united states – certain country of origin labelling ... · from india, wt/ds436/ab/r, circulated...

UNITED STATES – CERTAIN COUNTRY OF ORIGIN LABELLING REQUIREMENTS

RECOURSE TO ARTICLE 21.5 OF THE DSU BY MEXICO (DS386/RW)

(AB-2014-10)

Other Appeal Submission of Mexico

12 December 2014

United States – Certain Country of Origin Other Appeal Submission of Mexico Labelling Requirements (AB-2014-10/DS386) 12 December 2014

i

TABLE OF CONTENTS

I. INTRODUCTION .................................................................................................. 1

II. FACTUAL BACKGROUND ..................................................................................... 2

III. ARTICLE 2.2 OF THE TBT AGREEMENT ................................................................. 5

A. The Approach of the Panel ......................................................................... 6

B. The Amended COOL Measure is More Trade Restrictive than Necessary under the “Relational Analysis” in the First Step of the Necessity Test ............................................................................... 10

1. The Panel Erred by Replacing the Weighing and Balancing of Relevant Factors in the Relational Analysis with an “Exceptional Circumstances” Test ................................................ 11

2. The Panel Erred by Failing to Include in the Relational Analysis the Design and Operation of Label E when Ascertaining the Level of Contribution to the Objective ................... 14

3. The Panel Erred in its Assessment of the Risks Non-fulfilment would Create, including the Gravity of the Consequences that would Arise from Non-fulfilment of the amended COOL measure’s objective ....................................... 18

a. The Panel Erred when it Restricted its Assessment of the Risks Non-fulfilment would Create to only Two Criteria ................................................................... 19

b. The Panel Erred by Omitting from its Assessment the Relative Importance of the Interests or Values furthered by the Amended COOL Measure ................ 21

(i) The Relative Importance of the Amended COOL Measure’s Objective is Highly Relevant to the Assessment of the Risks Non-Fulfilment would Create ................................... 21

(ii) The Relative Importance of the Amended COOL Measure’s Objective is Very Low ..................... 23

(iii) Taking into Account the Very Low Relative Importance of the Amended COOL Measure’s Objective, the Gravity of the Consequences that would arise from Non-Fulfilment are Objectively Not Significant .................. 24

c. The Panel Erred by Omitting the Design, Structure and Architecture of the Amended COOL Measure from its Assessment of the Risks Non-Fulfilment would Create .................................................................. 25


ii

d. The Panel Erred in Failing to Ascertain the Gravity of the Consequences that Would Arise from Non-Fulfilment of the Amended COOL Measure’s Objective ....................................................................... 26

(i) The Critical Impact on the Panel’s Entire Article 2.2 Analysis as a Result of its Error in Not Making a Finding regarding the Gravity of the Consequences that Would Arise from Non-Fulfilment ....................................... 26

(ii) The Panel Erred by Not Finding on the Merits that the Gravity of the Consequences of Non-fulfilment is Very Low ..................................................................... 28

(A) The Panel Applied an Erroneous Approach to its Assessment of the Evidence regarding Consumer Demand ..................................................... 28

(B) The Panel Erred in Accepting the Inherent Assumption in the Design of a Telephone Poll Question as Evidence of “Some Consumer Interest” in COOL Information ....................... 30

(C) The Panel Erred in Finding that Willingness to Pay for a “Product of North American Label” is Evidence of Willingness to Pay for Country of Origin Information ....................................... 31

(D) The Panel Failed to Find that the “Small” Economic Benefits of receiving the COOL Information Indicates that the Demand for COOL Information is “Small” ......................... 31

(E) An Objective Assessment of the Evidence on the Record Establishes that the Gravity of the Consequences arising from Non-Fulfilment are Very Low ................................ 32

4. The Panel Erred by Not Completing the “Relational Analysis” and Not Drawing a Conclusion as to the “Necessity” of the Trade-Restrictiveness of the Amended COOL Measure .......................................................................... 32

5. A Properly Conducted, Complete Relational Analysis Demonstrates that the Amended COOL Measure is More


iii

Trade Restrictive than Necessary and an Unnecessary Obstacle to International Trade ................................................... 33

C. The amended COOL measure is More Trade Restrictive than Necessary under the “Comparative Analysis” in the Second Step of the Necessity Test ........................................................................... 34

1. The Panel’s Failure to Make any Finding Regarding the Gravity of the Consequences that would Arise from Non-fulfilment Rendered Unworkable the Entire “Comparative Analysis” in the Second Step of the Necessity Test ........................ 35

2. The Correct “Equivalent Contribution” Analysis ................................. 35

3. Mexico’s Proposed Alternative Measures Demonstrate that the Amended COOL Measure is More Trade Restrictive than Necessary ......................................................................... 37

a. Mexico’s First Alternative Measure ......................................... 37

(i) Mexico’s First Alternative Measure Achieves an Equivalent Degree of Contribution in Light of the Risks of Non-Fulfilment of the Relevant Objective ................................................ 37

(ii) Mexico’s First Alternative Measure is Less Trade Restrictive ................................................... 40

(iii) Mexico’s First Alternative Measure is Reasonably Available ............................................. 41

b. Mexico’s Second Alternative Measure ..................................... 43

(i) Mexico’s Second Alternative Measure Achieves an Equivalent Degree of Contribution in Light of the Risks of Non-Fulfilment of the Relevant Objective ......................... 43

(ii) Mexico’s Second Alternative Measure is Less Trade Restrictive ................................................... 45

(iii) Mexico’s Second Alternative Measure is Reasonably Available ............................................. 46

c. Mexico’s Third and Fourth Alternative Measures ...................... 47

(i) Description of Third Alternative Measure ...................... 47

(ii) Description of Fourth Alternative Measure ................... 48

(iii) Standard of Proof for Identification of Alternative Measures ............................................. 48


iv

IV. ARTICLE 2.1 OF THE TBT AGREEMENT ............................................................... 52

A. Label E is Relevant to the Issue of Whether the amended COOL measure Lacks Even-handedness or Legitimacy of Regulatory Distinctions under the Article 2.1 Legal Analysis ...................................... 52

V. CONDITIONAL APPEAL CONCERNING ARTICLE XXIII:1(B) OF THE GATT 1994 ........................................................................................................... 55

A. Approach of the Panel and Reasons for Exercise Judicial Economy of the Panel ....................................................................................... 56

B. There are Sufficient Factual Findings to Conclude that the Amended COOL Measure Causes Nullification or Impairment within the meaning of Article XXIII:1(b) of the GATT 1994 ....................... 57

VI. CONCLUSIONS ............................................................................................... 58


v

CASES CITED IN THIS SUBMISSION

TABLE OF WTO DISPUTES CITED IN THIS SUBMISSION

Short Title Full Case Title and Citation

Brazil – Retreaded Tyres Appellate Body Report, Brazil – Measures Affecting Imports of Retreaded Tyres, WT/DS332/AB/R, adopted 17 December 2007, DSR 2007:IV, 1527

China – Publications and Audiovisual Products

Appellate Body Report, China – Measures Affecting Trading Rights and Distribution Services for Certain Publications and Audiovisual Entertainment Products, WT/DS363/AB/R, adopted 19 January 2010, DSR 2010:I, p. 3

China – Publications and Audiovisual Products

Panel Report, China – Measures Affecting Trading Rights and Distribution Services for Certain Publications and Audiovisual Entertainment Products, WT/DS363/R and Corr.1, adopted 19 January 2010, as modified by Appellate Body Report WT/DS363/AB/R, DSR 2010:II, p. 261

EC – Seal Products Appellate Body Reports, European Communities – Measures Prohibiting the Importation and Marketing of Seal Products, WT/DS400/AB/R / WT/DS401/AB/R, adopted 18 June 2014

EC – Seal Products Panel Reports, European Communities – Measures Prohibiting the Importation and Marketing of Seal Products, WT/DS400/R / WT/DS401/R / and Add.1, adopted 18 June 2014, as modified by Appellate Body Reports WT/DS400/AB/R / WT/DS401/AB/R

Korea – Various Measures on Beef

Appellate Body Report, Korea – Measures Affecting Imports of Fresh, Chilled and Frozen Beef, WT/DS161/AB/R, WT/DS169/AB/R, adopted 10 January 2001, DSR 2001:I, 5

US – Carbon Steel (India)

Appellate Body Report, United States – Countervailing Measures on Certain Hot-Rolled Carbon Steel Flat Products from India, WT/DS436/AB/R, circulated to WTO Members 8 December 2014 [adoption pending]

US – COOL Appellate Body Reports, United States – Certain Country of Origin Labelling (COOL) Requirements, WT/DS384/AB/R / WT/DS386/AB/R, adopted 23 July 2012, DSR 2012:V, p. 2449

US – COOL Panel Reports, United States – Certain Country of Origin Labelling (COOL) Requirements, WT/DS384/R / WT/DS386/R, adopted 23 July 2012, as modified by Appellate Body Reports WT/DS384/AB/R / WT/DS386/AB/R, DSR 2012:VI, p. 2745

US – COOL

(Article 21.5 – Canada and Mexico)

Panel Reports, United States – Certain Country of Origin Labelling (COOL) Requirements – Recourse to Article 21.5 of the DSU by Canada and Mexico, WT/DS384/RW / WT/DS386/RW / and Add.1, circulated to WTO Members 20 October 2014 [appeal in progress]


vi

US – Gambling Appellate Body Report, United States – Measures Affecting the Cross-Border Supply of Gambling and Betting Services, WT/DS285/AB/R, adopted 20 April 2005, DSR 2005:XII, 5663 (Corr.1, DSR 2006:XII, 5475)

US – Tuna II (Mexico) Appellate Body Report, United States – Measures Concerning the Importation, Marketing and Sale of Tuna and Tuna Products, WT/DS381/AB/R, adopted 13 June 2012, DSR 2012:IV, p. 1837

TABLE OF ACRONYMS USED IN THIS SUBMISSION

Acronym Full Name 2002 Farm Bill The Farm Security and Rural Investment Act of 2002

2008 Farm Bill Food, Conservation, and Energy Act of 2008

2009 Final Rule (AMS) Final Rule on Mandatory Country of Origin Labelling of Beef, Pork, Lamb, Chicken, Goat Meat, Perishable Agricultural Commodities, Peanuts, Pecans, Ginseng, and Macadamia Nuts, published on 15 January 2009 as 7 CFR Part 65

BCI Additional procedures for the protection of business confidential information

AMS Agriculture Marketing Service (of the United States Department of Agriculture)

C.F.R. Code of Federal Regulations

COOL Country of Origin Labelling

DSB Dispute Settlement Body

DSU Understanding on Rules and Procedures Governing the Settlement of Disputes

FSIS Food Safety and Inspection Service (of the United States Department of Agriculture)

GATT 1994 General Agreement on Tariffs and Trade 1994

GATS General Agreement on Trade in Services

Interim Final Rule (AMS)

Interim Final Rule on Mandatory Country of Origin Labelling of Beef, Pork, Lamb, Chicken, Goat Meat, Perishable Agricultural Commodities, Peanuts, Pecans, Ginseng, and Macadamia Nuts, published on 1 August 2008 as 7 CFR Part 65

Interim Final Rule (FSIS)

Interim Final Rule on Mandatory Country of Origin Labelling of Muscle Cuts of Beef (including Veal), Lamb, Chicken, Goat, and Pork, Ground Beef, Ground Lamb, Ground Chicken, Ground Goat, and Ground Pork, published on 28 August 2008 as 9 CFR Parts 317 and 381

NAFTA North American Free Trade Agreement

SPS Agreement Agreement on the Application of Sanitary and Phytosanitary Measures


vii

TBT Agreement Agreement on Technical Barriers to Trade

U.S.C. United States Code

USDA United States Department of Agriculture

WTO World Trade Organization


1

I. INTRODUCTION

1. On 22 December 2008 Mexico initiated its challenge against the mandatory country of origin labelling system (COOL) adopted by the United States. This measure disrupted the integrated North American market for cattle and beef to the detriment of exports of Mexican cattle to the United States. Almost six years later, Mexican cattle producers continue to face the higher costs and additional burdens created by the COOL measure compared to those faced by U.S. domestic cattle producers.

2. The United States has had multiple opportunities to eliminate the discriminatory aspects of the COOL measure (comprising the “COOL Statute” passed by the U.S. Congress, and its implementing regulation, the “2009 Final Rule” issued by the U.S. Department of Agriculture (USDA)). Instead of complying with its WTO obligations, the United States decided, by amendment of the implementing regulation, to increase the incentive in favour of processing exclusively domestic cattle and against handling imported Mexican cattle in the production of beef.1

3. In the compliance proceedings under Article 21.5 of the Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU), the Panel rejected the United States’ arguments that the detrimental impact of the amended COOL measure stems exclusively from a legitimate regulatory distinction within the meaning of Article 2.1 of the Agreement on Technical Barriers to Trade (TBT Agreement). It found that Article 2.1 had been violated. The Panel also found that the amended COOL measure violated Article III:4 of the General Agreement on Tariffs and Trade 1994 (GATT 1994). In an attempt to reverse these findings, the United States filed a Notice of Appeal on 28 November 2014 and its Appellant’s Submission on 5 December 2014. The United States’ appeal is without merit and should be rejected. In its Appellee’s Submission, Mexico will address in detail the arguments advanced by the United States.

4. By denying Mexico’s claims under Article 2.2 of the TBT Agreement, the Panel interpreted and applied that provision in a manner that renders elements of the obligation meaningless, an outcome that is inconsistent with the principle of effectiveness in the interpretation of treaties. Mexico believes it is of great systemic importance that the Appellate Body clarifies the interpretation and application of each element of the test to be applied under Article 2.2 and, for that reason, is submitting this Other Appeal.

5. This dispute is of great importance to Mexico from an economic point of view, because it concerns substantial exports of Mexican cattle to the United States for further production into beef. It is also important from a systemic point of view. A proper interpretation of the WTO provisions raised in this dispute will balance the right of a WTO Member to adopt technical regulations with the obligation to ensure that those regulations are not prepared, adopted or applied with a view to or with the effect of creating unnecessary obstacles to international trade. If a proper balance is not reached, all WTO Members, particularly developing country Members such as Mexico, will be at the highest risk of being adversely affected by non-tariff measures such as the mandatory labelling measure at issue in this dispute. It is therefore essential for Mexico that the Appellate Body carefully review the measure at issue and correctly and cumulatively apply, in a strict manner, the WTO provisions raised by Mexico, in favour of the multilateral trading system.

1 Panel Reports, US – COOL (Article 21.5), para. 7.176.


2

6. By virtue of its design, architecture, structure and operation, the mandatory country of origin labelling regime implemented in the amended COOL measure is discriminatory, more trade restrictive than necessary, and creates an unnecessary obstacle to international trade.

7. The general grounds raised by Mexico in this Other Appeal are as follows: (i) the Panel erred in concluding that Mexico did not make a prima facie case that the amended COOL measure is more trade restrictive than necessary within the meaning of Article 2.2 of the TBT Agreement and the Panel failed to make an objective assessment of the matter before it as required under Article 11 of the DSU; (ii) the Panel erred in finding that Label E (the ground meat label) is not relevant to the legal analysis under Article 2.1 of the TBT Agreement; and (iii) in the event that the Appellate Body overturns the Panel’s findings that the COOL measure is inconsistent with Article 2.1 of the TBT Agreement and Article III:4 of the GATT 1994, Mexico conditionally appeals the Panel’s decision to exercise judicial economy with respect to Mexico's claim of non-violation nullification or impairment under Article XXIII:1(b) of the GATT 1994.

II. FACTUAL BACKGROUND

8. At its meeting on 23 July 2012, the Dispute Settlement Body (DSB) adopted the Appellate Body Reports and the Panel Reports as modified by the Appellate Body Reports, pertaining to the COOL dispute. The Panel and Appellate Body Reports found that the COOL measure2, particularly in relation to the muscle cut meat labels, was inconsistent with Article 2.1 of the TBT Agreement because it accorded less favourable treatment to imported livestock than to like domestic livestock. The DSB recommended that the United States bring the COOL measure into conformity with the United States’ obligations under the covered agreements.

9. On 23 May 2013, the USDA published the “Final Rule on Mandatory Country of Origin Labeling of Beef, Pork, Lamb, Chicken, Goat Meat, Wild and Farm-Raised Fish and Shellfish, Perishable Agricultural Commodities, Peanuts, Pecans, Ginseng, and Macadamia Nuts” (the “2013 Final Rule”).3 The Final Rule purportedly entered into force on that same date, although the notice published with the rule indicated that the new regulation would not actually be enforced for an additional six months.4 Although the COOL Statute was part of the measure found by the Panel and the Appellate Body to be inconsistent with Article 2.1 of the TBT Agreement, the United States did not modify the Statute.

10. In this dispute the “measures taken to comply with the recommendations and rulings” of the DSB comprise the COOL Statute and the 2009 Final Rule, as amended by the 2013 Final Rule, which collectively are the “amended COOL measure”.

11. The features of the relevant market remain unchanged. Mexico generally exports “feeder cattle” to the United States immediately after the cow/calf stage to U.S. backgrounding and feeding operations. The average age of the Mexican cattle when

2 Comprising the “COOL Statute” passed by the U.S. Congress, and its implementing regulation, the “2009 Final Rule” issued by the USDA. 3 Panel Reports, US –COOL (Article 21.5), para. 7.8. 4 Panel Reports, US – COOL (Article 21.5), footnote 41.


3

exported is six to seven months, and the cattle are slaughtered in the United States at approximately 22 months.5 The muscle cuts derived from this process are sold as a final product in the United States or exported to Mexico and other countries for final consumption. The amended COOL measure adversely affects the entire production process in the United States for beef derived from cattle born in Mexico, resulting in lower cattle prices throughout Mexico and fewer opportunities for the sale of such cattle in the United States.

12. It is undisputed that the amended COOL measure is not a food safety measure.

13. The amended COOL measure mandates that consumers be informed of the country of origin of certain categories of covered meat products. For muscle cuts of meat, origin must be reported not based on the final country of processing, but rather on the basis of stages of production of the animal from which the meat was processed: birth, raising, and slaughtering. Other covered categories of beef products – in particular ground beef – have different labelling rules that do not require information on stages of production, and a large portion of meat products sold for consumption in the United States are exempt entirely from country of origin labelling requirements.

14. In fact, rather than adopting a measure to comply with the recommendations of the DSB and eliminate the discriminatory impact of the measure, the changes to the 2009 Final Rule introduced by the United States in May 2013 created more burdensome restrictions that further discourage the purchase of Mexican cattle.

15. The original COOL measure established five categories, consistent with the requirements of the COOL Statute:

• Category A muscle cuts, defined as from animals exclusively born, raised and slaughtered in the United States;

• Category B muscle cuts, which are from animals born, raised and slaughtered in more than one country;

• Category C muscle cuts, which are from animals born and raised in another country and imported into the United States for immediate slaughter;

• Category D muscle cuts, which are from animals slaughtered and processed in a foreign country before importation, and for which the regular customs origin rule is used (i.e., “substantial transformation”); and

• Category E, which covers ground meat, for which the origin is identified as all countries of origin contained therein or are which are reasonably contained therein, specifically meaning that the origins of any meat in a processor’s inventory for the prior 60 days are deemed contained therein.6

The COOL Statute was not amended and these categories remain in effect.

5 Panel Reports, US – COOL, para. 7.141; Panel Reports, US – COOL (Article 21.5), paras. 7.241-7.242. 6 Panel Reports, US – COOL (Article 21.5), paras. 7.10-7.16.


4

16. With respect to the information required to be provided on the labels, the original COOL measure required only the names of countries. Thus, Category A labels were required to state “product of United States” or similar. For Category B, a meat product made from an animal born and raised in another country and slaughtered in the United States could be labelled “product of United States and Mexico” or similar. For Category C, the designation was “Product of X and United States.” Category D could be labelled as declared to the U.S. customs authorities upon importation. The label for Category E was required to list all countries of origin based on the 60-day “inventory allowance.”7

17. Under the new regulations adopted as part of the 2013 Final Rule, the labelling requirements changed for Categories A, B and C. The 2013 Final Rule requires the labels for those categories to indicate the place where each production step of the animal (born, raised, slaughtered) occurred. The labels for Categories D and E remain unchanged.8

18. Neither the 2009 Final Rule nor the 2013 Final Rule prescribed specific requirements for the location or size of the origin declaration, and the 2013 Final Rule provides that abbreviations could be used such as “brn” for “born,” “raisd” for “raised”, and “slghtrd” for “slaughtered.”9 The 2013 Final Rule also allows the use of the word “harvested” in place of “slaughtered”.10

19. The amended COOL measure retained the original COOL measure’s three main exemptions from coverage.

• First, the measure does not apply to entities not meeting the applicable definition of “retailer,” which is defined as an entity (i) that has shipped, received, or contracted to be shipped or receive in a single day perishable agricultural commodities in a quantity of at least 2,000 pounds (one ton) and (ii) that purchases perishable agricultural commodities with a cost over US$230,000 per year.11 Thus, entities that do not sell any fruits and vegetables (and therefore do not buy them) or who sell fruits and vegetable but do not purchase them in the required amount are not covered by the COOL provisions. For example, “butcher shops” that specialize in the sale of meat products are not covered by the amended COOL measure.

• Second, the measure does not apply to any meat product that is “an ingredient in a processed food item.”12

• Third, the measure does not apply to any meat products sold in a “food service establishment”, which is defined as “a restaurant, cafeteria, lunch room, food stand, saloon, tavern, bar, lounge, or other similar facility

7 Panel Reports, US – COOL (Article 21.5), para. 7.17. 8 Panel Reports, US – COOL (Article 21.5), paras. 7.18-7.20. 9 Panel Reports, US – COOL (Article 21.5), paras. 7.22-7.23. 10 Panel Reports, US – COOL (Article 21.5), footnote 78 to para. 7.23. 11 Panel Reports, US – COOL (Article 21.5), para. 7.28 and footnote 88. 12 Panel Reports, US – COOL (Article 21.5), para. 7.29.


5

operated as an enterprise engaged in the business of selling food to the public.”13

These exemptions remain in effect under the amended COOL measure.

20. Under the original COOL measure, it was possible to use the Category B label (“e.g., Product of United States, Mexico”) if meat products made from cattle born in Mexico and the United States were commingled on the same production day (similarly if cattle born in the two countries were commingled in processing on the same production day). The amended COOL measure eliminated that flexibility, and therefore all situations involving muscle cuts (with the exception of those covered by Category D) require precise tracking of the birth country of the cattle through all stages of production, if any of the resulting production will be muscle cuts to be sold in a covered retail establishment.14

21. The amended COOL measure also left in place the recordkeeping and verification rules of the original COOL measure, including the prohibition on the establishment of a traceback system.15

22. In the original proceedings, the Panel found and the Appellate Body affirmed that the original COOL measure had a detrimental impact on competitive opportunities for imported livestock.16 Specifically, the original Panel held that “in the context of muscle cut labels, the COOL measure de facto discriminate[d] against imported livestock by according less favourable treatment to … Mexican cattle, especially Mexican feeder cattle, than to like domestic livestock.”17 It is undisputed that the amended COOL measure was not intended to remove the original measure’s detrimental impact on imports.18

III. ARTICLE 2.2 OF THE TBT AGREEMENT

23. With respect to this legal claim, the Panel concluded that Mexico did not make a prima facie case that the amended COOL measure is more trade restrictive than necessary within the meaning of Article 2.2 of the TBT Agreement.19 The Panel’s conclusion is based on a number of erroneous findings on issues of law and related legal interpretations. Having erred in its interpretation and application of the legal analysis under Article 2.2, the Panel improperly excluded relevant facts from the analysis and declined to make key legal and factual findings. As a consequence, the Panel’s analysis is both legally incorrect and factually erroneous. In this regard, with respect to certain facts, the Panel failed to make an objective assessment of the matter before it as required by Article 11 of the DSU. Had the Panel applied the correct legal approach and made an objective assessment of the evidence

13 Panel Reports, US – COOL (Article 21.5), para. 7.30. 14 See Panel Reports, US – COOL (Article 21.5), paras. 7.32-7.34. 15 Panel Reports, US – COOL (Article 21.5), para. 7.45. 16 Appellate Body Reports, US – COOL, para. 292; Panel Reports, US – COOL, paras. 7.372, 7.381, and 7.420. 17 Panel Reports, US – COOL, para. 7.420. 18 Panel Reports, US – COOL (Article 21.5), para. 7.63. 19 Panel Reports, US – COOL (Article 21.5), paras. 7.612-7.613 and 8.3(c).


6

and legal arguments, it would have concluded that the amended COOL measure is inconsistent with Article 2.2 of the TBT Agreement.

A. The Approach of the Panel

24. The Panel began by presenting its legal interpretations and findings with respect to the elements of the necessity test under Article 2.2 of the TBT Agreement. To start with, the Panel acknowledged the description of the analytical approach set out by the Appellate Body in US – Tuna II (Mexico), which “should begin” with an analysis of “a number of factors” relating to the challenged measure “that include” its degree of contribution to the fulfilment of the relevant legitimate objective, its trade-restrictiveness, and the nature of the risks at issue and the gravity of consequences that would arise from non-fulfilment of the objective.20

25. The Panel also acknowledged that, “[a]ccording to the Appellate Body, in addition to the above three factors, ‘in most cases’ an Article 2.2 analysis needs to also entail a comparison of the challenged measure with possible alternatives”.21

26. With respect to the phrase “in most cases”, the Panel considered the relevant reasons of the Appellate Body in both US – Tuna II (Mexico) and the original US – COOL dispute, which identified “at least two instances” in which it may not be necessary to undertake a comparative analysis of the challenged measure with proposed alternative measures:22

We can identify at least two instances where a comparison of the challenged measure and possible alternative measures may not be required. For example, it would seem to us that if a measure is not trade restrictive, then it may not be inconsistent with Article 2.2. Conversely, if a measure is trade restrictive and makes no contribution to the achievement of the legitimate objective, then it may be inconsistent with Article 2.2.23 (emphasis original)

27. Although the Appellate Body expressly indicated that the “two instances” it identified were examples, the Panel interpreted these examples to suggest that “only in exceptional circumstances” would it be possible to determine an Article 2.2 claim without undertaking a

20 Appellate Body Report, US – Tuna II (Mexico), para. 322 (“[A]n assessment of whether a technical regulation is “more trade-restrictive than necessary” within the meaning of Article 2.2 of the TBT Agreement involves an evaluation of a number of factors. A panel should begin by considering factors that include: (i) the degree of contribution made by the measure to the legitimate objective at issue; (ii) the trade-restrictiveness of the measure; and (iii) the nature of the risks at issue and the gravity of consequences that would arise from non-fulfilment of the objective(s) pursued by the Member through the measure”), cited in Panel Reports, US – COOL (Article 21.5), paras. 7.287, 7.292 and 7.303. See also Appellate Body Reports, US – COOL, para. 378. 21 Panel Reports, US – COOL (Article 21.5), para. 7.289. 22 Panel Reports, US – COOL (Article 21.5), para. 7.297 (“As Mexico points out, the Appellate Body foresaw this possibility in ‘at least two instances’. In US – Tuna II (Mexico), the Appellate Body identified the above two scenarios in a footnote to a statement foreseeing the need for the comparative analysis ‘in most cases’. In the original US – COOL dispute, the Appellate Body confirmed both exceptions as well as the need for a comparative analysis ‘in most cases’”). 23 Appellate Body Report, US – Tuna II (Mexico), footnote 647 to para. 322. See also Appellate Body Reports, US – COOL, footnote 748 to para. 376, footnote 929 to para. 461.


7

“comparative analysis”.24 Further, the Panel specifically defined such “exceptional circumstances” to mean “where consistency or inconsistency with Article 2.2 may be deduced by looking solely at certain aspects of the challenged measure.”25 As discussed below, the Panel’s definition and requirement for “exceptional circumstances” indicates a legal analysis that is different from, and inconsistent with, the “relational analysis” described by the Appellate Body as the first step of the necessity test under Article 2.2. In this regard, the Panel found that Mexico had neither argued that the amended COOL measure “falls within either of the two exceptional scenarios identified by the Appellate Body” nor sufficiently identified what a “third exceptional scenario entails in the context of the amended COOL measure”.26

28. The Panel also considered the Appellate Body’s finding in the original US – COOL dispute that the original panel erred “by finding the COOL measure to be inconsistent with Article 2.2 of the TBT Agreement without examining the proposed alternative measures.”27 Specifically, the Panel interpreted the Appellate Body’s reasons to indicate that “[t]he Appellate Body took issue with the original panel stopping short of the ‘comparative analysis’.”28

29. In the original US – COOL dispute, the panel found the original COOL measure to be inconsistent with Article 2.2 on the basis that it did not fulfil its objective within the meaning of Article 2.2 because it failed to convey meaningful origin information to consumers.29 The Appellate Body found this approach erroneous, as “a panel’s assessment should focus on ascertaining the degree of contribution achieved by the measure, rather than on answering the questions of whether the measure fulfils the objective completely or satisfies some minimum level of fulfilment of that objective.”30

30. The Appellate Body also found that the panel had erred by ignoring its own findings that the labels under the original COOL measure contributed to the objective of providing

24 Panel Reports, US – COOL (Article 21.5), para. 7.298 (“Our reading of the relevant Appellate Body statements suggests that a ‘comparative analysis’ would be redundant only in exceptional circumstances where consistency or inconsistency with Article 2.2 may be deduced by looking solely at certain aspects of the challenged measure. Mexico has not explained why the Panel is faced with such exceptional circumstances in this case”). 25 Panel Reports, US – COOL (Article 21.5), para. 7.298 (emphasis added). 26 Panel Reports, US – COOL (Article 21.5), para. 7.298. Mexico notes that while the Panel found that “Mexico contends that [the] amended COOL measure represents a third scenario under which it should be found inconsistent without looking at alternatives”, Mexico raised no argument that the amended COOL measure represents a “third exceptional scenario” (emphasis added). Rather, Mexico argued that the “relational analysis” of all of the relevant factors relating to the amended COOL measure was sufficient to show that the measure is more trade restrictive than necessary because its trade-restrictiveness is disproportionate to the risks non-fulfilment of its objective would create. 27 Panel Reports, US – COOL (Article 21.5), para. 7.299 (emphasis original), citing Appellate Body Reports, US – COOL, para. 469. 28 Panel Reports, US – COOL (Article 21.5), para. 7.300, citing Appellate Body Reports, US – COOL, para. 469. 29 See original Panel Reports, US – COOL, paras. 7.719-7.720; Panel Reports, US – COOL (Article 21.5), para. 7.337. 30 Appellate Body Reports, US – COOL, para. 468.


8

consumer information.31 Having found that the original Panel had erred in its Article 2.2 analysis, the Appellate Body reversed the ultimate finding of inconsistency with Article 2.2.32 It is within this context that the Appellate Body found that the original Panel should have proceeded with the “comparative analysis” in order to determine whether or not the original COOL measure was more trade restrictive than necessary. Stated another way, the Appellate Body found a “comparative analysis” was required “in this case” not because a “comparative analysis” must always be undertaken, in all cases, before any conclusions can be drawn under Article 2.2, but rather because the original measure’s imperfect degree of contribution to its objective was not a sufficient basis, on its own, for the original Panel’s finding that the measure was an unnecessary obstacle to international trade. Thus, in the original proceeding, additional evidence and arguments were required to establish a prima facie case.33

31. Further, the Panel considered that, in the original US – COOL dispute, the Appellate Body “did not draw any conclusions on Article 2.2 consistency at the end of its ‘relational analysis’,” and referred to this as a “preliminary assessment of the COOL measure”.34 It should be noted, however, that the Appellate Body was not performing a “relational analysis” in the first instance. Rather, having reversed the original panel’s conclusion of inconsistency under Article 2.2, as discussed above, the Appellate Body was attempting to “complete the legal analysis” under Article 2.2 to the extent that it was able to do so on the basis of the factual findings of the panel and the undisputed facts in the panel record.35 While the Appellate Body found that the original panel’s factual findings “suggest[ed]” that the original COOL measure made some contribution to its objective; that it had a considerable degree of trade-restrictiveness; and that the consequences that may arise from non-fulfilment of its objective would not be particularly grave”, the Appellate Body stressed that it lacked “clear and precise Panel findings with regard to these factors”.36 The Appellate Body found that, in particular, it was unable to identify the degree of contribution made by the original COOL measure to its objective.37 In this context, the Appellate Body was unable to draw conclusions from the “relational analysis” and, accordingly, it proceeded to attempt to complete the next step of the necessity test, i.e., the “comparative analysis”. The following passage from the Report indicates that the Appellate Body considered its assessment of the “relational analysis” to be “preliminary” because it had been unable to “complete the legal analysis” of this first step of the necessity test:

We stress, however, that we lack clear and precise Panel findings with regard to these factors, and, in particular, findings that would enable us to identify the degree of contribution made by the COOL measure to the United States’ objective. Against this preliminary assessment of the COOL measure, we proceed to examine the alternative measures proposed by Canada and by Mexico in order to see whether we are able to complete our assessment of

31 Appellate Body Reports, US – COOL, para. 468. 32 Appellate Body Reports, US – COOL, para. 468. 33 Appellate Body Reports, US – COOL, paras. 469 and 471. 34 Panel Reports, US – COOL (Article 21.5), para. 7.301. 35 Appellate Body Reports, US – COOL, para. 470. 36 Appellate Body Reports, US – COOL, para. 479. 37 Appellate Body Reports, US – COOL, para. 479.


9

whether the COOL measure is “more trade-restrictive than necessary to fulfil a legitimate objective”.38 (emphasis original)

32. The Appellate Body ultimately found that it was also unable to complete the “comparative analysis” of the necessity test. Accordingly, it concluded that: “[o]verall, due to the absence of relevant factual findings by the Panel, and of sufficient undisputed facts on the record, we are unable to complete the legal analysis under Article 2.2 of the TBT Agreement and determine whether the COOL measure is more trade restrictive than necessary to fulfil its legitimate objective.”39

33. Although this was the logical overall outcome of the Appellate Body’s attempt to complete both steps of the Article 2.2 necessity test in the original dispute, the compliance Panel incorrectly interpreted this statement as support for the proposition that “overall” conclusions with regard to Article 2.2 may only be made after undertaking the “comparative analysis” and may not be made on the basis of the “relational analysis” alone.40

34. Finally, the Panel considered that the panel in EC – Seal Products “adopted a similar approach, noting that ‘[t]he Appellate Body stated that all these factors [including a comparison with alternative measures] provide the basis for the determination of what is to be considered ‘necessary’ in the sense of Article 2.2 … in a particular case.”41 It should be noted, however, that the panel in EC – Seal Products cited this statement to findings made by the Appellate Body in US – Tuna II (Mexico) specifically with respect to the “relational analysis” only, and not in relation to the comparative analysis:

In the context of Article 2.2, the assessment of “necessity” involves a relational analysis of the trade-restrictiveness of the technical regulation, the degree of contribution that it makes to the achievement of a legitimate objective, and the risks non-fulfilment would create. We consider, therefore, that all these factors provide the basis for the determination of what is to be considered “necessary” in the sense of Article 2.2 in a particular case.”42 (emphasis added)

38 Appellate Body Reports, US – COOL, para. 479. 39 Appellate Body Reports, US – COOL, para. 491. 40 Panel Reports, US – COOL (Article 21.5), para. 7.302 (“The Appellate Body undertook the ‘comparative analysis’ before trying to draw ‘overall’ conclusions with regard to Article 2.2”). 41 Panel Reports, US – COOL (Article 21.5), para. 7.302, citing Panel Report, EC – Seal Products, para. 7.356. 42 Appellate Body Report, US – Tuna II (Mexico), para. 318, cited in Panel Report, EC – Seal Products, para. 7.356. Further, while the Panel relied upon what it considered to be the “similar approach” adopted by the panel in EC – Seal Products, Mexico notes that the latter also conflated the “relational analysis” of the challenged measure with the “comparative analysis” of the challenged measure with possible alternatives. Specifically, as made clear in the following passage from its report, the panel in EC – Seal Products included an element of the “comparative analysis” (i.e., “the availability of alternative measures”) as an “element” of the “relational analysis”: “As noted above, an examination of the obligations under Article 2.2 requires a relational analysis of all of the following elements: (a) trade-restrictiveness of the EU Seal Regime; (b) degree of the measure’s contribution to the identified objective; and (c) availability of alternative measures. Panel Report, EC – Seal Products, para. 7.422.


10

35. On the basis of the foregoing considerations, the Panel determined that it would “draw conclusions as to the consistency of the amended COOL measure with Article 2.2 of the TBT Agreement only after having considered all relevant factors.”43 In this respect, the Panel effectively consolidated the factors of the “relational analysis” with the elements of the “comparative analysis” into a single six-factor approach:

Like the Appellate Body in both US – Tuna II (Mexico) and the original US – COOL dispute, we address the following six factors before reaching an overall conclusion on the complainants’ Article 2.2 claims:

a. the amended COOL measure's degree of contribution to a legitimate objective;

b. the trade-restrictiveness of the amended COOL measure;

c. the nature of the risks at issue and the gravity of consequences that would arise from non-fulfilment of the objective pursued by the United States through the amended COOL measure;

d. whether the alternatives proposed by the complainants are less trade restrictive than the amended COOL measure;

e. whether the proposed alternatives would make an equivalent contribution to the relevant legitimate objective, taking account of the risks non-fulfilment would create; and

f. whether the proposed alternatives are reasonably available.44

36. The Panel therefore rejected the possibility that the “relational analysis” in the first step of the necessity test under Article 2.2 of the TBT Agreement could independently demonstrate that the measure is more trade restrictive than necessary and an unnecessary obstacle to international trade within the meaning of Article 2.2.

B. The Amended COOL Measure is More Trade Restrictive than Necessary under the “Relational Analysis” in the First Step of the Necessity Test

37. Although the Panel’s reasons suggest that it rejected a two-step approach – combining the “relational analysis” and the “comparative analysis”, conceptually, into a single analysis comprising six factors – the Panel subsequently recognized the “relational analysis” (i.e., factors a, b, c) as a distinct step from the “comparative analysis” (i.e., factors d, e, f). Specifically, the Panel held that: “we conclude our relational analysis of the amended COOL measure without drawing definitive conclusions on the complainants’ Article 2.2 claims. … We turn to the comparison with the complainants’ proposed alternative measures.”45

43 Panel Reports, US – COOL (Article 21.5), para. 7.303. 44 Panel Reports, US – COOL (Article 21.5), para. 7.303. 45 Panel Reports, US – COOL (Article 21.5), paras. 7.424-7.425.


11

38. The Panel acknowledged that “neither Mexico, nor the United States, question that the ‘relational analysis’ of the contested measure precedes the “comparative analysis” of the contested measure and suggested alternatives.”46 However, the Panel made a number of errors in its legal interpretation of the first step and then applied an incorrect approach to determining whether the amended COOL measure is more trade restrictive than necessary within the meaning of Article 2.2. Had the Panel applied the correct legal approach, it would have undertaken a weighing and balancing of all relevant factors and would have provided conclusions after completing the relational analysis. It would have found that, under a relational analysis, Mexico had presented a prima facie case that the amended COOL measure is more trade restrictive than necessary and, thus, an unnecessary obstacle to international trade that is inconsistent with Article 2.2.

1. The Panel Erred by Replacing the Weighing and Balancing of Relevant Factors in the Relational Analysis with an “Exceptional Circumstances” Test

39. The Panel’s interpretation erroneously narrows the “relational analysis” step of the Article 2.2 necessity test to a determination of whether “exceptional circumstances” exist. In doing so, the Panel failed to recognize the essential importance of the weighing and balancing process, in which all of the relevant factors relating to the challenged measure are considered in relation to one another for the purpose of determining whether the measure itself is, on its own merits, more trade restrictive than necessary within the meaning of Article 2.2.

40. In US – Tuna II (Mexico) and the original US – COOL dispute, the Appellate Body stated that “[i]n the context of Article 2.2, the assessment of ‘necessity’ involves a ‘relational analysis’ of the trade-restrictiveness of the technical regulation, the degree of contribution that it makes to the achievement of a legitimate objective, and the risks non-fulfilment would create”,47 and that “a panel’s determination of what is considered ‘necessary’ will be based on a consideration of all these factors”.48 It elaborated that “[s]imilarly, in the context of Article XX of the GATT 1994 and Article XIV of the General Agreement on Trade in Services (GATS), ‘necessity’ is determined on the basis of ‘weighing and balancing’ a number of factors”.49 It can be concluded from these statements that the foundation of the “relational analysis” step at the beginning of the Article 2.2 necessity test involves a process of “weighing and balancing” together “all factors” relevant to the necessity of the trade-restrictiveness of the measure. This can be contrasted against the “comparative analysis” step, wherein the challenged measure is compared against alternative measures to determine whether or not a less trade restrictive alternative is reasonably available.

46 Panel Reports, US – COOL (Article 21.5), para. 7.296. 47 Appellate Body Report, US – Tuna II (Mexico), para. 318. See also Appellate Body Reports, US – COOL, para. 374. 48 Appellate Body Reports, US – COOL, para. 374. See also Appellate Body Report, US – Tuna II (Mexico), para. 318 (“We consider, therefore, that all these factors provide the basis for the determination of what is to be considered “necessary” in the sense of Article 2.2 in a particular case”). 49 Appellate Body Report, US – Tuna II (Mexico), footnote 643 to para. 318; Appellate Body Reports, US – COOL, footnote 745 to para. 374.


12

41. As a footnote to its statement in US – Tuna II (Mexico) that, “[i]n most cases, a comparison of the challenged measure and possible alternative measures should be undertaken”,50 the Appellate Body identified “at least two instances where a comparison of the challenged measure and possible alternative measures may not be required”.51 In other words, in these two instances, the question of necessity could be answered without having to undertake a comparative analysis. These instances were expressly provided as examples, and the words “at least” clearly indicate that these examples were not intended to be exclusive or exhaustive.

42. It is notable that the two instances identified by the Appellate Body comprise factual situations occupying extremes on the spectrum of possibilities.52 The first instance relates to a threshold requirement for the applicability of Article 2.2. If a measure is not trade restrictive, then by definition it cannot be inconsistent with Article 2.2 because that provision disciplines measures only if they are more trade restrictive than necessary. The second instance relates to a measure that is trade restrictive, but makes no contribution to the achievement of the relevant legitimate objective. Such a measure would inherently be more trade restrictive than necessary.

43. Between these two extreme examples, there is the range of more realistic possibilities where the challenged measure is trade restrictive to some degree and contributes to the achievement of its legitimate objective to some degree. Within this range of possibilities, the process of weighing and balancing all of the relevant factors in relation to one another must be undertaken to determine whether the degree of the measure’s trade-restrictiveness is “necessary” in order to fulfil the relevant legitimate objective.

44. For example, where the challenged measure makes only a small contribution to the achievement of its objective, the “necessity” of its trade-restrictiveness might be outweighed by the other factors, such as the risks that non-fulfilment would create, the relative importance of the interests or values furthered by the measure, and the degree of its trade-restrictiveness. In another example, where the degree of contribution is low, the degree of trade-restrictiveness is severe, and the risks that non-fulfilment would create are

50 Appellate Body Report, US – Tuna II (Mexico), para. 322. See also para. 320 (“The use of the comparative ‘more … than’ in the second sentence of Article 2.2 suggests that the existence of an ‘unnecessary obstacle[] to international trade’ in the first sentence may be established on the basis of a comparative analysis of the above-mentioned factors. In most cases, this would involve a comparison of the trade-restrictiveness and the degree of achievement of the objective by the measure at issue with that of possible alternative measures that may be reasonably available and less trade restrictive than the challenged measure, taking account of the risks non-fulfilment would create”). 51 Appellate Body Report, US – Tuna II (Mexico), footnote 647 to para. 322 (“We can identify at least two instances where a comparison of the challenged measure and possible alternative measures may not be required. For example, it would seem to us that if a measure is not trade restrictive, then it may not be inconsistent with Article 2.2. Conversely, if a measure is trade restrictive and makes no contribution to the achievement of the legitimate objective, then it may be inconsistent with Article 2.2” (emphasis original)). See also Appellate Body Reports, US – COOL, footnote 748 to para. 376, footnote 929 to para. 461. 52 By using the words “at least” in identifying the “two instances” at extreme ends of the spectrum, the Appellate Body indicated that there may be other instances within the range of possibilities between these examples where conclusions with respect to Article 2.2 may be drawn from the outcome of the “relational analysis”, such that a “comparative analysis” may not be required to resolve the claim.


13

insignificant or trivial in relation to the severity of the trade-restrictiveness, the weighing and balancing process in the “relational analysis” may lead to the conclusion that the technical regulation is more trade restrictive than necessary and, thus, an “unnecessary obstacle to international trade”.

45. The Panel erred in its interpretation of the Appellate Body’s statements, treating the two examples discussed above as the basis for a new “exceptional circumstances” requirement that must be satisfied before any “overall” conclusions with respect to Article 2.2 may be drawn from the “relational analysis”. Specifically, the Panel found that “a ‘comparative analysis’ would be redundant only in exceptional circumstances”, which it defined to mean “where consistency or inconsistency with Article 2.2 may be deduced by looking solely at certain aspects of the challenged measure.”53 The language introduced in this interpretation is unique to the Panel’s own reasoning and is not based on the Appellate Body’s statements or reasons in US – Tuna II (Mexico) or the original US – COOL dispute.54

46. The weighing and balancing process in the “relational analysis” is crucial because it determines whether the trade-restrictiveness of the measure, in itself and on its own merits, is “necessary” in the first place to fulfil the legitimate objective. Where this weighing and balancing process is sufficient to demonstrate that the challenged measure is more trade restrictive than necessary, taking into account the risks non-fulfilment would create, a “comparative analysis” may not be required in order to find the measure inconsistent with Article 2.2. Critically, the “comparative analysis” does not question the underlying “necessity” of the challenged measure’s trade-restrictiveness; rather, it inherently assumes that the challenged measure’s trade-restrictiveness is “necessary” in order to fulfil the legitimate objective, and seeks to determine whether or not there are less trade restrictive alternatives that are reasonably available. Eliminating the step of reaching a conclusion under the “relational analysis” creates a substantial gap in the necessity test under Article 2.2, as it leaves unresolved the question of whether the trade-restrictiveness of the challenged measure is “necessary” to fulfil the legitimate objective in the first place. In fact, this interpretation reads the term “unnecessary obstacles” out of Article 2.2 entirely.

47. This gap is avoided in the necessity tests in Article XX of the GATT 1994 and Article XIV of the GATS by employing a two-step test which first examines whether the measure is

53 Panel Reports, US – COOL (Article 21.5), para. 7.298. 54 The Appellate Body did not refer to “exceptional circumstances”; rather, it referred to “at least two instances”, which were expressly provided as examples. It did not state that the comparative analysis would be “redundant”; rather, it explained that the comparative analysis “might not be required”. Moreover, “deduction” of the consistency or inconsistency of the challenged measure with Article 2.2 by examining only “certain aspects of the challenged measure” is not consistent with the Appellate Body’s description of the “relational analysis” of the challenged measure in the first step of the necessity test, which involves a process of weighing and balancing all of the relevant factors in relation to one another. See e.g., Appellate Body Report, US – Tuna II (Mexico), para. 318 (“In the context of Article 2.2, the assessment of ‘necessity’ involves a relational analysis of the trade-restrictiveness of the technical regulation, the degree of contribution that it makes to the achievement of a legitimate objective, and the risks non-fulfilment would create. We consider, therefore, that all these factors provide the basis for the determination of what is to be considered ‘necessary’ in the sense of Article 2.2 in a particular case”). See also footnote 643 to para. 318 (“Similarly, in the context of Article XX of the GATT 1994 and Article XIV of the GATS, “necessity” is determined on the basis of “weighing and balancing” a number of factors. (Appellate Body Report, Brazil – Retreaded Tyres, para. 178; Appellate Body Report, US – Gambling, paras. 306-308)). See also Appellate Body Reports, US – COOL, para. 374.


14

necessary by examining all relevant factors and then, if a panel makes a preliminary conclusion that the measure is necessary, this result is confirmed by comparing the challenged measure with possible alternatives.55

48. The necessity tests in the GATT 1994, GATS and the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS Agreement)56 provide consistent and time-tested models whereby the necessity of the measure itself is first assessed and then, only after that assessment is completed, is a comparison with less trade restrictive alternatives made. This approach is sufficiently flexible and rigorous to accommodate all potential factual scenarios.

49. The Panel’s interpretation of the legal test under Article 2.2 completely eliminates the purpose and the value of the “relational analysis”, substituting the crucial process of weighing and balancing all of the relevant factors in this first step with a restrictive approach that was neither described nor suggested by the Appellate Body.

50. As a consequence of its incorrect interpretation and application of the first step of the Article 2.2 necessity test, the Panel erred in asserting that Mexico was required to explain why the Panel was faced with “exceptional circumstances” or a “third exceptional scenario” in the context of the amended COOL measure.

51. As discussed below, Mexico presented detailed arguments in its submissions demonstrating why the holistic weighing and balancing process in the “relational analysis” was sufficient to establish a prima facie case that the amended COOL measure is more trade restrictive than necessary, such that the measure could be found to be inconsistent with Article 2.2 under the first step of the “necessity analysis”, and a “comparative analysis” under the second step would not be required.57 Requiring Mexico to further demonstrate the existence of an “exceptional scenario” based on a separate legal test that “deduces” the challenged measure’s inconsistency with Article 2.2 by “looking solely at certain aspects of the challenged measure” is not only entirely unnecessary where inconsistency with Article 2.2 has already been established by the “relational analysis”, but it is also inconsistent and cannot be reconciled with the legal test established by the Appellate Body.

2. The Panel Erred by Failing to Include in the Relational Analysis the Design and Operation of Label E when Ascertaining the Level of Contribution to the Objective

52. Accurately determining the degree of contribution made by a measure to its objective is a key component of the relational analysis that comprises the first step of the necessity test under Article 2.2.

55 See, for example, Appellate Body Report, EC – Seal Products, para. 5.169 and Appellate Body Report, China – Publications and Audiovisual Products, paras. 241-242. 56 The necessity test in Article 5.6 of the SPS Agreement establishes an obligation similar to the second (i.e., comparative analysis) step of the necessity analyses under Article XX of the GATT 1994, Article XIV of the GATS and Article 2.2 of the TBT Agreement. However, this obligation is cumulative with other obligations in the SPS Agreement, in particular Articles 2.1 and 2.2, which impose necessity requirements on SPS measures, and Article 5.1, which ensures that Members’ SPS measures have an appropriate scientific basis. 57 Mexico’s first written submission, US – COOL (Article 21.5), para. 178.


15

53. The Panel identified the objective of the amended COOL measure to be “to provide consumer information on origin”.58 With respect to the contribution of the measure to this objective, the parties submitted substantial data on the portion of U.S. beef products that were subject to the amended COOL measure’s labelling requirements, including in response to specific questions from the Panel on that subject.59 Mexico did not claim that Label E (the ground beef labelling rule) is inconsistent with the covered agreements as part of the compliance proceedings.60 However, Label E remains an integral component of the amended COOL measure because it is applied to beef products that are consumed in the United States for the purpose of providing consumer information on origin.61

54. Mexico argued that the Label E rules were an important factor in several respects. In particular, in measuring the degree to which the amended COOL measure supplies point of production information on beef to U.S. consumers, Mexico argued that it was relevant that Label E did not provide consumers with such information, just as the categories that are completely exempt from the amended COOL measure (sales at food service establishments, sales of beef as an ingredient in processed foods, and sales by non-“retailers”) do not provide such information.62 The Panel, however, refused to treat Label E as relevant in this context, stating as follows in the section of the Reports entitled “Scope of review of the amended COOL measure's degree of contribution”:

In addition to Labels A-C, the complainants include references to Category D muscle cuts and the ground meat label (Label E) in support of their Article 2.2 claims. In the context of analysing legitimate regulatory distinctions under Article 2.1 of the TBT Agreement, we addressed the parties’ disagreement as to which aspects of the amended COOL measure were relevant to our analysis. In that context, we noted the Appellate Body’s clarification that “in answering the question of whether the measure gives accurate information to consumers, all distinctions drawn by the measure are potentially relevant. By contrast, in an analysis under Article 2.1, we only need to examine the distinction that accounts for the detrimental impact”. The Appellate Body made this statement in relying on findings under Article 2.2 of the TBT Agreement to complete the legal analysis of “whether the measure is calibrated for the purposes of Article 2.1”.

In principle, Label D (muscle cuts from foreign-slaughtered animals) and Label E (ground meat) could potentially be relevant for our assessment of Article 2.2 – including for the amended COOL measure’s degree of contribution – irrespective of whether these labels account for any detrimental impact or constitute relevant regulatory distinctions under Article 2.1. However, the complainants have unequivocally stated that in this compliance dispute they are not bringing claims with respect to Labels D and E. As explained below, the alternative measures they have proposed under Article

58 Panel Reports, US – COOL (Article 21.5), para. 7.331. 59 Panel Reports, US – COOL (Article 21.5), para. 7.258, and footnotes 592 and 593. 60 Mexico challenged the Label E rules in the original proceedings. 61 Label D, which applies to imported muscle cuts, is also part of the amended COOL measure, but the share of Label D meat of U.S. consumption of beef products is minor. 62 Mexico’s first written submission, paras. 137 and 169; Mexico’s second written submission para. 65 and 128; Mexico’s responses to Panel’s questions, paras. 195 - 202.


16

2.2 specifically apply only to US-slaughtered muscle cuts that would be eligible for Labels A-C. The complainants provide no indication of how the conclusions from a relational analysis based on all distinctions of the amended COOL measure could be meaningfully compared to alternative measures pertaining only to Labels A-C.

Thus, including Labels D and E in our relational analysis under Article 2.2 would not be consonant with the manner in which the complainants have presented their arguments, and could prejudice a comparative analysis under the same provision. As the Appellate Body held in US –Tuna II (Mexico), examining alternative measures that do not align with the scope of a relational analysis can result in “an improper comparison” amounting to legal error under Article 2.2 of the TBT Agreement. Accordingly, we do not examine aspects of the measure in our relational analysis which the complainants do not challenge and which they exclude in their arguments under the comparative analysis.63 (emphasis added)

55. There are several errors in this reasoning and in the Panel’s application of this reasoning to the facts.

56. First, the imposition of a requirement for factual “alignment” between the measures in the relational and comparative analyses is inconsistent with the necessity analysis under Article 2.2. In particular, the Panel did not explain how or why the “conclusions” from a relational analysis would be “compared” to alternative measures. The first and second steps of the necessity test involve different types of analysis which are not “compared.” In any event, Mexico’s alternatives are “aligned” with the challenged measure, in that Label E would continue to exist under all of Mexico’s alternatives. Accordingly, to the extent that the Appellate Body finds that there must be some form of factual “alignment” between the measures in the relational and comparative analyses, such alignment exists under Mexico’s arguments. Omitting Label E from the relational analysis would create the exact misalignment that the Panel is seeking to avoid.

57. Second, the relational analysis must be conducted taking into account all relevant factors related to the challenged measure. Clearly Label E is one of these factors because it is directly relevant to determining the degree to which the challenged measure contributes to the legitimate objective. The objective of the amended COOL measure relates to consumer information on origin, in Mexico’s case the origin of beef. It does not relate to specific beef products. The Panel and the Appellate Body in the original proceedings accepted Mexico’s arguments regarding the relevance of the exemptions despite the fact that Mexico had not claimed that the exemptions constituted violations of Article 2.2 and that the exemptions related to various types of beef products including processed beef products. The existence of Label E is incontrovertible evidence that the amended COOL Measure includes ground beef in its objective to provide consumer information on origin. Accordingly, there is no basis to exclude from the contribution analysis Label E beef.

58. Finally, after deciding that Label E should be excluded from the relational analysis, the Panel went on to conduct that analysis in a contradictory manner. In its discussion of the amended COOL measure’s degree of contribution to its objective, the Panel stated:

63 Panel Reports, US – COOL (Article 21.5), paras. 7.343-7.345.


17

As noted, the amended COOL measure only covers between 33.3% and 42.3% of all beef consumed in the United States, and between 15.9% and 16.5% of all pork muscle cuts. Conversely, between 57.7% and 66.7% of all beef consumed in the United States, and between 83.5% and 84.1% of all pork muscle cuts, are either sold in a food service establishment, as an ingredient in a processed food item, or by an entity not subject to be licensed as a “retailer”.64

The Panel went on to discuss Labels A, B and C, and stated that the amount of information provided for Labels B and C had been increased under the amended COOL measure. However, the Panel failed to mention Label E at all. The Panel’s final conclusion was that “the amended COOL measure thus makes a considerable but necessarily partial contribution to its objective of providing consumer information on origin.”65

59. Thus, in making its finding that the amended COOL measure makes a “considerable” contribution to its objective, the Panel relied on the figures of 33.3% to 42.3% of all types of beef products consumed in the United States.66 Notwithstanding the Panel’s finding that Label E should not be taken into consideration in the contribution analysis, those figures include beef sold under Label E. The Panel separately found that "[o]f total US beef consumption, between 16.3% and 24.5% are muscle cuts carrying Labels A-D, and between 16.6% and 17.8% are ground meat carrying Label E"67. Accordingly, Label E applies to about half of the beef products subject to the labelling requirements under the amended COOL measure.68 Had the Panel excluded the share of beef products sold under Label E from its calculation of the degree of contribution, the coverage of the muscle cut Labels A, B, C and D would have been only 16.3 percent to 24.5 percent of total beef products consumed in the United States, and the share of Labels A, B and C would have been somewhat lower because Label D would have to be excluded (Label D does not provide point of production information).69 In Mexico’s view, contrary to the finding of the Panel, a degree of contribution below one quarter is not a “considerable” contribution to the objective. The Panel therefore erred in its evaluation of the contribution of the muscle cut labelling provisions to the objective.

60. There are further contradictions in the Panel’s analysis. Later, in considering the Label E rules as a potential alternative measure for the purposes of the comparative analysis, the Panel found that the Label E rules did not provide sufficient information to fulfil the objective. This was inconsistent with crediting Label E as contributing to the fulfilment of the objective of providing consumers with information on origin in order to determine the 33.3%-42.3% degree of contribution.

64 Panel Reports, US – COOL (Article 21.5), para. 7.347 and footnote 785. This footnote references the data presented in paras. 7.258 (beef) and 7.262 (pork) of the Report. 65 Panel Reports, US – COOL (Article 21.5), para. 7.356. 66 Panel Reports, US – COOL (Article 21.5), paras. 7.347 and 7.356. 67 Panel Reports, US — COOL (Article 21.5), para. 7.258. 68 Panel Reports, US – COOL (Article 21.5), para. 7.258. Collectively, Labels A-E account for between 33.3% and 42.3% of total (muscle cut and ground) beef consumption. Label E ground beef accounts for between 16.6% and 17.8% of total (muscle cut and ground) beef consumption. Thus, Label E accounts for approximately half of the beef labelled. 69 This is derived from the figures in para. 7.258 of the Panel’s Reports by deducting the range for ground meat 16.6%-17.8% from the range for total labelled products 33.3%-42.3%.


18

61. Accordingly, the Panel erred by failing to include in the relational analysis the design and operation of Label E when ascertaining the level of contribution to the objective.

3. The Panel Erred in its Assessment of the Risks Non-fulfilment would Create, including the Gravity of the Consequences that would Arise from Non-fulfilment of the amended COOL measure’s objective

62. The Panel held that “[a]ccording to the Appellate Body, reviewing ‘the risks non-fulfilment [of a legitimate objective] would create’ entails assessing ‘the nature of the risks at issue and the gravity of the consequences that would arise from non-fulfilment of the legitimate objective’.”70 In this respect, the Panel acknowledged that an assessment of the risks non-fulfilment would create comprises an examination of two elements. The first element, the “nature of the risks at issue”, involves an evaluation of the outcome – that is, the resulting circumstances – of non-fulfilment of the legitimate objective. The second element, the “gravity of the consequences”, involves an assessment of the severity of the effects of the circumstances resulting from non-fulfilment of the legitimate objective. Put simply, the first step is to identify the consequences of non-fulfilment, and the second step is to ascertain the gravity of those consequences. The assessment of the risks non-fulfilment would create is based on the findings resulting from these two steps.

63. As discussed below, the Panel entirely failed to ascertain the “gravity of the consequences that would arise from non-fulfilment”. Specifically, the Panel erred in failing to find on the record before it that the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective is very low. Further, in failing to make any overall finding with respect to the gravity of the consequences, the Panel not only erroneously declined to properly complete the “relational analysis”, but it also prevented itself from properly completing the subsequent “comparative analysis” in the second step of the necessity test.

64. Despite these errors, Mexico notes that the Panel did make an overall finding on the risks non-fulfilment would create, concluding that “we find that there is some risk associated with the non-fulfilment of the amended COOL measure’s legitimate objective.”71 This finding should have been taken into account in the weighing and balancing process with the other relevant factors in the “relational analysis”, and then further taken into account in the “comparative analysis” of the amended COOL measure to find that the first and second proposed alternative measures make an equivalent degree of contribution, but are less trade restrictive, taking into account the degree of only “some risk” that non-fulfilment would create. The Panel did neither.

65. As discussed below, the Panel made the following errors in undertaking its analysis of the risks non-fulfilment would create and, specifically, in ascertaining the gravity of the consequences that would arise from non-fulfilment:

• The Panel erred in its decision to strictly limit its assessment of the risks that non-fulfilment would create to only two criteria – namely, (i) consumer

70 Panel Reports, US – COOL (Article 21.5), para. 7.374, citing Appellate Body Report, US – Tuna II (Mexico), para. 321; Appellate Body Reports, US – COOL, para. 377. 71 Panel Reports, US – COOL (Article 21.5), para. 7.417 (emphasis added).


19

interest in country of origin information, and (ii) the willingness of consumers to pay the costs of obtaining country of origin information on product labels – to the exclusion of all other relevant considerations;

• The Panel erred by omitting the relative importance of the interests or values furthered by the amended COOL measure from its assessment of the risks non-fulfilment would create;

• The Panel erred by omitting the design, structure and architecture of the amended COOL measure from its assessment of the risks non-fulfilment would create; and

• The Panel erred in not making a finding on the gravity of the consequences of non-fulfilment. In this respect, the Panel failed to make an objective assessment of the matter and the facts before it, and thereby acted inconsistently with Article 11 of the DSU. Specifically, the Panel erred by failing to find on the merits of the evidence before it that the gravity of the consequences that would arise from non-fulfilment is very low.

66. The Appellate Body recently summarized the interpretation and application of Article 11 of the DSU in its Report in US – Carbon Steel (India). It stated that: (i) Article 11 requires a panel to consider all the evidence presented to it, assess its credibility, determine its weight, and ensure that its factual findings have a proper basis in that evidence; (ii) panels may not make affirmative findings that lack a basis in the evidence contained in the panel record; (iii) within these parameters, it is generally within the discretion of the panel to decide which evidence it chooses to utilize in making findings; (iv) the mere fact that a panel did not explicitly refer to each and every piece of evidence in its reasoning is insufficient to establish a claim of violation under Article 11; (v) an appellant must explain why such evidence is so material to its case that the panel's failure to address explicitly and rely upon the evidence has a bearing on the objectivity of the panel's factual assessment; (vi) an appellant must identify specific errors regarding the objectivity of the panel's assessment; and (vii) not every error allegedly committed by a panel amounts to a violation of Article 11 of the DSU, but only those that are so material that, taken together or singly, they undermine the objectivity of the panel's assessment of the matter before it.72

67. As explained below, with respect to certain factual findings, the Panel did not meet the standard set out in Article 11.

a. The Panel Erred when it Restricted its Assessment of the Risks Non-fulfilment would Create to only Two Criteria

68. The Panel erred in its decision to strictly limit its assessment of the risks that non-fulfilment would create to only two criteria73 – namely, (i) consumer interest in country of origin information, and (ii) the willingness of consumers to pay the costs of obtaining country of origin information on product labels – to the exclusion of all other relevant

72 Appellate Body Report, US – Carbon Steel (India), paras. 4.78-4.80. 73 Panel Reports, US – COOL (Article 21.5), para. 7.379 (“We confine our analysis to the express terms of the TBT Agreement, as clarified by the Appellate Body, instead of defining the precise relationship between terms in the TBT Agreement and the GATT 1994”).


20

considerations.74 The Panel incorrectly determined this to be “in accordance with the Appellate Body’s approach in the original dispute.”75

69. As previously discussed, the Panel’s interpretation of the Appellate Body’s findings in the original US – COOL dispute is mistaken. The Appellate Body was not establishing an “approach” to the assessment of the risks that non-fulfilment of the amended COOL measure’s objective would create. Rather, in the context of attempting to complete the Article 2.2 legal analysis, the Appellate Body was merely working with the limited factual findings made by the original panel. Notably, the Appellate Body’s first observation in this respect was that the original panel had made no findings regarding the risks that non-fulfilment would create under the Article 2.2 necessity test.76 The Appellate Body was therefore constrained in its examination to whatever factual findings were available to it from the original panel’s record that might be helpful in its attempt to assess the risks non-fulfilment would create. In this respect, the Appellate Body referenced findings made by the original panel in the disparate contexts of: (i) its examination of the identified objective’s legitimacy under Article 2.2;77 and (ii) its assessment of de facto discrimination under Article 2.1.78 As none of the findings considered by the Appellate Body were made in the course of, or for the purpose of, undertaking an assessment of the risks non-fulfilment would create, these findings do not necessarily reflect the most relevant or appropriate criteria for the purposes of such an assessment, and certainly not to the exclusion of all other relevant considerations. Finally, at no point did the Appellate Body make any ruling or statement to the effect that the findings of the original panel which it “briefly” considered (in the space of a single paragraph) – i.e., (i) consumer interest in origin information; and (ii) the willingness of consumers to pay the costs necessary to obtain such information – represent the exclusive criteria that should be considered in an assessment of the risks that non-fulfilment of the objective would create.

74 Panel Reports, US – COOL (Article 21.5), para. 7.381 (“we review the risks non-fulfilment of the amended COOL measure’s objective would create by assessing the nature of the risks and the gravity of the consequences. We do this by assessing consumer interest in, and willingness to pay for, country of origin information, in accordance with the Appellate Body’s approach in the original dispute”). 75 Panel Reports, US – COOL (Article 21.5), para. 7.381. See Appellate Body Reports, US – COOL, para. 478. Specifically, the Panel held that: “In the original dispute, the Appellate Body briefly addressed the risks non-fulfilment of the original COOL measure’s objective would create by taking into account consumer interest in, and willingness to pay for, country of origin information. Like the Appellate Body, we address these factors for the risks non-fulfilment of the amended COOL measure’s objective would create.” See Panel Reports, US – COOL (Article 21.5), para. 7.375. 76 Appellate Body Reports, US – COOL, para. 478 (“The Panel did not make findings regarding the risks that non-fulfilment of the objective pursued by the United States through the COOL measure would create”). 77 See Appellate Body Reports, US – COOL, footnote 983 to para. 478, citing Panel Reports, US – COOL, para. 7.647 (“The absence of evidence independent of the legislative process showing consumer desire for the information on origin as defined by the United States, however, is not in itself sufficient to prove that providing such consumer information on origin is not a legitimate objective within the meaning of Article 2.2”). 78 See Appellate Body Reports, US – COOL, footnote 984 to para. 478, citing Panel Reports, US – COOL, para. 7.354 and footnote 498 thereto (“The fact that consumers are not ready to bear all the costs of country of origin labelling of beef and pork is also demonstrated by the lack of interest in a voluntary COOL regime”).


21

70. Moreover, the Panel failed to explain why only these criteria should be determinative of the risks non-fulfilment would create, to the exclusion of any and all other relevant criteria.79 Mexico does not dispute that evidence relating to consumer demand for country of origin information is relevant to evaluating the risks non-fulfilment of the amended COOL measure’s objective would create. The Panel erred, however, in arbitrarily restricting its analysis to the above-referenced criteria only, thereby rejecting other highly relevant criteria that were raised specifically in relation to this issue. This constitutes a legal error on the part of the Panel.

71. The other relevant criteria the Panel should have considered are discussed below.

b. The Panel Erred by Omitting from its Assessment the Relative Importance of the Interests or Values furthered by the Amended COOL Measure

72. The Panel erred in declining to consider the relative importance of the values or interests furthered by the amended COOL measure in assessing the risks that non-fulfilment would create.80 The Panel’s justification for rejecting this as a relevant consideration was that: (i) it had already decided that the relative importance of the interests or values protected by the challenged measure would not be considered as a separate factor in the “relational analysis”; and (ii) it would “confine” its analysis of the risks non-fulfilment would create “to the express terms of the TBT Agreement, as clarified by the Appellate Body”, such that “this factor of an Article 2.2 analysis is to be examined according to two criteria: the nature of the risks and the gravity of the consequences.”81 The Panel did not explain, however, why the “relative importance” of the amended COOL measure’s objective could not be considered for the purposes of ascertaining the “nature of the risks” or measuring the “gravity of the consequences”. As discussed below, this factor is particularly relevant to assessing the gravity of the consequences that would arise from non-fulfilment of the objective.

(i) The Relative Importance of the Amended COOL Measure’s Objective is Highly Relevant to the Assessment of the Risks Non-Fulfilment would Create

79 See Panel Reports, US – COOL (Article 21.5), para. 7.375 (“In the original dispute, the Appellate Body briefly addressed the risks non-fulfilment of the original COOL measure’s objective would create by taking into account consumer interest in, and willingness to pay for, country of origin information. Like the Appellate Body, we address these factors for the risks non-fulfilment of the amended COOL measure's objective would create”), citing Appellate Body Reports, US – COOL, para. 478 (“The Panel did, however, cast doubt on the probative value of evidence presented by the United States in order to show that consumers want information on the countries of birth, raising, and slaughter of livestock from which meat is derived. The Panel also took note of US consumers’ unwillingness to “bear all the costs of country of origin labelling of beef and pork”. That most US consumers are not prepared to pay to receive information on origin as defined in the COOL measure with respect to the meat products they purchase suggests that obtaining such information is not a high priority for such consumers. This in turn seems to indicate that the consequences that may arise from non-fulfilment of the objective would not be particularly grave”). 80 Panel Reports, US – COOL (Article 21.5), paras. 7.378-7.379. 81 Panel Reports, US – COOL (Article 21.5), para. 7.379.


22

73. The relative importance of the interests or values furthered by the amended COOL measure is a consideration that defines the risks that would arise in the event of non-fulfilment of the objective. This consideration is especially relevant to an assessment of the gravity of the consequences that would arise from non-fulfilment. For example, where the interests or values furthered by a measure are relatively important (i.e., the protection of human life or health), then the gravity of the consequences of non-fulfilment of the measure’s objective will be correspondingly significant (i.e., injury or death). Similarly, where the interests or values advanced by a measure are relatively unimportant (i.e., the provision of consumer information on the origin of beef products), then the gravity of the consequences of non-fulfilment will be correspondingly insignificant (i.e., consumers will merely be uninformed as to the origin of beef products, with no further or collateral effects).

74. The Panel refused to consider the relative importance of the interests or values furthered by the amended COOL measure as a “separate factor” of the Article 2.2 legal test.82 Instead, the Panel found that the relative importance of the amended COOL measure’s objective of providing consumer information on origin was not relevant under Article 2.2 on the basis that the Appellate Body had not already identified “relative importance” as such a factor when setting out the overall Article 2.2 test in US – Tuna II (Mexico) and the original US – COOL dispute.83

75. However, it is of no consequence or relevance to the present dispute that the Appellate Body did not specifically identify or consider, as a relevant factor in the “relational analysis”, the relative importance of the interests or values furthered by the measure at issue in either US – Tuna II (Mexico) or the original US – COOL dispute. This factor was not raised as an issue in US – Tuna II (Mexico), which involved a different type of measure. In the case of US – COOL, the Appellate Body was attempting to “compete the analysis”, and it was both limited to, and hindered by, the “lack [of] clear and precise Panel findings with regard to [the relevant] factors.”84

76. Mexico argued before the Panel that the Appellate Body did not intend to restrict the “evaluation of a number of factors”85 in the “relational analysis” to only the three factors that it identified.86 By using the phrase “evaluation of a number of factors”, the Appellate Body indicated an indefinite “number” of factors, rather than exactly “three” factors. Further, in the phrase “considering factors that include”, the word “include” indicates that the three factors identified by the Appellate Body are not closed or exclusive, but rather open and inclusive of other relevant factors. While the three identified factors should be considered as relevant to the evaluation of the “necessity” of the measure’s trade-restrictiveness, there is nothing in the Appellate Body’s reasoning to indicate that other factors, if also clearly relevant to this assessment, should be excluded from the “relational analysis”.

77. In Mexico’s view, the “relative importance” of the interests or values furthered by a trade restrictive measure is highly relevant to the question of whether or not the measure is

82 Panel Reports, US – COOL (Article 21.5), paras. 7.310-7.311. See also para. 7.379. 83 Panel Reports, US – COOL (Article 21.5), para. 7.308 (“In neither dispute did the Appellate Body identify ‘relative importance’ as a factor when setting out the overall test under Article 2.2”). 84 Appellate Body Reports, US – COOL, paras. 479 and 470. 85 See Appellate Body Report, Tuna II (Mexico), para. 322. 86 Mexico’s first written submission, US – COOL (Article 21.5), para. 157.


23

more trade restrictive than necessary within the meaning of Article 2.2. Accordingly, this factor should be taken into consideration in the weighing and balancing process of the “relational analysis”. It is noteworthy that this factor is considered at the beginning of a “necessity” analysis under Article XX of the GATT 1994 or Article XIV of the GATS,87 which demonstrates its relevance and importance to determining the issue of “necessity” in relation to a trade restrictive measure. To the extent that the “relative importance” of the amended COOL measure’s objective is not considered as a “separate factor” in the weighing and balancing process, it should be considered for the purpose of evaluating the risks non-fulfilment would create.

(ii) The Relative Importance of the Amended COOL Measure’s Objective is Very Low

78. The interests or values furthered by the amended COOL Measure – i.e., the provision of consumer information on origin for the covered beef products88 – are substantially lower on the spectrum of “relative importance” than, for example, protecting the environment or protecting human beings from health risks, both of which are vital and important in the highest degree,89 or protecting public morals, which has been ranked among the most important values or interests pursued by Members as a matter of public policy.90 In comparison to the protection of human life or health, or the protection of the environment, or the protection of public morals, there is nothing vital about the mere provision of origin information to consumers for a limited range of beef products.

79. Further, in the context of consumer information, one way to assess the “relative importance” of the interests or values furthered by the amended COOL measure is to consider the actual demand by U.S. consumers for the origin information in question. Low demand or general indifference among the public regarding consumer information is an indication of lower relative importance. In this regard, Mexico notes that the interests or values furthered by the amended COOL measure are not “common” among U.S. consumers.91 Rather, these interests or values serve a niche demand with particular preferences, as evidenced in the USDA’s acknowledgement that there is only “interest by certain U.S. consumers in information disclosing the countries of birth, raising and slaughter on muscle cut product labels”.92

80. For the foregoing reasons, the interests and values furthered by the amended COOL measure are not common, are clearly non-vital and are of very low “relative importance”,

87 Mexico's first written submission, para. 156. 88 See Panel Reports, US – COOL (Article 21.5), para. 7.331 (“we adopt the Appellate Body’s first formulation (without the footnote) as the identification of the objective of the amended COOL measure: ‘to provide consumer information on origin’”). 89 Appellate Body Report, Brazil - Retreaded Tyres, para. 179. 90 Panel Report, China – Publications and Audiovisual Products, para. 7.817. The Appellate Body acknowledged this finding of the Panel but did not rule upon it. See Appellate Body Reports, China – Publications and Audiovisual Products, para. 243. 91 See e.g., Appellate Body Report, Korea – Various Measures on Beef, para. 162 (“The more vital or important those common interests or values are, the easier it would be to accept [a trade restrictive measure] as “necessary”). 92 2013 Final Rule, p. 31376 (emphasis added) (Exhibit MEX-3 in original dispute).


24

especially in relation to the other relevant factors, such as the “considerable degree of trade-restrictiveness” that is exacerbated under the amended measure.93

(iii) Taking into Account the Very Low Relative Importance of the Amended COOL Measure’s Objective, the Gravity of the Consequences that would arise from Non-Fulfilment are Objectively Not Significant

81. A fair, objective and realistic evaluation of the gravity of the consequences of a non-fulfilment scenario must acknowledge and take into account the very low relative importance of providing consumer information on origin. Taking this very low relative importance into account, it simply cannot be said that the consequences of non-fulfilment would be grave, severe or even significant. The consequences of non-fulfilment of the amended COOL measure’s objective would simply be that consumers would receive no origin information on the limited range of covered beef products.94 These consequences would not endanger human or animal life or health, or the protection of the environment, or the integrity of public morals. Therefore, it must be found that the consequences of non-fulfilment would not be particularly significant. Under the circumstances, such a finding is not only objectively accurate and realistic, but also entirely consistent with the finding made by the Appellate Body in the original US – COOL dispute: “in our view, the Panel’s factual findings suggest ... that the consequences that may arise from non-fulfilment of the objective would not be particularly grave.”95

82. A finding that the consequences of non-fulfilment would not be significant leads to the corresponding conclusion that the risks non-fulfilment would create are not significant. This finding alone, considered in relation to the other relevant factors in the weighing and balancing process of the “relational analysis”, would weigh and balance against the “necessity” of the measure’s “considerable degree of trade-restrictiveness”.96 In other words, the “necessity” of the amended COOL measure’s considerable degree of trade-restrictiveness cannot be justified where the gravity of the consequences that would arise from non-fulfilment of the objective are very low. It is only where the risks of non-fulfilment are found to be substantial or grave that this factor would weigh and balance in favour of the “necessity” of a “considerable degree of trade-restrictiveness” under Article 2.2.

83. As discussed below, it is Mexico’s view that an objective assessment of the evidence before the Panel, on its merits, demonstrates that the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective is very low.

93 Panel Report, US – COOL (Article 21.5), para. 7.370. 94 Panel Reports, US – COOL (Article 21.5), para. 7.417. 95 Appellate Body Reports, US – COOL, paras. 478-479 (emphasis added). 96 Panel Reports, US – COOL (Article 21.5), para. 7.370.


25

c. The Panel Erred by Omitting the Design, Structure and Architecture of the Amended COOL Measure from its Assessment of the Risks Non-Fulfilment would Create

84. A consideration of the design, architecture, revealing structure, operation and application of the amended COOL measure is also relevant to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment.

85. As previously discussed, more than half of all beef products that are sold for consumption in the United States are entirely exempt from the country of origin labelling requirements under the amended COOL measure.97 Consumers of these exempted beef products do not receive country of origin information pursuant to the COOL measure, and may receive no origin information at all. Similarly, as discussed above, ground beef products are required to be labelled, but not with the same level of detailed information as muscle cuts.

86. If the gravity of the consequences that would arise from non-fulfilment of the legitimate objective were significant, then the amended COOL measure would not expressly exempt over half of all beef products sold for consumption in the United States from the requirement to provide country of origin information to consumers, and subject another major portion of beef products (ground beef) to less detailed and more flexible requirements. The very existence of these exemptions and different rules for ground beef in the design, architecture and revealing structure of the amended COOL measure, and the substantial effects of their operation and application in the U.S. market, constitutes demonstrative evidence that the gravity of the consequences of providing little or no information on the origin of beef products to consumers is very low to the point of being insignificant or trivial. The Panel erred in declining to take this analysis into account at all in its evaluation of the risks non-fulfilment would create.

87. The Panel’s reasons for rejecting any consideration of the design, architecture, revealing structure, operation and application of the amended COOL measure included speculation that “there may be a variety of possible reasons unrelated to risks for exempting or treating differently certain product categories under a Member’s technical regulation, such as regulatory or compliance costs.”98 While Mexico does not disagree with the standalone supposition that the reasons underlying the exemptions may be unrelated to the risks non-fulfilment of the objective would create, this has no relevance or effect on the above analysis. The fact that the exemptions and different rules for ground beef have been designed and applied to the great majority of beef products sold for consumption in the United States, regardless of the underlying reasons, is evidence that the gravity of the consequences of not providing consumer information on the origin of beef products is very low. Whether the gravity of the consequences is very low in relation to the burden of

97 Panel Reports, US – COOL (Article 21.5), para. 7.258. These exemptions cover any beef product that is used as “an ingredient in a processed food item”; or sold by an entity – such as a butcher shop – that does not meet the measure’s definition of “retailer”; or sold by a “food service establishment”, which is defined to include “a restaurant, cafeteria, lunch room, food stand, saloon, tavern, bar, lounge, or other similar facility operated as an enterprise engaged in the business of selling food to the public”. The exempted beef products constitute between 57.7 and 66.7 percent of all beef products sold for consumption in the United States. 98 Panel Reports, US – COOL (Article 21.5), para. 7.380 (emphasis added).


26

“regulatory or compliance costs” – that is, on the basis of a cost/benefit assessment – does not alter the outcome of the above analysis. Such a hypothetical scenario merely indicates that the gravity of the consequences that would arise from non-fulfilment of the legitimate objective is so low that it does not merit (i.e., it is not “worth”) incurring the “regulatory or compliance costs” of providing origin information. The Panel’s rejection of this clearly relevant consideration, particularly on the basis of such speculative and hypothetical grounds, constitutes a legal error.

88. In support of its decision to reject evidence based on the design, architecture, revealing structure, operation and application of the amended COOL measure for the purposes of assessing the risks non-fulfilment would create, the Panel also found that “the amended COOL measure’s treatment of different categories of meat products is more directly connected to the degree of contribution under Article 2.2 and the legitimacy of regulatory distinctions under Article 2.1.”99 The Panel did not explain on what basis this evidence is “more directly connected” to other factors or why the same evidence cannot be considered for the purposes of resolving different legal issues or tests in their respective contexts. Specifically, Mexico does not agree that the relevance of the exemptions under the amended COOL measure is “more directly connected” to the degree of contribution under Article 2.2 or legitimacy under Article 2.1. This evidence is clearly relevant to the issue of ascertaining the gravity of the consequences that would arise from not providing U.S. consumers with country of origin information on beef products. That it might also be relevant to other legal issues and tests does not disqualify it from being properly considered in this context, in order to resolve the issue of the risks non-fulfilment would create.

89. In rejecting this evidence, the Panel committed a legal error. Moreover, it failed to make an objective assessment of the matter before it, including an objective assessment of the facts, and therefore acted inconsistently with Article 11 of the DSU. This evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

d. The Panel Erred in Failing to Ascertain the Gravity of the Consequences that Would Arise from Non-Fulfilment of the Amended COOL Measure’s Objective

(i) The Critical Impact on the Panel’s Entire Article 2.2 Analysis as a Result of its Error in Not Making a Finding regarding the Gravity of the Consequences that Would Arise from Non-Fulfilment

90. While stating that it had “established the nature of the risks, and the consequences, of not fulfilling the amended COOL measure’s objective,”100 the Panel erred by finding that, based on the record of evidence before it, it was “unable to ascertain” the gravity of the

99 Panel Reports, US – COOL (Article 21.5), para. 7.380. 100 Panel Reports, US – COOL (Article 21.5), para. 7.424.


27

consequences that would arise from non-fulfilment.101 This error had the effect of rendering both steps of the Article 2.2 necessity test unworkable under the Panel’s approach.

91. First, on the basis that it could not ascertain the gravity of the consequences of non-fulfilment of the amended COOL measure’s objective, the Panel declined to consider all of the relevant factors in relation to one another in the weighing and balancing process of the “relational analysis” and draw definitive conclusions on the “necessity” of the amended COOL measure’s trade-restrictiveness.102

92. Second, the Panel’s failure to determine the gravity of the consequences of non-fulfilment also compromised the “comparative analysis” in the second step, where it prevented the Panel from comparing the degrees of contribution between the challenged measure and the first and second proposed alternative measures. In this respect, the Panel correctly acknowledged that “‘the risks non-fulfilment would create’ may be a relevant factor in assessing whether an alternative measure fulfils the legitimate objective to an equivalent degree as the challenged measure.”103 Recognizing “the potential relevance of the ‘risks non-fulfilment would create’ in comparing degrees of contribution”, the Panel considered that an alternative measure that provides consumers with less origin information or less accurate origin information for a significantly wider range of products “might achieve an equivalent degree of contribution as the amended COOL measure.”104 However, the Panel found that it was unable to determine “the specific implications of risks of non-fulfilment” because it had been unable to ascertain the gravity of the consequences that would arise from non-fulfilment in the “relational analysis”.105 As a consequence, the Panel held that it was unable to take into account “the specific implications of risks of non-fulfilment” in order to resolve the comparisons of the different degrees of contribution between the amended COOL measure and the first and second proposed alternative measures.106 This effectively ended the comparative analyses of the first and second alternative measures before any conclusions could be drawn.

101 Panel Reports, US – COOL (Article 21.5), para. 7.423. 102 Panel Reports, US – COOL (Article 21.5), para. 7.424. 103 Panel Reports, US – COOL (Article 21.5), paras. 7.486-7.487, citing Appellate Body Report, US – Tuna II (Mexico), para. 321 (“the obligation to consider ‘the risks non-fulfilment would create’ suggests that the comparison of the challenged measure with a possible alternative measure should be made in the light of the nature of the risks at issue and the gravity of the consequences that would arise from non-fulfilment of the legitimate objective. This suggests a further element of weighing and balancing in the determination of whether the trade-restrictiveness of a technical regulation is ‘necessary’ or, alternatively, whether a possible alternative measure, which is less trade restrictive, would make an equivalent contribution to the relevant legitimate objective, taking account of the risks non-fulfilment would create, and would be reasonably available”); Appellate Body Reports, US – COOL, para. 377. 104 Panel Reports, US – COOL (Article 21.5), para. 7.488. 105 Panel Reports, US – COOL (Article 21.5), paras. 7.488 and 7.501. See also para. 7.374, in which the Panel confirmed that: “According to the Appellate Body, reviewing ‘the risks non-fulfilment [of a legitimate objective] would create’ entails assessing ‘the nature of the risks at issue and the gravity of the consequences that would arise from non-fulfilment of the legitimate objective.” 106 See Panel Reports, US – COOL (Article 21.5), paras. 7.488-7.491; 7.501-7.503.


28

(ii) The Panel Erred by Not Finding on the Merits that the Gravity of the Consequences of Non-fulfilment is Very Low

93. For the purposes of assessing the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure, the Panel considered “consumer demand for origin information as a relevant indicator”.107 However, the Panel erred in its approach to the assessment of the evidence adduced by Mexico and the United States in relation to consumer demand for origin information. In Mexico’s view, an objective assessment of the evidence on the record before the Panel would establish a prima facie case that consumer demand for origin information on the covered products is very low. This, in turn, supports the conclusion that the gravity of the consequences that would arise in a non-fulfilment scenario is very low.

94. Further, while evidence of low or insignificant consumer demand supports a finding that the gravity of the consequences that would arise from nonfulfillment of the objective is low, evidence of significant consumer demand would not be sufficient, on its own, to support a finding that the gravity of the consequences of non-fulfilment is somehow significant. As previously discussed, there is nothing vital about the mere provision of country of origin information to consumers, and the relative importance of this objective is very low. The absence of country of origin information on the labels of beef products would not endanger the lives or the health of U.S. consumers, or the U.S. environment, or U.S. public morals. Thus, the gravity of the consequences of a non-fulfilment scenario – in which the only consequence would be that consumers simply do not receive country of origin information on beef products – cannot be said to be grave, severe or even significant. This is a reality that remains unaffected even in hypothetical scenarios of significant or high consumer demand.

(A) The Panel Applied an Erroneous Approach to its Assessment of the Evidence regarding Consumer Demand

95. The Panel considered that there are “two facets of consumer demand for origin information” that should be assessed separately: (i) consumers’ interest in country of origin information; and (ii) consumers’ willingness to pay for country of origin information.108 In Mexico’s view, consumers’ willingness to pay for origin information should not be treated as a separate “facet” of consumer demand. Rather, a consumer’s willingness to pay depends upon the degree of the consumer’s interest and provides an indicator of the significance of the degree of interest. In this respect, consumers’ willingness to pay is not a separate “facet” or element of consumer demand for origin information.

96. Second, the Panel made a fundamental error by requiring the evidence in relation to consumer demand for country of origin information to directly and narrowly address consumer interest in, and willingness to pay for, the specific kinds of “general” or “point-of-production” COOL information provided in accordance with the original and amended COOL



29

measures.109 Under this approach, the Panel rejected evidence regarding consumer demand for origin information because it was based on contexts that were not exactly equivalent to, or consistent with, the specific kinds of information provided under the COOL measures,110 or because it would not be relevant under certain hypothetical scenarios contemplated by the Panel.

97. For example, the Panel rejected a study about food values in relation to beef, chicken and milk that concluded that, compared to other values, such as food safety, consumers do not value origin highly.111 Rather than objectively assessing this evidence on its merits, the Panel disregarded it on the basis that the origin information referenced in the study – i.e., as “the extent to which the locations and identities of producers and processors are known” – was “very different” from the “definition” of origin information in the original and amended COOL measures. The Panel did not explain how the reference to origin information in the study was conceptually incompatible or impossible to reconcile with the amended COOL measures’ “definition” or, moreover, why such a comparison was even necessary for the purposes of assessing consumer demand for general origin information on beef products. Further, the Panel criticized the study on the speculative basis that “consumers might have answered differently had origin been defined more consistently with the original or amended COOL measures.”112 In rejecting this evidence, the Panel failed to make an objective assessment of the evidence on its own merits and thereby acted inconsistently with Article 11 of the DSU. This evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed above and below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

98. Similarly, the Panel rejected the evidence adduced in the KSU Revealed Demand Study on the basis of speculation that hypothetical circumstances “could” or “might” explain why the origin information on meat labels under the original COOL measure had no impact on consumer demand: “There could be various reasons for which relative demand for the covered and exempted products has remained steady. For example, consumers might have found the labels confusing, and demand for both the covered and exempted products might be inelastic for reasons unrelated to the measure, such as consumer preferences for specific types of meat or other foodstuffs.”113 Again, by assessing the evidence on the basis of 109 See e.g. Panel Reports, US – COOL (Article 21.5), paras. 7.383, 7.407, 7.413 (“The parties’ limited evidence concerning willingness to pay does not address consumers’ willingness to pay for origin information specifically according to point-of-production information”), 7.416 (“we have not been given probative evidence showing consumer willingness to pay for country of origin information according to point-of-production”). 110 See e.g. Panel Reports, US – COOL (Article 21.5), para. 7.393 (“The questions posed in the working paper referenced origin as follows: ‘the extent to which the locations and identities of producers and processors are known’. We are not inclined to rely on this evidence as proof of consumer general interest in covered products’ country of origin, given that origin has been referenced very differently from its definition in the original and amended COOL measures. For us, it cannot be excluded that consumers might have answered differently had origin been defined more consistently with the original or amended COOL measures”). 111 Panel Reports, US – COOL (Article 21.5), paras. 7.392-7.393. 112 Panel Reports, US – COOL (Article 21.5), para. 7.393. 113 Panel Reports, US – COOL (Article 21.5), para. 7.387 (emphasis added).


30

speculative reasons why it might not be relevant, the Panel failed to make an objective assessment of the evidence on its own merits and acted inconsistently with Article 11 of the DSU. Again, this evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed above and below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

(B) The Panel Erred in Accepting the Inherent Assumption in the Design of a Telephone Poll Question as Evidence of “Some Consumer Interest” in COOL Information

99. The Panel rejected Mexico’s evidence regarding consumer demand for origin information on the basis that it was not directly responsive to the level of consumer interest in, or willingness to pay for, the specific origin information provided under the COOL measures – holding the evidence to a very precise standard of relevance and probity. Inconsistently, the Panel accepted telephone opinion poll results adduced by the United States as “relevant” evidence that “suggests that there is some consumer interest in country of origin information in general”, even though the poll in no way tested for the level of consumer interest, demand, or willingness to pay for origin information.114

100. The poll question addressed only consumer preference for the specific kind of country of origin information received on supermarket beef products, after giving the consumer a range of possible options, none of which was “none” or “no origin information”.115 The Panel itself acknowledged that “the poll, in particular the question and the possible categories of responses, assumes that consumers would like some form of origin labelling.”116 The Panel further recognized that “the poll does not explicitly ask whether or not consumers would like to have meat carry labels with origin information.”117 In other words, the poll did not test consumer demand for origin information on beef products at all. The poll question at issue merely assumed “some” level of demand, and on the basis of this assumption in the design of the poll question, the Panel found that “there is some consumer interest in country of origin information”.118 This demonstrates a failure on the part of the Panel to make an objective and fair assessment of the evidence before it, and therefore was inconsistent with Article 11 of the DSU. This evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

114 See Panel Reports, US – COOL (Article 21.5), para. 7.399. 115 Panel Reports, US – COOL (Article 21.5), para. 7.397. 116 Panel Reports, US – COOL (Article 21.5), para. 7.399. 117 Panel Reports, US – COOL (Article 21.5), para. 7.399. 118 Panel Reports, US – COOL (Article 21.5), para. 7.399.


31

(C) The Panel Erred in Finding that Willingness to Pay for a “Product of North American Label” is Evidence of Willingness to Pay for Country of Origin Information

101. One of the other findings made by the Panel was that “consumers show some willingness to pay for general country of origin information.”119 However, the Panel appears to have based this conclusion on its earlier finding that the willingness of consumers to pay approximately the same amount for “Product of North America” origin information as for “product of United States” origin information on meat product labels shows that “consumers are indeed willing to pay something for origin information – whether country-specific or not.”120 “Product of North America” does not indicate “country” of origin, but could indicate one or more of three countries: Mexico, the United States and/or Canada. The Panel erred by conflating “product of North America” origin information with “country of origin” information for the purposes of its overall finding, and therefore failed to objectively assess the matter before it in accordance with Article 11 of the DSU. This error in assessing the evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed above and below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

102. An objective assessment of this evidence would result in the conclusion that consumers have no greater willingness to pay for specific country of origin information in comparison to the much broader, non-country-specific origin information on the “product of North America” label. Thus, in terms of willingness to pay, this evidence indicates that consumers are indifferent to country-specific origin information, suggesting that there is little or no apparent demand for country of origin information relative to more general origin information.

(D) The Panel Failed to Find that the “Small” Economic Benefits of receiving the COOL Information Indicates that the Demand for COOL Information is “Small”

103. The Panel held that the “benefits accruing to consumers from receiving origin information may also be a determinant of consumer demand for such information.”121 The Appellate Body then acknowledged that the USDA itself “concluded that ‘the economic benefits will be small and will accrue mainly to those consumers who desire country of origin information.’”122 Further, the Appellate Body recognized that the USDA subsequently found that the “incremental economic benefits of [the amended COOL measure] will be

119 Panel Reports, US – COOL (Article 21.5), para. 7.416. 120 Panel Reports, US – COOL (Article 21.5), para. 7.410. 121 Panel Reports, US – COOL (Article 21.5), para. 7.419. 122 Panel Reports, US – COOL (Article 21.5), para. 7.420.


32

comparatively small relative to those afforded by the [original] COOL requirements.”123 An objective assessment of this evidence would find that because the economic benefits of the amended COOL measure for U.S. consumers are “small”, consumer demand for such information is “small”. The Panel failed to make this conclusion, however, focusing instead on the isolated statement that the “expected benefits … are difficult to quantify.”124

104. The Panel erred in failing to objectively assess the evidence on its merits, declining to make a finding for the reason that “even the USDA found it difficult to quantify the continued ‘small’ consumer benefits under both the original and the amended COOL measures”.125 The Panel therefore failed to objectively assess the matter before it in accordance with Article 11 of the DSU. This error in assessing the evidence was material to the assessment of the risks that non-fulfilment of the legitimate objective would create, specifically with respect to measuring the gravity of the consequences that would arise from non-fulfilment. Taken singly or together with the other evidentiary errors of the Panel discussed above and below, this error undermines the objectivity of the Panel’s assessment of the matter before it.

(E) An Objective Assessment of the Evidence on the Record Establishes that the Gravity of the Consequences arising from Non-Fulfilment are Very Low

105. For the foregoing reasons, Mexico requests that the Appellate Body overturn the Panel’s findings, re-assess the evidence regarding consumer demand for origin information on the covered products in the U.S. market, and make a finding on the gravity of the consequences that would arise from non-fulfilment of the legitimate objective of providing consumer information on origin for the covered products.

106. An objective assessment of the evidence on the record, on its merits, establishes a prima facie case that consumer demand for country of origin information on beef products is very low. Taking into account the very low relative importance of the objective of providing consumer information on origin, very low consumer demand for country of origin information on beef products supports the conclusion that the gravity of the consequences that would arise from a non-fulfilment scenario, in which consumers receive no origin information at all, is very low.

4. The Panel Erred by Not Completing the “Relational Analysis” and Not Drawing a Conclusion as to the “Necessity” of the Trade-Restrictiveness of the Amended COOL Measure

107. As previously discussed, the Panel erred by misinterpreting and misapplying the Appellate Body’s reasoning in US – Tuna II (Mexico) and US – COOL. At no point in either US – Tuna II (Mexico) or the original US – COOL dispute did the Appellate Body find or rule that conclusions regarding Article 2.2 should not be drawn until after both steps of the

123 Panel Reports, US – COOL (Article 21.5), para. 7.420. 124 Panel Reports, US – COOL (Article 21.5), para. 7.420. 125 Panel Reports, US – COOL (Article 21.5), paras. 7.420, 7.422.


33

“necessity” test have been undertaken, including the “comparative analysis” of the challenged measure with possible alternative measures. Such an interpretation would be contrary to the reasons provided by the Appellate Body in US – Tuna II (Mexico) and followed by the Appellate Body in US – COOL, as outlined above. Regardless, a panel’s analysis of a challenged measure under Article 2.2 at first instance is entirely different from the Appellate Body’s analysis of such a decision in the context of an appeal. The latter simply cannot be applied as a model for the purposes of undertaking the former.

108. The Panel should have undertaken a rigorous and complete “relational analysis” in the first step of the Article 2.2 necessity test and rendered a conclusion as to the “necessity” of the trade-restrictiveness of the challenged measure. Its failure to do so amounts to a legal error and, as discussed above, creates a substantial gap in the necessity test under Article 2.2.

5. A Properly Conducted, Complete Relational Analysis Demonstrates that the Amended COOL Measure is More Trade Restrictive than Necessary and an Unnecessary Obstacle to International Trade

109. In conducting the “relational analysis” in the first step of the necessity test, the Panel must weigh and balance all of the relevant factors in relation to one another in a holistic process. The relevant factors can be summarized as follows:

• Disregarding the Panel’s separate error in taking into account the coverage of ground beef with Label E in assessing the degree of contribution, the Panel found that the amended COOL measure makes a “considerable but necessarily partial” degree of contribution to the objective of providing consumer information on origin;126

• Correcting the Panel’s separate error regarding Label E, as discussed above, had the Panel excluded the share of beef products sold under Label E from its calculation of the degree of contribution, the coverage of the muscle cut Labels A, B, C and D would have been only 16.3 percent to 24.5 percent of total beef products consumed in the United States. In Mexico’s view, contrary to the finding of the Panel, a degree of contribution below one quarter is not a “considerable” contribution to the objective.

• The amended COOL measure has increased the “considerable degree of trade-restrictiveness” found by the Appellate Body in respect of the original COOL measure in the original US – COOL dispute;127 and

• The nature of the risks at issue and the consequences that may arise from non-fulfilment of the objective, that is to say, the risks non-fulfilment would create, are very low. In the alternative, the Panel itself found that there is “some risk associated with the non-fulfilment of the amended COOL measure’s legitimate objective.”128 In Mexico’s view, “some risk” would not

126 Panel Reports, US – COOL (Article 21.5), paras. 7.356, 7.611. 127 Panel Reports, US – COOL (Article 21.5), paras. 7.370, 7.611. 128 Panel Reports, US – COOL (Article 21.5), para. 7.417.


34

change the outcome of the weighing and balancing process of the “relational analysis”, considering the highly disproportionate weight of the very considerable trade-restrictiveness of the measure.

110. When weighed and balanced in a holistic manner, it is clear from these factors that the very considerable trade-restrictiveness of the amended COOL measure is entirely disproportionate to the insignificant or trivial risks that non-fulfilment would create. As the gravity of the consequences that would arise from non-fulfilment is very low, the risks non-fulfilment would create are insufficient to justify the measure’s very considerable trade-restrictiveness. The fact that the amended COOL measure makes a “considerable but necessarily partial” contribution to the fulfilment of the objective is not sufficient to outweigh the profoundly disproportional relationship between the other relevant factors.

111. Accordingly, the trade-restrictiveness of the amended COOL measure is clearly unnecessary, with the effect of creating an unnecessary obstacle to international trade. On the basis of the “relational analysis” in the first step of the Article 2.2 necessity test, Mexico has thus established a prima facie case that the amended COOL measure is inconsistent with Article 2.2 of the TBT Agreement.

C. The amended COOL measure is More Trade Restrictive than Necessary under the “Comparative Analysis” in the Second Step of the Necessity Test

112. The comparative analysis forming the second step of the two-part “necessity test” under Article 2.2 of the TBT Agreement involves the “weighing and balancing” of a number of factors relating to possible alternative measures that may be reasonably available to the responding Member as a means of achieving the objective pursued by the impugned measure.129 The Appellate Body has clarified that this exercise “will involve a comparison of the trade-restrictiveness of, and the degree of achievement of the objective by, the measure at issue, with that of possible alternative measures that may be reasonably available and that are less trade restrictive than the challenged measure, taking account of the risks non-fulfilment would create.”130 Accordingly, where an alternative measure is proposed which satisfies these criteria, namely one that is less trade restrictive than the impugned measure, achieves an equivalent contribution to the relevant objective taking account of the risks non-fulfilment would create, and is reasonably available, the impugned measure will have been shown to be more trade restrictive than necessary, in contravention of Article 2.2 of the TBT Agreement.

113. Mexico proposed four alternative measures that fulfil these criteria. However, and as addressed below, the Panel erred in its interpretation and application of key components of the “relational analysis,” which in turn had the effect of compromising its ability to engage in a fulsome comparative analysis of the amended COOL measure and Mexico’s first and second alternative measures. Furthermore, the Panel erred in its interpretation and application of the standard of proof in relation to the “adequate identification” of Mexico’s

129 Appellate Body Reports, US – COOL, footnote 745 to para. 374 (referring to Appellate Body Report, US – Tuna II (Mexico), footnote 643 to para. 318 (footnotes omitted)). 130 Appellate Body Reports, US – COOL, para. 376 (citing Appellate Body Report, US – Tuna II (Mexico), para. 320 (footnote omitted) (emphasis original).


35

third and fourth alternative measures, resulting in a flawed assessment of their respective reasonable availability.

1. The Panel’s Failure to Make any Finding Regarding the Gravity of the Consequences that would Arise from Non-fulfilment Rendered Unworkable the Entire “Comparative Analysis” in the Second Step of the Necessity Test

114. As explained above, the Panel incorrectly found that, based on the evidence before it, it was “unable to ascertain” the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective.131 Mexico has described how this finding prevented the Panel from properly drawing conclusions on the complainants’ Article 2.2 claims following the completion of the “relational analysis” in the first step of the necessity test.132 The Panel’s failure to determine the gravity of the consequences of non-fulfilment also compromised the “comparative analysis” in the second step, where it prevented the Panel from comparing the degrees of contribution between the challenged measure and Mexico’s first and second proposed alternative measures.

115. As a consequence of its failure to ascertain the gravity of the consequences that would arise from non-fulfilment of the relevant objective, the Panel held that it was unable to take into account “the specific implications of risks of non-fulfilment” in order to resolve the comparisons of the different degrees of contribution between the amended COOL measure and the first and second proposed alternative measures.133 This effectively ended the comparative analyses of Mexico’s first and second alternative measures before any conclusions could be drawn, wrongly depriving Mexico of a full comparative analysis under the second step of the Article 2.2 necessity test.

116. As Mexico has explained, a proper analysis of the evidence before the Panel leads to the conclusion that the gravity of the consequences that would arise from non-fulfilment of the objective of the amended COOL measure’s objective can be ascertained, and that it is very low. As such, the Panel ought to have completed its comparative analysis of Mexico’s first and second alternative measures.

2. The Correct “Equivalent Contribution” Analysis

117. As confirmed by the Appellate Body, the comparative analysis to be carried out under the second step of the Article 2.2 “necessity test” is, as a whole, informed by the assessment of the “risks non-fulfilment would create” conducted pursuant to the relational analysis. In US – Tuna II (Mexico), the Appellate Body characterized the comparative analysis in the following way:

…the comparison of the challenged measure with a possible alternative measure should be made in the light of the nature of the risks at issue and

131 Panel Reports, US – COOL (Article 21.5), para. 7.423. 132 Panel Reports, US – COOL (Article 21.5), para. 7.424. 133 See Panel Reports, US – COOL (Article 21.5), paras. 7.488-7,491; 7.501-7.503.


36

the gravity of the consequences that would arise from non-fulfilment of the legitimate objective.134

118. The assessment of whether a proposed alternative measure makes an “equivalent contribution” to the achievement of the relevant legitimate objective is central to this comparative analysis. Within the context of this proceeding, the Panel identified the legitimate objective of the amended COOL measure as the provision of origin information to consumers.135

119. The Appellate Body in the original proceeding described the three elements comprising the comparative analysis under Article 2.2 of the TBT Agreement in the following terms:

…it will be relevant to consider whether the proposed alternative is less trade restrictive; whether it would make an equivalent contribution to the relevant legitimate objective, taking account of the risks non-fulfilment would create; and whether it is reasonably available.136

120. This articulation of the “equivalent contribution” requirement makes explicit the relevance of ‘the risks that non-fulfilment of the objective would create’ to the analysis. The evaluation of whether a proposed alternative measure makes an “equivalent contribution” can only be properly carried out where consideration is given to the risks that non-fulfilment would create. The evaluation of whether a proposed alternative measure makes an “equivalent contribution” can only be properly carried out where consideration is given to the risks that non-fulfilment would create.

121. The term “equivalent contribution’ derives from the jurisprudence concerning Article XX of the GATT 1994 and Article XIV of the GATS. Neither of these provisions contains the phrase “taking account the risks non-fulfilment would create.” The phrase “equivalent contribution to the relevant legitimate objective, taking account of the risks non-fulfilment would create” must therefore be interpreted as having a meaning distinct from a mere “equivalent contribution.”

122. Consequently, in the context of this proceeding, the Panel was required to assess the contribution of Mexico’s proposed alternative measures to the consumer information objective, as qualified by a consideration of the risks non-fulfilment of that objective would create. It is only by undertaking this analysis that each alternative measure’s contribution to the consumer information objective can be assessed as “equivalent” to the amended COOL measure’s contribution, or not.

123. The Panel appears to have acknowledged the validity of this approach. Commenting on the Appellate Body’s analysis in US – Tuna II (Mexico),137 the Panel correctly stated that:

134 Appellate Body Report, US – Tuna II (Mexico), para. 321. 135 Panel Reports, US – COOL (Article 21.5), para. 7.332. 136 Appellate Body Reports, US – COOL, para. 471 (citing Appellate Body Report, US – Tuna II (Mexico), para. 322). 137 In US – Tuna II (Mexico), the Appellate Body explained at paragraph 377 the relevance of the phrase “the risks non-fulfilment would create” in the context of a comparative analysis under Article 2.2 as follows:

Footnote continued on next page


37

This suggests that “the risks non-fulfilment would create” may be a relevant factor in assessing whether an alternative measure fulfils the legitimate objective to an equivalent degree as the challenged measure.138

124. To the extent that any of Mexico’s proposed alternative measures make a somewhat lesser contribution to the consumer information objective, they should nonetheless be assessed as fulfilling the “equivalent contribution” requirement where the risks of non-fulfilment associated with the lesser contribution are insignificant so as not to outweigh the alternative measure’s less trade restrictive effect. This approach is consistent with the Appellate Body’s articulations of the “equivalent contribution” requirement in the context of Article 2.2 of the TBT Agreement.

125. While the Panel appears to have agreed with Mexico’s legal arguments to this effect, the Panel’s incorrect finding that it was unable to ascertain the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective prevented it from completing the “equivalent contribution” analysis in respect of Mexico’s first and second alternative measures.

3. Mexico’s Proposed Alternative Measures Demonstrate that the Amended COOL Measure is More Trade Restrictive than Necessary

a. Mexico’s First Alternative Measure

(i) Mexico’s First Alternative Measure Achieves an Equivalent Degree of Contribution in Light of the Risks of Non-Fulfilment of the Relevant Objective

126. As its first alternative measure, Mexico proposed a combination of mandatory labelling using the substantial transformation rule (which would provide consumers with information related to the slaughter production step) and voluntary labelling of more specific origin information (including information on the birth and raising production steps). This alternative measure also proposed to remove the three exemptions established in the original COOL measure and maintained by the amended COOL measure.139 Finally, this alternative measure would not affect Labels D and E, which would remain in effect.

Footnote continued from previous page

[T]he obligation to consider ‘the risks non-fulfilment would create’ suggests that the comparison of the challenged measure with a possible alternative measure should be made in the light of the nature of the risks at issue and the gravity of the consequences that would arise from non-fulfilment of the legitimate objective. This suggests a further element of weighing and balancing in the determination of whether the trade-restrictiveness of a technical regulation is ‘necessary’ or, alternatively, whether a possible alternative measure, which is less trade restrictive, would make an equivalent contribution to the relevant legitimate objective, taking account of the risks non-fulfilment would create, and would be reasonably available.

138 Panel Reports, US – COOL (Article 21.5), para. 7.487. 139 Mexico’s first written submission, US – COOL (Article 21.5), para. 182.


38

127. Pursuant to this alternative, consumers who place value on knowing in which country certain steps of production took place when making purchasing decisions would have the option of seeking out (and paying a higher price for) beef products that voluntarily provide additional information on the birth and raising production steps.

128. While the voluntary nature of the provision of origin information pertaining to the birth and raising production steps might be said to result in “less” origin information being provided to consumers, Mexico argued before the compliance Panel that market mechanisms would match the degree of labelling specificity to the consumer demand for such information.140 The United States acknowledged that “evidence suggests that market mechanisms could ensure that the optimal level of country of origin information would be provided to the degree valued by consumers.”141 Accordingly, the trade-restrictiveness of Mexico’s first alternative measure is adjusted and is proportionate to the very low or insignificant risks non-fulfilment would create. Furthermore, Mexico’s first alternative removes the three exemptions maintained by the amended COOL measure, being the exemptions for processed food items, beef sold in food service establishments, and beef sold by entities not meeting the definition of a “retailer”. This is significant. As found by the Panel, these exemptions covered between 57.7% and 66.7% of total beef consumption.142 In this sense, by virtue of its significantly broader coverage, this alternative measure makes a greater contribution to the provision of origin information to consumers than the amended COOL measure.

129. In assessing Mexico’s first alternative measure, the Panel determined that it would potentially provide consumers with less detailed origin information, but that this information would be provided in relation to a broader subset of beef products, which might achieve an equivalent degree of contribution as the amended COOL measure. The Panel observed that, in some circumstances, less origin information might reduce consumer confusion or misinformation. For Mexico’s first alternative, the Panel found that it had difficulty in establishing the exact implications of providing less information on origin, although it recognized that the coverage of products was substantially wider than under the amended COOL measure.143 Specifically, the Panel stated the following:

Given the potential relevance of risks of non-fulfilment in comparing degrees of contribution, we consider that providing less origin information to consumers for a significantly wider range of products through a measure like the complainants’ first alternative measure might achieve an equivalent degree of contribution as the amended COOL measure.144

130. In doing so, the Panel seemed to recognize the explicit link between an alternative measure’s degree of contribution to the relevant legitimate objective and the risks of non-fulfilment, giving credence to Mexico’s argument that an alternative measure may satisfy the requirements of the comparative analysis under Article 2.2 where it provides a lesser contribution to the legitimate objective than the impugned measure, but where the risks of

140 Mexico’s first written submission, US – COOL (Article 21.5), para. 185. 141 2013 Final Rule, p. 31377 (Exhibit MEX-3); 2009 Final Rule, p. 2682 (Exhibit MEX-12). 142 Panel Reports, US – COOL (Article 21.5), para. 7.258. 143 Panel Reports, US – COOL (Article 21.5), para. 7.489. 144 Panel Reports, US – COOL (Article 21.5), para. 7.488.


39

non-fulfilment associated with this lesser contribution are very low so as not to outweigh the less trade restrictive effect of the alternative measure.145

131. However, and as discussed above, the Panel proceeded to find that it was unable to determine whether this is the case for either the first or second alternative measures proposed by Mexico on the basis that it was unable to “ascertain the gravity of the consequences of not fulfilling the objective of providing consumer information on origin”146 and that, accordingly, it could not determine the specific implications of risks of non-fulfilment for the interplay between less information and more extensive coverage, or for either alternative measure’s degree of contribution to the relevant objective.147 As Mexico has explained, the Panel’s finding that it was unable to ascertain the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective constituted an error.

132. By improperly failing to ascertain the gravity of the consequences that would arise from non-fulfilment of the amended COOL measure’s objective, the Panel precluded itself from completing a fulsome comparative analysis under the second step of the Article 2.2 necessity test, thereby compounding the adverse effect of its error on the relational analysis falling under the first step of the “necessity” test. Had the Panel correctly completed the analysis of the gravity of the consequences that would arise from the non-fulfilment of the amended COOL measure’s objective, the Panel would likewise have proceeded to correctly complete the assessment of the degree of contribution of Mexico’s first alternative measure.

133. Mexico’s first alternative measure achieves a degree of contribution to the consumer information objective that is at least equivalent to the amended COOL measure. By eliminating the exemptions for processed food items and beef sold in food service establishments and by retailers, the mandatory labelling requirement based on the substantial transformation rule has a significantly more broad scope of application than the amended COOL measure, over double the coverage.148 In short, information on origin (based on the substantial transformation rule) is provided to consumers in respect of a much wider range of beef products.

134. With respect to the voluntary component of the measure, the Panel found that “[o]n the whole, the evidence does not suggest that the voluntary option would be exercised on a wide scale”,149 and that consequently the voluntary component “would provide less information on origin than the amended COOL measure for muscle cuts derived from US-slaughtered livestock covered by Labels A-C.”150 While this may be the case, it is a result of insufficient consumer demand for more detailed information on origin – it is in no way a function of the design or structure of the measure. The design of Mexico’s first alternative measure does not preclude or limit the provision of information on origin to consumers at a level equivalent to the amended COOL measure. To the contrary, given sufficient consumer

145 See Mexico’s first written submissions, paras. 185-187. 146 Panel Reports, US – COOL (Article 21.5), para. 7.501. 147 Panel Reports, US – COOL (Article 21.5), para. 7.488 and para. 7.501. 148 As found by the Panel at para. 7.258 of its Reports, between 57.7% and 66.7% of beef consumption was excluded by the exemptions. 149 Panel Reports, US – COOL (Article 21.5), para. 7.478. 150 Panel Reports, US – COOL (Article 21.5), para. 7.479.


40

demand for such information, detailed origin information could be available to consumers pursuant to this alternative measure for 100% of the covered beef products. In light of the elimination of the amended COOL measure’s three exemptions, Mexico’s first alternative measure is designed so as to make possible the achievement of an even greater contribution to the consumer information objective than the amended COOL measure.

135. By implementing a two-pronged labelling system, consisting of mandatory labelling based on the substantial transformation rule and the provision of more detailed origin information on a voluntary basis, Mexico’s first alternative measure in fact replicates the consumer demand for such information. As explained above, the gravity of the consequences, as indicated by consumer demand, constitutes the measure of the severity of the risk of non-fulfilment – that is, the severity of the effect of non-fulfilment of the consumer information objective. For these reasons, the very low gravity, or “not grave” consequences flowing from non-fulfilment translate into a very low risk created by non-fulfilment.

136. Accordingly, to the extent that Mexico’s first alternative measure provides “less” information on origin, this is a function of low consumer demand for that information. Given the very low risk created by the non-fulfilment of the consumer information objective, as borne out by the low consumer demand for such information, this alternative measure is in fact directly proportionate to the risks non-fulfilment would create. Should consumer preference for detailed information on the origin of beef products increase, market forces would shift to increase the provision of such information on a voluntary basis. As such, Mexico’s first alternative measure makes an equivalent contribution to the amended COOL measure’s objective, in light of the gravity of the consequences of non-fulfilment and the very low risks non-fulfilment would create.

(ii) Mexico’s First Alternative Measure is Less Trade Restrictive

137. Mexico’s first alternative measure, consisting of mandatory labelling of the country of origin of beef products based on a substantial transformation rule, combined with voluntary labelling of beef products with more specific information regarding the born, raised and slaughtered production steps, is not trade restrictive. By employing the substantial transformation rule as the basis for mandatory country of origin labelling, the country of origin of a given beef product will be the country in which slaughter took place. Consequently, the significant segregation costs that result from the amended COOL measure’s labelling requirements are eliminated pursuant to this alternative measure. As noted by the Appellate Body in the original proceeding, such an alternative measure would impose no “restriction or limitation” on imported livestock; all meat products derived from cattle slaughtered in the U.S. would be labelled as “Product of US.”151

138. With regard to the voluntary labelling of beef products with more specific information pertaining to where an animal was born, raised and slaughtered, it is clear that no trade restrictive effect is entailed. The provision of this additional information to consumers would be carried out on an entirely voluntary basis, permitting market forces to recognize and

151 Appellate Body Reports, US – COOL, para. 485.


41

cater to consumer demand for such information. In short, there is no governmental measure that would have the effect of restricting trade.

139. As explained above, the Panel’s incorrect finding that it was unable to ascertain the gravity of the consequences associated with the risks of non-fulfilment of the amended COOL measure’s objective had the effect of preventing it from completing its assessment of whether Mexico’s first alternative measure made an equivalent contribution to the relevant objective. Having found that Mexico failed to make a prima facie case that the first alternative measure would make an equivalent contribution to the consumer information objective, the Panel declined to engage in an analysis of its trade-restrictiveness.152 In the Panel’s view, the “conjunctive relationship” between the three elements of the comparative analysis permitted it to end its analysis without considering the trade-restrictiveness or reasonable availability of Mexico’s first alternative measure.153

140. The Panel’s failure to engage in an assessment of trade-restrictiveness is a direct result of its incorrect finding that it could not ascertain the risks of non-fulfilment, and its consequent inability to assess the first alternative measure’s degree of contribution to the relevant objective. Mexico has addressed these errors in detail. A correct application of the comparative analysis in these circumstances leads to the conclusion that Mexico’s first alternative measure achieves at least an equivalent contribution to the consumer information objective, in light of the risks of non-fulfilment. Consequently, an assessment of this alternative measure’s trade-restrictiveness is required.

141. As described above, Mexico’s first alternative measure is not trade restrictive. The use of the substantial transformation rule as the basis for mandatory country of origin labelling eliminates the need for segregation, which is a principal source of the amended COOL measure’s trade-restrictiveness. Furthermore, the voluntary provision of additional detail with respect to the location of the born, raised and slaughtered production steps entails no trade restrictive effects. Accordingly, Mexico’s first alternative measure is less trade restrictive than the amended COOL measure.

(iii) Mexico’s First Alternative Measure is Reasonably Available

142. The last factor to be considered pursuant to the comparative analysis is whether Mexico’s first alternative measure is reasonably available to the U.S. The Appellate Body has stated that, in the context of Article XX of the GATT 1994 and Article XIV of the GATS, “[a]n alternative measure may be found not to be ‘reasonably available’ … where it is merely theoretical in nature, for instance, where the responding Member is not capable of taking it, or where the measure imposes an undue burden on that Member, such as prohibitive costs or substantial technical difficulties.”154 In the Panel’s view, this commentary was instructive for the purposes of determining whether a technical regulation is “more trade –restrictive than necessary” under the TBT Agreement.155

152 Panel Reports, US – COOL (Article 21.5), para. 7.490. 153 Panel Reports, US – COOL (Article 21.5), para. 7.491. 154 Appellate Body Report, Brazil – Retreaded Tyres, para. 156 (quoting Appellate Body Report, US – Gambling, para. 308). 155 Panel Reports, US – COOL (Article 21.5), footnote 1232.


42

143. Mexico’s first alternative measure involves three principal components: the mandatory labelling of beef products using the substantial transformation rule, the implementation of voluntary labelling of more specific point-of-production origin information, and the elimination of the three exemptions maintained by the amended COOL measure.

144. With respect to the mandatory labelling component based on the substantial transformation rule, it is clear that this element of the alternative measure is reasonably available to the United States. Prior to the implementation of the original COOL measure, the USDA permitted meat products to be labelled “Product of US” if they were substantially transformed (i.e., if the animals were slaughtered) within the United States.156 As such, Mexico’s first alternative measure would change this prior rule only by requiring that the country of origin be always listed on the label. This requirement cannot be said to impose an “undue burden” on the United States.

145. Having regard to the voluntary labelling component, Mexico has identified successful and long-standing voluntary labelling programs currently in existence in the United States. For example, the Certified Angus Beef program, which has been active since 1978, and the voluntary USDA Grade Labels both have long histories of successful use.157 Furthermore, the voluntary component of Mexico’s first alternative measure involves no substantial obstacles to implementation, nor would it impose any costs greater than the amended COOL measure.

146. The reasonable availability of this alternative is most clearly demonstrated by the fact that the amended COOL measure currently employs a similar approach to origin labelling for imported muscle cuts of beef. Pursuant to Section 65.300(f)(2), muscle cut covered commodities derived from an animal that was slaughtered in another country retain its origin, as declared to U.S. Customs and Border Protection at the time of entry into the United States, through retail sale. This includes animals born and/or raised in the United States and slaughtered in another country.158 These requirements are supplemented by a voluntary component, according to which “the origin declaration may include more specific location information related to production steps (i.e., born, raised and slaughtered)”.159 Accordingly, the amended COOL measure itself contains labelling requirements consisting of both mandatory and voluntary components. The existence of such a provision in the amended COOL measure confirms that Mexico’s first alternative measure is reasonably available to the United States.

156 USDA Food Safety and Inspection Service, Product Labeling: Defining United States Cattle and United States Fresh Beef Products, Advance Notice of Proposed Rulemaking, 66 Fed. Reg. 41160-41161 (August 7, 2001) (Exhibit MEX-47). 157 See discussion in U.S. Answers to Panel’s First Set of Questions, October 4, 2010, Response to Question 43, paras. 77-78. 158 Section 65.300(f)(2) of the 2009 Final Rule, as amended by the 2013 Final Rule (Exhibit MEX-3). 159 Section 65.300(f)(2) of the 2009 Final Rule, as amended by the 2013 Final Rule (Exhibit MEX-3).


43

b. Mexico’s Second Alternative Measure

(i) Mexico’s Second Alternative Measure Achieves an Equivalent Degree of Contribution in Light of the Risks of Non-Fulfilment of the Relevant Objective

147. As its second alternative measure, Mexico proposed to extend the ground beef labelling rules to muscle cuts from US-slaughtered animals, while again eliminating the three exemptions maintained by the amended COOL measure.160 Pursuant to the existing labelling requirements for ground beef, the origin label must list all countries of origin contained therein, or that may be reasonably contained therein.161 When a raw material from a specific country of origin is not in a processor’s inventory for more than 60 days, that country will no longer be included as a possible country of origin. This flexibility is referred to as the 60-day inventory allowance.

148. In its analysis of the implications of the extension of the ground beef 60-day inventory allowance, the Panel determined that “the second alternative measure could thus result in muscle cuts from US-slaughtered animals born or raised in different countries carrying the same label, which could possibly be affixed on muscle cuts that do not originate in at least one of the countries shown on the label.”162 As was the case with Mexico’s first alternative measure, the Panel determined that Mexico’s second alternative measure could potentially provide consumers with “less” information on origin, but for a significantly wider range of products.163 Specifically, with respect to Mexico’s second alternative measure, the Panel found that:

…we consider that by providing less accurate origin information to consumers for a significantly wider range of products, the second alternative measure might achieve an equivalent degree of contribution as the amended COOL measure.164

149. In doing so, the Panel acknowledged that, despite providing consumers with “less” information on origin (in the sense of less accurate information), it was possible that Mexico’s second alternative measure could nonetheless achieve an equivalent degree of contribution to the consumer information objective as the amended COOL measure, taking into account the risks non-fulfilment would create.165

150. However, and as explained above, the Panel’s finding that it was unable to “ascertain the gravity of the consequences of not fulfilling the objective of providing consumer information on origin”166 led to the conclusion that it could not determine the specific

160 Mexico’s first written submission, para. 192. 161 2009 Final Rule, CFR§65.300(h) (Exhibit MEX-3) 162 Panel Reports, US – COOL (Article 21.5), para. 7.500 (footnote omitted). 163 Panel Reports, US – COOL (Article 21.5), para. 7.501. 164 Panel Reports, US – COOL (Article 21.5), para. 7.501 (emphasis original). 165 Panel Reports, US – COOL (Article 21.5), para. 7.488 and para. 7.501. 166 Panel Reports, US – COOL (Article 21.5), para. 7.501.


44

implications of risks of non-fulfilment for the interplay between less information and more extensive coverage, or for the second alternative measure’s degree of contribution to the relevant objective.167 Had the Panel correctly completed the analysis of the gravity of the consequences that would arise from the non-fulfilment of the amended COOL measure’s objective, the Panel would then have proceeded to correctly complete the comparative assessment of the degree of contribution of Mexico’s second alternative measure.

151. In designing both the original and amended COOL measure, the United States determined that the more flexible labelling rules for ground beef satisfied the consumer information objective of the amended COOL measure for those products. In the absence of any legitimate and objective basis upon which to distinguish between ground beef and other beef products in this context, these same rules can be used to satisfy the amended COOL measure’s consumer information objective for muscle cuts of beef and processed food items.

152. To the extent that the amended COOL measure’s ground beef labelling rules could be argued to provide less origin information to consumers than the labelling rules presently applied to muscle cuts of beef, Mexico’s second alternative measure might likewise be argued to provide less accurate origin information to consumers than the amended COOL measure. However, as was the case with Mexico’s first alternative measure, the second alternative measure would remove the three exemptions contained in the amended COOL measure. The result is, once again, a much broader scope of application: all beef products would be covered by this alternative measure.

153. Furthermore, the Panel found that ground beef comprises a substantial proportion of covered beef commodities.168 It must therefore be concluded that the United States was satisfied, for the purposes of achieving the consumer information objective pursued by the amended COOL measure, to have a substantial proportion of the beef products purchased by U.S. consumers subject to a labelling requirement that provides this “less accurate” information on origin. There is no legitimate justification for applying a higher standard of accuracy to muscle cuts of beef.

154. An additional factor to be considered is that in reaching its conclusion that Mexico’s second alternative measure potentially provided “less accurate” information, the Panel stated that “the second alternative measure as described by the complainants does not seem capable of making an actual contribution to the objective of providing consumer information on origin at least equivalent to the actual contribution of the amended COOL measure.”169 In reaching this conclusion, the Panel failed to address Mexico’s point that the Label E rules are actually already part of the amended COOL measure, and therefore should be considered to fulfil the U.S. objective of providing consumer information on origin. Instead, the Panel addressed the Category E ground beef labelling rules as though they were part of a separate measure, or did not exist at all. The Panel did not provide any reason as to why the objective of providing consumers with origin information on the cattle used to produce ground beef is different than the objective of providing origin information on cattle used to produce muscle cuts – especially when meat from the same animal may be used for both types of products. Moreover, as discussed above, in evaluating the amended

167 Panel Reports, US – COOL (Article 21.5), para. 7.488 and para. 7.501. 168 Panel Reports, US – COOL (Article 21.5), para. 7.258. 169 Panel Reports, US – COOL (Article 21.5), para. 7.500.


45

COOL measure’s contribution to fulfilling the objective, the Panel gave credit to the Category E rules for increasing the percentage of meat products it deemed covered by the amended COOL measure. This is inconsistent; if the Category E rules are considered to fulfil the consumer information objective in the context of the assessment of the amended COOL measure’s contribution to achieving this objective, they must also be treated as fulfilling the consumer information objective when considered as an alternative to the point-of-production labelling scheme under the amended COOL measure.

155. As explained above, Mexico has established that the risk associated with non-fulfilment is very low in light of the low gravity of consequences. Pursuant to the amended COOL measure, the consumer information objective is fulfilled for ground beef using the 60-day inventory allowance system. In the absence of any basis for distinguishing between ground beef and other beef products, this same system must be judged to fulfil the consumer information objective for muscle cuts of beef, particularly in light of the very low risks associated with the non-fulfilment of the objective: consumers are not demanding more accurate origin information for ground beef.

156. Like Mexico’s first alternative measure, and as explained below, the second alternative measure is also less trade restrictive than the amended COOL measure. The less trade restrictive nature of this alternative measure, when weighed against its contribution to the consumer information objective, has the result of striking a proportionate balance between trade restrictive effects and the very low risks non-fulfilment of the objective would create. By providing country of origin information for a significantly more broad range of beef products through the elimination of the three exemptions maintained by the amended COOL measure. Mexico’s second alternative measure yields an equivalent contribution to the consumer information objective in light of the low risks associated with the non-fulfilment of this objective.

(ii) Mexico’s Second Alternative Measure is Less Trade Restrictive

157. As was the case with Mexico’s first alternative measure, the Panel’s incorrect finding that it was unable to ascertain the gravity of the consequences associated with the risks of non-fulfilment of the amended COOL measure’s objective had the effect of preventing it from completing its assessment of whether Mexico’s second alternative measure made an equivalent contribution to the relevant objective. As a result, the Panel declined to continue the comparative analysis and engage in an assessment of the second alternative measure’s trade-restrictiveness.170

158. For the reasons described above, this constituted an error on the Panel’s part. A correct application of the comparative analysis leads to the conclusion that Mexico has demonstrated that its second alternative measure makes an equivalent contribution to the consumer information objective, taking account of the risks non-fulfilment would create. Accordingly, an assessment of this alternative measure’s trade-restrictiveness is required as the next step in the comparative analysis.

159. As explained, Mexico’s second alternative measure extends the mandatory country of origin labelling rule applicable to ground beef under the amended COOL measure to muscle



46

cuts of beef and to processed food items made of beef. The ground beef labelling rules provide a 60-day “inventory allowance”, whereby the country of origin of a given beef product can be listed as any country from which the processor has had raw materials in its inventory in the past 60 days. The extension of this 60-day inventory allowance to muscle cuts and processed food items provides significant flexibility to producers and meat processors, and reduces the degree to which processors would be required to segregate beef products.

160. This was affirmed by the panel in the original proceeding, which found that “the 60-day ‘inventory allowance’ provides for significant flexibility with respect to the segregation entailed by the ground meat label under the COOL measure”.171 The panel further found that the inventory allowance “limits any additional costs of implementing the COOL measure with regard to ground meat.”172 Accordingly, Mexico’s second alternative measure would have the effect of limiting the costs associated with segregation for an even broader subset of beef products, thereby further reducing the incentive to discriminate against imported beef products.

(iii) Mexico’s Second Alternative Measure is Reasonably Available

161. Mexico’s second alternative measure involves the extension of the amended COOL measure’s ground beef labelling requirements, which currently apply to approximately half of covered beef products, to muscle cuts of beef and processed beef products. Given that the ground beef labelling rules are designed to provide processors with a 60-day inventory allowance, as described above, these requirements are both more flexible and more easily implemented than the current labelling requirements for muscle cuts.

162. The Panel found that the flexibility inherent in the 60-day inventory allowance has the result of reducing the segregation entailed by the ground beef label, and “thus limits any additional costs of implementing the COOL measure with regard to ground meat.”173 Accordingly, an extension of this labelling system to other beef products would have the effect of similarly limiting the need for the segregation, and consequently limiting the attendant implementation costs, for a wider range of beef products.

163. Given the fact that this labelling system already forms part of the amended COOL measure, it is reasonably available in respect of ground beef products. Because there is no legitimate basis upon which to distinguish between ground beef and other beef products in this context, there is no reason why these labelling requirements would be reasonably available in respect of ground beef, but not reasonably available in respect of muscle cuts of beef. The United States’ own use of this labelling system in the amended COOL measure is the clearest possible indication that Mexico’s second alternative measure is reasonably available.

164. In light of the increased flexibility inherent in the 60-day inventory allowance, the associated reduction in segregation and the attendant costs, and the fact that the United

171 Panel Reports, US – COOL, para. 7.435. 172 Panel Reports, US – COOL, para. 7.435. 173 Panel Reports, US – COOL, para. 7.435.


47

States already applies this labelling system to a large proportion of covered beef products, it is evident that Mexico’s second alternative measure is reasonably available to the United States.

c. Mexico’s Third and Fourth Alternative Measures

165. Throughout its comparative analysis of Mexico’s third and fourth alternative measures, the Panel erred in its interpretation and application of the correct standard of proof required to “adequately identify” a proposed alternative measure. In both cases, this error resulted in the Panel’s failure to complete the comparative analysis under the second step of the Article 2.2 necessity test.

(i) Description of Third Alternative Measure

166. Mexico’s third alternative measure applies a mandatory trace-back system to muscle cuts from U.S.-slaughtered animals. As described by the Panel, “[t]he essence of the third alternative measure would be to preserve the link between the animal (or group of animals) and the resulting meat by requiring that a retailer be able to trace a muscle cut piece of meat back to the original animal.”174 This alternative measure establishes a system by which an individual animal can be “followed” during all stages of its life, from the time it was born through the time it was slaughtered. As such, Mexico’s proposed trace-back system would provide precise and detailed information with respect to each of the born, raised and slaughtered productions steps.

167. In compliance with U.S. animal health measures, each Mexican animal bears a blue metallic ear tag which identifies the origin of the cattle and information about the producer and exporter. This ear tag information is linked to the Export Zoosanitary Certificate of Animals” issued by Mexico’s Ministry of Agriculture, Livestock, Rural Development, Fisheries and Food. The Export Zoosanitary Certificate provides information as to the age, sex, breed and colour of the cattle, as well as its State of origin and the ranch to which the cattle belongs.175 This existing ear tag mechanism employed by the Mexican cattle industry is one of the available methods by which to implement the proposed trace-back system, as it already provides all of the information required to fulfil recordkeeping and verification requirements from the point of birth of the animal to the point of importation into the United States.

168. As explained by Mexico, the trace-back system would apply the same tracing requirements equally to both domestic and imported animals. Consequently, the same costs are associated with the use of both domestic and imported animals in the production of beef products. As a result, the provenance of an animal would become irrelevant to U.S. beef processors’ costs, eliminating any economic incentive for excluding imported Mexican cattle.

169. In support of its position on the third alternative, Mexico submitted evidence on traceback programs already implemented in Japan, Uruguay and elsewhere.176

174 Panel Reports, US – COOL (Article 21.5), para. 7.505 (footnote omitted). 175 Mexico’s first written submission, para. 91. 176 See, e.g., Exhibits MEX-41, 42, 43, and 44.


48

(ii) Description of Fourth Alternative Measure

170. As its fourth alternative measure, Mexico proposed mandatory labelling of muscle cuts from U.S.-slaughtered animals indicating the place of birth, raising and slaughter according to states and/or provinces, in addition to the country of origin designations imposed by the amended COOL measure.

171. Specifically, this measure would require that covered beef products bear labels identifying the state(s) and/or province(s) in which each of the three production steps occurred. Mexico explained that the United States could rely on records kept pursuant to the 2013 Final Rule on Traceability for Livestock Moving Interstate, or could adopt a mandatory national identification and traceability system as a means of satisfying the associated recordkeeping requirements.177

172. This state and/or province-based trace-back system would impose the same requirements on both domestic and imported livestock, and would therefore avoid the lowest cost compliance problem created by the amended COOL measure. As was the case with Mexico’s third alternative measure, this measure would eliminate any incentive to exclude imported Mexican cattle or shift the cost of compliance solely to Mexican animals.

(iii) Standard of Proof for Identification of Alternative Measures

173. Although the “adequate identification” of a proposed alternative measure is a precondition to the meaningful assessment of its degree of contribution to the relevant objective, reasonable availability and trade-restrictiveness, the Panel incorrectly required the complainants to meet a standard of proof requiring that precise and detailed explanations be provided with respect to the third and fourth alternative measures’ implementation and the costs associated with implementation. In short, the Panel erred by requiring the complainants to provide unnecessarily precise and complete explanations of how the third and fourth alternative measures would be implemented in the United States and precise estimates of the associated costs in order to “adequately identify” the alternative measures for the purposes of the comparative analysis.

174. In this regard, the Panel considered that “adequate identification of alternative measures requires more precision than the sometimes vague and in some respects incomplete description of the implementation and ultimate magnitude of the associate [sic] costs provided by the complainants in this case.”178 The Panel found that the complaints’ explanations and arguments were insufficient to meet its standard, resulting in a “lack of adequate identification”179 of the third and fourth alternative measures that prevented the Panel from assessing the “required elements” under the legal test.180 On this basis, and as explained below, the Panel concluded that the complainants had failed to establish a prima

177 Mexico’s response to Panel’s Question 71, para. 150. 178 Panel Reports, US – COOL (Article 21.5), para. 7.556. 179 Panel Reports, US – COOL (Article 21.5), para. 7.564. 180 Panel Reports, US – COOL (Article 21.5), para. 7.564 (regarding the third alternative measure) and 7.609-7.610 (regarding the fourth alternative measure).


49

facie case that the alternative measures were “reasonably available”,181 would be “less trade restrictive”,182 and would make at least an equivalent contribution to the relevant legitimate objective.183

175. With respect to the implementation of Mexico’s third and fourth alternative measures, the Panel found that “adequately explaining how an alternative measure would be implemented is an essential part of the complainants’ burden to identify an alternative measure that is not limited to a ‘concept[]’ but entails an ‘actual, concrete proposal[]’.”184 The Panel also held that “[h]aving a basic understanding of how an alternative measure would be implemented is a prerequisite for comparing this alternative with the challenged measure.”185 In doing so, the Panel effectively imposed a new and unjustifiable requirement on the complainants to explain exactly how a proposed alternative measure would be implemented by the respondent, thereby fundamentally misconstruing the standard of proof to be met in adequately identifying the measure. The Panel’s approach was neither correct nor reasonable.

176. Having regard to the costs associated with the implementation of Mexico’s third and fourth alternative measures, the Panel held that the costs of an alternative measure may be relevant to assessing whether the complainants have met their burden of “adequately identifying” the measure. In support, the Panel cited the Appellate Body in EC – Seal Products for the proposition that the issue of costs may be relevant to the question whether an alternative measure is “reasonably available” under Article 2.2.186

177. In fact, there is no authority for the Panel’s determination that precise and complete cost estimates are a prerequisite to the “adequate identification” of an alternative measure. Though the Panel cited reasons provided by the Appellate Body in EC – Seal Products, it

181 Panel Reports, US – COOL (Article 21.5), paras. 7.554-7.557 and 7.602. 182 Panel Reports, US – COOL (Article 21.5), paras. 7.558-7.560 and 7.602. 183 Panel Reports, US – COOL (Article 21.5), paras. 7.561-7.563 and 7.602. 184 Panel Reports, US – COOL (Article 21.5), para. 7.586 and footnote 1285, citing the United States’ comment on Canada’s response to Panel question No. 74. See also Panel Reports, US – COOL (Article 21.5), para. 7.506 (“An alternative measure may be found not to be reasonably available, for example ‘where it is merely theoretical in nature’. Likewise, adequate identification of the alternative is essential for determining the alternative measure’s trade-restrictiveness and degree of contribution in comparison with the challenged measure. We therefore consider the complainants’ explanations of the implementation of trace-back by the United States to be of central importance to their burden of proof and our examination of the complainants’ third alternative measure” (italics emphasis original)). 185 Panel Reports, US – COOL (Article 21.5), para. 7.586. However, the words “basic understanding” are inconsistent with the detailed and precise explanations that the Panel actually required from the complainants. See e.g., Panel Reports, US – COOL (Article 21.5), para. 7.556. 186 Panel Report, US – COOL (Article 21.5), footnote 1129 to para. 7.509, citing Appellate Body Report, EC – Seal Products, para. 5.277 (“As we see it, if there are reasons why the prospect of imposing an alternative measure faces significant, even prohibitive, obstacles, it may be that such a measure cannot be considered ‘reasonably available’. We would not exclude a priori the possibility that an alternative measure may be deemed not reasonably available due to significant costs or difficulties faced by the affected industry, in particular where such costs or difficulties could affect the ability or willingness of the industry to comply with the requirements of that measure. We therefore consider that an assessment of the reasonable availability of an alternative measure could potentially include the burden on the industries concerned”).


50

misinterpreted and misapplied this jurisprudence. In the context of Article XX of the GATT 1994, the Appellate Body stated in EC – Seal Products:

In Brazil – Retreaded Tyres, the Appellate Body stated as follows:

As the Appellate Body indicated in US – Gambling, “[a]n alternative measure may be found not to be ‘reasonably available’ ... where it is merely theoretical in nature, for instance, where the responding Member is not capable of taking it, or where the measure imposes an undue burden on that Member, such as prohibitive costs or substantial technical difficulties” [para. 156, quoting Appellate Body Reports, US – Gambling, para. 308 (in the context of Article XIV of the GATS].

This passage suggests that the “prohibitive costs or substantial technical difficulties” are indeed those associated with the burden placed on a Member. At the same time, however, this language does not foreclose the possibility that there may be other indications that the alternative measure is “merely theoretical in nature”. As we see it, if there are reasons why the prospect of imposing an alternative measure faces significant, even prohibitive, obstacles, it may be that such a measure cannot be considered “reasonably available”. We would not exclude a priori the possibility that an alternative measure may be deemed not reasonably available due to significant costs or difficulties faced by the affected industry, in particular where such costs or difficulties could affect the ability or willingness of the industry to comply with the requirements of that measure. We therefore consider that an assessment of the reasonable availability of an alternative measure could potentially include the burden on the industries concerned.187

178. The language used by the Appellate Body above refers to evidence and arguments that the respondent would raise in its rebuttal – that is, to demonstrate “significant, even prohibitive obstacles” facing the implementation of a possible alternative measure, such as “significant costs or difficulties faced by the affected industry” ‒ rather than evidence and explanations that would be adduced by the complainant in identifying a possible alternative measure in the first instance. Consequently, the Appellate Body’s reasons do not support the Panel’s suggestion that such “obstacles”, “significant costs”, or “difficulties” should be presumed or inferred where the complainant has not explained precisely how an alternative measure would be implemented or provided a complete estimate of the magnitude of the associated costs. Similarly, where the United States failed to articulate such obstacles, costs or difficulties in its own submissions, its mere allegation that the complainants have failed to meet their burden of proof is not sufficient to satisfy its own burden of rebuttal.

179. In fact, in EC – Seal Products, the Appellate Body was critical of the panel’s finding that the complainants had not clearly defined the separate components of their proposed alternative measure or the cumulative capacity of those components to contribute to the achievement of the relevant objective. The Appellate Body’s concern is a clear indication that a panel should not impose too high a standard of proof in respect of alternative measures:

187 See Panel Report, US – COOL (Article 21.5), para. 7.555, quoting Appellate Body Report, EC – Seal Products, paras. 7.276-7.277.


51

Finally, we note the Panel’s finding that, although the contribution of the EU Seal Regime to the fulfilment of its objective is lowered by the implicit and explicit exceptions of the measure, “the complainants have not clearly defined an alternative measure in respect of its separate components and their cumulative capability to address the moral concerns of the EU public”. Under the circumstances of this case, it is not clear to us how much more could have been required of the complainants in articulating the nature and scope of the alternative measure. We have certain doubts about the Panel’s criticism, particularly given what seems to be substantial engagement between the Panel and the parties regarding the features of such a certification scheme and its ability to contribute to the objective. In any event, however, our concerns are particular to the circumstances of this case, and do not alter our assessment of the Panel’s overall finding.188

180. By requiring precise explanations and complete cost estimates with respect to the implementation of proposed alternative measures in the foreign jurisdiction of the responding Member for the purposes of “adequate identification”, the Panel required a standard of proof that was impossible for the complainants to satisfy. As a consequence, the Panel effectively relieved the United States of its burden to adduce evidence and arguments in rebuttal. It should not be sufficient for the United States to merely argue that the complainants have failed to meet their burden because they have not set out an “actual, concrete proposal” or exact explanations regarding the implementation of the alternative measure. Rather, as discussed previously, the United States was obligated to demonstrate specific obstacles, costs, or difficulties facing the implementation of each proposed alternative measure.

181. It is neither reasonable nor appropriate to require the complainants to adduce detailed evidence or precise explanations regarding exactly how a proposed alternative measure will be implemented in the foreign jurisdiction of the responding Member, by the responding Member, or a complete estimate of the associated costs. The only Member who is in a position to comment meaningfully on the specific issues of implementation and the associated costs – including any significant obstacles or difficulties – is the responding Member itself. This is expressly recognized in the Appellate Body’s description of the shifting burden of proof under Article 2.2 in US – Tuna II (Mexico) and US – COOL, where the respondent bears the burden of rebutting a prima facie case by “presenting evidence and arguments showing that the challenged measure is not more trade restrictive than necessary … by demonstrating, for example, that the alternative measure identified by the complainant is not, in fact, ‘reasonably available’, is not less trade restrictive, or does not make an equivalent contribution to the achievement of the relevant legitimate objective.”189

182. The standard of proof in establishing a prima facie case should be attuned accordingly. In the absence of any real and objective doubt as to the capacity of the responding member to implement the alternative measure, it is not reasonable or appropriate to require that the implementation or the costs associated with the measure be described with the precision demanded by the Panel.

188 Appellate Body Report, EC – Seal Products, para. 5.280. 189 See Appellate Body Report, US – Tuna II (Mexico), para. 323; Appellate Body Reports, US – COOL, para. 379.


52

183. Given Mexico’s adequate identification of its third and fourth alternative measures and the establishment of a prima facie case, the Panel ought to have looked to the U.S. as respondent to adduce evidence and arguments demonstrating specific obstacles, cost issues, or other difficulties that would render the alternative measures impractical, inappropriate, or otherwise not reasonably available. The Panel’s interpretation of the appropriate standard of proof required to establish a prima facie case in this context constitutes a clear error with serious systemic implications for the comparative analysis to be carried out under the second step of the Article 2.2 “necessity test.”

184. For these reasons, Mexico requests that the Appellate Body reverse the Panel’s finding that Mexico has not sufficiently and adequately identified the third and fourth alternative measures to enable it to complete the analysis of the measures’ respective reasonable availability, trade-restrictiveness and degree of contribution, as compared against the amended COOL measure. Specifically, Mexico requests that the Appellate Body reverse the Panel’s finding that precise and complete cost estimates are a prerequisite to the “adequate identification” of an alternative measure for the purposes of establishing prima facie that an alternative measure is reasonably available.

IV. ARTICLE 2.1 OF THE TBT AGREEMENT

A. Label E is Relevant to the Issue of Whether the amended COOL measure Lacks Even-handedness or Legitimacy of Regulatory Distinctions under the Article 2.1 Legal Analysis

185. In its Appellant Submission, the United States extensively appeals the Panel’s findings and conclusions under Article 2.1 of the TBT Agreement. It is Mexico’s view that the Panel was correct in its conclusion that the amended COOL measure is inconsistent with the United States’ obligations under Article 2.1. Mexico’s Appellee Submission will respond in detail to the arguments advanced by the United States. In this Other Appeal, Mexico challenges a discrete finding made by the Panel regarding the relevance of Label E to the Panel’s assessment of the even-handedness of the relevant regulatory distinction under Article 2.1. Mexico raises this ground of other appeal so that the Appellate Body may fully correct the Panel’s legal analysis under Article 2.1.

186. As was the case with its analysis under Article 2.2, the Panel erred in improperly declining to consider the amended COOL measure’s Label E rules in its analysis of Mexico’s claim under Article 2.1. Specifically, the Panel noted that Label E (being the ground meat label) had not been shown to result in less favourable treatment for imported livestock, and that it “does not constitute a relevant regulatory distinction of the amended COOL measure for the purposes of Article 2.1.”190 For these reasons, the Panel refused to recognize Label E as relevant to the analysis carried out under Article 2.1.191

187. This approach is entirely inconsistent with the Panel’s earlier explicit acknowledgement that the Article 2.1 analysis must take into account “the ‘overall architecture’ of the measure, and encompass[es] aspects of the measure that [are] not themselves ‘relevant regulatory distinctions’ or independent sources of detrimental



53

impact.”192 In fact, it was on this basis that the Panel correctly determined that the exemptions maintained by the amended COOL measure were relevant factors to be considered in the course of the Article 2.1 analysis, despite the fact that they did not constitute a relevant regulatory distinction, and were not found to give rise to any detrimental impact.193 In deciding that the amended COOL measure’s exemptions should be considered in the analysis of even-handedness under Article 2.1, the Panel stated that

This incorporation of the exemptions into the Article 2.1 analysis is not at odds with the findings of the original panel that the exemptions do not in themselves “alter the distribution of compliance costs for livestock and meat producers and processors in a way that would modify the incentives created by the COOL measure. […] While they do not modify incentives to the detriment of imported livestock, they are relevant to the distinct question of whether that detriment stems exclusively from legitimate regulatory distinctions.194

188. The Panel further stated that, “although the exemptions under the amended COOL measure are not ‘relevant regulatory distinctions’ as such, we nevertheless take them into account as part of our examination of the ‘overall architecture’ of the amended COOL measure, insofar as they are relevant to whether ‘the detrimental impact reflects discrimination in violation of Article 2.1.’”195

189. These very same statements hold true for the amended COOL measure’s Label E rules. As a central component of the measure, and one that applies to a substantial proportion of covered beef products, the Label E rules are relevant to the question of whether the detrimental impact that results from the amended COOL measure “stems exclusively from legitimate regulatory distinctions.”196 In short, the Panel on the one hand recognized that aspects of the measure that neither constitute a relevant regulatory distinction nor give rise to the detrimental impact are relevant to the Article 2.1 analysis (i.e. the amended COOL measure’s exemptions), and on the other hand refused to consider Label E in its Article 2.1 analysis because it neither constituted a relevant regulatory distinction, nor was it found to give rise to any detrimental impact. This is contradictory and, for the reasons that follow, led to an incomplete and partial assessment of the discriminatory effect of the amended COOL measure.

190. The Panel was correct in acknowledging that the appropriate approach to the evaluation of a claim pursuant to Article 2.1 requires not only the identification of the detrimental impact and relevant regulatory distinctions, but also a “broad appraisal of a measure’s design and application,” which includes careful scrutiny of such factors as the measure’s architecture, revealing structure and operation.197 It is well-established that a regulatory distinction that is not designed and applied in an even-handed manner cannot be considered legitimate, and will therefore constitute discrimination within the meaning of

192 Panel Reports, US – COOL (Article 21.5), para. 7.202. 193 Panel Reports, US – COOL (Article 21.5), para. 7.202. 194 Panel Reports, US – COOL (Article 21.5), footnote 480 to para. 7.202. 195 Panel Reports, US – COOL (Article 21.5), para. 7.203. 196 Panel Reports, US – COOL (Article 21.5), footnote 480 to para. 7.202. 197 Panel Reports, US – COOL (Article 21.5), paras. 7.202 and 7.203.


54

Article 2.1.198 It is only by undertaking this “broad appraisal” of the measure’s design and application that its even-handedness (or the degree to which it lacks even-handedness) can be properly and fully assessed.

191. The Panel attempted to validate its exclusion of Label E from the assessment of the amended COOL measure’s even-handedness on the basis that “the production of ground meat entails the processing of ‘trimmings’ of diverse origin that are ground into a final products, and the ground meat labelling rules were adapted to the purchasing, inventory, and production practices of US beef grinders,” as distinguishable from the production processes applied to other covered beef products.199 This is an untenable premise, as this reasoning deliberately ignores the fact that different muscle cuts of beef themselves undergo distinct processing methods, as do processed beef products and beef products served in food service establishments. Accordingly, the processing steps that make up the production process for ground meat cannot in and of themselves constitute a legitimate justification for excluding Label E from the analysis under Article 2.1.

192. While it is true that that “the complainants do not refute the different forms of processing undergone by muscle cuts and ground meat,”200 the relevance of the various elements comprising the amended COOL measure to the assessment of its even-handedness is in no way dictated by or dependent upon the production process associated with the various covered commodities. This is made evident in the fact that muscle cuts of beef, processed beef products and products served in food service establishments were all considered by the Panel to be relevant for the purposes of the Article 2.1 analysis, despite the fact that they all involve distinct production processes.

193. The Panel also makes reference to the Appellate Body’s decision in the original dispute, which it interprets as implying that the Appellate Body specifically excluded Label E from consideration in its analysis under Article 2.1.201 This is not the case. At no point in its reasons did the Appellate Body indicate that the COOL measure’s Label E rules had no relevance to the assessment of the measure’s even-handedness. The passage relied upon by the Panel in this respect describes the discriminatory effect of the measure as arising “because the prescribed labels do not expressly identify specific production steps and, in particular for Labels B and C, contain confusing or inaccurate origin information, or because the meat or meat products are exempt from the labelling requirements altogether.”202 This finding cannot properly be interpreted as explicitly determining that the Label E rules are irrelevant for the purposes of the evaluation of the even-handedness of the amended COOL measure.

194. As explained, the amended COOL measure’s Label E rules are an integral component of the measure’s design and architecture, given that they establish the labelling requirements applicable to ground beef – which accounts for approximately one half of all covered beef products. The “broad appraisal” of the amended COOL measure’s design and application which is required in order to correctly assess its even-handedness cannot be

198 Appellate Body Reports, US – COOL, para. 271. 199 Panel Reports, US – COOL (Article 21.5), para. 7.280. 200 Panel Reports, US – COOL (Article 21.5), para. 7.280. 201 Panel Reports, US – COOL (Article 21.5), para. 7.280. 202 Appellate Body Reports, US – COOL, para. 349.


55

carried out if the labelling rules that govern approximately half of the covered beef products are excluded from consideration. For the same reasons that the Panel determined the amended COOL measure’s exemptions to be relevant to the inquiry into whether the detrimental impact stems exclusively from a legitimate regulatory distinction, the labelling requirements imposed upon Category E beef products are directly relevant to an assessment of the even-handedness (or in this case, lack thereof) of the amended COOL measure as a whole.

195. The labelling requirements applied to ground beef pursuant to Label E allow a processor to reference a country of origin on its label even if the processor has not had ground meat from that particular country in its inventory for the last 60 days or less. The flexibility built into this 60-day inventory allowance is not replicated in the labelling rules that apply to muscle cuts of beef, thereby creating a significant imbalance in the accuracy of the origin information provided to consumers of the two forms of beef products. This imbalance begs the question of why this lower standard of accuracy is acceptable for beef that happens to be ground, but is not acceptable for muscle cuts of beef, particularly in light of the fact that meat from the same animal may be used in both. The reality is that the United States has provided no explanation that could serve to legitimize the amended COOL measure’s arbitrarily imbalanced standards of accuracy. The absence of any legitimate justification for this asymmetry is a clear indication of a lack of even-handedness and, consequently, it indicates arbitrariness in the amended COOL measure.

196. Like the exemptions maintained by the amended COOL measure, the Label E rules properly constitute a “key element” of the evaluation of the legitimacy of that measure’s regulatory distinctions.203 By improperly failing to recognize the relevance of Label E as an integral component of the design and architecture of the amended COOL measure, the Panel engaged in an incomplete and partial assessment of the measure’s even-handedness. This in turn resulted in a flawed and incomplete assessment of the degree to which the amended COOL measure discriminates against imported livestock contrary to Article 2.1.

197. A correct assessment of the amended COOL measure’s even-handedness must necessarily include consideration of its ground beef labelling rules as a component of the measure’s overall design and architecture, thereby providing further evidence of the measure’s lack of even-handedness, arbitrariness and the discriminatory nature of its detrimental impact.

V. CONDITIONAL APPEAL CONCERNING ARTICLE XXIII:1(B) OF THE GATT 1994

198. Before the Panel, Mexico presented a claim under Article XXIII:1(b) of the GATT 1994 that the amended COOL measure nullifies and impairs tariff concessions made by the United States and inscribed in its WTO tariff bindings. The Panel declined to rule on that claim, exercising judicial economy on the basis that compliance by the United States with the Panel’s finding of violation under Article III:4 of the GATT 1994 would remove the basis of the non-violation claims of nullification or impairment.

199. Should the Appellate Body conclude that the Panel erred in its findings of violation under Article III:4 of the GATT 1994 and Article 2.1 of the TBT Agreement and further



56

determine that the amended COOL measure is not inconsistent with these provisions, Mexico appeals the Panel’s decision to exercise judicial economy in respect of Mexico’s claim under Article XXIII:1(b) of the GATT 1994, and requests that the Appellate Body complete the analysis of this claim and find that the amended COOL measure is inconsistent with Article XXII:1(b).

200. In relation to the principle of judicial economy, the Appellate Body has cautioned that “panels may refrain from ruling on every claim as long as it does not lead to a ‘partial resolution of the matter’.” In this regard, the Appellate Body has found that “[a] panel has to address those claims on which a finding is necessary in order to enable the DSB to make sufficiently precise recommendations and rulings ... ‘in order to secure effective resolution of disputes to the benefit of all Members.’”

A. Approach of the Panel and Reasons for Exercise Judicial Economy of the Panel

201. In its review of the non-violation nullification and impairment claim brought by Mexico, the Panel recognized that

[t]he text of Article XXIII:1(b) of the GATT 1994 clearly stipulates that a claim under that provision may arise whether or not the contested measure conflicts with the GATT 1994. It is therefore possible for a measure to simultaneously violate a provision of the GATT 1994 or another covered agreement, and still give rise to a cause of action under Article XXIII:1(b).204 (emphasis original)

202. The Panel engaged in an analysis of whether exercising judicial economy in relation with this claim would be appropriate for this particular dispute. The Panel determined that the findings in this Article 21.5 proceeding differ from the findings in the original panel for two main reasons: (i) the finding of violation under Article 2.1 of the TBT Agreement reflects the fact that Article 2.1 of the TBT Agreement does not prohibit all detrimental modifications of the conditions of competition, but only those that do not stem exclusively from a "legitimate regulatory distinction"; and (ii) the finding of violation under Article III.4 of the GATT 1994 implies that effective equality of opportunities for imported products to compete with like domestic products was upset.

203. Accordingly, the Panel compared the scope of the findings under Article III.4 and XXIII:1(b). The Panel concluded that the “compliance by the United States with [the] finding of violation of Article III:4 would require eliminating the detrimental impact on competitive opportunities for Canadian and Mexican livestock altogether. Hence, restoring ‘effective equality’ between foreign and domestic products in this regard would necessarily remove the basis of nullification or impairment claimed under Article XXIII:1(b).”205 On that basis, the Panel determined to exercise judicial economy with respect to the complainants' non violation claims.

204. Despite the decision to exercise judicial economy, the Panel proceeded to analyse the complainants' non-violation claims in the eventuality “were the Appellate Body to



57

disagree with our findings on violation or with our approach to judicial economy, it will have the benefit of our factual findings under Article XXIII:1(b).”206 The Panel characterized this examination as “conditional and primarily factual.”207 The Panel made several conditional factual findings that support the conclusion that the amended COOL measure causes nullification or impairment in the sense of Article XXIII:1(b) of the GATT 1994.

205. The Panel erred in exercising judicial economy; particularly considering that the Appellate Body could reverse the Panel’s findings regarding Article 2.1 of the TBT Agreement and Article III.4 of the GATT 1994.

B. There are Sufficient Factual Findings to Conclude that the Amended COOL Measure Causes Nullification or Impairment within the meaning of Article XXIII:1(b) of the GATT 1994

206. The Panel made sufficient factual findings to support the conclusion that the amended COOL measure is contrary to Article XXIII:1(b) of the GATT 1994. The Panel correctly recognized the three-part test followed by Mexico in support of its claim was correct. The Panel also made the following factual findings:

• It was uncontested that the amended COOL Measure is being applied by the United States;208

• The United States’ WTO tariff concessions for cattle are within the meaning of the term "benefit accruing" in the sense of Article XXIII:1(b) of the GATT and Article 26.1 of the DSU.209 The Panel clarified that “it would be an impermissibly narrow reading of "benefit accruing" in Article XXIII:1(b) of the GATT 1994 and Article 26.1 of the DSU to limit that term to concessions that are actually enjoyed or applied to trade in the relevant goods.”

• There was no compelling evidence to confirm that the amended COOL measure could have been reasonably anticipated before its adoption. Additionally, the Panel stated that the amended COOL measure introduced a significant degree of regulatory novelty to the labelling of the products relevant in this dispute.210

207. In light of the Panel’s conditional factual findings, the Appellate Body has sufficient facts to complete the analysis and find that the amended COOL measure results in nullification or impairment that is inconsistent with Article XXIII:1(b).

206 Panel Reports, US – COOL (Article 21.5), para. 7.672. 207 Panel Reports, US – COOL (Article 21.5), para. 7.672. 208 Panel Reports, US – COOL (Article 21.5), para. 7.675. 209 Panel Reports, US – COOL (Article 21.5), para. 7.690. 210 Panel Reports, US – COOL (Article 21.5), para. 7.712.


58

VI. CONCLUSIONS

208. On the basis of the foregoing, Mexico respectfully requests that the Appellate Body:

a. Find that the Panel erred in its interpretation and application of Article 2.2 of the TBT Agreement for the reasons set out above;

b. Find that the Panel erred in finding that Mexico did not make a prima facie case that the amended COOL measure is more trade restrictive than necessary within the meaning of Article 2.2 of the TBT Agreement;

c. Overturn the Panel’s findings and conclusions under Article 2.2 of TBT Agreement, complete the analysis under Article 2.2 of the TBT Agreement, and make a finding that the amended COOL measure is more trade restrictive than necessary and has the effect of creating unnecessary obstacles to international trade in violation of Article 2.2 of the TBT Agreement;

d. Find that the Panel erred in paragraph 7.207 of its Report by determining that Label E (the ground meat label) is not relevant to the legal analysis under Article 2.1 of the TBT Agreement; and

e. In the event that the Appellate Body determines that the Panel erred in its findings of violation under Article III:4 of the GATT 1994 and Article 2.1 of the TBT Agreement, and further determines that the amended COOL measure is not inconsistent with these provisions, Mexico appeals the Panel’s decision to exercise judicial economy in respect of Mexico’s claim under Article XXIII:1(b) of the GATT 1994, and requests that the Appellate Body finds that the amended COOL measure is inconsistent with Article XXII:1(b).

united states – certain country of origin labelling ... · from india, wt/ds436/ab/r, circulated...

Documents