December 2010 brought a tidal wave of regulatory proposals by the US Department of Commerce (Commerce) that would significantly change the rules governing the administration of US antidumping and countervailing duty (AD/CVD) laws. Fueled by winds from all points of the compass - a request from Congress to review the retrospective duty assessment system, the administration’s National Export Initiative, a string of losses before the World Trade Organization’s Appellate Body, court decisions questioning Commerce’s methodologies in AD/CVD cases, and various legislative initiatives asking Commerce to strengthen trade laws - Commerce poured its energy in an unusual year-end display of regulatory proposals.

1. Commerce to Give Up Zeroing in Most Reviews

The most anticipated and the most significant of the proposals, due to the large number of proceedings that it would impact and its direct lowering effect on dumping margins, is Commerce’s announcement on December 28, 2010 that it intends to discontinue its practice of zeroing in administrative reviews by allowing negative margins of dumping to offset positive margins in the calculation of weighted average dumping margins.1 The announcement of this change in practice comes after a long series of losses before the WTO Appellate Body finding zeroing inconsistent with the WTO Antidumping Agreement in all reviews of AD/CVD orders: periodic administrative reviews, sunset review, new shipper reviews (collectively, “reviews”). While under Commerce’s zeroing practice all individual negative margins were set to zero when calculating the weighted average margin (a.k.a. “zeroing”), offsetting rather than zeroing will finally allow for both negative and positive margins to be taken into account, thus lowering overall margins of dumping in future proceedings. The extent of the reduction will depend on the facts and circumstances of individual cases.

Offsetting and zeroing are different calculation methodologies applied by Commerce when determining weighted average dumping margins, neither of which is endorsed in the statute or the regulations. To calculate a weighted average dumping margin in an administrative review, for each US entry of goods Commerce compares the normal value (generally, the price a producer charges in its home market) and the export price (the price of the product in the United States).2 The amount by which the normal value exceeds the export price is the dumping margin.3 Then Commerce aggregates these values over all entries in the period of review to calculate a weighted-average dumping margin.4 The weighted-average dumping margin is expressed as a percentage by dividing the aggregate dumping margins of a specific exporter or producer by the aggregate export prices of that same exporter or producer.5 With offsetting, the numerator in the weighted average dumping margin calculation is the sum of all dumping margins, both positive and negative values. 75 Fed. Reg. 81534. Adding the negative and positive margins reduces the aggregate amount of dumping compared to what would have been obtained under zeroing.

Commerce’s proposal introduces another important change to the calculation of margins in reviews by replacing the “transaction-to-average” method of calculating margins with the comparison of monthly weighted average US export prices to monthly weighted average normal values. This method, known as the “average-to-average” method, will become Commerce’s “preferred” methodology for calculating margins in reviews.

In 2006, under pressure from the WTO, Commerce discontinued its practice of zeroing in investigations, in the context of average-to-average comparisons.6 The recent proposal intends to harmonize Commerce’s practice in reviews and investigations, although some differences remain between investigations and reviews as to the time period over which Commerce may weight-average export prices and normal values in determining dumping margins.

Zeroing, while not entirely abandoned, seems to survive in very limited circumstances. Currently, in investigations Commerce is not prohibited from using zeroing in individual-to-individual transaction comparisons, or for weighted average-to-individual transaction comparisons where Commerce finds targeted dumping.7 Similarly, in reviews, Commerce proposes to discontinue zeroing only in the context of average-to-average comparisons, leaving open the theoretical possibility for zeroing should Commerce ever resort to the weighted average-to-individual transaction method to calculated margins in reviews.

Public comments on Commerce’s proposal to (largely) abandon zeroing may be submitted to the agency until February 18, 2011. The final modification will then take effect when it is published in the Federal Register, which is expected anytime in mid to late 2011. Please click here for a more detailed analysis of Commerce’s proposal regarding offsetting of margins.

2. Proposed Changes to the Respondent Selection Process in AD/CVD Reviews

A wave of proposals in December 2010 aimed at respondent selection procedures, sunset review procedures and separate rates practice in cases involving non-market economy countries (NME), traces back to Commerce’s announcement on August 26, 2010 of a series of 14 specific measures designed to strengthen AD/CVD enforcement and support the administration’s National Export Initiative. Please click here for a more detailed analysis of Commerce’s proposed changes to the landscape of AD/CVD rules.

On December 16, 2010, Commerce issued a proposed new methodology for the selection of respondents in AD proceedings based on sampling. 75 Fed. Reg. 78,678 (Dep't Commerce Dec. 16, 2010). In investigations or reviews involving multiple producers and exporters, Commerce has been limiting its review to a small number of exporters and producers accounting for the largest volume of subject merchandise from the exporting country. That typically resulted in Commerce selecting as few as two respondents in a large number of cases. The practice of limiting the pool of respondents to such a small number was rejected in 2009 by the Court of International Trade (the Court). In Zhejiang Native Producer & Animal By-Products Imp. & Exp. Co. v. U.S., 637 F. Supp. 2d 1260 (Ct. Int’l Trade 2009), the Court rejected Commerce’s conclusion that four was a “large” number of respondents in an administrative review, so that individual review of all four respondents in that case was not practicable.8

Under the proposed new rules, Commerce will use sampling to select respondents rather than selecting the companies accounting for the largest import volume that can be reasonably examined. The new selection method proposes a random sampling method that is stratified, and uses probability-proportional-to-size samples. As a result, even companies with smaller import volumes, that have typically not qualified for selection as mandatory respondents, could stand a better chance of being selected. However, Commerce will be relieved from sampling in the following cases:

  • Commerce is unable to examine at least three companies because of resource constraints;
  • the largest companies by import volume account for at least 75 percent of total imports, or
  • characteristics of the underlying population make it highly likely that results obtained from the largest possible sample, given resource constraints, would be unreasonable to represent the population.

Consistent with its current practice, for non-selected companies, Commerce will calculate a sample rate that will be the average of all selected respondent rates, weighted by the import share of their corresponding strata.

Commerce’s proposal also leaves a series of questions unanswered: a) How should Commerce address a case in which a selected respondent needs to be replaced, due to withdrawal or disqualification for any reason? b) How should Commerce treat voluntary respondents in the sampling context? and c) How should Commerce treat adverse-facts-available, de minimis, and zero antidumping duty rates in its calculation of the sample rate? Commerce opened the floor for public comments on these matters until January 31, 2011.

3. Proposed Changes to Separate Rates Practice in NME Cases

Half of Commerce’s set of 14 proposals announced this summer are directed at NME cases. For NMEs, the proposals expand Commerce’s authority to account for certain additional costs and taxes in a manner that will bias antidumping margins higher than under current practice.

Similarly, the proposed changes to the separate rates practice announced on December 16, 2010 could also make it more difficult for NME respondents to avoid high AD/CVD measures. 75 Fed. Reg. 78676. In NME cases, Commerce applies a rebuttable presumption that the export activities of all companies within the country are subject to government control. Thus, Commerce assigns all exporters in the NME country the antidumping duty rate of the NME entity, unless they can demonstrate that they are independent of government control and should receive a separate antidumping duty rate. An exporter can demonstrate its entitlement to a separate rate by proving the absence of both de jure and de facto governmental control over its export activities.

Commerce’s de jure and de facto tests have been the cornerstone of the separate rates practice. The December 16th proposal suggests a supplementation of the four traditional de facto criteria: a) whether the export prices are set by or are subject to the approval of a governmental agency; b) whether the exporter has authority to negotiate and sign contracts; c) whether the exporter makes decisions regarding management selection independent of government control; and d) whether the exporter can keep the proceeds of its export sales and makes independent decisions regarding disposition of profits. Commerce explained that its emphasis on direct government involvement in export activities may not sufficiently account for the government’s role in the NME and how that role may impact export activities and setting prices.

Specifically, Commerce is considering modifying the de facto criteria to look beyond direct government control of export activities in assessing separate rate status and incorporating additional de facto criteria into its analysis. However, no such potential additional factors are identified.

Rather than offering a solution, Commerce leaves the question hanging, inviting public comments on such additional de facto criteria and the documentation that would support such factors. Comments on this proposed change in practice were accepted up to January 31, 2011.

4. Conclusion

Commerce has started 2011 on the crest of this wave of change in AD/CVD practice. Adoption of the most advanced proposals such as those relating to zeroing, the use of sampling for selecting respondents or the changes to the separate rates practice for NME respondents, could happen in 2011 already, while others may need more time for Commerce to refine and unveil for public comment. Expect more changes in 2011 – all of the 2010 proposals will play out in 2011 and beyond.