other direct and indirect costs also
include some costs related to nuclei production. ACA claims that it
appears that the hypothetical beekeeper is buying nuclei from himself and
allocating the COP and the cost of purchasing nuclei to the cost of
producing the honey. According to ACA, the appropriate treatment of a
joint product, such as nuclei, is to allocate the costs and expenses
between the joint products using the Department's co-product methodology
or to apply the by-product methodology used prior to the introduction of
the 1999 cost studies. 

ACA also asserts petitioners' cost studies rely on erroneous assumptions
regarding Argentine beekeeping and careless mathematical calculations. ACA
contends that, based on an affidavit from a government expert on Argentine
beekeeping, the petitioners' studies overestimate the percentage of queens
replaced each year by a typical beekeeper (see ACA's Letter of June 11,
2001). ACA also notes the studies overstate wax costs by assuming that
beekeepers would annually renew one-third of the wax in each hive and
would pay to purchase the full amount of that wax. This is compared to
ACA's expert's affidavit, which indicates that normal Argentine beekeeping
practice is to replace only one-third of the frame wax in the breeding
boxes and to renew wax in the honey boxes every 8 years. 

ACA maintains that petitioners' studies also appear to include errors in
the treatment of costs associated with producing nuclei. According to ACA,
nuclei are actually less than one-quarter the size of a hive. ACA notes,
however, that the studies provided by petitioner attribute identical
feeding and health care to nuclei and full hives. Hence, argues ACA, the
studies do not account for the cost differences attributable to feeding
and providing health care for hives versus nuclei. ACA also observes that
petitioners' studies overestimate the cost of drums based on its own
experience in purchasing drums, which is supported by an expert's
affidavit. ACA further states that petitioners' cost studies mysteriously
create a line item in administrative expenses and overestimate the number
of labor hours needed to build each hive box. 

ACA contends petitioners have presented the Department with cost studies
which include accounting mistakes, erroneous assumptions, and faulty
logic. ACA argues that the CIT and the Federal Court have made clear that
the Department may not base facts available on data that are not credible.
See Borden, Inc. v. United States, 4 F. Supp. 2d 1221, 1246 (CIT 1998).
ACA states the Department should replace the petitioners' information with
corroborated evidence on the record, either by using ACA's acquisition
price to calculate COP or by correcting the errors in petitioners' studies.

ACA contends that its acquisition price provides the Department with a
credible, verifiable and corroborated proxy for beekeeper cost of
production. According to ACA, unless there is reason to believe that the
price the exporter pays the producer does not fully reflect the cost to
the producer, the statute authorizes use of acquisition price in
calculating COP (see section 771(28) of the Act). ACA's cites the Notice
of Final Determination of Sales at Less Than Fair Value: Live Cattle from
Canada, 64 FR 56739 (October 21, 1999) (Live Cattle from Canada), in
arguing that in previous cases when an attempt to gather producer costs
has gone awry, the Department used acquisition price as a proxy for
producer price. ACA notes that it has provided the Department with its
acquisition cost during the POI, which can be easily verified in ACA's
annual report. 

ACA states that if the Department elects to use the petitioners' studies
it then must correct them to properly treat the nuclei as co-products of
the honey production process. ACA argues that the two most important
factors in the Department's analysis for determining co-production are the
relative sales value of the products and the way the company normally
accounts for the products. According to ACA, the relative sales value of
the nuclei compared to honey is significant in the March 1999 study (40
percent). ACA states that, although the factor as to how the company
treats the product in its normal accounting procedures cannot be analyzed
because the study does not reflect an actual company with normal
accounting procedures, it is evident from the studies that the entirety of
nuclei production is considered a revenue stream for the business. ACA
maintains that the nuclei reported in petitioners' studies satisfy all the
relevant criteria for treatment as co-products, and the cost of production
of honey should reflect this reality. ACA argues the costs of producing
honey and nuclei should therefore be allocated according to their relative
sales value. 

ACA asserts that if the Department determines not to treat nuclei as co-
products, it should at least recognize the nuclei as by-products of honey
production. According to ACA, by-products inevitably result from the joint
production processes. ACA contends that whether by-product or co-product,
it is appropriate for the Department to consider nuclei revenue as
relevant to the cost of producing honey. 

ACA argues that the Department's use of 18.06 percent profit to calculate
constructed value (CV) is incorrect. ACA states that the cost studies
relied upon by petitioners have so inflated costs that only one of the
five studies shows a profit. According to ACA, correcting only the basic
errors in the cost studies would result in positive profits in additional
studies. ACA notes that, with respect to the November 1999 study, for
example, correcting for the double counting of nuclei costs, the incorrect
drum price, nuclei feeding costs, nuclei health costs, or eliminating the
inexplicable administrative costs, would produce positive profit for the
hypothetical beekeeper. According to ACA, petitioners' studies assume that
all nuclei are sold for purposes of calculating profits, resulting in an
extraordinarily high profit percent. 

Petitioners state that, while ACA asserts that nuclei are a co-product or
a by-product of the production process for honey, ACA has failed to submit
any evidence to support its claim. Petitioners indicate that, according to
the declaration filed by the President of Adee Honey Farms (one the
largest honey producers in the United States), the standard accounting
procedure for all honey producers in the United States is to treat nuclei
as a direct materials cost (see petitioners' Letter of June 21, 2001,
Attachment 1). Petitioners state that honey producers in the United States
account for nuclei in this manner because nuclei are required to replace
dead and queenless colonies. Therefore, according to petitioners, honey
producers must produce nuclei in order to maintain their bee population
and to continue producing honey. Petitioners maintain that the record
demonstrates that the treatment of nuclei in the 1999 Gestion Apicola cost
studies are consistent with the treatment of nuclei by U.S. honey
producers, and that nuclei are not considered to be a co-product of honey
within the domestic industry.

Petitioners find it surprising that ACA failed to submit accounting
records for any of the many beekeepers from which ACA receives honey to
support its claim that nuclei are properly accounted for as a co-product
or as a by-product within the honey industry. Petitioners state that ACA's
proposal that the Department revise the Gestion Apicola cost studies to
assume that a beekeeper sells each year a number of nuclei equal to half
the number of existing hives in operation is unreasonable and contrary to
reality. Petitioners assert that ACA's proposal would result in a country-
wide growth rate of about 50 percent each year, far greater than the
actual growth rate in the total number of hives. 

Petitioners contend that ACA's assertions that the Gestion Apicola cost
studies rely on erroneous assumptions about beekeeping in Argentina are
based almost entirely on an affidavit from an Argentine government
official who claims no personal experience in keeping bees or operating a
honey production business. Petitioners assert the statements in the
affidavit have little or no independent support, and that the statements
are clearly biased, because they are from an employee of the Argentine
government. 

Petitioners state that ACA provides no support for its claim that the
1999 cost studies overstate the value of queen bees (by assuming that a
typical beekeeper replaces 50 percent of his queens each year). According
to petitioners, the 50 percent queen replacement rate experienced by the
Gestion Apicola cost studies approximates the 60 percent replacement rate
used by U.S. beekeepers. Petitioners question how ACA's expert's ten
percent queen replacement rate accounts for the queen bees needed for the
production of nuclei. Petitioners also contend that ACA's assertions that
the 1999 studies overstate the cost of drums for honey storage are
unsupported, except by the single invoice for a single sale provided by
ACA. Petitioners note ACA's claim that the cost studies erroneously
attribute identical feeding and health care costs to full hives and to
nuclei is similarly unsupported. Petitioners assert in ACA's eagerness to
disclaim the cost studies, ACA overlooks the fact that nuclei consume more
feed per bee than full hives. 

Petitioners state ACA's assertions with respect to costs in the Gestion
Apicola studies for wax replacement, honey extraction, and administration
expenses are all unsupported by record evidence and claim that statements
in ACA's expert's affidavit are self-serving. Petitioners also argue ACA's
claim regarding the labor hours needed to build each hive box distorts the
information in the studies. Petitioners maintain that none of ACA's
allegations detracts from the fact that the 1999 cost studies remain the
only source of independent, unbiased and contemporaneous cost information
in this case.

Petitioners argue the Department should not reallocate costs in the
manner requested by ACA because nuclei are not co-products or even by-
products of the production process for honey. Petitioners note that the
domestic industry's treatment of nuclei costs as direct material costs is
consistent with the fact that nuclei are necessary to maintain production
of honey from year to year. According to petitioners, a honey producer
must constantly produce nuclei in order to maintain a consistent bee
population and to maintain production of honey at existing levels. For
this reason, nuclei are appropriately considered to be a direct material
cost. Petitioners also state that, although ACA claims that the cost
studies must be adjusted to conform to GAAP, ACA failed to provide any
citations to GAAP that require nuclei to be treated as a co-product or by-
product; they also failed to provide any information to show that
beekeepers anywhere in the world actually record nuclei as co-products or
as by-products in their accounting records. 

Petitioners rebut ACA's assertions that the author of the 1997 cost
studies did not apply the same methodology in the 1999 cost studies by
asserting that the author subtracted an amount for revenue from
"subproducts" in both the 1997 and 1999 studies. Petitioners note that the
Department should also keep in mind that it was ACA that intervened in the
Department's attempt to obtain actual COP data from the selected Argentine
beekeepers, and that it was ACA's counsel who decided unilaterally that
the beekeepers's cost data were not reliable. According to petitioners,
ACA withheld from the Department the data that may have revealed whether
Argentine beekeepers treat nuclei as co-products or by-products, or as
cost of direct materials. For these reasons, petitioners assert, the
Department should rely on the 1999 cost studies published in Gestion
Apicola for the final determination.

Petitioners also argue that the 1997 cost studies do not undermine the
credibility of the 1999 studies, as asserted by ACA. Petitioners state
that despite ACA's criticism of the 1999 studies, the 1997 cost studies
confirm the reliability of the 1999 cost studies. According to
petitioners, not only is the average COP reported in the 1997 cost studies
greater than the average COP reported in the 1999 studies, but the highest
monthly cost figure calculated in the 1997 studies exceeds the similar
figures for 1999. Petitioners state that the Department should continue to
rely on the 1999 studies not only because the findings in those studies
are corroborated by the 1997 studies, but because the 1999 studies
represent cost information from an independent, reputable source that is
the most contemporaneous with the POI. 

Petitioners note that having failed to submit requested cost information,
respondents have no standing to propose that the Department apply an
alternative methodology to calculate the cost of producing honey in
Argentina. Petitioners state that ACA's request that the Department rely
on its acquisition prices as a surrogate cost of producing honey should be
rejected. Petitioners note that the Department has already declined to
adopt the suggestion that it rely on the exporters' acquisition price as
the COP. They state the Department should continue to reject this proposal
because a decision to rely on acquisition prices would reward the
respondents for their refusal to respond to the Department's
questionnaire, and ACA's assertion that the Department's determination in
Live Cattle from Canada provides precedent in support of a switch in
methodologies is baseless. Contrary to the impression left by ACA in its
case brief, petitioners argue that the Department relied on acquisition
price as a surrogate for COP in only very limited circumstances in Live
Cattle from Canada. Petitioners maintain that the respondents' failure to
submit any independent cost data contemporaneous with the POI suggests
that the 1999 cost studies are the most favorable cost studies to the
respondents available. 

Department's Position: We agree with petitioners. We calculated the cost
of production for honey in the preliminary determination based on the 1999
Gestion Apicola cost studies submitted by petitioners in the original
petition. The cost studies were obtained from a publicly available
publication for the Argentine honey industry and prepared by the same
individual who prepared the 1997 cost studies submitted by respondent ACA.
We note that the cost studies are based on information regarding honey
production in different regions in Argentina. The cost studies provide
cost of production data which includes direct materials, labor expenses,
overhead expenses, and general expenses. In addition, the cost studies
provide average sales revenue and net income (or loss) figures.

Regarding ACA's claim that the Department should revise the cost studies
to reflect by-product treatment for joint product costs, we disagree.
While the studies prepared by the same author in 1997 may have relied on
by-product treatment (that is, reducing the total costs of the primary
product by the revenues from the sales of the by-product) for nuclei, the
studies systematically used a co-product allocation for 1999. The fact
that the author changed the cost allocation methodology does not appear to
be unreasonable as the value of nuclei production changed significantly
from 1997 to 1999. The relative significance of the joint products
produced is one of the primary factors used by the Department in
determining whether to treat a product as a by-product or co-product. See,
Memorandum from Richard W. Moreland to Faryar Shirzad, September 14, 2001,
(Pages 9 and 10).

The methodology used to allocate total production costs, as explained in
the proprietary attachment to the studies, demonstrates that the author
reduced the honey and nuclei to the same basis using a common denominator.
According to Management Accountants' Handbook, Fourth Edition, by Keller,
Bulloch, and Shultis, (Chapter 11, page 11) "joint costs are distributed
to products on the basis of some physical measure. That is, the joint cost
is broken up in proportion to the raw materials contained in each
product...If the joint products are not directly measurable, some common
denominator may be used." We do not have any reason to conclude that the
costs provided and the allocation methodologies used in the cost studies
are not reflective of the honey producers' costs in Argentina, or the
industry in Argentina as a whole.

The Department's longstanding practice is to rely on a respondent's
normal books and records if they reasonably reflect the cost of production
of the subject merchandise. See, Final Determination of Sales at Less Than
Fair Value: Canned Pineapple Fruit from Thailand, 60 FR 29553 (June 5,
1995) (Comment 6). In doing so, we look first to how the company maintains
its own books and records and determine the reasonableness of their
allocation methodologies. However, in this case, the honey producers did
not provide the Department with the requested honey cost of production
information. In the absence of each producer's actual costs and allocation
methodologies, we continue to believe that these studies represent the
most relevant, contemporaneous, and specific data available on the record
of this proceeding for purposes of determining the cost of production for
honey. As the study appears to capture all costs related to producing
honey, and appears to allocate these costs to all products produced using
a reasonable methodology, we have relied on the study as non-adverse facts
available for the final determination. See the facts available section. 

We disagree with respondents' renewed proposals to use acquisition price
as a surrogate for the cost of producing honey in Argentina. Consistent
with our findings in Final Determination of Sales at Less Than Fair Value:
Fresh and Chilled Atlantic Salmon from Norway, 56 FR 7661 (February 25,
1991) (Salmon from Norway), we do not believe that acquisition prices are
appropriate to our COP analysis. In determining whether exporters' sales
are made at less than the COP, the Department examines the "cost of
producing the merchandise," in accordance with section 773(b) of the Act.
In this case, we have found that the "cost of producing" honey in
Argentina includes the sum of the beekeepers' COP, plus an amount for the
exporters' selling, general, and administrative expenses. ACA's claim that
the Department's determination in Live Cattle from Canada provides
precedent for the use of acquisition price in this case is misguided. In
Live Cattle from Canada, and in contrast to the present investigation, the
producers of traded cattle responded to the Department's questionnaire,
and three of the five were verified. The Department determined as a result
of verification that the selected cattle producers did not keep books and
records that allowed the Department to confirm the accuracy of their
product-specific costs. In addition, the Department used the average loss
of those producers to adjust the traded cattle acquisition price. In the
present case, the beekeepers failed to provide the requested cost
information. Therefore, the Department was never able to reach the issue
of whether or not reported costs were fully verifiable. 

In the preliminary determination, we based CV profit on the September
1999 cost study contained in the petition. We have further reviewed our
methodology with respect to profit, and revised our calculation to include
information from all five cost studies which were used to calculate COP.
Specifically, we have calculated an average profit. See Cost Analysis
Memorandum, September 26, 2001 (on file in the CRU, room B-099 of the main
Commerce building). 

Consequently, for purposes of the final determination, we continue to
base COP and profit for CV on the average of the 1999 studies from Gestion
Apicola. 

Comment 3: Middlemen Reseller Expenses

Petitioners assert that the Department's preliminary determination
understates the average COP and constructed value of Argentine honey
because the Department failed to include the selling, general and
administrative expenses and profit of the "middleman" resellers, often
referred to as brokers and collectors, or acopiadores. Petitioners contend
that, according to the SAA accompanying the Uruguay Round Agreements Act
of 1994, Commerce should calculate constructed value based on the total
profit and total SG&A expenses incurred by producers and exporters, and
failure to do so would result in an understatement of the true COP and
constructed value of the merchandise. Citing the January 11, 2001 Radix
Letter at 3 and the "ACA Verification Report", Exhibit 10 at 1 dated July
27, 2001 (on file in the CRU, room B-099 of the main Commerce building),
petitioners assert that the small number of exporters in Argentina in
comparison to the vast number of honey producers confirms the exporter's
reliance upon "middlemen". Petitioners contend that the "middlemen"
purchase the majority of honey produced directly from honey producers, and
transport it to the small number of exporters in Buenos Aires. Petitioners
maintain that the Department should calculate a COP and CV which reflect
all expenses actually incurred on honey exported to the United States and
Germany, including the expenses and profit of these middlemen. 

ACA contends that petitioners greatly exaggerate the role of middlemen in
the production and sale of honey. ACA asserts that honey does not undergo
a refining process performed by "middlemen", although it does undergo a
heating process performed by honey packers after the raw honey has been
sold by ACA. ACA maintains that the "middlemen" do not remove the honey
that is already contained in a drum. ACA states that, according to
petitioners' studies, the cost of the drum purchase is included in the
beekeepers' costs. (4) The respondent insists that the role of the
"middleman" is to negotiate a sale to the exporter, which often means that
the middleman does not bring the honey to the exporter, but merely
instructs the beekeeper where to deliver a certain quantity of honey. ACA
further asserts that the Department failed to request data regarding
middlemen, even though ACA recommended that the Department obtain cost
information from one of ACA's member cooperatives which would have
provided verifiable data regarding its use of middlemen. ACA contends
that, if the Department imposes an unverified commission cost, it would
constitute adverse facts available. 

Department's Position: We agree with petitioners that "middlemen" costs
should be included in our final analysis in order to construct a COP
inclusive of all expenses incurred by producers and exporters of honey. In
fact, in allocating a portion of the G&A and selling expenses of ACA and
Radix to COP, we believe that any commissions paid to brokers of honey are
captured in our final margin calculations. Because we also believe that
there is no conclusive evidence on the record that "middlemen" costs are
not already reflected in the COP information (i.e., the cost studies), we
have not made an additional adjustment in our final margin calculations
for "middlemen" costs. In doing so, we note that we intend to examine this
issue fully in any subsequent proceedings.


Radix

Comment 4: Facts Available

Petitioners object to the Department's decision to grant Radix's request
to re-enter the investigation, to verify the information placed on the
record by Radix, and to apparently issue a final determination for Radix
in which this information is used to calculate a dumping margin.
Petitioners express their concerns that the Department's issuance of a
preliminary margin analysis shortly before the hearing highlights the need
for the Department not to establish a precedent that would enable
respondents to withdraw and subsequently re-enter investigations following
the issuance of a preliminary determination. Petitioners argue that the
Department should therefore continue to calculate a dumping margin for
Radix on the basis of facts available for purposes of the final
determination.

Radix states that petitioners' request ignores the fact that the
Department has already made its decision on this matter, and that the
statute and the Department's regulations require it to use Radix's
verified information (see section 782(e) of the Act and 19 CFR 351.308).
For these reasons, according to Radix, petitioners' request should be
rejected.

Department's Position: We disagree with petitioners. On May 1, 2001,
three days before the preliminary determination was issued in this case,
Radix informed the Department that it wished to withdraw from the
investigation. However, on May 11, 2001, ten days following this request,
Radix asked the Department's permission to re-enter the investigation,
which the Department granted on June 12, 2001. Our decision to allow Radix
to re-enter the proceeding is consistent with the Department's general
preference to calculate margins using the parties' proprietary
information, which serves the statutory goal of transparency, accuracy and
fairness. Moreover, we note that Radix's request to re-enter the
investigation came only ten days following the company's withdrawal,
thereby giving the Department ample time to prepare for and conduct
verification, and to perform the preliminary margin calculations using
Radix's data. See Preliminary Determination of Sales at Less Than Fair
Value: Butt-weld Pipe Fittings from the Phillipines, 65 FR 47393 (August
2, 2000) (the Department calculated the margin of a voluntary respondent
after the preliminary determination was issued). As a result, in
accordance with the statute and our regulations, we will rely on all
verified information for purposes of calculating Radix's dumping margin in
this investigation.

Comment 5: Testing Expenses 

Petitioners argue that Radix's testing expenses, claimed as direct
selling expenses in the German market, should be treated as indirect
selling expenses in all markets. Petitioners contend that Radix failed to
demonstrate that its claimed testing expenses were incurred only on sales
of honey exported to Germany. Petitioners assert that although Radix
claimed that it incurred testing costs on its German sales, all honey of
50 mm or darker that is received during the summer months is tested, and
the honey which fails the tests is sold to other markets (i.e., the United
States, France, etc). Petitioners claim that Radix did not provide the
Department with any evidence indicating that specific testing expenses
were directly related to specific German sales. According to petitioners,
the verification report demonstrates that Radix could not directly trace
particular tests to corresponding invoices because of the manner in which
the testing and inventory procedures are conducted within the warehouse.
See Radix's Verification Report at 25. Petitioners cite Radix's
Verification Report, however, claiming that the cash account sub-ledgers
for a particular testing provider indicate that payment for testing was
made by Radix during and prior to a month(s) in which no honey sales were
made to Germany. 

Petitioners assert that as Radix does not accept orders more than 90 days
in advance of the shipment date, the testing expenses recorded must have
been for honey sales to markets other than Germany. Petitioners also claim
that as the number of testing invoices and the number of honey invoices
are not comparable, some of the testing invoices must have included honey
that was eventually sold to markets other than Germany, resulting in an
overstatement of Radix's German testing expenses. Consequently, according
to petitioners, the Department must continue to treat these expenses as
indirect selling expenses in the final determination.

Radix asserts that as only German importers require special antibiotic
and phenol testing, the expensive tests related to these residues are not
performed on products destined for other markets. Radix maintains that it
correctly allocated the antibiotic and phenol testing expenses to German
sales. Radix cites the Department's Verification Report, claiming that the
Department verified that the testing expenses were related to honey sales
to Germany and that the tests at issue were only required by German
customers. Radix claims that the Department in referring to the tests as
"German testing" and by stating that "the costs of the testing are
directly related to German sales because the tests are not required for
other markets", confirmed that Radix had demonstrated the direct nature of
the testing expenses. See Radix Verification Report at 25. With respect to
the petitioners' argument that honey which fails the German standards
tests is then sold to other markets, Radix states purity testing is always
intended to certify qualifying products. Radix also notes that the testing
process does not improve Radix's ability to sell the honey to other
markets. Radix contends that even though all darker honey is tested
immediately in the summer months, only the honey which passes the tests is
then sold to Germany. Radix explains that, during the busy season, it
cannot wait to perform the testing until it receives an order from a
German customer. Radix maintains that the essential issue here is that the
testing is required only for sales to Germany, and that only drums which
pass the standards tests are sold to the German market. Radix insists that
drums which pass the standard tests are never sold to other markets
because the tests are considerably expensive and the selling of such drums
to other markets would in effect be "wasting" the qualifying drums and
testing expenses. Radix disagrees with petitioners' claim that Radix paid
the laboratory for German testing in July, even though there were no sales
to Germany in July, by asserting that the ledger petitioners refer to
shows the date of payment of the testing invoice and not the date of
rendered services. Radix explains that it is not unusual for Radix to pay
in July for testing provided in April. Radix notes that, although
petitioners allude to a discrepancy between the number of testing invoices
and the number of sales invoices to Germany, Radix explains that there is
not a one-to-one relationship between testing invoices and sales invoices,
since each sales invoice may contain 90 to 400 drums of honey, and drums
are testing in lots of 60. 

Radix argues that the Department should, if it continues to determine
that testing expenses are indirect selling expenses, allocate the testing
expenses over all sales and not only over sales to Germany.

Department's Position: We agree with petitioners. Although Radix reported
expenses associated with sampling and/or testing required by German
customers as direct selling expenses attributable to German sales only,
the Department has determined that the testing expenses at issue are more
appropriately classified as indirect selling expenses. 

Direct selling expenses are typically expenses that are incurred as a
direct and unavoidable consequence of the sale (i.e., in the absence of
the sale these expenses would not be incurred). In other words, while
indirect expenses constitute fixed expenses that are incurred whether or
not a sale is made, direct selling expenses are expenses which can vary
from sale to sale, and result from and bear a direct relationship to the
particular sale in question. See 19 CFR 351.410(c); Oil Country Tubular
Goods From Mexico; Final Results of Antidumping Duty Administrative
Review, 66 FR 15832 (March 21, 2001); and Canned Pineapple Fruit From
Thailand: Final Results of Antidumping Duty Administrative Review, 65 FR
77851 (December 13, 2000).

The evidence on the record indicates that Radix's testing expenses are
more properly classified as indirect selling expenses, given that the
expenses are often incurred whether or not the product being tested is
eventually sold to the German market. During verification Radix explained
that for the first months of summer, it performs German standard tests on
all lots of honey darker than a certain color (i.e., 50 mm on the pfund
scale). See Radix Verification Report at 26 dated July 26, 2001. It is
clear from the record evidence, including the verification report, that
honey which is tested but which does not meet German standards is shipped
to other markets. Under Radix's reporting methodology, those testing
expenses would still be allocated only to German sales even though the
merchandise was sold to another market. 

We find that the information on the record regarding expenses incurred in
sampling and/or testing honey for German customers does not unequivocally
demonstrate that these expenses result from and bear a direct relationship
to the sales in question within the meaning of 19 CFR 351.410(c) and the
Department's practice. Rather, the evidence indicates that these expenses
appear to have the characteristics of indirect selling expenses associated
with sales in all markets in that they are attributable to sales other
than sales to Germany. Accordingly, for the final determination, we will
continue to treat Radix's sampling and/or testing expenses as indirect
selling expenses. 

The Department, however, does agree with respondents that the expenses
should be allocated over sales to all markets and has done so for this
final determination.

Comment 6: General and Administrative and Indirect Selling Expenses

Petitioners request that the Department revise Radix's general and
administrative (G&A) indirect selling expenses to include all such
expenses incurred during the POI. Petitioners claim that such expenses
incurred by Radix should be included within the calculation of COP or
subtracted from the reported gross sales prices. Petitioners assert that
Radix failed to include "other product-specific indirect selling expenses"
and "interest expenses" in its reported sales databases. (5) These
expenses, which were recorded in Radix's audited income statement, were,
according to petitioners, not accounted for in Radix's sales databases.
Petitioners contend that Radix's attempt to exclude those expenses was
improper because the SAA requires that total profit and total expenses be
used to calculate constructed value. Petitioners request that the
Department revise the reported general and administrative expenses and
determine an indirect selling expense ratio which accounts for all of
Radix's selling expenses.

Petitioners assert that Radix's allocation methodology shifts costs away
from the subject merchandise to other products, resulting in an
understatement of indirect selling expenses. Petitioners contend, based on
the Radix Verification Report at 6 (on file in the CRU, room B-099 of the
main Commerce building), that the percentage of total sales accounted for
by honey should be used to allocate a portion of G&A expenses and indirect
selling expenses to sales of honey. Petitioners claim that Radix's
allocation of expenses is not appropriate because the Department's normal
practice is to calculate a G&A expense ratio by dividing total company-
wide G&A expenses by the total cost of goods sold, and apply the G&A ratio
to the value of individual sales. Petitioners also claim that Radix's
allocation methodology does not accurately reflect operations, asserting
that purchase commissions, for example, should be allocated to honey based
on the percentage of honey sales and not based on the exact product for
which Radix paid the purchase commission. Petitioners request that the
Department revise the G&A and indirect selling expense ratio used in the
calculation of CV.

Radix argues that all expenses are accounted for in the sales response,
as movement expenses, direct expenses, or indirect expenses. Radix claims
that it has allocated expenses as specifically as possible over honey and
other products. Radix notes that although petitioners have objected to
Radix's practice of identifying, when possible, indirect expenses related
to particular products, the Department's questionnaire instructs parties
to allocate expenses on as specific basis possible. Radix states that
petitioners' complaint that Radix's "other product-specific" expenses are
not allocated over total sales, is inherently incorrect. Radix further
notes that petitioners' complaint that interest expenses are not included
in indirect selling expenses is similarly incorrect, as these expenses are
already included in its imputed credit expense.

With respect to petitioners' arguments regarding the G&A expenses, Radix
states that these expenses relate to COP and CV, and CV is strictly a
normal value calculation, not a U.S. price. According to Radix, the G&A
portion of normal value is based on facts available and has no
relationship to Radix's U.S. sales response or its U.S. price calculation.
Radix maintains it is a selling company, and not a manufacturing company;
therefore, all of its expenses are selling expenses, and all of its
general expenses are indirect selling expenses. Radix argues that all of
those expenses have been allocated to Radix's sales, and there is no
additional expense left over to constitute G&A in need of further
allocation. 

Department's Position: We agree, in part, with both petitioners and
Radix. We have further examined Radix's reported G&A expenses since
verification, and found that Radix has accounted for the majority of G&A
expenses in its reported indirect selling expense calculation. Based on
Radix's fiscal year 2000 expenses, we have calculated a G&A ratio using
the cost of goods sold derived from Radix's audited financial statement,
and applied this ratio to COP in order to account for all remaining G&A
expenses. See "Final Cost Analysis Memorandum", September 26, 2001 (on
file in the CRU, room B-099 of the main Commerce building). With respect
to the petitioners' concerns regarding Radix's allocation methodology for
indirect selling expenses, based on our findings verification, we believe
that Radix's calculation methodology is reasonable because Radix allocated
its indirect selling expenses to honey and to other products where it was
able within its own accounting records. Therefore, for purposes of the
final determination, we have continued to rely on Radix's reported amounts
for these expenses.

Comment 7: Sales Reconciliation

Petitioners claim that the prices reported in Radix' sales listing do not
reconcile with its audited financial statements. Petitioners assert that
the Department attempted to reconcile the gross sales prices Radix
reported for its U.S. and German sales to Radix's audited financial
statement. Petitioners claim, however, based on their interpretation of
the verification report, that the gross sales prices reported by Radix are
not consistent with total revenues recorded in its audited income
statement, resulting in inaccurate reported sales prices during the POI.
Petitioners contend that, if Radix's reported FOB value reconciles to its
audited income statement, the FOB value represents the actual C&F sales
price because Radix sells on a C&F basis. Therefore, according to
petitioners, the C&F price should be shown on the audited income
statement, representing actual sales revenue. Citing Verification Exhibit
at 5, petitioners claim that Radix incorrectly reported its sales prices
in its sales listing. Petitioners assert that Radix did not explain the
discrepancy and has thus failed to complete a fundamental requirement of
verification. As a result, petitioners argue, the Department should use
adverse facts available in calculating Radix's dumping margin for the
final determination.

Radix maintains that petitioners' arguments result from a misreading of
the verification report. Radix notes that, at verification, total FOB
sales prices reported to the Department matched exactly with the total
values in Radix's sales ledgers. Radix contends that, although there was a
calculation error regarding certain freight charges that resulted in an
apparent variance between the C&F sales values and the ledger, the error
was identified and corrected during verification. According to Radix, a
slight difference, the result of a rounding error, amounting to less than
0.0096 percent of the U.S. sales value, is clearly within a reasonable
tolerance for rounding errors. 

Department's Position: We disagree with petitioners. At verification, we
found that Radix's C&F sales prices, less international freight charges,
equals total FOB sales value. As we note in Radix's Verification Report at
7, we tied the total FOB value of all sales during the period directly to
Radix's audited financial statement. Because Radix sells on a C&F basis,
freight income (and expense) are recorded separately in its accounting
records. During verification, we selected numerous sample sales over
multiple periods, and tied freight income directly to the general ledger
(see Radix's Verification Report at 9). We noted that freight income in
all instances was properly accounted for in Radix's general ledger. We
also noted that freight income was correctly accounted for in Radix's
audited income statement. Therefore, we do not agree that any adverse
inference against Radix is warranted with respect to the reconciliation of
total quantity and value of reported sales. 


Comment 8: Reembolso/Reintegro Reimbursements

Radix argues that the Department should increase Radix's U.S. price by
the amount of the reintegro benefit provided by the Argentine government.
Radix states that, in the preliminary determination the Department
adjusted ACA's sales prices for reembolso/reintegro reimbursements, and
that the Department should make the same adjustment to Radix's U.S. price
in the final determination. 

Radix indicates that the Department is familiar with the reintegro
program. It notes that, in the preliminary countervailing duty (CVD)
determination of honey from Argentina, the Department described the
program as one which entitles Argentine exporters of honey produced in
Argentina to a rebate of many internal or domestic taxes that are levied
during the production and distribution process in Argentina on the
finished export product (see Honey from Argentina: Preliminary Affirmative
Countervailing Duty Determination and Alignment with Final Antidumping
Duty Determination on Honey from the People's Republic of China, 66 FR
14521, 14524 (March 13, 2001)). Radix also notes that the Department
indicated that prior to April 2000 the reintegro reimbursement rate was
4.1 percent for bulk honey and from April 2000 the rate increased to 5.4
percent. Radix states that, in the preliminary margin determination the
Department properly added the reintegro benefit received by ACA to ACA's
U.S. price. It contends that the Department should make a similar
adjustment to Radix's U.S. price.

Radix maintains that it did not report a variable for the reintegro
benefit in its questionnaire response, as did ACA, because the
Department's questionnaire does not request such information. Radix notes,
however, that all the information necessary to calculate the amount for
Radix's reintegro reimbursement is available on the record. Radix also
indicates that the Department confirmed at verification that the reintegro
was applied to Radix in the same manner as described in the preliminary
determination. Radix states that, if the Department does not adjust
Radix's U.S. price to account for the reintegro reimbursement it must
reduce Radix's overall antidumping rate by the amount of the reintegro
benefit determined in the CVD investigation. Radix cites the Department's
Antidumping Duty Manual (January 22, 1998), Chapter 7, part C.1.c., and
Notice of Final Determination of Sales at Less Than Fair Value: Stainless
Steel Wire Rod from Italy, 63 FR 49327 (July 29, 1998), where the
Department indicated: "Where the product under investigation is also
subject to a concurrent countervailing duty investigation, we instruct the
Customs Service to require a cash deposit or posting of a bond equal to
the weighted-average amount by which the NV exceeds the EP, as indicated
below, minus the amount determined to constitute an export subsidy."
Therefore, according to Radix, rather than adding the reintegro amount to
the U.S. price, the Department could alternatively subtract the amount of
the reintegro determined to constitute a subsidy from the cash deposit or
bonding rate. 

Petitioners agree with Radix that the Department should apply a
methodology that ensures that the reintegro benefits received by Radix
with respect to its exports of honey to the United States are not double-
counted. Petitioners note that, while Radix's proposal to increase its
U.S. price by the amount of the reintegro benefit was used by the
Department in the preliminary determination with respect to ACA, the
Department's past practice is to reduce the antidumping duty cash deposit
rate by the amount of the reintegro benefit. Petitioners state that, while
both methodologies proposed by Radix will ultimately accomplish the goal
of ensuring that the benefits received by Argentine honey exporters are
not double-counted, the methodology adopted by the Department should be
consistent for all respondents and facilitate the collection of cash
deposits by the U.S. Customs Service. 

Department's Position: We agree with both Radix and petitioners that for
purposes of the final determination an adjustment should be made for the
reembolso/reintegro benefits received by the Argentine honey exporters.
Upon further review of our methodology used in the preliminary
determination, we agree with petitioners that the more appropriate means
through which to prevent double counting in instances where the product
under investigation is also subject to a concurrent countervailing duty
investigation is to reduce the antidumping duty cash deposit by the amount
of the benefit (e.g., see Notice of Final Determination of Sales at Less
Than Fair Value: Stainless Steel Wire Rod from Italy, 63 FR 49327 (July
29, 1998)). Therefore, for purposes of the final determination, we have
not made an adjustment to U.S. price for either ACA or Radix in the margin
calculation program. Instead, for all Argentine honey exporters, including
ACA and Radix, we will instruct Customs to require a cash deposit or
posting of a bond equal to the weighted-average amount by which the NV
exceeds the EP for both companies, less the amount determined to
constitute an export subsidy in the concurrent CVD investigation. 


Comment 9: Interest Expenses

Petitioners state that the Department should revise its calculation of
constructed value to include interest expenses incurred by Radix.
According to petitioners, in accordance with the SAA at 841, the
Department's calculation of constructed value should include the expenses
and profits of Radix, as well as the expenses and profits of the honey
producers. Petitioners assert that Radix did not include interest expenses
in its reported indirect selling expenses or in the expenses accounted for
in its sales database. Petitioners therefore maintain that the Department
should include interest expenses in calculating constructed value for
Radix in the final determination.

Radix argues that, in the preliminary analysis, the Department used
punitive facts available to determine normal value, and that the adverse
information used by the Department was provided by the petitioners
themselves. Radix contends that, the petitioners' request that the
Department revise their own punitive data should be rejected, given that
one entire year has passed since the submission of the data on which the
Department relied.

Department's Position: We agree with petitioners that the Department's
preliminary calculation of constructed value (CV), which was based on the
producers' COP plus the exporter's SG&A expenses and an amount for profit,
did not include an amount for interest expenses incurred by Radix. Based
on our findings at verification, we also agree with petitioners that
interest expenses are not included in Radix's reported indirect selling
expenses or elsewhere in its reported sales databases. As stated in the
SAA, the Department's calculation of CV should include the expenses and
profits of exporters, as well as the profits of the producers (see SAA at
841, "....Commerce may continue to calculate constructed value on the
total profit and total SG&A expenses realized and incurred by both
[producer and exporter]"). Therefore, for purposes of the final
determination, we have calculated an interest expense ratio for Radix
using its March 31, 2000, audited financial statement (see Final Cost
Analysis Memorandum, dated September 26, 2001). 


Comment 10: Inventory Carrying Costs

Petitioners state that the Department should include inventory carrying
costs in CV. Petitioners reference Import Administration Policy Bulletin
94.2 in arguing that inventory carrying costs should be reported and
included in calculating CV in cases where CV is used. Petitioners maintain
that Radix failed to report inventory carrying costs in its questionnaire
responses and the Department did not include these expenses in its
calculation of CV in the preliminary determination. According to
petitioners, the Department should therefore use information in Radix's
financial statement and questionnaire response to calculate inventory
carrying costs.

Radix argues that the Department does not include imputed inventory
carrying costs in CV. Radix states that petitioners' cite an outdated
policy memorandum from 1994. Radix maintains that the Department's
Antidumping Manual states that, because CV selling costs are based on
actual amounts, imputed expenses (e.g., credit and inventory carrying
costs) are not included in the CV (see Antidumping Manual, Chapter 8, page
57). Radix further notes that this policy has been followed by the
Department in previous cases (see, e.g., Notice of Final Determination of
Sales at Less Than Fair Value: Stainless Steel Plate in Coils From
Belgium, 64 FR 15476, 15489 (March 31, 1999)) (Stainless Steel from
Belgium).

Department's Position: We agree with Radix, and with petitioners, in
part. In the present case, NV for Radix is based on CV. Section
773(e)(2)(A) of the Act specifies that the calculation of CV shall include
"the actual amounts incurred and realized by the exporter or producer
being examined in the investigation or review for selling, general, and
administrative, and profits ...." Although petitioners are correct in
noting that Radix did not report inventory carrying expenses in its sales
listings to the Department, we agree with Radix that it is the
Department's practice not to include imputed expenses in the calculation
of CV (see Stainless Steel from Belgium). Therefore, for our final
determination, we have not included inventory carrying expenses in the
calculation of CV for Radix.


ACA

Comment 11: Indirect Selling Expenses 

Petitioners argue that ACA applied a volume-based methodology in
calculating indirect selling expenses, instead of a value-based
methodology, which is the Department's normal practice. Petitioners claim
that ACA calculated a percentage of honey sales by volume and allocated
indirect selling expenses based on that percentage, with the exception of
salaries. Petitioners claim that ACA has provided no reason for deviating
from the Department's normal practice. Petitioners also contend that by
using a volume-based methodology, the respondent is allocating expenses
away from honey, while a value-based methodology results in a more
reasonable and fair match between expenses and sales revenues. Petitioners
urge that the Department recalculate ACA's indirect selling expenses based
on a value-based methodology using information contained in ACA's audited
financial statement.

ACA argues that it calculated its indirect selling expenses over volume,
and not value, because a significant amount of the work performed by the
export department is volume-dependent (i.e., arranging for and documenting
shipments, arranging inspections of merchandise, coordinating warehousing
and movement from warehouse to dock). ACA notes that it based its
calculation on the amount of honey purchased and received by the export
department during the POI. ACA indicates that petitioners calculated an
indirect selling ratio based on the total sales volume recorded in ACA's
accounting system, and as provided in ACA's Verification Report at Exhibit
7. ACA states that, although this data is a less accurate reflection of
the true economic situation, it is not inaccurate or false. Assuming that
Department rejects its volume-based allocation of indirect selling
expenses in favor of a value-based methodology, ACA urges the Department
to use the value-based calculation reviewed by the Department at
verification in Exhibit 24 of ACA's Verification Report. 

Department's Position: We agree with petitioners. The Department's normal
practice is to calculate indirect selling expenses based on a value-based
rather than volume-based methodology. See Notice of Final Determination of
Sales at Less Than Fair Value: Stainless Steel Round Wire From Canada, 64
FR 17324, 17329-30 (April 9, 1999). Therefore, we have recalculated ACA's
indirect selling expenses based on the reported selling expenses of ACA's
export department and the total sales value of honey during the POI (see
Final Sales Analysis Memorandum, dated September 26, 2001). 


Comment 12: General and Administrative Expenses

Petitioners claim that ACA's country-wide G&A expenses were not included
in the Department's preliminary analysis. See Preliminary Determination at
6 and ACA's Verification Report at 33. Petitioners cite the SAA, which
instructs the Department to include total profit and SG&A expenses
incurred by producers and exporters to calculate a fully-absorbed COP.
Petitioners argue that the Department should calculate a G&A expense from
the audited financial statement because, according to petitioners, ACA
informed the Department that its country-wide expenses could be obtained
from the audited financial statement. Petitioners also contend that ACA
did not submit its financial statement in its entirety, although ACA
claimed that it had. 

Petitioners state that ACA's submitted information in its case brief
regarding the calculation of G&A is not in accordance with the
Department's normal practice, nor in accordance with Generally Accepted
Principles of Accounting (GAAP). According to petitioners, the
Department's normal practice is to calculate the amount for G&A by
dividing the company-wide G&A expenses by the company-wide total cost of
goods sold to obtain a G&A ratio, and to apply the G&A ratio to total cost
of manufacturing. Petitioners state that the Department should calculate a
G&A ratio of 3.66 percent, and an interest expense ratio of 7.72 percent,
of ACA's total cost of goods sold, based on ACA's financial data. 

ACA argues that at no point in the questionnaire process did the
Department seek general and administrative (G&A) expenses from ACA. ACA
states that the data was not requested because it was not necessary. ACA
notes that, while it disputes the credibility of the data used for the
calculation of COP, it does not dispute the basic methodology used. ACA
indicates that, in the preliminary determination, the Department correctly
noted that G&A expenses were included in the total cost calculated by
Gestion Apicola in each study. The Department determined the average COP
of the five studies, and then added ACA's selling expenses, including
indirect selling expenses. ACA notes that, as demonstrated at
verification, indirect selling expenses include administrative expenses
related to sales, such as real estate, secretarial support, gas,
electricity, telephone and supplies. According to ACA, the G&A expenses of
ACA, a large company that sells and manufactures many products, are not
appropriately included in COP or CV. ACA notes nonetheless that the
Department requested at verification that ACA provide data necessary to
calculate G&A expenses. ACA states that the proper calculation of a G&A
ratio per its annual report is as follows: total G&A, less financing
expenses and income taxes, divided by the total economic activity invoiced
to ACA for the fiscal year. 

Department's Position: We agree with petitioners. At verification we
found that ACA's company-wide G&A expenses were not included in its
reported indirect selling expenses, or elsewhere in its databases. See
ACA's Verification Report at 33. Because the SAA instructs the Department
to include total profit and SG&A expenses incurred by producers and by
exporters to calculate a fully-absorbed COP, see SAA at 841, we have
calculated a G&A ratio based on ACA's audited financial statement for
purposes of the final determination (see Final Cost Analysis Memorandum,
dated September 26, 2001). 


Comment 13: Interest Expenses

Petitioners argue that the Department should include an amount for
interest expenses and calculate an interest expenses ratio using
information from ACA's audited financial statement.

ACA argues that interest expenses are already accounted for in its
inventory carrying costs and credit expenses. According to ACA, its
company-wide interest expenses consist mainly of short-term loans or
revolving credit lines to provide pre-financing for sales. To calculate
inventory carrying costs, ACA states that it determined the average
interest rate on these short-term loans and applied it to the average
number of days the honey was in the warehouse. To calculate the credit
costs, ACA states that it determined a daily interest rate based on the
loan most proximate to the sale and multiplied this rate by the average
number of days between invoicing and payment. ACA asserts that allocating
additional interest expenses to honey would count interest expense twice.

Department's Position: We agree with petitioners that we should calculate
an interest expense ratio for ACA. At verification, we found that ACA's
company-wide interest expenses were not included in its reported indirect
selling expenses (see ACA's Verification Report at 33). Regarding ACA's
concerns that interest expense is already included in the calculation of
inventory carrying costs and credit expenses. We note that NV for ACA is
based on CV, and that it is the Department's practice in calculating CV to
use actual interest expense, and not include imputed credit expenses or
inventory carrying costs (see Final Results of Antidumping Duty
Administrative Review: Certain Cold-Rolled and Corrosion-Resistant Carbon
Steel Flat Products From Korea, 63 FR 13170 (March 18, 1998) (Corrosion-
Resistent Steel Products from Korea). Therefore, for purposes of the final
determination, we have calculated an interest expense ratio for ACA based
on its audited financial statement (see Final Cost Analysis Memorandum,
dated September 26, 2001) and excluded by ACA's reported credit expenses
and inventory carrying costs. 

Comment 14: Testing Expenses

ACA argues in its case brief that the Department erred in the preliminary
determination in treating the cost of testing for antibiotics and phenol
as an indirect cost. ACA contends the Federal Circuit courts have analyzed
this issue in depth (see United States v. Torrington, 254 F.3d 1022,
1050), and noted that indirect expenses are "not related to a particular
sale." According to ACA, the cost of testing is only incurred as a direct
result of a German sale. ACA states that testing requirements are spelled
out in contracts, and given that contracts with German customers are open,
as ACA waits for the results of testing from the laboratory prior to
loading the containers for shipment, there can be no doubt that all costs
associated with testing for antibiotics and phenol are incurred as a
direct result of a sale to Germany. ACA maintains that, at verification,
it demonstrated that for each German sale in the POI, testing always
occurred after contracting, and that such expenses can only be considered
direct costs. ACA indicates that should the Department continue to treat
antibiotic and phenol testing as an indirect selling expense, it must
correct the calculation used in the preliminary determination and
distribute testing expenses overall all worldwide sales, rather than over
German sales only. 

ACA notes that in each German contract the exact antibiotic and phenol
contamination tolerances are spelled out in detail. According to ACA, when
it finalizes a German contract, it then orders the laboratory to send a
mixed sample of honey for a lot that satisfies the color/grade
requirements of the contract, and the sample is tested for antibiotics and
phenol. ACA notes that, if the test results show no contamination, the lot
is designated for shipment to the German customer. Consequently, if the
lot shows contamination it is marked as "contaminated" and eventually sold
and shipped to another destination. ACA states that, as the Department
found at verification, testing was performed after the date of contracting
for sale. ACA also notes that all honey darker than grade 34 was sent for
testing during the POI, because during the POI there was always an open
contract for dark honey to Germany which required honey to be tested in
order to fill the contract. ACA asserts that, at verification, it linked
the testing expenses to invoices by demonstrating that the lot
identification number on the test results was the same lot identification
number on the loading instructions to the warehouse and the same number on
the loading report from the warehouse. 

ACA asserts that the petitioners' argument that the testing costs are
incurred on U.S. sales because honey tested and found to be contaminated
with residues is sometimes sold to the United States is incorrect. ACA
maintains that the tests are not performed because the U.S. customers want
antibiotic and phenol laden honey, but because the German customers do not
want such honey. According to ACA, the fact that more than one lot must be
tested to identify a lot suitable for sale to Germany does not diminish
the fact that the tests are performed as a direct result of ACA making a
sale to a German customer. ACA states that if it made no German sales it
would incur no testing costs.

Petitioners state that, in the preliminary determination, the Department
noted that direct expenses are typically expenses incurred as a direct and
unavoidable consequence of the sale (i.e., in the absence of the sale
these expenses would not be incurred). Petitioners maintain that while
indirect expenses constitute fixed expenses that are incurred whether or
not a sale is made, direct selling expenses are expenses which can vary
from sale to sale, and result from and bear a direct relationship to the
particular sale in question. Petitioners indicate that, during
verification, ACA informed the Department that since October 1999 it has
tested honey it purchases by removing two samples from each drum upon
receipt of the honey at its warehouses, and that if a particular drum was
considered acceptable, one of the samples was placed on the lid, and this
sample was then blended with others in the same lot for testing of
antibiotic and phenol levels. Petitioners note that, as a general rule,
after October 1999, if honey was determined to be darker than a certain
color, samples were always drawn and tested. Petitioners contend that the
Department correctly considered the testing expenses as indirect expenses
in the preliminary determination, because ACA incurred testing costs
regardless of whether there was a German sale, as long as the honey was of
a certain color or darker. Therefore, petitioners' assert, in the final
determination, the Department should continue to treat expenses incurred
by ACA to test its honey for antibiotic and phenol contamination as
indirect expenses.

ACA argues that the Department should, if it continues to determine that
testing expenses are indirect selling expenses, allocate the testing
expenses over all sales and not only over sales to Germany.

Department's Position: We agree with petitioners. Although ACA reported
sampling and/or testing expenses, required only by German customers, as
direct selling expenses attributable to sales to the third country market
(Germany), the Department has determined that these expenses are more
appropriately classified as indirect selling expenses. 

Direct selling expenses are typically expenses that are incurred as a
direct and unavoidable consequence of the sale (i.e., in the absence of
the sale these expenses would not be incurred). In other words, while
indirect expenses constitute fixed expenses that are incurred whether or
not a sale is made, direct selling expenses are expenses which can vary
from sale to sale, and result from and bear a direct relationship to the
particular sale in question. See 19 CFR 351.410(c); Oil Country Tubular
Goods From Mexico; Final Results of Antidumping Duty Administrative
Review, 66 FR 15832 (March 21, 2001); and Canned Pineapple Fruit From
Thailand: Final Results of Antidumping Duty Administrative Review, 65 FR
77851 (December 13, 2000).

The evidence on the record indicates that ACA's sampling and/or testing
expenses are more properly classified as indirect selling expenses, given
that these expenses are often incurred whether or not the product being
tested is eventually sold to the German market. In ACA's submission, on
March 26, 2001, ACA reports that since October 1999 it has performed
testing according to German standards on all lots of honey darker than a
certain color ( i.e., 34 mm on the pfund scale). It is also clear from the
record evidence, including the verification report, that honey, which is
tested but which does not meet German standards, is shipped to other
markets (see ACA Verification Report, at 37). Although we found that
testing expenses could be linked to specific invoices by lot
identification number, several lots of honey may be tested before the test
results for one lot is deemed acceptable for sale to Germany (see ACA
Verification Report, at 36). Under ACA's treatment of these expenses as
direct selling expenses, the expenses would be allocated to German sales
when in fact some of the merchandise was sold and shipped to other
markets. 

The record evidence regarding sampling and/or testing required by German
customers does not unequivocally demonstrate expenses associated with
these activities result from and bear a direct relationship to the sales
in question, within the meaning of 19 CFR 351.410(c) and the Department's
practice. Rather, the evidence indicates that these expenses appear to
have the characteristics of indirect selling expenses. Accordingly, for
the final determination, we will continue to treat ACA's sampling and/or
testing expenses as indirect selling expenses. 

The Department, however, agrees with ACA that the expenses should be
allocated over sales to all markets and has done so for purposes of this
final determination.


Comment 15: German Warranty Expenses

Petitioners argue that in its calculation of German warranty expenses,
ACA incorrectly included expenses that are not related to returns. They
maintain that the warranty calculation worksheet provided by ACA at
verification indicates that it included freight and demurrage expenses
related to the re-sale of the rejected honey in the calculations.
Petitioners state that the resale to the United States should be treated
no differently than any other honey ACA sold to its U.S. customers.
Petitioners indicate that, like its other U.S. sales, ACA would incur
transportation expenses for the shipment of the products to its U.S.
customers. According to petitioners, the inappropriate combining of
expenses associated with sales in one market to warranty expenses in the
other market violates the principle accounting rule that requires expenses
to be matched to sales revenues. Petitioners contend that ACA's inclusion
of U.S. movement expenses in the warranty calculations is even more
inappropriate, given that part of the expenses are estimates, rather than
actual amounts. Petitioners argue that the Department should exclude ACA's
estimated expenses in the calculation of warranty expenses for German
sales.

ACA argues that petitioners have misinterpreted ACA's warranty worksheet.
ACA contends that its claimed warranty expenses are limited to those
additional expenses incurred as a direct result of rejection of the
merchandise by the German customer. ACA notes that the freight charges
claimed as warranty expenses were calculated by summing the original
freight to Germany with the cost of transporting the honey from Germany to
the United States, and then subtracting the estimated Argentina-to-United
States freight cost for the same period. Similarly, according to ACA, the
demurrage charges were additional costs directly related to the
rejections. ACA states that these additional expenses were incurred as a
direct result of the German customer's rejection of the honey.

Department's Position: We agree with petitioners. At verification, we
found that ACA included in its German warranty calculation certain
expenses associated with honey rejected by ACA's German customers, and
later resold to ACA's customers in United States, or returned to
Argentina. These expenses include freight and movement expenses for
transporting the honey to its final destination (i.e., the United States
or Argentina) and U.S. inland trucking and demurrage expenses incurred at
the U.S. port. See ACA's Verification Report at 29, and Verification
Exhibit 27. For our final determination, we have re-calculated ACA's
German warranty expenses to include only those expenses associated with
ACA's original sales of honey to Germany and returned to Argentina, and
excluded freight and movement expenses associated with transporting the
resold honey to the United States. Consistent with our calculation of
warranty expenses in the preliminary determination, we have allocated
German warranty expenses over the volume of total sales to Germany. See
Final Sales Analysis Memorandum, September 26, 2001.


Comment 16: International Freight Expenses

Petitioners argue that ACA's reported ocean freight expenses are
understated for certain U.S. sales, and that the Department should adjust
the expenses for these sales in the final determination. According to
petitioners, as the Department found at verification, the invoiced
quantity for certain U.S. sales was adjusted because of a quantity
shortage or overage, and the adjustment amount was reported in the field
QTYADJU. Petitioners indicate that because it is the Department's normal
practice to allocate freight expenses based on net (rather than gross)
weight, ACA should have calculated the unit freight amounts for these
sales based on the actual quantities shipped. 

ACA states that, in theory, it does not contest petitioners' proposed
correction to its U.S. freight costs. ACA maintains, however, that when it
used a quantity adjustment to account for rejection of adulterated honey
by U.S. customers, petitioners' calculation is inappropriate. In those
instances, according to ACA, the quantity adjustment accounted for
merchandise that was actually shipped but turned out not to be the product
under investigation. ACA contends that the freight costs on those
shipments that involved adulterated honey were incurred by ACA for the
entire shipment and are thus appropriately allocated over the weight of
the entire shipment, including the adulterated honey.

Department's Position: We agree with ACA and petitioners, in part. At
verification, we found that ACA made adjustments (reported in field
QTYADJU) to the invoiced weight for certain U.S. sales (see ACA's
Verification Report at 24). However, as petitioners correctly note, ACA
did not account for the quantity adjustments in reporting U.S. freight
expenses for these sales. For the final determination, in order to account
for all U.S. freight expenses, we have re-calculated ACA's per unit
freight expenses for those sales which incurred a quantity adjustment. We
agree with ACA with respect to its sales of adulterated honey, and have
not included these sales in our re-calculations, because the freight
expenses on these shipments are allocated over the weight of the entire
shipment (see Final Sales Analysis Memorandum, dated September 26, 2001). 


Comment 17: Differences in Physical Characteristics of Merchandise

ACA argues that the statute provides for a DIFMER adjustment to NV when
identical products are not sold in the United States and the comparison
market. According to ACA, as reported in its questionnaire responses and
demonstrated at verification, German customers require honey free of
antibiotic and phenol contamination. ACA maintains that honey that meets
the strict sanitary standards of German customers has different physical
characteristics (i.e., it does not contain antibiotics and phenol) and
involves additional variable direct selling costs (i.e., laboratory
testing) than honey sold in the United States. ACA also notes that
uncontaminated honey that meets the standards required by German customers
is relatively scarce and, therefore, more valuable in comparison to honey
that it sold in the United States. ACA claims that honey which meets the
standards of its German customers is therefore not an identical product to
that sold by ACA to its U.S. customers. 

ACA contends that the Department 's regulations require it make a
reasonable allowance for differences in the physical characteristics of
merchandise to the extent the Secretary is satisfied that the amount of
any price differential is due to such differences. According to ACA,
correspondence with German customers examined at verification demonstrates
that the price difference between the German market price and the U.S.
market price is driven by differences in the physical characteristics of
the honey sold, i.e., German customers require rare antibiotic free honey
and demand steep discounts for even slightly contaminated honey. ACA
states that in cases where honey is rejected by German customers because
of higher contamination levels, ACA refunds the purchase price and may
resell the honey in the U.S. market where customers are not concerned
about antibiotic and phenol contamination levels. ACA argues that its
experience with German customers demonstrates that the market value of
antibiotic and phenol free honey is, at a minimum, $80 per metric ton
higher than the market value of honey that is not certified antibiotic and
phenol free. ACA contends that there is no evidence on the record to the
contrary as to why the Department should not grant a DIFMER adjustment in
this case, and asserts that when, as in the present case, the price
differential between different merchandise is value-driven, the Department
must consider the market value of the goods in calculating the adjustment. 

ACA maintains that, as it demonstrated at verification, laboratory tests
of honey reveal that some honey is contaminated with residual antibiotics
and phenol and some is not. ACA notes that its German customers have
confirmed their belief that the honey is physically different by rejecting
the lots that are contaminated with these substances. ACA contends that
petitioners' observation that honey is produced by bees does not prove
anything. According to ACA, the contamination with antibiotic and phenol
residues is a result of action by the beekeeper. 

Petitioners argue that, under the statute, the Department may increase or
decrease NV for physical differences when making comparisons of "similar"
or "like" merchandise. They maintain that, in this case, however, the
Department made no comparisons of "similar" or "like" merchandise, and
therefore, by law may not adjust NV as requested by ACA. Petitioners
indicate that because the Department relied on CV, and because CV is by
definition a construction of the product "as sold in the United States,"
no adjustment for physical differences is necessary or proper. 

Petitioners state that the Department's regulations require it to
determine that the merchandise sold in the United States or foreign market
has (1) different physical characteristics, and (2) that these physical
differences have an "effect on price" (see section 351.411 of the
Department's regulations). They argue that ACA has not satisfied either of
these conditions, and that in making such an adjustment, the Department
will consider only differences in variable costs of manufacturing that
result in differences in physical characteristics of the products being
compared. 

According to petitioners, honey is produced by bees and is identical in
physical characteristics regardless of where it is produced or sold. They
maintain that ACA itself acknowledged in its questionnaire response that
there can be no systemic physical differences in honey sold in the United
States vis-a-vis Germany, and that there is no resulting differences in
sales prices. Petitioners contend that there are no differences in the
variable costs of manufacturing the Department relied upon in calculating
COP for purposes of the preliminary determination. They state that the
basis of ACA's claim for an adjustment is an alleged difference in
quality, based on testing procedures, and whether the honey was found to
be adequate for the German market. However, because ACA has not identified
any physical differences, price differences, or cost differences for honey
sold to Germany, its claim must be rejected.

Petitioners also argue that ACA's DIFMER claim hinges solely on whether
certain laboratory tests were performed on honey being sold and whether,
as a result of the tests, the honey was found to be free of antibiotics
and phenol. They maintain that there is no record evidence to establish
that each of ACA's shipments was tested and found to be free of
contamination nor that the honey sold to the United States by ACA was
consistently not free of antibiotic and phenol. According to petitioners,
even if honey with different amounts of antibiotics and phenol were
accepted as physically different, ACA has failed to establish that the
alleged difference had an effect on sales prices. Petitioners state that
because ACA has failed to establish that the honey it sold in Germany is
physically different from the honey it sold in the United States, and that
such differences had an effect on prices, the Department should reject
ACA's claim for a DIFMER adjustment.

Department's Position: We agree with petitioners. Section
773(a)(6)(C)(ii) of the Act directs the Department to adjust NV for
physical differences when making comparisons of "similar" or "like"
merchandise. In this case, we attempted to compare ACA's U.S. sales with
sales of the foreign like product in the third country market. However,
because all third country sales were sold at prices below the COP, we
based NV on CV. Because CV is, by definition, a construction of the
product "as sold in the United States," no adjustment for physical
differences is necessary or proper. See Import Administration Policy
Bulletin 91/2, July 18, 1991.


Comment 18: Level of Trade

ACA argues that, as reported in its questionnaire responses and verified
by the Department, ACA provides important additional selling services to
its German customers, which are honey packers. According to ACA, its sales
to Germany occurred at a more remote stage in the honey distribution chain
(i.e., packing stage) and, as a result, ACA took over the functions
normally performed by importers. ACA maintains that it provides
significantly more warranty services to German packers than to U.S.
importers, and that, as demonstrated at verification, ACA provides to
packers a level of warranty protection on honey that is typically provided
by importers in the U.S. market. ACA states that this occurs because
German customers cannot easily sell honey that does not meet
specifications to another purchaser. This is in contrast to U.S.
importers, who may have many customers and can re-direct honey that is the
wrong color to a different customer or who may blend slightly contaminated
honey with uncontaminated honey. Additionally, according to ACA, the
strict German requirements related to contamination make resale of such
honey impossible for German customers. ACA claims that its reported LOT
adjustment of five percent accounts for the fact that importers in the
United States must earn profit and cover expenses associated with their
services, while in the German market there is no importer profit and ACA
must charge more to account for the additional selling services it
provides. ACA states, therefore, that importers and packers cannot be
considered at the same LOT. 

ACA maintains that section 773(a)(1)(B) of the Act requires that the
Department, to the extent practicable, establish NV based on comparison
sales at the same LOT as the export price. ACA states that if, as in the
present case, there are no sales at the same LOT, the Department compares
U.S. sales to comparison market sales at a different LOT and makes a LOT
adjustment if the difference in LOT affects price comparability. ACA
contends that, because all of ACA's customers in Germany are packers and
all customers in the U.S. market are importers, the impact on price
comparability cannot be seen by examining ACA's prices at different LOTs
in a single market. ACA notes that, as alternative proof that the LOT
difference affects price comparability, it submitted affidavits in its
questionnaire responses from U.S. customers attesting to their normal mark-
up. ACA claims that the affidavits provide an adequate proxy for the LOT
differential.

ACA further notes that its reported warranty expenses do not negate its
LOT adjustment. ACA argues that a LOT adjustment is required when the
sales in question are made at a remote stage in the channel of
distribution, the seller provides additional selling services, and there
is a difference in the price at the two levels. ACA contends that the fact
that the seller incurred additional services related to the additional
selling expenses does not negate the price differential. ACA asserts that
the reported warranty expenses in its sales listings account for the
actual expense incurred by ACA, the LOT adjustment accounts for the price
differential charged, in part, because the warranty service is provided.

Petitioners maintain that ACA misunderstands the Department's practice,
and that its sales to German and U.S. customers are at the same marketing
stages because all of these sales were direct sales to these customers.
According to petitioners, ACA is reporting its direct sales to U.S.
purchasers and direct sales to German purchasers, notwithstanding its
claims that U.S. customers are importers and German customers are packers.
Petitioners contend that even if this were true, ACA assumed no additional
selling functions that would have been assumed by the German importer. 

Petitioners note that the Department's regulations indicate that the
nominal reference to a distribution channel, such as "importers" in one
market and "packers" in another, as claimed by ACA, is not sufficient to
establish different levels of trade. Rather, the exporter must establish
"substantial differences in selling activities that result in a "pattern
of consistent price differences" in order to quality for a LOT adjustment
(see section 351.412 of the Department' s regulations). According to
petitioners, ACA has not established that there are any differences in
selling functions, with the possible exception of warranty claims for
which a separate adjustment is already being made, or any pattern of
consistent price differences between markets. As a result, petitioners
argue that the Department should reject ACA's claim for a LOT adjustment.

Petitioners maintain that, as reflected in the Department's verification
report, ACA also incurred warranty expenses for its U.S. sales during the
POI. According to petitioners, ACA is therefore the party that ultimately
bears the responsibility for warranty claims irrespective of the nominal
title that ACA assigns to its customers. Petitioners state that, likewise,
ACA would have been responsible for warranty claims for its German
customers whether the customers were packers or importers. Petitioners
state that ACA has thus failed to demonstrate that its German sales were
at a more remote stage than its U.S. sales. Petitioners also assert that
the Department examine in their entirety selling functions performed by a
respondent with respect to sales in different markets in conducting its
LOT analysis. Petitioners state that a difference in a single selling
function, even if one existed, is not sufficient to establish a different
LOT. Petitioners additionally state that ACA has provided no indication
that there is a consistent pattern of price differences between its German
sales, noting that ACA could not provide such price differences because it
sold only to packers in the German market. According to petitioners, any
differences in warranty services provided by ACA between its German and
U.S. markets are taken into account by the difference in warranty expenses
calculated by ACA for the sales in question.

Department's Position: We agree with petitioners. As we stated in the
preliminary determination, in accordance with section 773(a)(1)(B) of the
Act, to the extent practicable, we determine NV based on sales in the
comparison market at the same level of trade (LOT) as the EP transaction.
The NV LOT is that of the starting-price sales in the comparison market
or, when NV is based on CV, that of the sales from which we derive SG&A
expenses and profit. For EP, the U.S. LOT is also the level of the
starting-price sale, which is usually from the exporter to the importer. 

To determine whether NV sales are at a different LOT than EP, we examine
stages in the marketing process and selling functions along the chain of
distribution between the exporter and the unaffiliated customer. If the
comparison-market sales are at a different LOT and the difference affects
price comparability, as manifested in a pattern of consistent price
differences between the sales on which NV is based and comparison-market
sales at the LOT of the export transaction, we make a LOT adjustment
pursuant to section 773(a)(7)(A) of the Act. See Notice of Final
Determination of Sales at Less Than Fair Value: Certain Cut-to-Length
Carbon Steel Plate from South Africa, 62 FR 61731 (November 19, 1997).

In our preliminary determination, in reviewing the channels of
distribution, we found that ACA made sales directly to unaffiliated
customers in both the third country and the U.S. markets. We also found
that the selling functions ACA provided to its reported channels of
distribution in the third country and U.S. markets were the same, varying
only by the degree to which warranty services were provided. We stated
that because we do not find the varying degree to which warranty services
are provided sufficient to determine the existence of different marketing
stages, we found that the LOT for all EP sales was the same as that for
third country sales. Because ACA's U.S. sales and third country sales were
at the same LOT, we did not grant it a LOT adjustment. 

At verification, we examined ACA's sales to the third country and U.S.
markets. We found that ACA made sales in both markets directly to
unaffiliated customers. ACA reported its category of customers in the
third country as packers, and customers in the United States as importers,
who resell to packers. As we stated in the preliminary determination,
while we consider the type of customer an important indicator in
identifying differences in LOT, the existence of different classes of
customers is not sufficient to establish a difference in the LOTs. Whereas
certain titles used to describe classes of customers (e.g., original
equipment manufacturer, distributor, wholesaler, retailer) may actually
describe LOTs, the fact that two sales were made by entities with titles
suggesting different stages of the marketing process is not sufficient to
establish that the two sales were made at different LOTs. (See Antidumping
Duties: Countervailing Duties, Preamble to 19 CFR, 351, FR 27296, 27371
(May 19, 1997).) 

In analyzing ACA's LOT claims, we reviewed information on the record
regarding ACA's selling functions. At verification, we found that ACA
performed activities relating to the arrangement of international freight
and delivery, for both the third country and U.S. markets (see ACA
Verification Report, at 27). We also found that ACA performed activities
relating to warranty services for sales to the third country and U.S.
sales (see ACA Verification Report, at 29-30). In reviewing ACA's warranty
expenses for both markets, we found that ACA incurred a higher level of
warranty expenses in Germany compared to the United States (see ACA
Verification Report, at 29-30); however, we note that this does not
indicate that additional services are sufficient to warrant a separate
LOT. 

Based on this information, and consistent with our findings in the
preliminary determination, we agree with petitioners that ACA's selling
functions in both markets, therefore, appear to be the same, and differ
only by the degree to which warranty services were provided. As we stated
in the preliminary determination, we do not believe the varying degree to
which warranty services are provided sufficient to determine the existence
of different marketing stages. Because ACA's German sales during the POI
were only to packers, we also agree with petitioners that ACA has not
provided any evidence that there is a pattern of consistent price
differences between the alleged different LOTs based on its German sales. 

Therefore, based upon the record information, and consistent with our
preliminary findings, we have continued to find that the LOT for all EP
sales is the same as that for third country sales. Accordingly, because
U.S. sales and third country sales are at the same LOT, no LOT adjustment
under section 773(a)(7)(A) of the Act is warranted. 


Recommendation

Based on our analysis of the comments received, we recommend adopting all
of the positions set forth above and adjusting all related margin
calculations accordingly. If these recommendations are accepted, we will
publish the final determination and the final weighted-average dumping
margins for all firms in the Federal Register.




AGREE____ DISAGREE____



Faryar Shirzad
Assistant Secretary
  for Import Administration


Date 



________________________________________________________________________
footnotes:

1. Petitioners in this case are the American Honey Producers Association
and the Sioux Honey Association (collectively, "petitioners"). 

2. See Chinsung Industrial Co., Ltd. v. United States, 13 C.I.T. 103,
105, 705 F. Supp. 598, 600 (1989). 

3. See F. LLI de Cecco Di Filippo Fara S. Martino v. United States, 216
F. 3d 1027, 1032 (Fed. Cir. 2000) (emphasizing that the use of adverse
facts available is to encourage respondents to cooperate, and it not be
used as a punitive measure). 

4. See Gestion Apicola March 1999 at line 62 (Petition attachment 4). 

5. See Radix Verification Report at 27, Exhibit 6 (on file in the CRU,
room B-099 of the main Commerce building).