In the recent case Argentina – Financial Services the Appellate Body (“AB”) established, for the very first time in the history of the WTO disputes settlement, the examination of the “likeness” under the General Agreement on Trade in Services (“GATS”). In particular, the dispositions at issue were Art. II:1 and Article XVII:1 of the GATS, respectively the Most-Favoured-Nation Treatment, and the National Treatment. Needless to say, the concept of “likeness” under the GATS has been explored by WTO panels, but without successfully achieving consistent interpretations. Will the AB’s reasoning in Argentina – Financial Services clarify the meaning of “likeness” under the GATS?
The alleged measure
Argentina – Financial Services is a dispute brought before the WTO dispute settlement by Panama against a measure issued by Argentina. The alleged measure affects services and service suppliers from countries or territories that do not share tax information with Argentina for purposes of tax transparency, prevention of money laundering and terrorist financing. Accordingly, Argentina made a distinction between services and services suppliers of countries cooperating for tax transparency purposes (“cooperative countries”) and countries not cooperating for tax transparency purposes (“non-cooperative countries”). To be considered as “cooperative”, a country must either (i) sign with Argentina an agreement/convention on exchange of tax or (ii) initiate with Argentina the negotiations necessary for concluding such an agreement/convention.[1]
On the basis of its “non-cooperative” and “cooperative” classification, Argentina laid down eight measures that affect the cross-border financial services trade between Argentine customers and tax payers and Panamanian financial service suppliers.
“Likeness” under the GATS vis-à-vis of “likeness” under the GATT
The AB commenced its analysis by clarifying the meaning of “likeness”- and it must be noted that the AB clarified its meaning by supporting its reasoning, to a certain extent, with previous case law applicable to the trade in goods -. Accordingly, the term “like” must be interpreted in the light of its context, the object and purpose of the agreement in which the relevant provision appears.[2] The term “like” generally means something that shares a number of identical or similar characteristics or qualities.[3] Thus, both in the context of trade in goods and trade in services, “likeness” refers to something being similar and requires to seek for a similar comparator.[4]
The AB noted that both Art. II:1 and Art. XVII:1 refer to Treatment No Less Favourable of like services and services suppliers; further, Art. XVII:3 states that a treatment is considered to be less favourable if it modifies the conditions of competition.
Turning to previous GATS case law, the AB noted that the Panel in China – Electronic Payment Services found that Article XVII is meant to ensure equal competitive opportunities for like services of other WTO Members; services are thus “like” when they are in a competitive relationship. In the aforementioned dispute, the Panel further established that a measure can modify the condition of competition, and therefore being held inconsistent with art. XVII:1, only when foreign and domestic services and suppliers at stake compete with each other.
The AB then turned to analyze “likeness” under the Most-Favoured-Nation Treatment obligation and observed that Art. II:3 refers to the term “advantage”, which is the “state of being in a better position with respect of another one”.[5] It consequently found that the fact of being in a better position as compared to another also relates to the concept of competitive relationship. Also in the context of Article II of the GATS, the analysis of “likeness” as concerning services and services suppliers is centered on the competitive relationship of the latters.[6] Thus, the concept of “likeness” under both Art. II:1 and Art. XVII:1 of the GATS is focused on an assessment of the competitive relationship of services and services suppliers; accordingly, as under the GATT, also under the GATS “likeness” is ultimately a determination about the nature and the extent of competitive relationships;[7] in other words, it is a market-based analysis.[8]
Similarly to the GATT,[9] the AB also found that under the GATS there is a spectrum of degrees of “competitiveness” or “substitutability” of the service and service suppliers. It is nearly impossible to precisely assess, in the abstract, where such spectrum of likeness falls. The likeness of services and services suppliers shall be determined on a case-by-case basis, taking into account the specific circumstances of the specific case.[10]
It is self-evident that “likeness” under the GATS closely resembled “likeness” under the General Agreement on Tariffs and Trade (“GATT”). Nevertheless, we should make a distinction between the “likeness” assessment under the two different agreements. Accordingly, under the GATS the “likeness” analysis entails an examination of the service and service providers, while under the GATT the analysis only focuses on products as such, and producers are therefore not relevant. Thus, the determination of “likeness” under the GATS might entail additional steps as compared to the one under the GATT.
The term “service suppliers”, according to Art. XXVIII (g) of the GATS, means “any person that supplies a service”, while “supplies of a service”, according to Art. XXVIII(b), “includes the production, distribution, marketing, sale and delivery of a service” (the word “service” is not defined in the text of the GATS).[11] In the AB’s view, separate assessments with respect to the “likeness” of services, on the one hand, and the “likeness” of service suppliers, on the other hand, are not required as both Art. II:1 and Art. XVII:1 refer to “service and service suppliers”.[12] Accordingly, the “likeness” assessment must be a holistic exercise, depending on the particular circumstances, that keep into account both service and service suppliers on a case-by-case basis.[13]
The obvious question to be asked is: are there situations where we can make a distinction between, on the one hand, the service itself, and, on the other hand, the service supplier?
Presumption of likeness
In the case at hand, the Panel tried to evaluate Argentina’s distinction in order to rule on the likeness of services and services suppliers. At this regard, its findings are unclear and raise several and serious doubts. The Panel first affirmed that whether a country was included or not in the list of cooperative countries was closely linked to origin.[14] Nevertheless, later, it found that,
“it is not origin per se which determines that certain countries are on the list and others not, but the regulatory framework inextricably linked to such origin.”[15]
Although the Panel had considered that the presumption of “likeness” applied, it later proceeded to undertake a “likeness” assessment. In so doing, the Panel tried to take into account the regulatory aspect of Argentine measure, that is the possibility to have access to tax information provided that it “reflected in the competitive relationship between services and service suppliers of cooperative and non-cooperative countries”.[17] According to the Panel, the burden of proving that “this other factor” affects the competitive relationship between services and service suppliers was upon Argentina.[18] Then, the Panel claimed, incredibly, that it was impossible to compare relevant services and service suppliers for examining “likeness”.[19] It ultimately concluded, without undertaking a “likeness” assessment, that Panama had proved that services and service suppliers of cooperative and non-cooperative countries are like due to origin reasons, omitting to refer to the notion of “based exclusively on origin”.[20] Had the panel recognized –implicitly– that measure was not “exclusively” based on origin?
On appeal, the AB was called to pronounce itself on the Panel’s findings regarding “likeness”. The AB found that the Panel had erred in finding “likeness” by reason of origin for the following reasons. First, the measure at hand did not provide for a distinction based exclusively on origin, so “likeness” could not have been presumed because the measure’s distinction between cooperative and non-cooperative countries was based on certain conditions, namely whether a tax agreement existed or was being negotiated. Second, the Panel did not engage in an analysis of “likeness”.[21]
When there is a distinction based exclusively on origin and, thus, “likeness” can be presumed?
Under the GATT, several panels have established that when a measure distinguishes between products based exclusively on the origin “likeness” can be presumed (the so-called “presumption approach”).[22] Accordingly, under the GATS, such presumption applies when the measure also distinguishes between services and services suppliers based exclusively on the origin. It is to be outlined that, under the GATT, the presumption of “likeness” is typically applied when a measure makes a de jure distinction between products of different origin.[23]
In order to clearly understand the presumption approach, it is relevant to compare the presumption approach in the context of the GATS. Accordingly, it is necessary to deepen the distinction exclusively based on origin. Such distinction has been ruled by panels in several cases, inter alia, China – Publications and Audiovisual Products, Argentina – Hides and Leather, and Argentina – Import Measures.
For instance, the measure at hand in China – Publications and Audiovisual Products, prohibited “foreign-invested enterprises” from engaging in the business of wholesale of imported reading materials, but allowed for wholly Chinese-owned enterprises to engage in the supply of such service.[24] In Argentina – Hides and Leather, the measure made a distinction on whether the products at stake were being imported into Argentine territory or sold within it. The tax rates on internal sales applied to both imported and domestic products, nevertheless imported products were subject to higher tax rates.[25] Another example is set in Argentina – Import Measures, where the measure required a local content requirement, that is the purchase or use by an enterprise of products of Argentine origin or from any Argentine source in a way that only the use of domestic products would enable a producer to comply with the local Argentine requirement.[26]
The presumption of “likeness” in trade in services might be more complex with respect to the presumption approach in trade in goods due to the existence of different modes of supply and their subsequent implication in the determination of the origin of services and services suppliers. Nevertheless, the AB clarified that such complexities do not make the presumption approach inapplicable to the trade in services, even though they can render the scope of the latter more limited than the scope of such approach in trade in goods.[27]
Concerning the burden of proof in establishing “likeness” according to the presumption approach, such burden rests upon the party that asserts the affirmative of a particular claim,[28] and thus the complainant must make a prima facie case that a measure distinguishes between services and services suppliers based exclusively on origin.[29] In the event that a Panel concludes that a complainant has failed to make such prima facie case, it must engage in an analysis of “likeness”.[30]
How a Panel shall analyze “likeness”
In the steps that a Panel should take when analyzing “likeness” under the GATS, the AB commenced by recalling how likeness is determined under the GATT. Accordingly, under the GATT, the AB in EC – Asbestos applied four general criteria – which later on have been embraced as guidelines in the “likeness test” -:
- the properties, nature, and quality of the products;
- the end-uses of the products;
- consumers’ tastes and habits or consumers’ perceptions and behaviour in respect of the products; and
- the tariff classification of the products.[31]
Even though each criterion examines a different aspect, the four categories should not be examined separately, since they are interrelated one influencing the other.[32] Nevertheless, it is well established that the four categories are only analytical tools that can be applied in the “likeness” assessment, categories that are not enforced by any agreements nor they are a closed list for determining the legal characterization of the products at issue.[33]
With the aforementioned considerations in mind, in present case the AB observed that:
“to the extent that the criteria for assessing “likeness” traditionally employed as analytical tools in the context of trade in goods are relevant for assessing the competitive relationship of services and service suppliers, these criteria may be employed also in assessing “likeness” in the context of trade in services, provided that they are adapted as appropriate to account for the specific characteristics of trade in services.”[34]
In the present case, the AB was of the same view. It found that “to the extent that evidence relating to the regulatory aspects has a bearing on the conditions of competition, it might be taken into account, subject to the particular circumstances of a case, and as an integral part of a panel’s analysis of whether the measure at issue modifies the conditions of competition to the detriment of like services or service suppliers of any other Member”.[40] The AB thus established that, in the “likeness” examination, the regulatory aspect of a measure might be taken into account. Yet, this is a rather open interpretation: how much weight should we give to the regulatory aspect? Is it a determinant element or rather a trivial one? The AB seems to both open and close the door for the regulatory aspect under the “likeness” test.
In conclusion, Argentina – Financial Services marked an important milestone of the WTO dispute settlements as it established, for the first time, how to engage in the analysis of the “likeness” in the context of trade in services. It has been outlined that such “likeness” under the GATS resembled the “likeness” under the GATT as, ultimately, under both is a matter of competitive relationship to be evaluated on a case-by-case basis. Moreover, under both, for establishing whether such services and services suppliers compete the one with the other, the four criteria set out in EC – Asbestos are also relevant. Some clarifications were made also regarding the regulatory aspect of a measure. Accordingly, it might be taken into account not only in the “exception” clauses, but also under the “likeness” examination.
Nonetheless, the GATS raises additional complexities as the latter is provided with different modes of supply, their intangible nature, as well as how each member described its commitments in their respective schedule. Likewise, other complexities arise as under the GATS not only services but also services suppliers must be analyzed and found “like”. Thus many questions arise. To what extent are the four criteria established under the GATT applicable to the GATS? Are there “like” services that can be distinguished from their “like” services suppliers and must be analyzed separately? Only future WTO disputes will tell.
(Matisse graduated from University in Geneva in International Relations with majors in law. Her areas of interest are International Trade and Investment law. She is currently working as a Legal Intern at Vázquez Tercero y Asociados, Mexico. She can be contacted at matisse.barbaro@gmail.com)
________________________________________
International Law Square 