65 FR 31514, May 18, 2000 A-570-828 AR:12/1/97-11/30/98 Public Document G2/O4: TPF, HS MEMORANDUM TO: Troy H. Cribb Acting Assistant Secretary For Import Administration FROM: Holly A. Kuga Acting Deputy Assistant Secretary For Import Administration SUBJECT: Issues and Decision Memorandum for the Antidumping Duty Administrative Review of Silicomanganese from the People's Republic of China -- December 1, 1997 through November 30, 1998 Summary We have analyzed the comments and rebuttal comments of the interested parties in the 1997-1998 administrative review of the antidumping duty order covering silicomanganese from the People's Republic of China ("PRC"). As a result of our analysis, we have made changes, including corrections of certain inadvertent programming and clerical 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 administrative review for which we received comments and rebuttals by parties: Factor Valuation Facts Available Clerical Errors Miscellaneous Issues 1. Classifying Electrode Paste as a Direct Material or Part of Overhead 2. Allocating Costs Over Production Quantities That Include Fines 3. Reducing Normal Value for Sales of Silicomanganese Slag 4. Recalculating Emei's Electricity Consumption Based on Verification Findings Background On November 8, 1999, the Department published its preliminary results of administrative review of the antidumping duty order on silicomanganese from the PRC, Silicomanganese from the People's Republic of China: Preliminary Results of Antidumping Duty Administrative Review, 64 FR 60784 (November 8, 1999)("Preliminary Results"). This review covers two manufacturers/exporters, Guangxi Bayi Ferroalloy Works ("Bayi"), and Sichuan Emei Ferroalloy Import and Export Co., Ltd. ("Emei"). The period of review ("POR") is December 1, 1997 through November 30, 1998. Since the publication of the preliminary results, the following events have occurred. On November 29, 1999 the respondents and the petitioner, Eramet Marietta Inc. ("Eramet"), submitted publicly available information and comments regarding factor valuation. On December 8, 1999 the respondents filed rebuttal comments regarding the petitioner's November 29, 1999 factor value submission. The Department issued supplemental questionnaires to the respondents on December 16, 1999, and received responses to those questionnaires on January 7, 2000. On January 11, 2000, the Department published in the Federal Register a notice extending the due date for the final results until May 6, 2000 (65 FR 1597). In January 2000, the Department conducted verifications of Bayi and Emei. Public versions of our verification reports, dated March 1, 2000, are on file in room B-099 of the main Department of Commerce building, under the appropriate case number. In response to the Department's invitation to comment on the preliminary results of review, the petitioner and the respondents filed case and rebuttal briefs in March 2000. The Department held a public hearing regarding this review on April 3, 2000. The Department has conducted this administrative review in accordance with section 751 of the Tariff Act of 1930, as amended ("the Act"). Discussion of the Issues I. Factor Valuation Comment 1: Surrogate Values Submitted by the Petitioner The respondents request that the Department disregard the surrogate values submitted by the petitioner because all of them are distorted and aberrational. To prove their point, the respondents compared the average per-metric ton CIF value of silicomanganese imported into the United States during 1998 to the normal value for silicomanganese that they calculated using their factors of production and the surrogate values submitted by the petitioner. According to the respondents, the normal value calculated by using the petitioner's factor values would equal 195% of the average CIF value of U.S. imports of silicomanganese during 1998. In addition, the respondents note that the average U.S. CIF value includes ocean freight and marine insurance while the calculated normal value does not. Moreover, the respondents claim that even if one were to exclude material inputs from the normal value calculation, calculating normal value using the surrogate values submitted by the petitioner would result in a 23% dumping margin if one were to compare normal value with the highest average CIF import value for any country that exported silicomanganese to the United States during 1998. The respondents question the use of surrogate values that indicate that all U.S. imports are being sold at less than fair value. In addition, the respondents suspect that if the Department were to use the respondents' reported factors of production and all of the petitioner's surrogate values to construct normal value and compare the constructed normal value to the petitioner's own U.S. prices, it would find that the average price at which the petitioner sold silicomanganese during the POR was also less than normal value. Based on the foregoing, the respondents conclude that the surrogate values submitted by the petitioner are inappropriate and result in an unfair comparison between normal value and U.S. price. The petitioner did not comment on this issue. DOC Position We disagree with the respondents that the petitioner's submitted factor values must be rejected in toto. When dealing with a non- market economy ("NME") country, section 773(c)(1) of the Act directs the Department to use the best available information on the record to value factors of production. When the Department has several surrogate values to choose from, its practice is to evaluate each surrogate value based upon its quality and contemporaneity, in an attempt to find the best available information (see, also the DOC Position to Comment 2.A). Quality, which is the characteristic in question here, is determined, in part, by deciding whether the surrogate value appears to be unreliable when compared to other available values. We have selected surrogate values for the final results by evaluating the quality and contemporaneity of each value submitted by the parties to this review, individually. We have addressed the respondents' concerns regarding specific surrogate values in our positions below. Nevertheless, in response to the respondents' request that we reject the petitioner's surrogate values in toto, we note that the fact that the respondents used the petitioner's surrogate values to calculate a normal value that is greater than U.S. import values for silicomanganese does not necessarily indicate that the petitioner's surrogate values are aberrational. Normal value may differ from U.S. import values for a number of reasons including differences in production efficiencies among producers which are reflected in the factors of production reported by the PRC producers. Comment 2: Selecting a Surrogate Value for Electricity Since a large portion of the comments received by the Department in this case deal specifically with the issue of which surrogate value is the most appropriate value for electricity, for the sake of clarity we divided Comment 2 into sections A through H. In addition, we divided section A into subsections A.1 through A.5. Each subsection heading is a direct quotation from the Department's preliminary decision which was specifically challenged by the respondents. The petitioner provided comments only in response to the issues in sections A.4., F., G., and H. In brief, the Department continues to rely on Indian electricity rates as the surrogate for this input because the Department has determined that such rates are reliable and are the best available information. A. The Department's Decision in the Preliminary Results is Arbitrary and Capricious The respondents argue that the Department's use of Indian electricity rates as a surrogate for the market value of Chinese electricity in the preliminary results is "arbitrary and capricious, and simply "result oriented." The respondents allege that the Department used "unfettered discretion" in selecting the surrogate value for electricity to arrive at a result the Department desired. The respondents, citing section 773(c)(1) of the Act, assert that the Department must value the factors of production based on the best available information in a market economy country and, accordingly, cannot select any surrogate value it desires. Moreover, the respondents, citing section 773(a) of the Act, emphasize that the Department must make a "fair comparison" between the "export price or constructed export price and normal value" in calculating antidumping margins in all proceedings. Finally, the respondents argue that the Department did not follow its established practice of selecting surrogate values in a surrogate market economy country that best reflect the value of the same factors in China, if Chinese prices were set by market economy forces. The respondents argue, as detailed in Comment 2.A. through 2.H. below, that the Department should change its determination with respect to electricity. DOC Position We disagree with the respondents that our use of Indian electricity rates in the preliminary results was arbitrary and capricious and was result oriented. On the contrary, our use of Indian electricity rates was reasonable and fair because the Indian rates represent the best available information. As described in detail in the Department's November 1, 1999 memorandum, Surrogate Values Used for the Preliminary Results of Administrative Review of Silicomanganese from China (September 1, 1997 through August 31, 1998) ("Preliminary Surrogate Value Memorandum"), India is the best choice for surrogate values in this case. Of the countries deemed to be economically comparable, India is the largest producer of comparable merchandise. Moreover, India has much published information on prices of factor inputs. Finally, we have used India as a surrogate country in numerous other antidumping cases involving the PRC, including products which are electricity intensive like silicomanganese (see, e.g.,Manganese Metal from the People's Republic of China; Final Results of Second Antidumping Administrative Review, 64 FR 49447 (September 13, 1999) ("Manganese Metal from the PRC: 2nd AR")). Once we have selected a country as our first choice surrogate, our preference is to use factor values solely from that country to value all the inputs. However, there are exceptions. For example, we have used other sources to value inputs when information in our first choice surrogate is: 1) not available, 2) not sufficiently contemporaneous, 3) of poor quality (i.e., not sufficiently specific to the input in question), and 4) otherwise unreliable. In this case, with the exception of domestic manganese ore and fluorspar (see Comments 3 and 6 below), we used Indian information because we find the Indian values to be reliable. Prior to the preliminary results, and in their briefs, the respondents have gone to great lengths to question the Indian industrial electrical rate and convince the Department to use an Indonesian rate. As discussed in the preliminary results, and in the detailed point-by-point discussion below, the respondents' arguments are not convincing. Indeed, the respondents' own factor value submissions support the Department's use of Indian electricity rates. A.1. We have used Indian electricity rates consistently for many PRC cases, including products for which electricity is a major input, see e.g., Manganese Metal from China. The respondents argue that, when selecting surrogate values, the Department's "well-established principle" is to "investigate and treat each case and factor separately." The respondents cite Tapered Roller Bearings and Parts Thereof, Finished and Unfinished, from the People's Republic of China, 62 Fed. Reg. 61,276 (Nov. 17, 1997), as an instance where the Department considered using a surrogate value outside the primary surrogate country (i.e., India), in favor of a value determined to be more appropriate from a different surrogate country. Moreover, citing Certain Helical Spring Lock Washers from the People's Republic of China, (61 Fed. Reg. 66,255, 66,259 (Dec. 17, 1996)), the respondents contend that "simply because a particular source was used in previous reviews... does not preclude the Department from relying on alternate sources if circumstances necessitate a change." The respondents assert that the results and the decision of the Department to use an Indian surrogate value for electricity in Manganese Metal from the PRC: 2nd AR have absolutely no bearing on the present review and that "the cost of electricity in silicomanganese production is far and away the highest cost of production" unlike in Manganese Metal from the PRC: 2nd AR. DOC Position We agree with the respondents that each case must be dealt with on the basis of the facts in that case. We also agree that simply because we have used a particular source in one case does not mean we cannot rely on alternate sources if circumstances so warrant. That said, the Department prefers to use one country as a surrogate and to use one source to value a particular input from segment to segment. The Department strives to be consistent, allowing for the facts of individual cases. In many cases involving the PRC, including this one, we have found India to be economically comparable to the PRC, and to be a significant producer of comparable merchandise. The facts in this case clearly support selecting India as the first choice surrogate, independent of any findings in previous cases. See the Memorandum from Director, Office of Policy, to Office Director, AD/CVD Group II, Office IV, dated June 24, 1999 and the Preliminary Surrogate Value Memorandum. The fact that Indian electricity rates were used and were found to be reliable in other recent cases is consistent with our finding in the instant case. A.2. The fact that the state controls electricity rates is not a basis to reject Indian rates as some degree of state control is common in many countries. The respondents acknowledge that there is some degree of state control over electricity sectors around the world. However, they argue that the Indian government "exercises a type of control that resembles a non-market economy country." The respondents refer to four features economists use to classify basic economic systems. They are: "(1) the decision-making process; (2) the mechanism for providing information and coordinating activities; (3) property rights; and (4) incentives to performance." Based on an analysis of India's electricity sector, the respondents conclude that market forces do not set electricity rates in India. The respondents cite the following features of the Indian electricity sector as evidence: 1) regulatory bodies on national and state levels that "can determine the tariff for electricity, wholesale, bulk grid, or retail;" 2) central and state government ownership and control of power plants and several power companies; 3) government ownership and control of "all national resources required for production of electricity, including coal deposits in India, which were nationalized in 1975;" 4) "private investment projects will be insulated from the market and the political risk of investing in India." Finally, the respondents propose that if the Department "believes that all governments set electricity rates and that the rates are not set by market forces," the Department should determine the cost of electricity by using the actual costs in China since the actual costs were verified by the Department. DOC Position We disagree that government regulation of electricity in India extends to such an extent as to create sufficient distortions in electricity rates that we should switch to another surrogate. The facts that the respondents set forth have existed in India for years, yet the Department has never found that India's state regulation of electricity makes its rates unreliable. The respondents have presented no new evidence that we find persuasive. A.3. In addition, a comparison of the ratio of industrial to residential rates between India and other countries is not necessarily meaningful for purposes of selecting sources for factor valuation. The respondents argue that "the ratio of industrial to residential rates is a yardstick with which an economist can determine whether a rate structure in a particular country is aberrational." Drawing on the economics of price theory, the respondents contend that the rate charged to an industrial user must be less than that charged to a residential user because: "(1) the electricity sector is either a monopolistic or an oligolopistic [sic] environment; (2) the demand function for a residential consumer is different than that for an industrial consumer in the sense that the residential consumer's demand function is more inelastic than the industrial consumer; (3) the markets for the residential sector and industrial sector are different in many ways (the consumption pattern, the voltage consumed, etc.); and, (4) electricity is not a type of product that the industrial user can resell to the residential user." The respondents assert that if the ratio of residential rates to industrial rates is less than one (residential rates less than industrial rates) then "the pricing structure is aberrational and contrary to the fundamental principles of economics." Furthermore, citing their November 29, 1999 submission regarding surrogate values and the Department's March 16, 2000 memorandum Additional Research on Electricity Rates and Silicomanganese Production (A-570-828) ("Additional Research Memorandum"), the respondents contend that all major market economy countries in the world have ratios greater than one, i.e., residential electricity rates that are greater than industrial electricity rates. Therefore, since the Indian ratio of electricity rates is below one, Indian rates are aberrant and should not be used for valuation. In addition, the respondents contest the Department's use of Indian electricity rates from 1995. The respondents argue that using rates outside the period of review, when more contemporaneous rates are available, is "unacceptable." The respondents further argue that "the Indian electricity rates for 1995 used by the Department are aberrational because they are higher than the electricity rates of those market economy countries with significantly higher levels of development, including the United States, Finland, France, Sweden, Greece, New Zealand, South Africa, South Korea; as well as in countries such as the Czech Republic, Hungary, Mexico, Poland, the Slovak Republic and Indonesia." The respondents propose that the Department use electricity rates from Indonesia because it is both a significant producer of subject merchandise and a market economy country with a level of economic development comparable to China. In addition, the respondents prepared a lengthy description of the electricity sectors in China, India and the United States to illustrate their points that: 1) there is some degree of state control in all electricity sectors; 2) the Indian electricity market is subject to greater state control than either the U.S. or Chinese sectors and resembles that of a NME; and, 3) the Chinese electricity sector has been restructured to resemble a market economy enterprise. DOC Position We do not find the ratio of industrial to residential rates to be relevant to our surrogate value selection for electricity. We only use the industrial rate in our calculations. Moreover, as discussed in more detail in section A.4. below, countries set their electricity rates differently. Thus, the fact that this ratio changes from one country to another is not necessarily proof of a distortion. See further discussion in section A.4. With respect to the respondents' argument that the Indian rates are aberrational because they are greater than those of other market economy countries, please see the DOC Position to section A.5. below. Finally, we note that the respondents' argument regarding the contemporaneity of the Indian electricity rates is moot because in the final results we valued electricity using Indian rates that are contemporaneous with the POR. A.4. Moreover, electricity is not generally a traded good. Thus, cross-country comparisons are inappropriate for purposes of factor valuation. The respondents assert that any form of surrogate value selection requires cross-country comparisons. According to the respondents, to find otherwise means that with respect to electricity, parties cannot challenge a potential surrogate rate which they believe to be aberrational because the only way to demonstrate that the rate is aberrational is to compare it to rates in other countries. Nevertheless, absent a cross-country comparison, the respondents contend that the fact that the residential to industrial ratio of India's electricity rates is less than one shows that the Indian electricity rate structure is "anomalous." However, the respondents note that if the Department is conceding that not only are PRC electricity rates set by the state but all countries' electricity rates are set by the state, then "the Department should follow the appropriate statutory language and use the Chinese rates since "available information" in China does permit the normal value of silicomanganese to be determined under the statutory standards used in determining costs in a market economy case." In a different section of their brief, the respondents expand upon their argument that the Department should use Chinese rates. The respondents argue that if the Department feels that some degree of state control is inherent in the electricity sector of every country, as evidenced by the Department's comments in the preliminary results, then the Department should consider using the actual cost of electricity incurred in the PRC in the cost of production. The respondents argue that if the Department rejects the ratios of residential to industrial rates, then the "best available information" would be actual cost under section 773(c)(1) of the Act. In addition, the respondents cite an excerpt of the Omnibus Trade Act of 1987 S. Rep. No. 100-71, 100th Cong., 1st Sess. at 108 (June 11, 1987) as legal justification for using Chinese prices: The Committee bill does not prohibit the Commerce Department from using its normal methodology for determining the foreign market value in cases regarding non-market economy countries. If information submitted by a non-market economy country to the Commerce Department permits foreign market value to be determined accurately using the normal methodology, then the Committee expects such methodology to be used by the Commerce Department (emphasis added by the respondents). The petitioner asserts that the statute does not suggest that the Department may use Chinese prices. The petitioner states that: 1) Chinese electricity prices are non-market prices denominated in a non- convertible currency; and, 2) section 773(c) of the Act requires the Department to value factors of production "based on the best information available" in a market economy country or countries. DOC Position In the preliminary results, we were not persuaded by the respondents' arguments for rejecting Indian electricity rates. We disagreed with the respondents' position regarding Indian rates because 1) there is some degree of state regulation in the electric sector in many countries, 2) the relevance of the ratio of industrial to residential rates has not been established, and 3) electricity is a non-traded good. These reasons for preliminarily rejecting the respondents' argument remain valid. Our reasons for rejecting the respondents' arguments must be viewed in context--the purpose of our analysis under the statute is to use the best available information to value the factors of production in an appropriate surrogate country. Thus, we followed our policy of using information from our first choice surrogate because we were not persuaded that these data are unreliable. Our reasons for rejecting the respondents' argument specifically address the respondents' extensive attempts to show that the Indian electricity rate is unreliable. While it is true that we often compare surrogate values from different sources to determine the most appropriate value, such comparisons typically involve materials and other inputs that are freely traded on the world market. As such, market forces tend to push prices for such goods towards each other, i.e., toward a world market price. By contrast, the circumstances surrounding electricity prices are quite different from those surrounding the other goods subject to this type of comparison. Electricity is not a good where world market forces would impact domestic prices. In fact, electricity prices vary quite significantly from country to country. This alone does not make the rates of electricity in any country unreliable. Furthermore, although we rejected the respondents' argument, we nevertheless conducted extensive electricity price comparisons and found the Indian rate to be comparable. See the Surrogate Values Used for the Final Results of the 1997-1998 Administrative Review of Silicomanganese from the People's Republic of China ("Final Surrogate Value Memorandum") and subsection A.5. below. We disagree with the respondents that we should use PRC electricity rates. No one has disputed our treatment of the PRC as an NME in this review. We have consistently determined that prices in an NME do not allow for fair comparisons. The only exception we have contemplated is for market-oriented industry ("MOI") situations. See e.g., Chrome- Plated Lug Nuts from the People's Republic of China, 57 FR 15054, (April 24, 1992). It is not the Department's policy to consider the use of PRC prices on a factor-by-factor basis. The respondents have not alleged that silicomanganese is a market-orientated industry and we, therefore, have not addressed this issue. A.5. Furthermore, unless the record convincingly demonstrates that factor values are unreliable, the Department generally prefers to stay within the same country for factor valuation wherever possible because it leads to more consistent results than picking and choosing factor values from different countries. The respondents reiterate that they submitted numerous articles from reliable sources that "convincingly demonstrate to any reasonable person that Indian electricity rates are truly aberrational." The respondents contend that the Department would find that all imports of silicomanganese during the POR were dumped if it compares the calculated normal value for Emei and Bayi to the prices of all U.S. silicomanganese imports. Accordingly, the respondents assert that the Department is "telling the Chinese exporters" that they must sell at prices greater than the rest of the world in order not to be guilty of dumping. Lastly, the respondents argue that using the Indian electricity rates would result in a finding that Indian exporters of silicomanganese are dumping in the U.S. market. DOC Position We disagree with the respondents that Indian electricity rates are unreliable. As discussed in detail in the Department's Final Surrogate Value Memorandum, there is nothing particularly unusual about Indian industrial electricity rates. Even if cross-country comparisons of electricity rates were relevant, these show that the Indian industrial rates are comparable to rates in a wide range of countries, with a wide range of levels of economic development, that are silicomanganese producers (see Final Surrogate Value Memorandum). Our analysis of the International Energy Agency ("IEA") data for industrial rates for 32 different countries during the POR shows that the Indian rate of $0.0673 used in the final results is in the middle of rates ranging from $0.0235 to $0.1572. The Indian rate is only slightly higher than the average industrial rates for all 32 countries of $0.0643. In short, the Indian rates are comparable to other rates. The Indian rates for this POR are also consistent with Indian rates from prior years, which were used by the Department as surrogate values in other cases, see Manganese Metal from the PRC: 2nd AR. B. The Department Utilized an Improper Selection Method The respondents argue that the Department's selection method for the surrogate value of electricity is improper. Citing Notice of Final Determination of Sales at Less Than Fair Value: Certain Preserved Mushrooms from the People's Republic of China, 63 Fed. Reg. 72255, 72265 (Dec. 31, 1998), the respondents state that the well established rule is to select surrogate values that are superior in terms of specificity, quality and contemporaneity. With regard to specificity, the respondents contend that the Indonesian industrial rate they provided is more specific since it is divided into three usage categories: low, medium and high usage. Additionally, the respondents state that Indonesia is a significant producer of silicomanganese. With regard to quality, the respondents argue that the Indian government manipulates the electricity sector for social purposes at the expense of the industrial sector and that its control resembles that of a NME. The respondents contend that the Indonesian electricity rates are more reflective of "true market economy conditions." Finally, with regard to contemporaneity, the respondents argue that the provided Indonesian rate is within the POR, as opposed to the 1995 Indian rate used in the preliminary determination. DOC Position We agree with the respondents that contemporaneity and specificity are relevant considerations in selecting surrogate values. Regarding contemporaneity, in the preliminary results we used Indian industrial electricity rates from 1995 because those were the most contemporaneous Indian rates on the record at that time. For these final results, we are using rates from the fourth quarter of 1997, which is contemporaneous with the POR. Regarding specificity, we disagree that the Indonesian rates submitted by the respondents are more specific. Both the Indian rate and the Indonesian rate apply to industries. Furthermore, these rates, if applied to the respondents in this case, would vary from month to month depending on the amount of electricity used during the month (i.e., there are different rates for different consumption ranges). For purposes of our calculation, we cannot reliably match the Chinese electricity utilization level to a specific Indonesian consumption range. The respondents use of electricity and production vary significantly from month to month. Moreover, the respondents produced other products during certain months. Thus, we would have to use an average industrial rate. Accordingly, for purposes of our calculation, the Indian rates (which are contemporaneous averages) are just as specific as the Indonesian averages. Because India is our first choice surrogate country, we continue to use the Indian rates. The Indian Electricity Policy and Sectoral Variations in Electricity Pricing in India and Other Market Economy Countries The respondents argue that Indian electricity rates are not the proper surrogate value because the government of India ("GOI") controls the electricity sector and artificially sets the rates to achieve social goals. They argue that the GOI's planned intervention to institute social and economic policies "invariably" leads to "unreliable and aberrational" rates. Further, the respondents contend that sectoral variations in agricultural, residential, and industrial rates are in direct contradiction of the laws of supply and demand, i.e., "the more you purchase the lower the cost." Accordingly, the Indian social policy is to subsidize residential users at the expense of industrial users. The respondents conclude that the Indian electricity pricing structure is more indicative of a NME. Thus, with respect to its electricity sector, the respondents contend that India should be treated the same as a NME country. DOC Position We disagree with the respondents that India should be treated as an NME with respect to its electricity sector because of the GOI's role in the electricity sector. We have treated India as a market economy country in numerous antidumping cases, see, e.g., Notice of Final Determination of Sales at Less Than Fair Value: Certain Cut-to-Length Carbon-Quality Steel Plate Products From India, 64 FR 73126 (December 29, 1999). Moreover, we do not make a determination on whether a country should be considered an NME based solely on an individual sector of the economy. Rather, section 771(18) of the Act prescribes factors we consider which cover the entire economy. See, DOC Position to Comment A.3. above. Large Indian Industrial Users of Electricity, Such as Silicomanganese Producers, Must Generate Their Own Electricity The respondents argue that Indian electricity rates are not the proper surrogate value because in order to avoid the high costs of electricity, many Indian industries established their own power generation plants. In fact, the respondents claim that most large industries in Indian that rely heavily on electricity as a main source of energy, such as the ferroalloy industry, must either internalize the cost of electricity (i.e., rely on in-house power plants), or cease operations. DOC Position We disagree with the respondents' argument that high electricity rates have caused Indian silicomanganese producers to produce their own power and that this is evidence that Indian electricity rates are not the proper surrogate value. Respondents' own submissions show that India has experienced a steady shortfall of electricity over the last several years. The GOI has made a concerted effort to promote independent power generation to supplement and complement the state grid. Thus, it appears, electricity intensive producers may be setting up their own plants to ensure a steady supply of power, not because of high rates. Moreover, as discussed below in response to section 2E, there is no credible evidence suggesting that the high regional rates are related to the country-wide shortage of electricity. Finally, there is no evidence that all Indian ferroalloy producers are self-producing electricity and not buying it from the state grid. See Final Surrogate Value Memorandum. Accordingly, there is no one specific reason for the internalization of electricity generation and the existence of such internal generation cannot be viewed as indicative of aberrational (i.e. high) electricity rates. The Indian Ferroalloys Industry The respondents argue that "the combination of electricity shortfalls and the inequitable allocation of electricity rates between residential and industrial consumers in India have all but eliminated the Indian ferro-alloys industry." The respondents contend that the GOI "opted to sacrifice" the ferroalloy industry in order to provide affordable electricity to other sectors of the economy. The respondents draw on articles published in newspapers and trade magazines to support their argument. They identify three states in India (Karnataka, Madhya Pradesh, and Maharastra) where the high cost of electricity has forced several ferroalloy companies to close their plants. DOC Position We disagree with the respondents' argument that high electricity rates are driving Indian silicomanganese producers out of business and that this is evidence of aberrationally high Indian industrial electricity rates. The respondents' submissions show that high regional (i.e., state-by-state) rates have caused difficulties for some producers. We, however, used the national industrial average, which is significantly lower than these high regional rates. In fact, the respondents' own submissions show that the average industrial rate we used ($0.0673) is close to the rate in West Bengal of $ 0.0565, which is low enough to attract investment in ferroalloy production (see Attachments 18 and 19 of the respondents' November 29, 1999 submission). The respondents' examples of plants being closed because of high electricity rates stem from Mahdya Pradesh and Karnataka, where the rates are $0.0836 and $0.10332, respectively-- rates that are significantly higher than the national average. Moreover, the respondents' argument that high electricity rates have made 1998 the worst year on record for the Indian ferroalloy industry are likewise inapposite--their own submissions show that the number one problem affecting the industry is overcapacity (see Attachments 18 and 19 of the respondents' November 29, 1999 submission). Overcapacity is proof of significant production volumes, not cutbacks due to prohibitively high input costs (i.e., electricity). Thus, the argument that high electricity rates caused silicomanganese producers to experience their worst year in 1998 is without merit. Indonesian Electricity Rates Provide the Proper Surrogate Value The respondents argue that the average Indonesian electricity rate is more similar to other market economies than the Indian electricity rate for a variety of reasons. First, they cite several cases where the Department used Indonesia as a surrogate country. Second, they argue that Indonesia produces and exports silicomanganese. Finally, using the industrial to residential ratio the respondents contend that, unlike India, Indonesia does not have a rate structure that favors the residential sector at the expense of the industrial sector. Further, the respondents argue that the economic crisis that affected Indonesia should not play a role in the Department's analysis. The respondents contend that the "major cause" of the financial trouble the Indonesian electricity sector experienced was "corruption and nothing more." They state that corruption and its costs should not be taken into consideration since they "were not part of true market forces" and that without the costs of corruption the Indonesian rates were "true market rates" during the POR. The respondents base their argument on three points. First, the Indonesian power company, Perusahaan Listrik Negara ("PLN"), has authority to make periodic adjustments to rates based on fluctuations in the price of inputs and the exchange rate of the rupiah against the dollar. The respondents contend that PLN constantly adjusts its rates to reflect market events. Second, two delays in rate increases in August and November 1998 did not affect the industrial sector and that if they did the delays "probably did not occur" during the POR. Third, financial distortions to the electricity sector were caused by corrupt policies that forced PLN to purchase electricity from foreign investors at "unusually high prices." The respondents contend that if PLN passed onto consumers these costs the electricity rates would have been aberrational, and, as a result, PLN suffered losses during the POR. The respondents conclude that their argument "shows that Indonesian electricity rates are actual market rates since the distortion caused by corruption was not part of the cost structure." The petitioner argues that the Indonesian electricity rate is not the proper surrogate value for several reasons. First, the petitioner contends that the Indonesian rate proposed by the respondents is not "more similar" to other market economies due to the "massive decline" in value of the Indonesian rupiah during the Asian economic crisis. The petitioner bases this argument on the fact that the Indonesian industrial rate of electricity was comparable to the Indian industrial rate for four years prior to the economic crisis, and, in fact, that the rates were identical for both countries in one of those years. Moreover, the petitioner maintains that the "300%" decline in electricity rates from first quarter 1997 to third quarter 1998 is directly the result of the 500% depreciation of the currency and, consequently, is indicative of an aberrationally low price. The petitioner further argues that the Indonesian rate provided by the respondents is "150% lower than the next lowest industrial rate" provided in the Department's Preliminary Surrogate Value Memorandum. In addition, the petitioner contends that the Indian rate is not unusual when compared to the average rates of countries that are identified in the Preliminary Surrogate Value Memorandum. Second, the petitioner argues that the Indonesian rates would not allow the Department to properly determine the market valued cost of producing and selling the subject merchandise, plus profit. The petitioner contends that the Indonesian government's control of PLN and the tariff structure is contrary to the Department 's practice of not using government-subsidized prices as factor values in NME cases. Further, the petitioner rejects the respondents' argument that "if costs related to corruption were subtracted from the equation" that Indonesian electricity rates during the POR were true market rates. The petitioner contends that PLN suffered huge losses during the POR as result of two factors: 1) PLN had pre-existing contracts that required it to purchase electricity in dollars and that those costs increased greatly due to the decline in value of the rupiah, and, 2) PLN's financial situation was exacerbated by the government's refusal to allow PLN to raise rupiah rates in successive quarters to cover the increased generation costs associated with aforementioned contracts. The petitioner concludes its argument by asserting that if corruption occurred in the Indonesian electricity sector, it is further reason for not using the Indonesian rate. DOC Position We agree with petitioners that the Indonesian rate is not an appropriate surrogate. This rate appears to have dropped precipitously during the Asian economic crisis and is a considerably lower rate than almost all other rates we have on the record. Moreover, the record also shows that Indonesian rates from years prior to the Asian economic crisis were comparable to Indian rates (see Additional Research Memorandum). Thus, to the extent that any electricity rates on the record of this review should be considered unreliable, it is the 1998 Indonesian rate that appears unreliable. G. The Department Should Use The RBI Electricity-Specific Price Index To Adjust The Surrogate Value For Electricity For Inflation The petitioner requests that the Department use an Indian wholesale price index ("WPI") from the Reserve Bank of India ("RBI") instead of the International Monetary Fund ("IMF") WPI used by the Department in the preliminary results. The petitioner argues that the RBI rate more accurately reflects the rate of inflation for electricity products because the RBI rate provides electricity-specific information, as opposed to the IMF's total commodity index. The respondents did not comment on this issue. DOC Position The Indian electricity rates we used in the final results are contemporaneous with the POR and therefore do not need to be inflated. H. The Department Should Use The More Specific And Contemporaneous Electricity Data Submitted By Eramet The petitioner argues that more contemporaneous Indian electricity rates are on the record of this review and should be used in the final determination. The petitioner contends that the rate they submitted from the Center for Monitoring the Indian Economy ("CMIE") is "significantly more contemporaneous" than the International Energy Agency ("IEA") rate used by the Department. The petitioner further argues that the CMIE data are more specific because they are separated by size of the industry: small, medium and large industrial users, and that the large industrial category is further subdivided into eleven classes of industrial user. The petitioner also contends that the Department can calculate each respondent's monthly usage and apply the appropriate applicable rate in order to accurately calculate a dumping margin. Finally, the petitioner recommends that if the Department does not use rates from the CMIE, the Department should use more contemporaneous rates provided by the IEA. The respondents argued that the Department should value electricity using an Indonesian surrogate. See the respondents' arguments above. DOC Position We disagree with the petitioner. For these final results, we are using the IEA average industrial rate from the fourth quarter of 1997, which is more contemporaneous with the POR than the CMIE data. The CMIE data, divided into specific utilization categories, are based on electricity rates in 1996 and 1997, largely outside the POR. Regarding specificity, we disagree that the Indian rates submitted by the petitioner are more specific. The petitioner calculated an average industrial rate that it asserts is more specific to the experience of the Chinese producers because it takes account of their monthly electricity usage. However, the petitioner's methodology involves picking and choosing different rates from different Indian states for different time periods. This methodology implies a high level of precision in matching the respondents' monthly usage amounts with these different Indian rates. One of the problems with this methodology is that it does not account for the fact that the respondents also produce non-subject merchandise in their facilities. Moreover, it is not clear that this detailed mixing and matching of rates would reasonably reflect the type of Indian industrial rates that would be available to ferroalloy producers. This is an important consideration given that the petitioner's calculated rate is appreciably higher than the IEA rate. In considering the reasonableness and reliability of the petitioner's calculated average rates, we considered the fact that the petitioner submitted 24 individual price lists from different Indian states, covering a two year period. These price lists were contained within a CMIE publication. However, because the petitioner did not submit the entire CMIE publication, we have no way of knowing whether there were other price lists that the petitioner omitted, or whether CMIE itself calculated an average industrial rate which was not reported. Because of this, we cannot evaluate whether the average of these prices calculated by the petitioner is truly reflective of the average industrial rate. By contrast, the IEA rate we have is clearly an average industrial rate. The IEA has routinely calculated and published average rates for OECD and non-OECD countries. It follows a consistent methodology in compiling these averages over time and across countries. (See IEA Energy Prices and Taxes, INTRODUCTION and GENERAL NOTE ON DEFINITIONS, METHODS AND SOURCES, appended to the Final Surrogate Value Memorandum at Attachment 4.) There is no reason to conclude that ferroalloy producers would incur the rate calculated by petitioners. Accordingly, we find that the IEA is the best available information because it is more reliable and contemporaneous. Comment 3: Selecting a Surrogate Value for Manganese Ore The petitioner maintains that the Department improperly valued domestic (PRC-produced) manganese ore because it based the value on the price of Indian ore that is not comparable to the domestic ore used by the respondents. Specifically, the petitioner contends that the Department erred when it determined that the Indian ore was comparable to the domestic ore based solely on the ore's manganese content. The petitioner notes that it submitted an affidavit from its employee, Donald Kozak, who has extensive technical experience in the manganese industry, which stated that manganese ores are differentiated by both the manganese content and the ratio of contained manganese to contained iron. According to the petitioner, the Indian and domestic ores are physically dissimilar because the manganese-to-iron ratio of the Indian ore (1.3 to 1) is significantly less than the manganese-to- iron ratios of the domestic ore used by the respondents (3.9 to 1 and 11.8 to 1 for Bayi and Emei, respectively). Moreover, the petitioner claims that if the respondents had produced silicomanganese using the Indian ore rather than the domestic ore, the manganese-to-iron ratio of the silicomanganese would have been far different from the manganese-to-iron ratio of the silicomanganese they actually produced and the ratio would not have met industry standards. This claim is based upon a calculation that the petitioner performed using the respondents' production information as well as information provided by Mr. Kozak (for details regarding the calculation, see Exhibit 1 of the petitioner's March 24, 2000 case brief). The petitioner points out that the result of its calculation is supported by Mr. Kozak's affidavit wherein he stated that a typical silicomanganese producer "could not use the high-iron Tata ore {Indian ore} as its manganese ore input." In addition, the petitioner asserts that its position is supported by decisions of the Department and the Court of International Trade ("CIT"). Specifically, the petitioner notes that in the administrative reviews of manganese metal from the PRC, the Department recognized the appropriateness of considering more than just manganese content in selecting a surrogate value for manganese ore. See Manganese Metal from the People's Republic of China; Final Results and Partial Rescission of Antidumping Duty Administrative Review, 63 FR 12440, 12443 (March 13, 1998)("Manganese Metal from the PRC"). The petitioner points out that in the manganese metal reviews, the Department selected a surrogate value for manganese ore with a manganese-to-iron ratio similar to that of the PRC ore. Also, the petitioner states that the CIT held that the Department may not select surrogates solely on the basis of concentration levels of various component chemicals where the record showed that the presence or lack of other elements that are impurities affect value. See Rhone- Poulenc, Inc. v. United States, 927 F. Supp. 451,456 (CIT 1996). Furthermore, the petitioner notes that the CIT has required the Department to compare a potential surrogate value with other potential surrogate values in determining whether the surrogate is appropriate. See Union Camp Corp. v. United States, 8F. Supp 2d 842, 848-50 (CIT 1998)("Union Camp"). The petitioner notes that the surrogate value used in the preliminary results is from 74% to 80% less than the other surrogate values for this input on the record of this review. Based on the foregoing, the petitioner concludes that the surrogate value used by the Department in the preliminary results is unreliable and does not reflect the price of manganese ore suitable for use in silicomanganese production. The petitioner urges the Department to value manganese ore using the price quote that it submitted for Brazilian manganese ore because this ore is more comparable to the PRC ore than any of the other ores for which prices were submitted. The Brazilian ore contains 31.00% to 34.00% manganese, 2.5% to 4.5% iron, and a manganese-to-iron ratio of 9.3 to 1. The respondents urge the Department to value manganese ore using a Brazilian price quote submitted by an interested party in the 1997- 1998 antidumping administrative review of manganese metal from the PRC. According to the respondents, this price quote is the most appropriate surrogate value for the PRC-produced (domestic) manganese ore that they use because: 1) the price quote is for manganese ore containing manganese and iron concentrations similar to those found in the domestic ore used to produce silicomanganese (i.e. manganese and iron concentrations of 29.5% and 4.7% respectively for domestic ore compared with manganese and iron concentrations of 30.35% and 4.57% for the Brazilian ore); 2) the Brazilian price quote has already been used by the Department to value manganese ore produced in the PRC (see Manganese Metal from the PRC: 2nd AR, also, see Manganese Metal from the People's Republic of China; Preliminary Results and Partial Rescission of Antidumping Administrative Review, 64 FR 68999 (December 9, 1999)); and 3) the price quote, which is for a shipment of manganese ore from Brazil to the PRC, is valid until June 9, 1998 which is well within the period of the instant review. However, the respondents note that before valuing manganese ore using the Brazilian price, the Department should adjust the price downward to reflect the much lower manganese concentration of the domestic manganese ore used to produce silicomanganese. The petitioner requests that the Department reject the respondents' Brazilian price quote based on the information it was able to obtain from the producer who supplied the quote. Specifically, the producer told the petitioner's Brazilian researcher that it had ceased mining ore at the end of 1997 and that in 1998 it was selling only remaining stock consisting of ore "totally out of official specifications." Moreover, the producer noted that the price of this low quality ore that does not meet official specifications "can{not} be considered for research purposes." In addition, according to the petitioner, the Department found that the U.S. importer that obtained the Brazilian price quote was engaged in submitting falsified documents to the U.S. Customs service. See Manganese Metal from the PRC: 2nd AR, 64 FR at 49454 . Therefore, the petitioner urges the Department to reject the respondents' Brazilian price quote. DOC Position We agree, in part, with both parties. The record in this review indicates that iron content and the manganese-to-iron ratio are important characteristics to consider in deciding whether different ores are comparable. Based on these characteristics, we have determined that the Brazilian manganese ore surrogate submitted by the respondents is more comparable to the manganese ore used by the respondents than the Indian ore which we used as a surrogate in the preliminary results. The composition of the respondents' Brazilian ore is similar to the average chemical specifications of the Chinese manganese ore inputs (see Final Surrogate Value Memorandum). We were also unable to find suitable Indian values for the appropriate ore type. Thus, for the final results, we valued domestic manganese ore using the Brazilian ore price submitted by the respondents because this is the best available value for this input. We did not reject the respondents' Brazilian price, which had been submitted in the 1997-1998 antidumping administrative review of manganese metal from the PRC, for the same reasons that the Department did not reject the price in that review where we stated: Moreover, given that the specifications stated for the 1998 price quotation were essentially the same as those for the 1993 price quotation (which was, undisputably, for a commercial grade ore), it would seem likely that the ore producer, a long-established seller of ore on the world market, would clearly indicate in the 1998 quotation that the ore grade on offer was not of commercial quality, if that were the case. There is nothing in the 1998 price quote, however, indicating that the merchandise on offer is not of normal commercial grade. Also, contrary to the information provided by the petitioners' researcher that "the remaining inventories of 1998 refers to the cleaning of stocks, with very low quantity ..." the quoted 1998 price is for a quantity of 35,000 to 44,000 metric tons, an amount which would generally be considered commercial. Additionally, despite the petitioners' general assertion to the contrary, there is no evidence on the record to suggest that in 1998 the Brazilian mine sold its ore at a discount merely because it was in the process of closing down its mining operations. Furthermore, we reject the petitioners' argument that we should not utilize information that was sent to a company accused by parties in this case of customs fraud. The price quotation was generated by the Brazilian producer and there is no evidence indicating that the producer was involved in any fraudulent activity. See Manganese Metal From the PRC: 2nd AR, 64 FR at 49454. We did not use the Brazilian ore value submitted by the petitioner because: 1) the petitioner's price quote, dated September 8, 1993, is not contemporaneous with the POR; 2) the petitioner did not provide sufficient details regarding the price quote (e.g., information regarding the parties involved in the transaction); and 3) the country of origin for the manganese ore described in the price quote is not clearly indicated. For these reasons, we find the petitioner's Brazilian price quote unreliable. See Exhibit 3 of the petitioner's November 29, 1999 surrogate value submission. Finally, we disagree with the respondents that the Brazilian surrogate value should be adjusted to reflect the actual manganese content of the domestic manganese ore used by the respondents. The chemical specifications of the respondents' manganese ore inputs are averages that cover a variety of domestic manganese ores with varying manganese concentrations. See Verification of the Sales Responses of Sichuan Emei Ferroalloy Import and Export Co., Ltd in the 1997-1998 Administrative Review of Silicomanganese from the People's Republic of China (A-570-828), ("Emei Verification Report") dated March 1 , 2000, at page 16; also See Verification of the Sales Responses of Guangxi Bayi Ferroalloy General Works in the 1997-1998 Administrative Review of Silicomanganese from the People's Republic of China (A-570- 828), ("Bayi Verification Report") dated March 1, 2000, at page 19. Therefore, we find that the Brazilian ore price quote is a reliable surrogate value, unadjusted, for the average chemical specifications of domestic manganese ore. Comment 4: Selecting a Surrogate Value for Manganese-Rich Slag The petitioner contends that the Department improperly valued manganese-rich slag because it based the value on the price of Indian ore that is not comparable to the manganese-rich slag used by the respondents. The petitioner notes that the Indian ore may consist of up to 25% iron and it has a manganese-to-iron ratio of 1.3 to 1. In contrast, the petitioner claims that the manganese-rich slag used by the respondents contains very little iron (0.85% and 2.0% for Bayi and Emei, respectively) and it has significantly greater manganese-to- iron ratios (38.5 to 1 and 15.0 to 1 for Bayi and Emei, respectively). In addition, the petitioner asserts that the record in this review contains evidence showing that the Indian ore cannot be used in place of the manganese-rich slag used by the respondents because the resulting silicomanganese would have too much iron and too little manganese. See Exhibit 1 of the petitioner's November 29, 1999 surrogate value submission. Furthermore, the petitioner contends that the Department's reason for using the price of Indian ore to value manganese-rich slag is flawed. The petitioner notes that the Department used the low-priced Indian ore to value slag because slag is "regarded as low grade manganese." See Preliminary Results, 64 FR at 60786. The petitioner maintains that the record does not support the Department's reasoning. Specifically, the petitioner notes that slag contains very little iron, which is a key determinant of the value of a manganese source. See Exhibit 1 of the petitioner's November 29, 1999 surrogate value submission. Moreover, the petitioner claims that manganese-rich slag is a "desirable" input in silicomanganese production because it will result in a more stable and improved operation. See Attachment A at Exhibit 5 of the petitioner's October 12, 1999 surrogate value submission. The petitioner urges the Department to value manganese-rich slag using the U.S. price that it submitted because this price is for the exact input used by the respondents. According to the petitioner, the Department must use the surrogate value most specific to the actual input being valued. Moreover, the petitioner argues that in order to accurately value slag it is necessary to use a value that is not from the primary surrogate country. According to the petitioner, the courts and departmental practice support using values outside the primary surrogate country if it is necessary to do so in order to accurately value an input. See Union Camp, 8 F. Supp.2d 842, 850; also See Manganese Metal from the PRC 63 FR at 12444. Alternatively, if the Department does not use the U.S. price, the petitioner requests that the Department value manganese-rich slag using the Brazilian ore price that it submitted. The petitioner claims that this value is appropriate because it is for an ore that is more comparable to the slag used by the respondents than the Indian ore used in the preliminary results. The respondents argue that manganese-rich slag has no market value in China. According to the respondents, the manganese-rich slag that they generate during ferromanganese production is a waste product with an extremely low iron content which was verified by the Department. Therefore, the respondents contend that it has no value to the Chinese other than as an input in silicomanganese production. The respondents claim that the U.S. price submitted by the petitioner comes from the petitioner's own invoices for highly concentrated ferromanganese slag. The respondents point out that these prices do not constitute publicly available information and have not been verified. Moreover, the respondents note that just because the petitioner sold ferromanganese slag at a certain price does not mean that Chinese ferromanganese slag should be valued based on that price. DOC Position We agree, in part, with the petitioner. There is no evidence on the record to support the respondents' assertion that manganese-rich slag has no market value in China. Thus, normal value should include a per- unit cost for manganese-rich slag that is based upon an appropriate surrogate value. The record in this review indicates that iron content and the manganese-to-iron ratio are important characteristics to consider in identifying a surrogate that is comparable to manganese-rich slag. Based on these characteristics, we have determined that the Brazilian manganese ore price quote submitted by the respondents is more comparable to the manganese-rich slag used by the respondents than the Indian ore which was used as a surrogate in the preliminary results. Although we used the same Brazilian price quote to value domestic manganese ore, which differs in content from manganese-rich slag, we nonetheless find that based on the characteristics noted above, the Brazilian ore price is the best information available on the record regarding the value of manganese- rich slag. We did not use the Brazilian ore value submitted by the petitioner for the reasons stated in our position to Comment 3. Moreover, we did not use the petitioner's U.S. prices for ferromanganese slag because: 1) there is no information on the record regarding the iron content or the manganese-to-iron ratio of the slag; and 2) the prices were from sales invoices issued in 1996 and 1999 and thus, they are not contemporaneous with the POR (see Exhibits 1 and 2 of section C of the petitioner's October 15, 1999 submission regarding factor values). Thus, for the final results, we valued manganese-rich slag using the Brazilian ore price submitted by the respondents. Comment 5: Selecting a Surrogate Value for Plastic Woven Bags The petitioner notes that in the preliminary results, the Department valued plastic woven bags using Indian import statistics for the material polyethylene. However, the petitioner requests that the Department value plastic woven bags using the Indian import statistics that it submitted for "sacks and bags...of polyethylene.." because these sacks and bags are more comparable to the plastic woven bags than polyethylene. The respondents did not comment on this issue. DOC Position We agree with the petitioner. Sacks and bags of polyethylene are more similar to the plastic woven bags used by the respondents than polyethylene. Therefore, for the final results we valued bags using the Indian import statistics for "sacks and bags...of polyethylene.." See Final Surrogate Value Memorandum. Comment 6: Including the Cost of Silicon Ore, Fluorspar, and Sintered Manganese Ore in NV The petitioner maintains that in the preliminary results, the Department erroneously failed to include the cost of silicon ore, fluorspar, and internally-produced sintered manganese ore in normal value. The petitioner notes that the Department excluded these costs because it was unable to find usable factor values for these materials. In the final results, the petitioner urges the Department to value silicon ore and fluorspar using the values it submitted for quartz (another term for silicon ore) and fluorspar, respectively. Because the respondents produced sintered manganese ore, the petitioner urges the Department to follow its established practice of valuing self-produced inputs using the value of the materials, energy and labor employed to manufacture the input. See Final Determination of Sales at Less Than Fair Value: Certain Cut-to-Length Carbon Steel Plate from the People's Republic of China, 62 FR 61964, 61976 (November 20, 1997). The respondents did not comment on this issue. DOC Position We agree with the petitioner. For the final results, we valued silicon ore and fluorspar using surrogate values submitted by the petitioner, while we valued sintered manganese ore using the factors of production reported by the respondents. See Final Surrogate Value Memorandum. In valuing Bayi's sintered manganese ore, we adjusted Bayi's labor factor based on information obtained at verification. For details, see Bayi Verification Report and the Department's May 8, 2000 memorandum Analysis for the Final Results of the 1997-1998 Administrative Review of Silicomanganese from the People's Republic of China: Guangxi Bayi Ferroalloy Works) ("Bayi Analysis Memorandum"). II. Facts Available Comment 1: Use of Partial Adverse Facts Available The petitioner claims that Bayi failed to provide the Department with necessary information regarding the transportation of imported manganese ore to its factory. Although Bayi reported that the ore arrived at two Chinese ports, it did not identify which shipments arrived at these ports or provide any information regarding the relative volumes arriving at each port. Further, the petitioner notes that Bayi failed to identify the mode of transportation used to move imported manganese from the ports to Bayi's factory. Therefore, the petitioner requests that the Department calculate Bayi's cost of transporting imported manganese ore to its factory using adverse facts available. Specifically, the petitioner requests that the Department calculate this cost by assuming that Bayi transported all imported manganese ore by truck over the greater of the two port distances.(1) Alternatively, if the Department does not use adverse facts available, the petitioner states that the Department should require Bayi to provide the missing information. The respondents did not comment on this issue. DOC Position We agree with the petitioner that Bayi failed to provide the Department with necessary information regarding the transportation of imported manganese ore to its factory. However, we have used facts available, rather than adverse facts available, to calculate Bayi's freight costs for imported manganese ore. In selecting from among the facts otherwise available, section 776(b) of the Act permits the Department to use inferences that are adverse to a party when it finds that the party has not acted to the best of its ability to comply with the Department's request for information. Throughout this review, Bayi has responded to each of the Department's requests for information. Although Bayi failed to provide certain freight information in its questionnaire response, this deficiency was not addressed in a supplemental questionnaire. Therefore, we have not used adverse facts available. As facts available, we valued Bayi's cost as if Bayi used trucks to transport imported manganese ore from the ports to its factory since this method of transportation was reported by Emei. As facts available, we calculated Bayi's freight cost for all imported manganese ore using a simple average of the distances between Bayi's factory and the two ports of importation. See Bayi Analysis Memorandum. III. Clerical Errors Comment 1: Clerical Errors in the Preliminary Results The petitioner alleges that the Department used incorrect distances in calculating the rail freight expense incurred to transport coke to Emei and Bayi and domestic manganese ore to Emei. Also, the petitioner alleges that the Department mistakenly used rail freight rates to calculate Emei's transportation expense for imported manganese ore. According to the petitioner, Emei stated that it transported imported manganese ore by truck. In addition, the petitioner claims that the Department calculated packing costs for both the respondents using an incorrect weight for plastic bags. Finally, the petitioner states that the Department inadvertently deducted the cost of manganese-rich slag from the production costs used to calculate normal value. The petitioner notes that in the preliminary results, the Department specifically identified manganese- rich slag as a major input in the production of silicomanganese and, thus, the cost of this input should be included in normal value. See Preliminary Results, 64 FR at 60787. The petitioner requests that the Department recalculate the above rail costs using the correct distance-specific rail rates from the Indian Railway Conference Association's publication Goods Tariff No. 42, Part II. In addition, the petitioner requests that the Department recalculate Emei's transportation expense for imported manganese ore using the Rahul Roadlines truck freight rate. Lastly, the petitioner notes that the correct weight of plastic bags can be found in Appendix C-1 of Bayi's and Emei's questionnaire responses. The respondents did not comment on these alleged errors. DOC Position We agree with the petitioner and have corrected each of these errors in the final results. For details regarding these corrections, see the Department's May 8, 2000 company-specific analysis memoranda (Bayi Analysis Memorandum and Analysis for the Final Results of the 1997-1998 Administrative Review of Silicomanganese from the People's Republic of China: Sichuan Emei Ferroalloy Import and Export Co., Ltd. ("Emei Analysis Memorandum"). IV. Miscellaneous Issues Comment 1: Classifying Electrode Paste as a Direct Material or Part of Overhead The petitioner contends that electrode paste is a direct material input in silicomanganese production and not part of factory overhead costs. According to the petitioner, the Department's practice is to consider materials as part of factory overhead only when they are infrequently used in the production process and small in value relative to the total cost of manufacturing. See Notice of Final Determination of Sales at Less Than Fair Value; Saccharin from the People's Republic of China, 59 FR 58818, 58824 (November 15, 1994)("Saccharin from the PRC"). The petitioner notes that electrode paste is routinely used in the production process because it is used to make electrodes which deliver electricity to the raw materials furnace. According to the petitioner, the paste is burned-off during production and must continually be replaced. Thus, the petitioner claims that use of electrode paste is significant in the production process and, based on the surrogate values used in the preliminary results, its cost is an important part of the cost of manufacturing silicomanganese. In addition, as stated in Mr. Kozak's affidavit, the petitioner maintains that the silicomanganese industry does not consider electrode paste to be part of factory overhead (as noted above, Mr. Kozak is an employee of the petitioner with extensive experience in the manganese industry). Based on the foregoing, the petitioner maintains that electrode paste is neither small in value relative to the total cost of manufacturing nor infrequently used in silicomanganese production and, thus, it is a direct material. The respondents maintain that electrode paste is not a direct material input since it is not incorporated into silicomanganese.(2) The respondents note that during silicomanganese production, the electrode paste, which is contained within an electrode (a cylindrical steel shell (i.e. a pipe)), is used to conduct electricity from an electrical source to the end of the electrode where an electric spark is generated. During this process, the electrode paste is sintered and cooked and heat is generated for smelting. After a period of time the sintered electrode paste is removed as a solid. Thus, the respondents contend that electrode paste is the part of the furnace used to convert electrical energy into heat energy and not a material incorporated into the final product.(3) According to the respondents, this fact was recognized by Department officials during verification. Specifically, the respondents note that analysts from the Department reported that "electrode paste is distributed directly into the steel pipes {electrodes}, separate from the raw materials" and that periodically "the steel pipes erode and the sintered electrode paste is removed as a solid." See Emei Verification Report at page 13; also see Bayi Verification Report at page 13. Although, the steel pipe erodes and traces of it enter the molten silicomanganese, the respondents contend that there is no evidence that any electrode paste enters the molten silicomanganese. In fact, the respondents argue that the melted electrode with the hardened electrode paste that the Department's analysts observed at Bayi, constitutes direct evidence that electrode paste remains entirely separate from molten silicomanganese. See Bayi Verification Report at page 13. In addition, the respondents note that in Saccharin from the PRC, the Department treated certain materials, that played a role similar to that of electrode paste, as part of factory overhead. Specifically, the respondents claim that, like the coolants in Saccharin from the PRC, which the Department treated as part of overhead, electrode paste is used as a medium through which energy is transferred. Thus, the respondents argue that electrode paste is simply a tool used to convert electrical energy into heat energy, and not a direct raw material input in the production of silicomanganese. Finally, the respondents point out that treating electrode paste as a direct material input would double count its cost because the cost is already included in overhead costs. DOC Position We agree, in part, with both parties. We agree with the petitioner that electrode paste is neither used in small quantities nor infrequently used. Rather, it is used regularly and in significant quantities as a necessary part of the production process. We agree with respondents that electrode paste is not physically incorporated within silicomanganese and, thus, would not normally be considered a direct material input. Electrode paste is more properly classified as a "process material" as it was called in the silicomanganese petition (see Petition in the Matter of Silicomanganese From Brazil, the People's Republic of China, Ukraine and Venezuela dated November 12, 1993 at Exhibit 6T which is included as Attachment 5 to the Final Surrogate Value Memorandum). Such materials are often included in factory overhead as "consumables" when they are used in production infrequently and in small quantities. However, when such materials constitute a significant portion of the cost of the finished product, companies may choose to trace the cost of the materials to the finished product, rather than allocating them over total production. We note that in other antidumping cases involving producers of ferroalloys, the respondent producers reported electrode paste as a separate element of cost rather than subsumed within factory overhead costs. See excerpts from Companhia Brasileira Carbureto De Calcio's November 10, 1998 supplemental questionnaire response (public version), in the seventh administrative review of silicon metal from Brazil (A-351-806), and Companhia Ferroligas Minas Gerais' July 13, 1998 supplemental questionnaire response (public version), in the fourth administrative review of ferrosilicon from Brazil (A-351-820) which are included as Attachment 6 to the Final Surrogate Value Memorandum.) In those cases, we valued electrode paste as a cost element separate from factory overhead. Comparisons to other ferroalloy producers are relevant to our analysis here because many ferroalloy products share substantially similar production processes which consume significant quantities of electrodes or electrode paste. In addressing this issue, we need to determine what is the best available information on the record of this review for valuing this significant factor of production. There are two choices: 1) assume that the cost of electrode paste is included in the overhead rate for the metals and chemicals industry from the Reserve Bank of India Bulletin (the valuation methodology used in the preliminary results), or 2) value electrode paste using the actual quantities consumed by the respondents and an Indian price quote for electrode paste. The problem with the first choice is that we would have to assume that the overhead rate for the Indian metals and chemical companies from the Reserve Bank of India Bulletin reflects the use of a significant amount of electrodes. Given that we do not have any information to support this assumption, we find that the second choice is clearly better because it ensures that the cost of electrode paste is included in normal value by specifically valuing electrode paste using the respondents' own consumption figures and an Indian price for electrode paste. The second choice is the best available information and results in the most accurate calculation of normal value. Although we are not satisfied that the overhead rate used in the preliminary results adequately accounts for the significant amount of electrode paste used by the respondents, we note that the overhead expenses used to calculate the rate include an expense category identified as "stores and spares." This expense category would typically include the cost of "consumable materials" similar to "process materials," like electrode paste, except that these materials are generally used infrequently and in small quantities. In order not to inadvertently overstate, or double-count these process materials, we recalculated the overhead rate by excluding the "stores and spares" expense category from the expenses used in our calculation. We valued electrode paste using the 1993 Indian price of electrode paste from the silicomanganese petition. We used the petition price, after adjusting for inflation, rather than the petitioner's Indian import price because the petition price is more specific to the input being valued. Comment 2: Allocating Costs Over Production Quantities That Include Fines The petitioner contends that the respondents incorrectly included fines in the production quantities used to calculate the factors of production. According to the petitioner, the Department verified this fact. See Bayi and Emei Verification Reports at page 12. The petitioner maintains that the Department's practice is to allocate production costs only to commercial-grade merchandise. The petitioner notes that in silicon metal from Brazil, the Department held that production costs should not be allocated to by-products such as ladle sculls, off-grades, and fines. See Silicon Metal from Brazil; Final Results of Antidumping Duty Administrative Review, 62 FR 1954, 1964 (January 14, 1997). Thus, the petitioner claims that the Department must recalculate the respondents' factors of production by dividing the quantity of each factor consumed during the POR by the volume of commercial-grade silicomanganese produced during the POR. Because this calculation uses information not on the record, the petitioner requests that the Department require the respondents to report the volume of fines that they generated during the POR. The respondents did not comment on this issue. DOC Position We disagree with the petitioner. The record in this review does not support the petitioner's claim that fines are by-products rather than commercial-grade silicomanganese. In determining whether a product is a by-product, the Department considers, among other things; 1) the relative sales value of the product compared to that of all other joint products produced during the same time period, 2) whether the product is an unavoidable consequence of producing another product, 3) whether management intentionally controls production of the other product, 4) whether the product requires significant further processing after the split-off point, and 5) how the company has historically accounted for the product. See Elemental Sulphur from Canada; Final Results of Antidumping Finding Administrative Review, 61 FR 8239, 8241 (March 4, 1996). These factors have not been addressed by the petitioner. Moreover, there is no information on the record regarding a number of these factors. Furthermore, the petitioner failed to explain why it considers silicomanganese fines, which are small pieces of the parent alloy containing the same chemistry as the parent, to be non-commercial grade silicomanganese. Finally, the scope of this review includes all compositions, forms, and sizes of silicomanganese, including fines. Thus, excluding fines from the production quantity used to calculate the reported factors would overstate the factors of production. Therefore, we have not recalculated the factors of production as suggested by the petitioner. Comment 3: Reducing Normal Value for Sales of Silicomanganese Slag The petitioner urges the Department not to reduce constructed value (which is normal value in this case) by surrogate revenue associated with slag sales for several reasons. First, although the respondents indicated in their supplemental questionnaire responses that they did not produce slag, at verification they reported for the first time that they had produced slag during the POR (see Antidumping Duty Administrative Review on Silicomanganese from the PRC: Supplemental Questionnaire Response dated July 30, 1999 ("Supplemental Questionnaire Response")). This is new information which the Department normally does not accept. See Certain Circular Welded Non- Alloy Steel Pipe from Mexico: Final Results of Administrative Reviews, 65 FR 6136, 6138 (February 8, 2000) ("Steel Pipe from Mexico"), which notes that verification is not an opportunity to provide substantial new data; also see Heavy Forged Hand Tools, Finished or Unfinished, With or Without Handles, from the People's Republic Of China; Final Results of Antidumping Duty Administrative Reviews, 63 FR 16758, 16761 (April 6, 1998) ("Hand Tools from the PRC"), where the Department noted that "{v}erification is not normally an appropriate venue for the submission of new factual information." Moreover, the petitioner contends that the Department should not use the information provided at verification to reduce constructed value when the respondents failed to provide this information in response to the Department's supplemental request. Second, the petitioner notes that at verification, Bayi used a formula to estimate the amount of slag that it produced. According to the petitioner, the statute and various rulings of the Court of Appeals for the Federal Circuit ("CAFC") and the CIT require the Department to calculate constructed value using actual factors of production. Third, the petitioner points out that the Department offsets costs only with revenue actually earned from secondary sales (i.e., sales of by-products, scrap, etc.). See Final Results of Antidumping Duty Administrative Review: Silicon Metal from Brazil, 65 FR 7497, 7500 (February 15, 2000). According to the petitioner, Bayi never reported the quantity of slag sold. Lastly, the petitioner states that parties to this proceeding never had an opportunity to submit surrogate values for slag since, prior to the verification, the respondents denied producing slag. Nevertheless, the petitioner asserts that slag is worth very little. The respondents disagree with the petitioner's argument. Specifically, the respondents claim that the Department's positions in Steel Pipe from Mexico and Hand Tools from the PRC do not apply here because the respondents in both of those cases submitted substantial changes to their data at verification. Moreover, the respondents note that in Hand Tools from the PRC, the Department decided to use the caps (estimates) reported for steel consumption rather than the actual steel consumption figures obtained at verification because use of the actual figures would impose an unreasonable burden with no significant increase in accuracy. Consistent with this position and the Department's responsibility to determine margins as accurately as possible (see Lasko Metal Products, Inc. v. United States 16 ITRD 2188 (CAFC 1994)), the respondents request that the Department grant the slag offset because it would result in a more accurate margin calculation without imposing an unreasonable burden on the Department. DOC Position We agree with the petitioner's conclusion and have not reduced normal value to account for sales of silicomanganese slag. In our July 12, 1999 supplemental questionnaire, we requested that Bayi and Emei report their production volumes of fines and alloy-poor slag. See Silicomanganese from the PRC: Supplemental Questionnaire dated July 12, 1999. Neither of the respondents reported that they produced alloy-poor slag during the POR. See Supplemental Questionnaire Response, question nine at page 3 of each respondent's response. However, at verification, the Department found that both the respondents generated a certain quantity of alloy-poor slag during silicomanganese production. In addition, the Department found that both of the respondents sold silicomanganese slag. The respondents never disclosed these facts prior to verification and, thus, the petitioner did not have an opportunity to comment on this information or provide a surrogate value for silicomanganese slag. This is particularly significant given the fact that an adjustment to normal value for sales of silicomanganese slag would benefit the respondents. Furthermore, because the respondents failed to provide this information in a timely fashion the Department had no opportunity to analyze the information prior to verification, nor to identify potential sources of surrogate value. Because of this, we lack sufficient information about this material necessary for identifying appropriate surrogate values. In addition, prior to verification, the Department had no opportunity to issue supplemental questions regarding this information. The Department has previously stated: It is the responsibility of respondents to provide an accurate and complete response prior to the preliminary determination and verification so that the Department may fully analyze the response and other parties may comment on it. See Final Determination of Sales at Less Than Fair Value: Certain Hot-Rolled Carbon Steel Flat Products and Certain Cold-Rolled Carbon Steel Flat Products from the Netherlands, 58 FR 37199, 37203 (July 9, 1993). While the respondents are correct in noting that the Department has a responsibility to determine margins as accurately as possible, the CIT has held that the burden for the creation of an adequate record rests with the respondents. See Chinsung Indus. Co. v. United States, 705 F. Supp. 598,601 (CIT 1989). The respondents, in the instant review, failed to create an adequate record by not submitting the information requested by the Department prior to verification. In light of the respondents' failures, it is inappropriate to use the information obtained at verification to reduce normal value. Comment 4: Recalculating Emei's Electricity Consumption Based on Verification Findings The petitioner requests that the Department reject the electricity consumption figure reported by Emei because the Department found that this figure understated electricity consumption. According to the petitioner, at verification, the Department found that Emei failed to report the amount of electricity consumed by supporting departments (e.g. repair shop, laboratory, etc.) and the amount of power lost on the line for workshop five and on the line for the supporting departments. The petitioner urges the Department to recalculate the amount of electricity consumed by Emei using the information obtained at verification. The petitioner maintains that the recalculated consumption figure should include 1) the amount of all direct and indirect electricity consumption recorded for workshops five and six; 2) all line losses for workshops five and six; and 3) an allocated share of total indirect electricity and line losses for all supporting departments. According to Emei, the Department should base its determination on the verified factors of production. However, Emei notes that factory overhead may include energy costs and, thus, it requests that the Department's methodology not double count electricity costs. DOC Position We agree with both parties. At verification, the Department found that Emei failed to report all of the indirect electricity and line losses related to silicomanganese production. Therefore, for the final results, we increased the reported electricity consumption to account for these unreported amounts. Although the respondents expressed concerns that factory overhead may already include the cost of indirect electricity and line losses, we determined that the surrogate value for factory overhead does not include energy costs. We calculated the surrogate value for factory overhead using the 1992- 1993 combined profit and loss statement for the Indian metals and chemicals industry from the Reserve Bank of India Bulletin. Factory overhead consists of expenses recorded in the following combined profit and loss statement accounts: 1) Other Manufacturing Expenses; 2) Stores and Spares Consumed; 3) Repairs to Buildings; 4) Repairs to Machinery; and 5) Depreciation Expense. There is no evidence to suggest that any of these accounts include energy costs. Therefore, adding indirect electricity and line losses to the reported electricity consumption should not double count electricity costs. Recommendation Based on our analysis of the comments received, we recommend adopting all of the above positions and adjusting the related margin calculations for corrections resulting from verification. If these recommendations are accepted, we will publish the final results of the review and the final weighted-average dumping margins for all reviewed firms in the Federal Register. AGREE _________ DISAGREE _________ Troy H. Cribb Acting Assistant Secretary For Import Administration ________________ Date _______________________________________________________________ footnotes: 1. The petitioner notes that Emei used trucks to transport imported manganese ore. 2. The respondents also maintained that electrode paste should be treated as part of factory overhead because the Department historically treats inputs that are infrequently used and small in value, relative to the total cost of manufacturing, as part of factory overhead. However, the respondents never argued that electrode paste itself was infrequently used in production nor that its value was small. 3. The respondents also contend that electrode paste should not be considered as fuel for the furnace since coke fuels their furnaces.