Ruling in asbestos case sows seeds of many future disputes

by Chakravarthi Raghavan

Geneva, 20 Mar 2001 - - An Appellate Body (AB) ruling in the Canada-EC dispute over a French ban on use of asbestos fibres and cement products with asbestos has perhaps widened the scope for using ‘consumer tastes’ in the ‘like product’ theory of competitivity under GATT.

The AB also upheld the ruling on the action of the French decree in banning the use, except in a few cases where no substitutes exist without risk of fires.

Of the three members of the division bench, one (not identified) has dissented from the wide meaning given for assessing  ‘likeness’ and the criteria of using ‘fundamentally’ and ‘exclusively’ economic views about ‘likeness’ and the probable difficulties it could cause in future.

If creating certainty in trade law is an important element of trade security, the ruling, particularly its 2-1 reasoning in a matter that was probably not necessary to give a finding, would perhaps result in creating a degree of uncertainty - possibly far from the intention of the AB.

The ruling, as of some others, makes clear that the WTO members must find a way of using their powers unambiguously reserved powers, to providing authoritative interpretations, or a way of separating final rulings in individual disputes from a generally applicable trade law for the future.

This is a far too serious an issue to be left to new rounds or the future. In this particular case, the asbestos ruling is such that it will easily command considerable civil society support. When applied to something else in future, it may not, jeopardising the WTO itself.

In the ruling, the AB reversed the panel finding that the French decree banning the imports and  marketing of asbestos fibres and products was not a technical regulation, and thus not covered by the WTO’s Technical Barriers to Trade (TBT) agreement.

The TBT agreement (Art. 1:2) defines a technical regulation as one that lays down product characteristics or their related processes and production methods, including applicable administrative provisions, with which compliance is mandatory.

The AB has taken the view that ‘product characteristiques’ could be presented in a positive or negative way (in this case, ‘shall not contain asbestos fibres’) and be made applicable to an ‘identifiable product’. There was nothing in the TBT, according to the AB, requiring the technical regulations to name or otherwise expressly identify a product or products.

Given the history of the TBT negotiations in the Uruguay Round - when the wording placing  ‘product’ and ‘process and production methods’ (PPMs) on an equal footing, capable of being regulated by technical standards, was slipped into the text during negotiations when some of the developing-country participants were not present, and was presented as a take-it-or-leave-it at the final stage - the expanded view of the TBT regulation now brought up by the AB is likely to create problems for developing countries in the future.

The PPM concept was brought in thus to satisfy the protectionist lobbies (in chemical and other industries) supported by some environment groups. With the expanded meaning to ‘product characteristique’ given by the AB, and when applied to PPMs - defining something in a positive or negative way and without naming or expressly identifying them - would give considerable leeway to the importing industrialized countries.

Since the TBT does not have the kind of language against ‘protectionist’ use found in other agreements and rules, a technical regulation banning or regulating may or may not stand scrutiny before panels in a dispute - but with the AB widening its remit by constantly talking of a case-by-case approach, even a successful challenge to a TBT may take considerable time and money, with the competing product losing the market.

Having reversed the panel’s findings on the French decree being a TBT, the AB notes that the panel had not examined the facts of Canada’s four claims, and that the meaning of different obligations under the TBT had not been the subject of any interpretation or application by either panels or the AB, nor in the old GATT under the Tokyo Round code. The AB also says that given the ‘novel’ character of the Canadian claim, the issues had not been explored in depth before the panel; the panel had not addressed the claims; and there were no ‘issues of law’ or ‘legal interpretation’ which alone could be agitated before the AB.

The AB ends its views on the TBT, after having seven pages devoted to it, with these words: “We also observe that the sufficiency of the facts on the record depends on the reach of the provisions of the TBT Agreement claimed to apply - a reach that has yet to be determined. With this particular collection of circumstances in mind, we consider that we do not have an adequate basis properly to examine Canada’s claims under Art.2.1, 2.2, 2.3, 2.4 and 2.8 of the TBT and, accordingly, we refrain from so doing.”

Considering that the Canada vs EC dispute has been resolved in upholding the French, and the EC bans as justified on ‘health’ grounds, one wonders why the AB did not practice its preaching (and sometimes, rebukes to panels) of the doctrine of judicial economy and why it did not say it was not going into this TBT issue.

Surely, they have done so in other appeals. And surely (as made out on their behalf to the DSB for making their jobs full-time or expanding the numbers), they are already overworked. So, why invite more?

The report has also dealt with the issue of ‘Like Products’ in Art.  III:4 of the GATT, devoting considerable space and words on products with asbestos fibres or ‘like’ products with ‘PCG fibres’ - with the outcome of further ‘expanding’ the reach of ‘like’ products, but leaving scope for case-by-case determinations in future.

(The term PCG is the acronym used in the report for polyvinyl alcohol or pva, cellulose and glass fibres.)

The AB (the report shows that two members favoured this approach, while the otherwise unidentified third member of the division (in paras 149-154), saw no need to go into such a expansive view of III:4 and reserved his position for the future. The division consisted of Florentino Feliciano, James Bacchus and Claus-Dieter Ehlermann. But the minority concurrant (in the final conclusion) is not identified (unlike the normal practice of dissenters and the minority being identified in any final judicial body pronouncing judgements). And given that the AB, via rules of procedure, has enabled each division to consult (without any authority from the DSU articles) the other members of the AB (who did not hear the appeal), were the others consulted and acquiesced in this ‘collegial view’?

The entire ‘like’ product question in Art. III (National Treatment on Internal Taxation and Regulations) goes back to the original London preparatory meeting, and even before that to the US-UK discussions and drafts, and the approach (cloaked in generalities about ‘free trade’) of the US concept (right from the Churchill-Roosevelt Atlantic Charter declaration and freedoms) of reciprocal exchange of tariff concessions for market access. The intention laid out in the London and Geneva meetings before Havana was to ensure that the UK and other Europeans under US pressure do not lower their tariffs, only to use other border taxes or local taxes to discriminate between the imported and domestic production.

Paragraph 1 of Art.III (of GATT) lays down that internal charges, taxes, regulations and requirements affecting internal sale, purchase, transportation, distribution or use of products, and internal “quantitative regulations” requiring mixture, processing or use of products in specified amounts or preparations (what later came to be known as the local content requirement) should not be applied to imported or domestic product so as to afford protection to domestic production.

Paragraph III:2, in two sentences, deals with internal taxes, and makes a specific reference to the principles set forth in paragraph 1.   Paragraph 5, dealing with the ‘local content’ issue also has specific reference to paragraph 1. None of the other paras refers to para 1 - though para 3 refers to para 2, and para 6 to para 5.

It was much later that the remit of the ‘like product’ theory was gradually expanded (even the working party on border tax adjustments did not go that far), with ‘apples’ and ‘oranges’ viewed as ‘like’ products, and carried much farther in the various disputes on spirits, and beer and wine, as like products in taxation cases, to strike down country differentiations of imported and domestic (native spirits) to create the concept of global level playing fields for competition to prevent protection.

But the AB has now in effect imported the general principles in paragraph 1 into paragraph 4, but retaining for the AB the freedom to differ on individual cases later on (with one member voicing doubts).

The panel that heard the dispute, examined and came to the view that chrysotile asbestos fibres and polyvinyl alcohol (PVA) fibres were ‘like products’, and that cement-based products containing chrysotile asbestos fibres were ‘like’ cement-based products containing one of the PCG fibres, and hence all these cement based products were ‘like’. In reaching this conclusion, the panel adopted, in the words of the AB, “an approach based on the Report of the Working Party on Border Tax adjustments”, which employed four criteria to judge ‘likeness’ - (i) properties, nature and quality of the products, (ii) the end-uses of the products,  (iii) consumers tastes and habits, and (iv) tariff classification.

The AB cites an earlier AB ruling (given in interpreting Art.III:2) that there could be no precise and absolute definition of ‘like’ and that the concept of ‘likeness’ is relative, and is like an accordion whose width in any one of the places (in the different WTO agreements where ‘like’ is used) is determined by the particular provision. The stretching of the width of the accordion to different agreements is now extended to different products and different cases that may come up under Art. III:4.

The AB argues that the ‘dictionary’ definition of ‘like’ does not indicate from whose perspective the ‘likeness’ is to be judged, and notes that “for instance, ultimate consumers may have a view about the ‘likeness’ of two products that is very different from that of inventors or producers of those products.”

Given the recent EU court ruling (against Sweden) about alcohol advertisements (can’t be banned on market principles), and the way WTO rulings mix obligations under GATT, GATS, TRIPS etc, current market theories, and product-differentiation among consumers shaped by mass advertising, where would this concept of ‘likeness’ lead into in international trade theory and practice, and convert the explicit ceding of sovereignty and commitments to implicit ones?

The AB notes that the texts of Art.III:2 and III:4 are different, and that this has considerable implications. The AB also notes an earlier ruling of it (over Japan's alcohol beverages case) that the competitive conditions between imported and domestic products in Art.III:1 (general principles) did not protect any particular trade volume, but ‘equal competitive relationship’.

Though Art. III:4 does not contain any reference to the general principles of Art. III:1, specifically mentioned in Art.III:2, the AB acknowledges this, but merely says that these are not “explicitly invoked’ (para 98), and goes on to use them to widen the meaning of Art. III:4.

It argues that ‘like’ in Art III:4 is to be interpreted to apply to products in such a ‘competitive relationship’, and a determination of ‘likeness’ be made by the nature and extent of competitive relationship. At the same time, it says in the next sentence, that there is a “spectrum of degrees” of ‘competitiveness’ or ‘substitutability’ of products in the market place, and that it is difficult, if not impossible, in the abstract to indicate precisely where this spectrum of ‘like’ in Art.III:4 falls. To add to the confusion (for trade policy-makers in countries setting out clearly and transparently, rules and regulations), the AB also says that in ruling on this particular measure, it was not attempting to define the precise scope of ‘like’ in Art.III:4 nor to decide if the precise scope of ‘like products’ is co-extensive with the scope of ‘like’ and ‘directly competitive or substitutable’ products in Art.III:2. However, the AB says it recognizes that the relationship between these two provisions is important because “there is no sharp distinction between fiscal regulation, covered by Art.III:2 and non-fiscal regulation covered by Art III:4.”

The AB recognizes, through its interpretation, that it has now expanded the meaning—“we give that provision a relatively broad product scope, although no broader than the product scope of Art.III:2...”

Clear? One is left wondering why then the drafters in 1947-48, who had a clearer command over English language, made the distinction at all?

The AB report then goes on to examine the ‘likeness’ issue and how ‘a treaty interpreter’ should proceed to determine ‘like products’, and then asserts the view that “an unavoidable element of individual, discretionary judgement’ has to be made on a case-by-case basis.

The AB then cites the four general criteria for ‘likeness’ in the Border Tax adjustment working party, and “followed and developed” since then by panels and the AB - properties, end uses, consumer tastes and tariff classification - and then goes on in effect to expand these further as having ‘characteristics’ that products might share: the physical properties of the products; extent to which they are capable of serving the same or similar end uses; extent to which consumers perceive and treat them as alternative means of performing particular functions to satisfy particular want or demand; and international classification for tariffs.

But then the AB qualifies this all, by claiming that these are simply tools, and “neither a treaty-mandated nor a closed list of criteria....”

And staking out a bigger say for consumer tastes, the AB says: “Consumer perceptions may similarly influence - modify or even render obsolete - traditional uses of products...”

Having thus set out a wide field for future AB interventions, the report then goes into the issue of ‘likeness’ between chrysotile asbestos and PCG fibres, and faults the panel for having relied on ‘end uses’ of the two (for the same cement products) and the references to the ‘small number’ of cases where the products have the same applications and replace each other. On this basis, the panel concluded, from the ‘properties’ criterion, that chrysotile asbestos fibres are like PVA, cellulose and glass fibres.

The panel did not apply the health risk criteria in judging likeness, though later in its reasoning, it upheld the ban on chrysotile fibres on the ground of the health risks exception.

The AB in reversing the panel view, says that the carcinogenicity or toxicity constitute a defining aspect of the physical properties of chrysotile asbestos fibres and hence, the AB could not see how this ‘highly significant physical difference’ could not be a consideration in examining the properties of a product as determining its ‘likeness’.

The AB disagreed with the panel that assessing the health risk, in relation to Art. III:4, nullified the effect of Art.XX (b), which allowed a member to adopt a measure necessary to protect human life or health even if inconsistent with another provision of GATT 1994.

The AB argues that the fact that its interpretation of Art.III:4 would result in a less frequent recourse to Art XX (b) did not deprive the exception of effet utile.

The AB also disagreed with the panel’s assessment of end uses, and said that only by forming a complete picture of the various end uses of different fibres, could a panel assess the significance of products sharing a limited number of end uses.

The AB argues that there would be ‘few’ situations where evidence on ‘likeness’ would lend itself to ‘clear results’ (about likeness) and many cases would give rise to conflicting indications - there may be some similar physical properties and some differing; or completely differing physical properties, yet with similar end uses and a high degree of substitutability from the perspective of the consumer. “A panel could not hence decline to hear evidence simply because it suspects “evidence may not be ‘clear’ or because parties agree that certain evidence may not be relevant. In any event, we have difficulty seeing how the Panel could conclude that an examination of consumers’ tastes and habits ‘would not provide clear results’, given that the panel did not examine any evidence relating to this criterion.”

The AB goes on to note that though consumers tastes and habits could be significant in determining ‘likeness’, at the oral hearing, Canada (the complainant) considered it an irrelevant criterion for the dispute. The AB disagreed with Canada about the irrelevance of this criterion, and said that a member may submit evidence of latent or suppressed consumer demand in that market, or evidence of substitutability from relevant third markets. The AB adds: “We do not wish to be taken to suggest that there is latent demand for chrysotile asbestos fibres. Our point simply is that the existence of the measure does not render consumers’ tastes and habits irrelevant, as Canada contends.”

The panel, the AB notes, decided that chrysotile and PCG fibres to be like products by disregarding their different properties, nature and quality, as well as the different tariff classifications, considered no evidence on consumer tastes, but found that for ‘a small number’ of many applications, the two were substitutable. It did not also consider the many other end uses for the fibre that are different.

For these reasons, the AB set aside the findings about the ‘likeness’ of chrysotile and PCG fibres.

Dealing with the panel view that cement products, containing chrysotile and PCG fibres to be like, the AB complained that the panel had not indicated to what extent the incorporation of one type of fibre for another affected the other physical properties of the cement-based product and hence affected its suitability for a specific end use. Nor did the panel examine the extent to which the presence of a particular fibre affected the ability of the cement-based product performing one or other functions.

And even if they were interchangeable, the AB said, the presence of a known carcinogenic in one would have an influence on consumers’ tastes and habits, and this would be true whether the consumer was a commercial party such as a construction firm or a do-it-yourself or someone who owns and lives or works in a building.

For these reasons, the AB also reversed the finding about likeness of cement products with Chrysotile or PCG fibres.

Basing itself on the available factual evidence on record, the AB analysed the likeness of the two fibres and said that physically, the two were different. And Canada had not proved that despite the physical differences there was a competitive relationship between the two products. While there was some evidence that chrysotile and PCG fibres shared “a small number” of  end-uses, there was no evidence of end uses that were not overlapping. And without this, it was difficult to judge the significance of the two fibres sharing “a small number of similar end uses”. The AB disagreed with Canada over its view that the consumers’ tastes were irrelevant to judge ‘likeness’.

Taken together, Canada had not discharged its burden of providing that chrysotile asbestos fibres and PCG fibres were ‘like’ products. The AB gave a similar finding about the cement-based products incorporating chrysotile or PCG fibres, and hence did not establish the likeness of the two products in terms of Art. III:4.

While agreeing with the overall findings and conclusion, one member in a minority view, notes that the scientific evidence before the panel was very clear about the carcinogenic risk associated with the inhalation of chrysotile fibres. Considering this, the minority member says that there was ample basis for concluding that chrysotile fibres and products with them were not ‘like’ PCG fibres and products with PCG fibres. Canada had not shown that PCG fibres had the same lethal properties.

The minority member has thus noted that there seemed to be no need going into the issues of ‘economic competitive relationships’.Any kind or degree of health risk associated with a particular product would a priori negate a finding of likeness. This suggestion was not a general proposition but restricted to the particular case.

“The necessity or appropriateness of adopting a ‘fundamentally’ economic interpretation of ‘likeness’ of products in Art. III:4 does not appear to me to be free from substantial doubt. In future concrete contexts, the line between ‘fundamentally’ and ‘exclusively’ economic view of like products may well prove to be very difficult, as a practical matter, to identify. It seems to me, the better part of valour to reserve one’s opinion on such an important, indeed, philosophical matter which may have unforeseeable implications...”

The AB disagreed with Canada’s contention that the panel had erred in holding chrysotile fibres and their carcinogenic effects as proving a health risk. The AB agreed that on the basis of the scientific opinion before it, the panel was well within its discretion in finding the products to pose a risk to human life or health.

The AB also turned down Canada’s contention that even if there was a health risk, banning its marketing was not necessary, and that France should not be expected to employ any ‘controlled use’ practices, since it would have prevented France from achieving its chosen level of health protection.

In the result, the AB reversed the panel rulings on the non-applicability of TBT to the particular regulation; reversed the panel view that health risks may not be taken into account in judging likeness; and that the two fibres and products were ‘like’. The panel ruling that the French measure was necessary to protect human life or health was upheld as also the view that the measure could be contested in terms of Art.XXIII:1 (b) -  a seldom evoked provision in the disputes at the GATT/WTO. –SUNS4860

The above article first appeared in the South-North Development Monitor (SUNS) of which Chakravarthi Raghavan is the Chief Editor.

© 2001, SUNS - All rights reserved. May not be reproduced, reprinted or posted to any system or service without specific permission from SUNS. This limitation includes incorporation into a database, distribution via Usenet News, bulletin board systems, mailing lists, print media or broadcast. For information about reproduction or multi-user subscriptions please e-mail < >