1. encourage small entrepreneurship. In essence governments normally

1.         Introduction


The phrase ‘like
products’ appears sixteen times throughout the GATT Agreement, however, the definition
of the concept may vary from one GATT provision to another1.
National Treatment (NT) violations are dealt with under Article III of the
GATT. A country will be deemed to be in breach of an NT violation where internal
measures adopted by countries inhibit trade. For instance, when an imported
product is deemed to be ‘like’ a domestic product it must be treated in the
same way once it has cleared customs, otherwise there will be a violation of Article
III. Article XX on the other hand deals with the general exceptions to the rules
imposed in Article III.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team


In international
agreements, fiscal powers remain within the realm of the government sovereignty
to the extent that they do not alter competitive opportunities of similar goods2.
Governments make several choices, many of which are manifested in the non-trade
fiscal policy of the state, such choices could be to increase or decrease raw
materials or encourage small entrepreneurship. In essence governments normally distinguish
between products for  ‘non-trade
However, an infinite number of non-trade measures can distort the market and
affect trade. Without the chance to show that measures fall within the
permissible realm of non-trade policies, governments will be wary of enacting
non-trade policies for fear that these policies may have unintended effects on
trade and may result in dispute proceedings4.


 This paper deals with Article III of the GATT
on internal measures and seeks to evaluate the various meanings of ‘like
products’ under the Article. The analysis of Article III will be done using GATT/WTO
case law and relevant academic articles to streamline the views of the Panel
and Appellate Body when interpreting ‘like products’. Furthermore the paper
will also assess whether the narrow definitions of ‘likeness’ within the GATT/WTO
Dispute Settlement Body limits the ability of its members to take into account
non-trade concerns within the ambits of Article III of the GATT.


2.         Article
III: 2


Article III: 2 of
the GATT is broken down into two sentences. In the first sentence of the
Article, two questions are posed, answers to which, determine whether a
violation has occurred. The first question relates to whether the imported and
domestic products are like products, and the second question relates to whether
the tax imposed on imported products is more than that imposed on domestic
If the answers to both questions are affirmative then there is a violation
under the first sentence of Article III: 2.


general, the GATT does not define or give any guidance about the
characteristics of products, which determine “likeness”. There are however a
plethora of cases analysing the meaning of ‘like products’ in the first
sentence of Article III. In Japan-
Alcoholic Beverages II6
the Appellate body stated that the concept of like products in relation to the
first sentence of Article III: 2 should be interpreted narrowly because of the
existence of the concept of ‘directly competitive or substitute products’ used
in the second sentence of Article III: 2. Consequently the Appellate body used
the basic approach for determining likeness set out in the Working Paper on Border
Tax Adjustments7
(“Working paper”). The working paper illustrated that the problems arising from
the interpretation of likeness should be treated on a case-by-case basis, an
approach believed to encourage a fair assessment in each case of the different
elements that constitute like products. The working paper further provided a criteria
for determining, on a case-by-case basis whether goods are ‘like products’,
these include: the products end uses in a given market; its potential consumers
tastes and habits; and the product’s properties, nature and quality8.


However in deviation
from the Working Paper, the GATT Panel (“the Panel”) in earlier case of US-Malt Beverages9
chose to interpret the ‘like product’ concept by considering, in addition
to the various likeness factors, the policy objective stated in paragraph 1 of
Article III. This addition of the policy objectives found in Article III: 1
would mean that internal measures should not be applied to afford protection to
domestic products10.
The Panel thus interpreted Article III: 1 and Article III: 2 together to the
effect that the legitimacy of internal measures should be determined on the basis
of their purpose and market effect. In doing so, it was suggested that there would
be a need to determine the presence of a bona fide regulatory purpose and
whether their effect on competition is protective. This approach served as a
method for Panels to consider the legitimacy of certain non-trade concerns when
assessing the purpose of the measures placed by governments. Where they were
considered bona fide and not
inherently protectionist, the measures would not be in breach of Article III:


 This novel approach was considered in two
issues that resulted from the US-Malt
Beverages case11.
The first involved a Mississippi tax on wine that imposed different tax rates
according to the type of grapes used in making the wine. The US could not show
any bona fide regulatory purpose for making the distinction between the
varieties of grapes; consequently the Panel concluded that the only glaring
purpose for the product distinction was to protect local producers12.
In finding the tax differentiation to be protective the panel concluded that
the product distinction had both the ‘aim and effect’ to protect trade.


The second issue involved
several state regulations, which had restrictions on the sale of beer with
alcohol levels exceeding 3.2 per cent. Canada argued that all beers where ‘like
products’ and the Panel noted various factors of ‘likeness’ both in favour for
and against Canada. The Panel further went beyond by analysing the regulatory
purpose and competitive effects of the regulations. The Panel noticed several
reasons of social welfare policy for making the product distinction and held
that the product distinction did not create adverse conditions of competition
for Canadian brewers because Canadian brewers produced other types of beers
that may not be caught by the measure. This further shows an attempt by the
Panel to take into account ‘non-trade’ concerns, in this case, social welfare
policy, in making decisions on the likeness of a product. In doing so, they extended
the meaning of ‘likeness’ as prescribed by the Working paper.


In dispute
concerning special tax levels of luxury cars, US- Taxes on Automobiles13,
the Panel elaborated on the above approach and stated that ‘likeness’ should in
most cases include the examination of the ‘aims and effects’ of the tax
measure. In the case at hand the Panel held that aim or effect of the luxury
tax on cars did not afford protection to the domestic production of cars.
Hence, cars above and below $30,000 could not, in respect to luxury products,
be considered like products. The Panel’s reasoning was that the regulations
were bona fide regulatory purposes
and found that the competitive effects where neither clear enough nor inherent
enough to be classified as protective. The ‘aim and effect’ test provided an
improvement to the original analysis of likeness. This is primarily because it
consigned the metaphysics of ‘likeness’ to a lesser role in its analysis by
depending on the trade effects of the regulation and whether or not the purpose
of the measure was bona fide. Thereby
taking into account non-trade measures to ascertain the intent of the
questioned regulation.

Furthermore the
aim and effect test allowed the consideration of the justification of the
regulation at the same time the issue of violation was determined. The aim and
effect test therefore avoided the premature dismissal of valid complaints on
grounds that goods were not alike and the rigorous treatments given to claims
of regulatory justification under Article XX whenever goods were deemed to be
‘like products’14.


the advantages mentioned above, the ‘aim and effect’ test for determining the
likeness of goods was explicitly rejected by the Panel in Japan- Alcoholic Beverages II15.
The Panel held that the ‘aim and effect’ test was not in line with the wording
of the first sentence of Article III: 2. They came to this analysis by drawing-on
the fact that the ‘aim and effect’ test emanated from the wording in Article
III: 1 – “so as to afford protection”. In the Panels view there were no words
in the first sentence of Article III: 2 that made reference to “so as to afford
protection” in Article III: 1 and consequently the ‘aim and effect’ of the


Furthermore the
Panel deemed it appropriate that a complainant will have the burden of not only
showing the effect of a particular measure, which is in principle discernable
but also its aim, which can be indiscernible. This appears to be the case where
there is a multiplicity of aims sought through enactment of legislation, it is expected
that difficulty would arise in determining the application of the aim and
effect test17.
Most importantly the list of exceptions contained in Article XX could be
rendered useless because the ‘aim and effect’ test does not contain a list of
grounds for the justification of the departure from obligations incorporated in
Article III. In buttressing this point it is important to note that the purpose
of Article XX was to provide a list of exceptions, subject to conditions, that
would otherwise have led to unjustifiable discrimination or disguised
restriction of international trade. Therefore if a country uses health policy
in the context of the ‘aim and effect’ test to justify derogation from Article
III, the standard of proof established in Article XX would be circumvented and
such a country will not have to prove that a health measure is ‘necessary’ to
achieve the health objective as prescribed in Article XX18.


 In light of the foregoing, the actions of the
Panel demonstrate that the ‘aim and effect test used in US Tax on Automobiles should be rejected. This ruling has served to
be the norm in relation to Article III: 2 ever since. Consequently the
‘likeness’ of products under Article III: 2 are to be treated narrowly and not
broadly by considering the non-trade policy objectives of the regulation. At
this point it is important to note that where the Dispute Settlement Body of
the WTO/GATT adopts the reports of Panel or Appellate Body, the conclusions and
recommendations become binding on the parties to the dispute. Therefore where a
Panel (and the Appellate Body) has found inconsistency, they recommend that the
defaulting Member concerned should bring it measure in conformity with WTO/GATT
However it is also important to note that although said decisions are binding
on the members to a particular dispute, the reports of the Dispute Settlement
Body are not binding precedents for other disputes between the same parties on
other matters or different parties on the same matter20.



3.         Article
III: 4


Article III: 4
states that products of one member State imported into another member State
shall be accorded treatment no less favourable than that accorded to like
products in the other state in relation to regulation, laws and requirements
affecting the internal sale, offering for sale, distribution or use. In Korea- Various Measures on Beef21
the Appellate Body provided a three step test for determining violations
under Article III: 4. In order to prove a violation one will need to show that
the measure at hand is a law, regulation or requirement covered by Article III:
4; the imported and domestic products are like products; and the imported
products are accorded less favourable treatment.


 In EC-
the Appellate Body sought to consider whether the concept of ‘like
products’ in Article III: 4 was to be treated with the same narrow
interpretation as the first sentence of Article III: 2. On this issue, it was noted
that Article III: 2 contained two separate sentences, each imposing a different
obligation; the first sentence deals with ‘like products’; and the second sentence
deals with ‘directly competitive or substitutable’ products23.
However Article III: 4 applies to only ‘like products’ and does not contain any
provisions equivalent to the second sentence of Article III: 2. They further
added that under Article III: 2, the interpretation of the first sentence
affects the second sentence. Therefore the scope of the term ‘like products’ in
the first sentence of Article III: 2, affects and is affected by the scope of
the second sentence on ‘directly competitive and substitutable’ products. As
such, following from the fact that the second sentence constitutes a broader
category of products, which are not ‘like products’ as contemplated by the
first sentence, they decided that the first sentence should be construed
narrowly to prevent a deviation from the measures its strict terms are expected
to uphold.


 In relation to Article III: 4, there is an absence
of the same interpretative consideration because the ‘general principle’
articulated in Article III: 1 is expressed in Article III: 4 in a single
sentence, (and not in two sentences as in Article III: 2) therefore, the
meaning of “like products” attributed to Article III: 2 is different from that
in Article III: 4. According to the Appellate Body the meaning of ‘likeness’
stretches in a different way in Article III: 4.


Due to the fact
that the decision in Japan- Alcoholic
Beverages did not express how far the Appellate Body’s rejection of the
‘aim and effect’ approach would reach, questions arose as to whether a
different approach to carrying out the policy stated in Article III: 1 would be
necessary. This was because the Panel in Banana
considered employing a version of the ‘aim and effects’ test known as the ‘protective
application’ concept to Article III: 4. In
doing so they did not propose an elaborate analysis, they merely considered it
appropriate to discern the protective application of a measure from its design,
architecture and the revealing structure under Article III: 4.

The Appellate
Body rejected and reversed this initiative by upholding that, like Article III:
2, the wording in Article III: 1 had no explicit reference to Article III: 4.
On this basis the Appellate body held that it would be inappropriate for a
Panel to make further inquiry on the protective application when applying the like
products test of Article III: 4.


In finding a distinction
between ‘like products’ under Article III: 2 and Article III: 4, the Appellate
Body in EC-Asbestos sought to
determine the definition of ‘like products’ under Article III: 4. The term was
to be interpreted in consonance with the objective pursued by Article III as
enunciated in the ‘general principles’ in Article III: 1 and the specific
interests expressed under Article III: 4. The reasoning of the Appellate Body
in EC- Asbestos25
was that products in competitive relationships in the market place could be
affected through treatment of imports in a ‘less favourable’ manner than that
ascribed to domestic products. On this basis, the determination of ‘likeness’
under Article III: 4 would be the extent of competitive relationship between
and among products. In holding this view they noted that although the scope of
‘like products’ under Article III: 4 would be broader than the concept of ‘like
products’ under the first sentence of Article III: 2, it would certainly not be
broader than the combined product of the two sentences of Article III: 2, which
also includes directly competitive and substitutable products in its second


 In respect of the criteria of determining
‘like products’ under Article III: 4, the Appellate body stated that as with Article
III: 2, no one approach will be appropriate for all cases, giving the
impression that the criteria will be determined on a case-by-case basis26.
They did however provide the criterion of ‘likeness’ as prescribed by the
Working Party on Border Tax Adjustment and included four characteristics to
complement the four criterion that the product in question could share, namely:
the physical properties; the extent to which products are capable of serving
the same or similar end-users; the extent to which consumers perceive and treat
the products as alternative means of performing particular functions in order
to satisfy certain wants and demands; and the international classification of
the products for tariff purposes.



4.         Conclusion


A common
characteristic on the scope of likeness is that it has expanded overtime27,
this can be seen in Brazilian Internal Taxes
where conhaque and cognac were deemed to be different products due to the
addictive and aromatic differences, whereas in Japan Alcoholic Beverages Shochu, Vodka, rum and liqueurs were all
considered like products. The aim and effect test brought about a breaker to
the narrow interpretation of like products through cases like US- Taxes on Automobiles and Malt Beverages. This era was short-lived
as the aim and effect test was rejected for being an approach that circumvented
the national treatment obligation. Now it seems tribunals have marched back
into the area of traditional means for determining likeness of product as
envisaged by the narrow definition in the Working paper. The issue here is that
there are many government policies which may be deemed as ‘non-trade’ concerns
that are taken for legitimate purposes and create merely incidental burdens on
the like products of imported goods that may be caught by this sweeping
There ought to be a better balance to fulfil the drive for free trade and the
legitimate regulatory autonomy of each WTO state30.
This balance can perhaps be achieved by interpreting the Article XX general
exception more generously. However, there is hope, because Panel/Appellate body
decisions are not binding on their subsequent counterparts. Therefore there is
a possibility that the law will evolve in order to take into consideration the non-trade
concerns of governments within the ambits of ‘likeness’ under Article III of
the GATT.

1 Report of the Working Party on Border
Tax Adjustments, BISD 18S/97, para. 18.

2 Serena B. Willie, Recapturing a Lost

III: 2 GATT 1994 Japan-Taxes on Alcoholic Beverages 1996 accessed 21 January 2018

3 Ibid

4 Ibid

5 Appellate Body
Report, Canada – Certain Measures Concerning Periodicals, WT/DS31/AB/R, adopted
30 July 1997, DSR 1997:I, p. 449

6 Appellate Body Report, Japan – Taxes on
Alcoholic Beverages, WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R, adopted 1
November 1996, DSR 1996:I, p. 97

7 Report of the Working Party on Border Tax
Adjustments, BISD 18S/97, para. 18.

8 Ibid

9 GATT Panel Report, United States –
Measures Affecting Alcoholic and Malt Beverages, DS23/R, adopted 19 June 1992,
BISD 39S/206

10 Ibid

11 Ibid

12 Ibid

13 GATT Panel Report, United States – Taxes
on Automobiles, DS31/R, 11 October 1994, unadopted

14 Robert E. Hudec, GATT/WTO Constriants
on National Regulation: Requiem for an Aim and Effects Test, 32 Int’L. 619

15 Appellate Body Report, Japan – Taxes on
Alcoholic Beverages, WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R, adopted 1
November 1996, DSR 1996:I, p. 97

16 Ibid

17 Van den Bossche,
P. and W. Zdouc, The Law and Policy of the World Trade Organization: Text,
Cases and Materials, 4th ed., Cambridge UP, 2017.

18 Appellate Body Report, Japan – Taxes on
Alcoholic Beverages, WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R, adopted 1 November
1996, DSR 1996:I, p. 97

19 Article 19.1
Dispute Settlement Understanding

20 WTO, Legal effect
of panel and appellate body reports and DSB recommendations and rulings, https://www.wto.org/english/tratop_e/dispu_e/disp_settlement_cbt_e/c7s2p1_e.htm,
accessed 21 January, 2017

21 Korea – Measures Affecting Imports of
Fresh, Chilled, and Frozen. Beef (complaint by the United States, WT/DS161/1,
and WT/DS169/1, complaint by Australia).

22 Appellate Body Report, European
Communities – Measures Affecting Asbestos and Asbestos-Containing Products,
WT/DS135/AB/R, adopted 5 April 2001, DSR 2001:VII, p. 3243

23 Ibid

24 EC — Bananas III. Appellate Body Report,
European Communities — Regime for the Importation, Sale and Distribution of
Bananas, WT/DS27/AB/R, 1997

25 Appellate Body Report, European
Communities – Measures Affecting Asbestos and Asbestos-Containing Products,
WT/DS135/AB/R, adopted 5 April 2001, DSR 2001:VII, p. 3243

26 Ibid

27 Won
Mog Choi, Overcoming the ‘Aim and Effect’
Theory: Interpreting of the ‘Like Product’ in GATT Article III, U.C. Davis
J. Int’L. & Pol’y 107 (2002)

28 Brazilian Internal Taxes, adopted on 30
June 1949 and 13 December 1950, BISD11/182,para.7: GATT Secretariat

29 Won
Mog Choi, Overcoming the ‘Aim and Effect’
Theory: Interpreting of the ‘Like Product’ in GATT Article III, U.C. Davis
J. Int’L. & Pol’y 107 (2002)

30 Ibid