Last year, a NAFTA Chapter 19 panel found that zeroing violates U.S. law. See this post for details. The panel thus remanded the matter back to Commerce to re-calculate the dumping margins without zeroing.
Commerce did take some action, but the respondents felt that it was insufficient, and went back to the NAFTA Chapter 19 panel. Here's a bit from the NAFTA Chapter 19 panel's new decision, which was issued last week. First, here's what was at issue:
This Panel will not entertain the relitigation of issues pertaining to zeroing. The issues of zeroing were addressed and concluded in our previous opinion. This Panel determination will only address the issue of whether Commerce followed the Panel’s instructions. That is, did Commerce follow the Panel’s Remand Order not to zero or should it be remanded accordingly.
Next, here's what Commerce did in its "remand determination":
The Remand Determination employs a new methodology which Commerce contends is both compliant with the remand and comports with the antidumping statute. The Remand Determination claims that Commerce did not use zeroing in recalculating Mexinox’s dumping margins. It argues that zeroing occurs at the aggregation stage when Commerce calculates a rate for cash deposit purposes or a rate for assessment purposes. Commerce reasons that absent the aggregation of dumping margins, it is conceptually impossible to zero.
The Remand Determination contends that this methodology is consistent with the antidumping statute. It holds that the antidumping statute defines “dumping margin” as the “amount by which the normal value exceeds the export price or constructed export price of the subject merchandise”. 19 U.S.C. 1677(35)(A). It notes that this term is distinct from the definition of the “weighted-average dumping margin” under 19 U.S.C. 1677(35)(B). In keeping with the definition of dumping margin, the Remand Determination explains that it calculated dumping margins pursuant to 19 U.S.C. 1677(35)(A) by comparing normal value (“NV”) with constructed export price (“CEP”). Where NV was not greater than CEP, Commerce did not assess antidumping duties. For U.S. sales, where NV was greater than CEP, Commerce calculated the duty owing on each sale as the absolute dollar amount by which the CEP of the U.S. sale was lower than NV. Since there was no aggregation of comparisons, the Remand Determination states that, by definition, there could not be zeroing. Rather, it merely examined individual sales separately and determined the duty for each sale.
And here's an excerpt from the Panel's reasoning:.
The Remand Determination goes to great lengths to liken this methodology to the prospective normal value system operated by Canada and other WTO members. It also notes that this is the same methodology for calculating dumping margins that the Treasury Department used in the past before zeroing was in use.
... While most other countries have adopted a prospective system, there are many variations in the application of the systems. There is, in fact, no single unitary antidumping duty collection system used by other WTO member countries. Prospective systems fall into three general categories-prospective ad valorem systems, prospective per unit systems and prospective normal value systems. In fact, many incorporate a mixture of systems. Given the variation in the application of these systems, references to the prospective system utilized by other WTO member countries is more likely to engender confusion than serve as useful reference points.
Importantly, the Panel did not remand on the basis of WTO jurisprudence. The Panel remanded strictly on the basis that the application of the original methodology was contrary to U.S. law. As such, to now suggest that this new methodology is comparable to that used by other WTO countries is not responsive to the Panel’s concerns. The fact that Commerce made much of the fact that Canada is one such country is equally not compelling. Similarly, the fact that the United States used a similar methodology some time ago does not justify the departure.
The Panel remains concerned that the dumping margin is not accurate. The Panel is mindful of the statute’s direction that as accurate a dumping margin as possible be calculated. This new methodology ignores the bulk of the transactions and sets a dumping margin on the basis of a fraction of the transactions.
Based on the foregoing and after review and consideration of all arguments raised by the participants, this Panel finds that Commerce failed to follow the Panel’s remand order to recalculate Mexinox’s dumping margins without zeroing, since such re-calculation excluded positive value sales.