REDETERMINATION ON REMAND - World Finer Foods Inc., Barilla Alimentari, S.p.A. and La Molisana Industrie Alimentari, S.p.A. v. United States - Consol. Court No. 99-03-00138 - CERTAIN PASTA FROM ITALY (A-475-818) - FINAL RESULTS OF ADMINISTRATIVE REVIEW (January 19, 1996-June 30, 1997)
September 15, 2000

REDETERMINATION ON REMAND

FINAL RESULTS OF ADMINISTRATIVE REVIEW (January 19, 1996-June 30, 1997)

CERTAIN PASTA FROM ITALY (A-475-818)

World Finer Foods Inc., Barilla Alimentari, S.p.A. and La Molisana Industrie Alimentari, S.p.A.

v. United States

Consol. Court No. 99-03-00138

REMANDED ISSUES

On June 26, 2000, the United States Court of International Trade ("CIT") in World Finer Foods Inc., Barilla Alimentari, S.p.A. and La Molisana Industrie Alimentari, S.p.A. v. United States ("World Finer Foods"), Consol. Court No. 99-03-00138, Slip. Op. 00-72, remanded to the Department of Commerce ("Commerce") the Notice of Final Results and Partial Rescission of Antidumping Duty Administrative Review: Certain Pasta from Italy, 64 FR 6615 (Final Results").

The Court remanded three issues: (1)

(1) The Court remanded to Commerce to determine an appropriate facts available rate for Arrighi S.p.A. Industrie Alimentari ("Arrighi").

(2) The Court remanded to Commerce to reconsider the adverse facts available margin with respect to Barilla Alimentari, S.p.A. ("Barilla"), and to assess a dumping margin, that while adverse, "bears a rational relationship to the probability of dumping."

(3) The Court remanded to Commerce to recalculate the assessment rate for La Molisana Industrie Alimentari, S.p.A. ("La Molisana").

REMAND RESULTS

I. APPLICATION OF TOTAL FACTS AVAILABLE TO ARRIGHI

In order to address the Court's concern regarding the reliability and use of the information that World Finer Foods ("WFF") provided, we articulated more fully the reasons underlying our initial rejection of this information in our Draft Redetermination at page 3. The data submitted by WFF did not provide any information regarding Arrighi's home market selling prices or cost of production. The information was also inadequate for purposes of determining Arrighi's U.S. prices during the period of review ("POR").

As explained in our Draft Redetermination (the reasoning of which is incorporated herein by reference with respect to Arrighi), and based on comments from WFF, in light of the Court's instructions and based on the facts of the particular situation affecting Arrighi, we have determined that an appropriate facts available rate for Arrighi is 19.09 percent, which was the verified weighted-average margin calculated for Arrighi in the less-than-fair-value ("LTFV") investigation less the portion of the margin attributable to the export subsidy from the countervailing duty investigation. In fact, this rate has served as the cash deposit rate for Arrighi since the LTFV. (2) We believe that applying the 19.09 percent margin as facts available for Arrighi is consistent with the Court's remand instructions.

The 19.09 percent margin is not "adverse" to Arrighi because it is based on Arrighi's own verified information. It is timely because it was provided by Arrighi during the course of the LTFV investigation, which immediately preceded the first antidumping administrative review. Absent other reliable information on the record for Arrighi, there is no reason to believe that this facts available rate is a not a reasonably accurate estimate of Arrighi's actual dumping margin for the first administrative review. Finally, this margin is consistent with the Court's holding in Borden I which arose out of the LTFV investigation. In that decision, this Court held that the 21.34 percent margin (3)

is applicable to a cooperative respondent and "fulfills the statutory purposes to provide an incentive to cooperate with Commerce without utilizing punitive, aberrational, or uncorroborated margins." (4)

In order to corroborate secondary information within the meaning of 19 U.S.C. § 1677e(c), Commerce will, to the extent practicable, examine the reliability and relevance of the information used. In the instant case, the 19.09 percent margin constitutes corroborated information because the source of the margin is an administrative determination from a prior segment of the proceeding and it is based on verified information provided by Arrighi in that proceeding. In addition, we have no record information which indicates that Arrighi's prior margin is not relevant for this period of review.

For these reasons, we revise our original decision in the Final Results and apply a non-adverse facts available margin of 19.09 percent in determining Arrighi's final antidumping duty margin in the first administrative review of certain pasta from Italy.

We address the comments from WFF below.

II. ADVERSE FACTS AVAILABLE RATE APPLICABLE TO BARILLA

In reviewing the record evidence available to use as a basis for an adverse facts available margin for Barilla, we reexamined the price information in the antidumping petition. The petition contains a price list from Barilla with an effective date of January 1, 1995 through March 31, 1995. The price list also was annotated with specific discounts that Barilla provided to its Italian customers. The Barilla price list contains the prices for [ ] pasta items and three unique prices per kg. (5) In an attempt to find additional information, we have reviewed other independent sources of information. We found that the petition contained the only information specific to Barilla. Therefore, we believe that it is appropriate to use this information as a basis for deriving Barilla's normal value.

With respect to U.S. price ("USP"), we reviewed U.S. Customs import statistics covering imports of pasta from Italy during the first administrative review period. From this information, we were able to calculate an average unit value ("AUV") specifically for Barilla. In so doing, we relied on only those entries which Barilla itself identified as subject to antidumping and/or countervailing duties. For this reason, we believe this AUV covers all of Barilla's exports of subject merchandise to the United States during the POR and is a reasonably reliable indicator of Barilla's U.S. pricing during the POR. This Customs data, however, does not identify the specific type of pasta being entered. Consequently, we calculated a single AUV for all types of pasta sold in the United States during the POR.

After reviewing all reasonably available sources of information regarding Barilla for the POR, we have used this information as the bases for NV and USP. This is the only information available on Barilla and it bears some relationship to Barilla's actual amount of dumping during the POR. We calculated margins by comparing the single AUV to each of the three home market price categories on the price list and found margins ranging from 39.63 to 63.63 percent with a simple average of 45.49 percent. (See Attachment I for a detailed discussion of these calculations).

In determining which of these margins to select, we considered the Court's instructions in two stages. First, we considered whether these margins bear a rational relationship to Barilla and the current level of dumping in the industry. Second, we considered which margin is appropriately adverse. Barilla's 1995 prices are reasonable to use as NV because they are taken directly from Barilla's own price list and are comparable to prices charged by other Italian pasta makers. (6) In addition, Barilla's prices, as quoted in the 1995 price list, are, at worst, a conservative estimate. The 1995 prices are likely to be lower than Barilla's POR prices due to inflation. In fact, we confirmed that other Italian pasta producers' market prices did increase from the LTFV (i.e., period of investigation ("POI") being May 1, 1994 through April 30, 1995) to the first administrative review (i.e., POR being January 19, 1996 through June 30, 1997). The USPs, which Commerce used for purposes of calculating the dumping margin, are reasonable because they are based on Barilla's actual entered values for the POR. Thus, the average of the margins calculated using Barilla's 1995 price list and the POR AUV, 45.49 percent, is a very conservative estimate of Barilla's margin of dumping. (7)

We next considered which margin is appropriately adverse. In so doing, we selected the highest of the calculated margins in Attachment II of the Draft Redetermination, 63.36 percent. In making this selection, we are mindful that the calculated margins in the first administrative review were appreciably lower. A similar distinction occurred in the LTFV investigation where the adverse facts available rate for De Cecco was significantly higher than the calculated rates. The LTFV investigation was reviewed by the CIT in Borden I, where the court found that the petition rates were unreliable as a basis for assigning De Cecco a margin, based on adverse facts available, because the facts developed during the LTFV investigation indicated that De Cecco's margin was unlikely to be as high as the estimated petition margins. However, there are significant differences between the facts of this situation and those in Borden I. Notably, the instant remand pertains to an administrative review, which does not necessarily cover all producers, exporters, or importers, whereas Borden I was related to an investigation, which covers all producers, exporters, or importers. (8) During the investigation, we limited the exporters reviewed because of large number of exporters shipping pasta to the United States; however, the Department identified which of the exporters should respond. As such, the margins calculated during the investigation generally are more indicative of market-wide selling practices.

In contrast, an administrative review covers only a selected portion of producers, exporters, or importers. In many cases, administrative reviews will cover only a self-selected group of exporters because only those exporters who believe the Department will calculate lower margins than those which they are currently assigned will request a review. Due to this self-selection process, margins calculated in administrative reviews are not necessarily indicative of market-wide selling practices. It is particularly instructive that, in this review, the lowest margins we calculated were for De Cecco and Indalco, each of which requested a review of its own exports. In contrast, petitioners requested reviews of Arrighi, Pagani, and Barilla, all of which failed to cooperate. In considering the extent to which the calculated margins in this review are reflective of market-wide selling practices, we note several factors. First, the companies that participated in this review (i.e., De Cecco, Indalco, La Molisana, Puglisi, and Rummo) accounted for only [ ] percent of imports during the POR. Second, we note that the AUV of all imports actually fell from $[ ] per pound during the period covering the LTFV investigation to $[ ] per pound (9) during the POR, which is indicative of declining export prices from Italy. Given these two factors, we cannot reasonably conclude that the size of the calculated margins call into question the higher adverse facts available rate we are attributing to Barilla.

As far as corroborating the revised adverse facts available rate, information available to Commerce indicates that a dumping margin for Barilla would likely be higher than the margins calculated for any fully cooperative respondent in this proceeding. First, Barilla increased its percentage of total U.S. imports considerably since the time period preceding the initiation of the LTFV investigation. Second, during this POR, the AUV of pasta imported from Barilla was $[ ] per pound, compared to an AUV for all imports from Italy (excluding Barilla's) of $[ ]. (10) Third, the Statement of Administrative Action to the Uruguay Round Agreements Act ("URAA") at 870 considers price lists from an interested party to be an independent source for corroboration, i.e., price lists are self-corroborating in the sense that they represent a company's own information. Inasmuch as we relied upon a Barilla price list to establish the home market price, it is unnecessary to corroborate further the information contained therein. Nevertheless, we reviewed the Italian-market price lists of companies that responded in the LTFV investigation and first administrative review. We observed that Barilla's home market prices as listed on the price list, are generally within the range of prices contained in the price lists of other Italian companies both for the POI and POR. (11)

Finally, we note that in selecting the 63.36 percent rate, we are making an inference adverse to Barilla. The information we had available regarding Barilla was a home market price list with three price categories and a single U.S. price based on the AUV. By failing to cooperate by not acting to the best of its ability, Barilla made it impossible for the Department to know exactly how Barilla's U.S. sales should be compared to its home market sales. Consequently, as an adverse inference, the Department has assumed that all of Barilla's U.S. sales during the POR were properly compared to home market sales in the highest price category. We believe this rate is appropriately adverse to Barilla and may serve as a reasonable deterrent to non-compliance.

We address the comments from Barilla below.

III. LA MOLISANA ASSESSMENT RATE

Remand Results

As explained in our Draft Redetermination, and upon re-examination of the record, we agree that we made an error in calculating only one importer-specific assessment rate. The U.S. sales tape used for the calculation of the assessment rates in the Final Results and the information provided by La Molisana in its October 27, 1998 letter clearly show that there were two importers, and that we should calculate two importer-specific rates. (12) Specifically, we have attributed all sales made up to [ ], and sales of certain specified private label products made thereafter to [ ] and, we have calculated an assessment rate for [ ] based on these transactions. (13) We continued to calculate an assessment rate for La Molisana based on all other sales. The assessment rates are now as follows:

IMPORTER ASSESSMENT RATE
La Molisana [ ]%
[ ] [ ]%

We received no comments on this issue.

World Finer Foods' Comments (14)

Comment 1: (Invalidity of Commerce's Observations)

WFF argues that information in its letter to Commerce, dated March 10, 1998, was probative in many respects. WFF proffers that the information in the letter was not submitted just for calculating NV, but it was also probative on the issue of the reasonableness of the facts available margin, and whether the margin selected by Commerce was corroborated by independent information. WFF asserts that Commerce made a number of false or unsubstantiated statements in the Draft Redetermination: 1) WFF claims that the Department's statement that WFF did not provide information regarding Arrighi's home market selling prices or cost of production is incorrect; WFF points out that it had supplied data on semolina costs which account for a substantial portion of total pasta production costs; 2) WFF notes that Commere could have easily calculated whether Arrighi's home market was viable had it requested that specific information, however, it chose not to; 3) WFF maintains that Arrighi's home market prices would be in alignment with those of other Italian producers as a result of market forces; 4) WFF asserts that the Department's comment in the Draft Redetermination that the pricing documents which WFF submitted took place prior to the first administrative review period is false. Finally, WFF notes that the Draft Redetermination does not attempt to corroborate use of the punitive 71.49 percent rate for Arrighi.

Commerce's Position:

WFF's assertions are without merit. As stated in the our Draft Redetermination, we could not use the information in WFF's March 10, 1998, letter to Commerce to determine normal value. Despite WFF's assertion, Commerce was not provided with Arrighi's home market pricing or cost information. (15) With regard to the cost of production, WFF provided general information for the cost of semolina (i.e., from the Milan Commodity Exchange), which is not specific to Arrighi. Moreover, WFF has not demonstrated how its reported cost is reflective of what the cost would be to Arrighi. Furthermore, WFF does not provide any cost data to account for variable overhead, fixed overhead, G&A, and interest expenses that would also have to be accounted for to reflect the full cost of production.

WFF suggested that, because Arrighi's home market prices would be in alignment with other Italian producers, Commerce could rely on the pricing information submitted by other respondents during the first administrative review. However, without company-specific information, we do not have a basis for calculating Arrighi's actual selling prices or production costs. With regard to the possibility of the alignment of home market prices, WFF has provided no support for its assertion that market forces would cause alignment of prices with other Italian producers and whether there may be variations in quality, selling practices, or other factors which might necessarily differentiate Arrighi's merchandise from that sold by others.

With regard to WFF's assertions that the pricing documents it submitted were contemporaneous with the POR, we agree in part. We actually stated that "most of the time periods for which WFF submitted this information took place prior to the first administrative review period." (16) Regardless of what portion of this information is within the POR, this does not change our determination that this pricing information is not a reliable basis for determining Arrighi's prices during the POR. These prices related only to general categories of pasta, i.e., longcuts for which no comparable home market data was submitted (and which is insufficiently specific for price-to-price comparison purposes). (17) Finally, with regard to corroboration, as we did not apply 71.49 percent to Arrighi in this remand determination, there is no need to corroborate this rate. Instead, we used 21.34 percent (19.09 percent when adjusted for the amount of the export subsidy), the verified weighted-average margin calculated for Arrighi in the original LTFV investigation.

Comment 2: (Calculation of the margin)

WFF states that when determining a facts available rate for Arrighi, Commerce should deduct the portion of the margin attributable to the export subsidy determined in the countervailing duty investigation.

Commerce's Position:

In calculating margins, Commerce normally adjusts an antidumping duty deposit rate to reflect export subsidies. Because the margin of dumping for Arrighi is being determined on the basis of non-adverse facts available, we have decided in this case to adjust the facts available rate to exclude the portion of the margin attributable to the export subsidy for the countervailing duty investigation.

Barilla's Comments (18)

Comment 1: (Home Market Price List Objections)

Barilla objects to Commerce's use of its home market price list from the petition as provided by the market researcher arguing that it is unreliable and not rationally related to the probability of dumping. First, Barilla argues that the price list covers merchandise not within the scope of the review or the investigation. Barilla contends that the price of non-subject merchandise such as "bulk pasta" does not reflect the price of subject merchandise. Furthermore, Barilla implies that the size limitation on the scope of subject merchandise indicates that the order was not intended to cover sales to caterers, citing Certain Pasta From Italy, 63 FR 54672, 54673 (October 13, 1998) in support of this position.

Second, Barilla objects to Commerce's use of a product line not sold in the United States, Barilla's "Ricetta d'Oro" line, as the basis for an adverse facts available margin. In addition, Barilla argues that the list was limited to certain geographical locations; that it was not applicable to all customers; and that the list was only valid for a short period. It also asserts that the use of the "Ricetta d' Oro" line is further evidence that Commerce based its adverse facts available margin on prices for non-subject merchandise, since this product line was sold [ ].

Third, Barilla points out that Commerce's corroboration of the market research included in the petition prior to the initiation of the original investigation did not indicate how that information was corroborated, and that Commerce ignored the findings of the court that the petition information was not reliable. Therefore, Barilla states that there is no basis for Commerce to rely on the market research information.

Commerce's Position

We disagree with Barilla that its price list is not rationally related to Barilla's probable dumping margin. However, we agree with Barilla in part that the price list does cover some non-subject merchandise. Many of the prices in the price list cover pasta sold in 5 kg packages, which is outside the scope of the order. The only prices for pasta sold in packages less than 5 kg (1 kg packages) result in the highest margin calculated, 63.63 percent. (19) Nevertheless, we have continued to rely on all of the prices on the price list for this remand determination. We find that this information has probative value for several reasons.

Although these prices cover so-called "bulk pasta" which is not included within the scope, we believe these prices are indicative of Barilla's home market pricing of subject merchandise. First, on the price list itself, the prices for 5 kg and 1 kg packages of "Ricetta d'Oro" are the same. Thus, subject and non-subject pasta are sold for the same price. Second, Barilla's 1995 prices for subject merchandise are comparable to the lower end prices for subject merchandise from the other producers' POR price lists. (20) Thus, because Barilla's price list prices are from an earlier time period and comparatively low, these prices are, if anything, a conservative estimate of Barilla's pricing of subject merchandise. Finally, this information is the best information available on what may constitute Barilla's "actual" dumping given that Barilla refused to provide any information during the proceeding.

As for sales to caterers, we disagree with Barilla's assertion that the price list Commerce relied upon is unreliable because it is intended for food service rather than retail sales. Pasta sold to the catering industry is not excluded from the scope of the order. The scope states, in relevant part, "[t]he Pasta covered by this scope is typically sold in the retail market, in fiberboard or cardboard cartons." Furthermore, it is our experience that companies which participated in our investigation and are participating in our reviews sell to caterers. In our investigation, De Cecco explained that caterers is a customer category which refers to restaurants, hotels, and institutions such as retirement homes. Pagani states, "pasta manufactured by Pagani is sold to catering companies which use it in restaurants, company cafeterias, university cafeterias and other food services. (21) In the first administrative review, three of the Italian companies reviewed during this proceeding had sales of subject merchandise to the food service industry (i.e., caterers and restaurants). (22)

Furthermore, we disagree with Barilla that the "Ricetta d'Oro" line of pasta is different in physical characteristics from the pasta it exported to the United States. Barilla exported a significant quantity of subject pastas to the United States. (23) There is no record evidence to suggest that all of this pasta is physically different or that it would not be comparable to the "Ricetta d'Oro" pasta line for purposes of comparisons in an antidumping analysis. Finally, regarding the time period for the price list and the other limitations to the price list's applicability, we disagree with Barilla that this renders the price list unreliable. This price list is the best, indeed the only, indicator of Barilla's home market pricing during the POR because Barilla refused to respond to Commerce's questions. Moreover, as discussed in detail in the Draft Redetermination, we corroborated this information.

We disagree with Barilla that information in the petition is unreliable in toto as a result of

any decisions of the CIT or the holding of the Court of Appeals for the Federal Circuit in F.lli. De Cecco Di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027 (Fed. Cir. June 16, 2000)("De Cecco"). In De Cecco, the Federal Circuit denied Commerce's request for a remand and upheld the CIT's finding that the petition rates were unreliable as a basis for assigning De Cecco a margin based on adverse facts available because the facts developed during Commerce's LTFV investigation indicated that De Cecco's margin was unlikely to be as high as the estimated petition margins. Id. at 1032-33. In addition, the Federal Circuit held that there was no need for a remand for further attempts to corroborate the petition rates because it agreed with the CIT that the administrative record developed during the LTFV investigation would not provide information necessary for corroboration of those petition rates. Id. at 1034-35.

The Federal Circuit made clear, however, that it affirmed the lower court's ruling on this issue based on the record before the CIT - that the petition rates were "unreliable" and "discredited" based solely on the information acquired by Commerce during the LTFV investigation. See De Cecco, 216 F.3d at 1032 ("all available evidence indicates that the [ ] was many times higher than De Cecco's actual margin") and at 1033 ("[u]nder the circumstances of this case," the petition rate could not be used on remand; and "sufficient evidence in the administrative record was not shown to support the application of the ... petition rate"). Contrary to Barilla's assertions, the Federal Circuit did not hold that the petition information could not be corroborated with information developed in future segments of the proceeding or that the petition information may not have relevance to an uncooperative respondent in the future. It simply stated that the petition rates were unreliable as adverse facts available for De Cecco based on the administrative record of the LTFV investigation. The Federal Circuit also held that the corroboration requirement of section 776(c) (19 U.S.C. 1677e(c)) means that "an adverse facts available rate [must] be a reasonably accurate estimate of the respondent's actual rate, albeit with some built-in increase intended as a deterrent to non-compliance." De Cecco, 216 F.3d at 1032. Commerce has attempted to calculate a reasonably accurate margin for Barilla using Barilla's own information in compliance with the Federal Circuit's holding in De Cecco.

Finally, section 776(b)(1) of the Act (19 U.S.C. § 1677e (b)(1)) plainly states that Commerce may rely "on information derived from . . . the petition" in making adverse inferences. Barilla's reading of De Cecco would effectively render section 776(b)(1) a nullity because petition information would be "discredited" in its entirety simply because Commerce's LTFV investigation yielded antidumping margins calculated for some of the exporters of the subject merchandise which were different from the antidumping rates estimated in the petition. Nothing in the statute, the legislative history to the URAA, the Statement of Administrative Action to the URAA, or the Federal Circuit's opinion in De Cecco suggests such a result is contemplated by the statute.

Given that Commerce is not precluded from attempting to corroborate the petition information in this administrative review, the only relevant question is whether the price list itself is reliable. It is significant that Barilla does not deny that this is its price list. Moreover, prior to initiating the investigation, we interviewed the market researcher who obtained the price list, and reviewed how this price list was consistent with other market research information. (24) For instance, the May 26, 1995, public summary of the market research report identifies 27 different Italian Pasta producers for which pricing data was collected. The public summary groups the prices from these companies into four price ranges (i.e., very high, high, medium, and low) in the retail pasta market in Italy.

Very High (more than 3600 L/kg)

High (from 2900 to 3600 L/kg)

Medium (from 2000 to 2900 L/kg)

Low (less than 2000 L/kg)

Barilla is listed in the medium price range, from 2000 to 2900 lira per kilogram, with 13 other producers or wholesalers.

Finally, we disagree with Barilla's assertion that the fact that its home market prices are lower than other Italian producers' renders it illogical that it should have a higher margin. As stated in the "Adverse Facts Available Rate Applicable to Barilla" section above, there are several reasons why Barilla would have a higher dumping margin. In addition, the other Italian producers who have higher price list prices but lower margins have identified many off-price list discounts in their questionnaire responses which serve to lower the NV. However, as Barilla did not respond to the questionnaire, we have no information about any such discounts. Moreover, the other Italian producers have different U.S. prices. Thus, it is not illogical or even unlikely that Barilla's margin would be higher.

Comment 2: (Use of AUV Data for the U.S. price)

On the issue of what data to use as a basis for U.S. price, Barilla argues that the use of Barilla-specific AUV data from Customs import statistics is an inappropriate basis for determining U.S. price. First, Barilla objects to the use of non-record evidence without the Court's instructions to re-open the record. Second, Barilla claims that the values reported to Customs reflect the transfer price to Barilla's affiliated importer, not the first unaffiliated customer, and therefore are inconsistent with the statutory requirement that the U.S. price be based upon prices between unaffiliated parties.

Commerce's Position:

We disagree with Barilla. The CIT ordered Commerce to "reconsider the adverse facts available margin with respect to Barilla and assess a dumping margin that, while adverse, bears a rational relationship to the probability of dumping" by Barilla. World Finer Foods, Slip Op. 2000-72 at 30. Nothing in the court's opinion suggests that Commerce could not seek additional information in an attempt to estimate Barilla's "actual" dumping margin. Indeed, it was in the interests of accuracy that Commerce has done so.

The AUVs are the best, and the only evidence we have of Barilla's U.S. prices during the POR because Barilla refused to participate in the review. The fact these AUVs are based on transfer prices between Barilla and its affiliated importers does not render these prices unusable because Commerce is merely attempting to estimate Barilla's "actual" dumping margin. It is impossible for Commerce to calculate Barilla's margin with the accuracy required by the statute because only Barilla has the information that would make such a calculation possible.

Although we agree with Barilla that we normally do not use transfer prices as the basis for U.S. price, these AUVs are the only evidence available and do reflect Barilla's U.S. pricing during the POR. To the extent that such AUVs are not reflective of commercial pricing, Commerce has no means to determine Barilla's actual U.S. prices and is using the AUVs as the best surrogate under the circumstances. We also disagree with Barilla that the sales to the first unaffiliated purchaser would likely result in lower margins. Such transactions would likely be treated as CEP and would be subject to the appropriate deductions.

Comment 3: (Circumvention of the order)

Barilla contends that, contrary to statements in the Draft Redetermination, Commerce's finding that Barilla was circumventing the order, is in no way indicative of Barilla's pricing practices nor the probability of high dumping margins. Rather, Barilla asserts that its failure to respond to Commerce's questionnaire in the circumvention investigation was based on its interpretation of the scope of the order and its belief that its practices did not constitute circumvention of the order.

Commere's Position:

We agree with Barilla, in part, that the affirmative circumvention determination is not necessarily dispositive of higher margins. Although the Department believes that an anti-circumvention finding may be indicative of the likelihood of dumping in certain circumstances, the Department has not used the affirmative anti-circumvention finding against Barilla as support for its determination in this remand.

SUMMARY

We hereby submit the final results of determination on remand pursuant to the CIT's order in World Finer Foods. Based on the above, we recommend applying a non-adverse facts available margin of 19.09 percent to Arrighi. For Barilla, we recommend applying an adverse facts available rate of 63.36 percent. With regard to La Molisana, we have corrected for clerical errors in calculating importer-specific assessment rates.

Should the Court affirm this final redetermination on remand, Commerce will publish in the Federal Register a notice announcing these results (pending final and conclusive results).

Troy H. Cribb
Acting Assistant Secretary
    for Import Administration

Date: September 15, 2000

Attachment I

Attachment II

Attachment III

Attachment IV

Attachment V

Attachment VI

Attachment VII


FOOTNOTES:

1. See Attachment I (Draft Redetermination) for details pertaining to background and analysis of each issue.

2. Where, as here, there is also a concurrent countervailing duty order, Commerce instructs the Customs Service to require a cash deposit or bond equal to the dumping margin calculated for a company, minus the amount determined to constitute an export subsidy. Such was the case here where Arrighi's 21.34 percent margin was reduced to a 19.09 percent deposit rate.

3. We note that, although the Court's decision referred to 24.31 percent, 24.34 percent, and a 24.37 percent margin, respectively, as the highest calculated margin from any segment of the proceeding (Borden I, Slip Op. 98-167 at 2, 4), the actual highest calculated margin was in fact 21.34 percent. See Notice of Antidumping Duty Order and Amended Final Determination of Sales at Less Than Fair Value: Certain Pasta From Italy, 61 FR 38547, 38548 (July 24, 1996).

4. Borden, Inc., Gooch Foods, and Hershey Foods Corp. v. United States, Consol. Court No. 96-08-01970, Slip Op. 98-167 (Ct. Int'l Trade Dec. 16, 1998) ("Borden I"), aff'd F.lli De Cecco di Filippo Fara S. Martino S.p.A. v. United States, 216 F.3d 1027 (Fed. Cir. June 16, 2000).

5. Several pasta types (i.e. shapes) are sold in two of three price categories [ ] (e.g. spaghetti.) The pasta sold at the price [ ] is identified by handwriting on the price list as the "Gold Ribbon Rigea Oro" line of pasta. The price list contains no description of the other two price categories.

6. For example, we obtained and reviewed proprietary home market prices from multiple Italian pasta producers in the investigation and first administrative review, and calculated simple average unit prices for spaghetti. Our calculated results supported Commerce's decision to rely on Barilla's 1995 price list because Barilla's unit price for spaghetti was in the lower range of prices offered by other Italian pasta producers. See Attachment II.

7. See Attachment II of the Draft Redetermination.

8. Section 735(c)(1)(i) and (ii) specify that the final determination of an investigation will affect all imports of subject merchandise (regardless of the number of exporters individually examined). In contrast Section 751(a)(2)(C) specifies that the final results of a review will affect only those entries subject to review (which normally does not cover all exporters).

9. The AUV of all imports excludes Barilla, which was appreciably lower. Including Barilla would result in an even lower AUV for all imports.

10. See Attachment III of the Draft Redetermination for the U.S. Customs import statistics used for this analysis.

11. See Attachment II.

12. We did not consider this an error originally because there was an inconsistency between the electronic data on the computer tape and the narrative portion of the questionnaire response. La Molisana's October 27, 1998 submission clarified this inconsistency. However, we did not consider it at that time because we treated the submission as untimely filed and disregarded it for the final results.

13. See Attachment V of the Draft Redetermination for the relevant SAS margin calculation program and output.

14. See Attachment III, Letter from World Finer Foods, Inc. to Secretary Mineta, dated August 24, 2000.

15. WFF did not supply home market prices. However, it did supply some U.S. pricing data.

16. Draft Redetermination at 3

17. Draft Redetermination at 4

18. See Attachment IV, Letter from Barilla Alimentare, S.p.A. to Secretary Mineta, dated August 24, 2000.

19. The price category including 1 kg packages also includes 5 kg packages resulting in the same margin.

20. See Attachment IV of the Draft Redetermination for a comparison of Barilla's price list with prices of similar subject merchandise sold by other Italian producers during the POR.

21. See Attachment V for responses identifying customers as caterers.

22. See Level-of-Trade Memorandum to Susan H. Kuhbach from John Brinkmann, dated July 31, 1998.

23. See Attachment III of the Draft Redetermination.

24. See Memo from John Brinkmann to The File, dated May 26, 1995, and Letter from Petitioners to The Secretary, dated May 26, 1995, in Attachments VI and VII, respectively.