14.167 The US approach to "like product" is narrower than that proposed by the European Communities. The United States contends that four US passenger cars sold or planned to be sold in Indonesia -- the Opel Vectra and Optima, the Ford Escort and the Chrysler Neon -- are "like products" to the Timor. In support of this conclusion, the United States contends that these products have similar end-uses (to transport passengers) and that, because the physical characteristics of these cars (e.g., size, weight, height, engine size) are similar or virtually identical to those of the Timor, they "closely resemble" that car. In support of its conclusions, the United States cites what it considers to be highly authoritative market segmentation analysis applied by the DRI Global Automotive Group and reflected in a number of DRI publications excerpts of which have been placed before the Panel, ⁷²⁵ under which the Timor, Optima, Escort and Neon are all classified as Segment C (Lower Medium Segment) passenger cars, while the Vectra is classified as Segment D (Upper Medium Segment) passenger cars.

14.168 Indonesia takes issue with both the EC and US approaches to the "like product" question. Indonesia considers that, because an affirmative finding of serious prejudice would deprive Indonesia of a conditional right to provide subsidies, the concept of "like product" must be very narrowly construed, and that the burden of proof on complainants is particularly high. Indonesia argues that the complainants have not met their burden of proving that products are like and of establishing acceptable like product categories. Indonesia does not appear to take issue with the principle that the analysis should be limited to passenger cars. However, it notes that while all passenger cars share certain basic characteristics, they are still highly differentiated on the basis of numerous other physical and non-physical characteristics and thus should not be treated as "like". In this case, Indonesia argues, because the Timor is highly differentiated from US and EC models and is thus non-substitutable, those products cannot be considered to be like products to the Timor. Rather, Indonesia considers that the Timor and the Bimantara Cakra are in a market segment by themselves, "small budget" cars, while even the US and EC cars closest to the Timor are in the "small normal/regular" market segment.

14.169 In assessing the arguments of the parties, we are cognizant that the complainants are required to demonstrate the existence of serious prejudice by positive evidence. Thus, we agree with Indonesia that the complainants bear the burden of presenting argument and evidence with respect to each element of their serious prejudice claims -- including the existence of effects on a "like product". This is consistent with the general principle, stated by the Appellate Body in Shirts and Blouses, that complainants must present evidence and argument sufficient to establish a presumption that a defending Member has acted in a manner inconsistent with its WTO obligations. ⁷²⁶ We do not agree, however, that the complainants bear a heavier than usual burden of proof in this dispute or that the concept of "like product" should be interpreted more narrowly than usual because Indonesia is a developing country Member. The special and differential treatment available to Indonesia is spelled out in Article 27 of the SCM Agreement, and it is substantial. But for that special and differential treatment, the subsidies in question would by Indonesia's own admission be prohibited under Article 3.1(b) of the SCM Agreement. Moreover, because Indonesia is a developing country Member, Article 27.8 requires complainants to demonstrate serious prejudice by positive evidence "in accordance with the provisions of paragraphs 3 through 8 of Article 6" rather than taking advantage of the rebuttable presumption of serious prejudice that otherwise would have applied under Article 6.1(a). Article 27 does not, however, impose a higher burden of proof on complainants than that normally applicable under Article 6, nor does it provide that the term "like product" is to be defined differently in the case of subsidization provided by a developing country Member.

14.170 Turning now to an analysis of the issue at hand, we note that the term "like product" is used in a variety of contexts within the WTO Agreement. ⁷²⁷ As it is used in most cases, the term is not defined, and numerous GATT and WTO panels have been required to apply the term to particular factual situations. The SCM Agreement, however, shares with the Agreement on Implementation of Article VI of GATT 1994 (ADP Agreement) a definition of the term "like product" that is not found elsewhere in the WTO Agreement. Note 46 to Article 5.1 of the SCM Agreement provides that:

Although this definition appears in Part V of the SCM Agreement (relating to countervailing measures), the use of the clause "throughout this Agreement" makes clear that this definition is equally applicable to the provisions of Part III, including Article 6 relating to adverse effects.

14.171 The definition of "like product" found in the SCM Agreement first appeared in the Kennedy Round Anti-Dumping Agreement ⁷²⁸ and, in spite of proposals for modifications during various negotiations, was carried virtually unchanged into the Tokyo Round Anti-Dumping and Subsidies Codes ⁷²⁹ and hence into the WTO SCM and ADP Agreements. Thus, the definition of "like product" applicable in this dispute has existed for thirty years. In light of this, and in spite of the fact that investigating authorities in anti-dumping and countervailing duty cases have been wrestling with the concept of "like product" for decades, it is surprising that no GATT or WTO panel has yet been required to apply this definition of "like product" to particular facts, or even to provide any detailed discussion regarding the meaning of the term or its clarification. Thus, we are operating in uncharted territory in our consideration of this issue.

14.172 In determining what criteria might be relevant to performing "like product" analysis under the SCM Agreement, our point of departure must of course be the text of the SCM Agreement. In this case, no party has argued that there is an "identical" product that should be treated as the "like product" to the Timor. Rather, the parties have focused on the question of which cars have "characteristics closely resembling" those of the Timor. On its face, this term is quite narrow. It is not enough that the products have characteristics which resemble the Timor; rather they must have characteristics which "closely" resemble the Timor. The parties have identified a wide range of physical characteristics which they consider may be relevant to this analysis. In addition, Indonesia considers certain "non-physical" characteristics, consumer perceptions and preferences to be relevant to the analysis. Further, there have been references to the uses to which a product may be put, to the substitutability of products, to price and to tariff classification principles.

14.173 In our view, the analysis as to which cars have "characteristics closely resembling" those of the Timor logically must include as an important element the physical characteristics of the cars in question. This is especially the case because many of the other possible criteria identified by the parties are closely related to the physical characteristics of the cars in question. Thus, factors such as brand loyalty, brand image/reputation, status and resale value reflect, at least in part, an assessment by purchasers of the physical characteristics of the cars being purchased. Although it is possible that products that are physically very different can be put to the same uses, differences in uses generally arise out of, and assist in assessing the importance of, different physical characteristics of products. Similarly, the extent to which products are substitutable may also be determined in substantial part by their physical characteristics. Price differences also may (but will not necessarily) reflect physical differences in products. An analysis of tariff classification principles may be useful because it provides guidance as to which physical distinctions between products were considered significant by Customs experts. However, we do not see that the SCM Agreement precludes us from looking at criteria other than physical characteristics, where relevant to the like product analysis. The term "characteristics closely resembling" in its ordinary meaning includes but is not limited to physical characteristics, and we see nothing in the context or object and purpose of the SCM Agreement that would dictate a different conclusion. ⁷³⁰

14.174 Although we are required in this dispute to interpret the term "like product" in conformity with the specific definition provided in the SCM Agreement, we believe that useful guidance can nevertheless be derived from prior analysis of "like product" issues under other provisions of the WTO Agreement. Thus, we note the statement of the Appellate Body in Alcoholic Beverages (1996) ⁷³¹ that, in this context as in any other, the issue of "like product" must be considered on a case-by-case basis, that in applying relevant criteria panels can only use their best judgement regarding whether in fact products are like, and that this will always involve an unavoidable element of individual, discretionary judgement. With this in mind, we now proceed to consider the application of these general principles to the case at hand.

14.175 Turning first to the argument of the European Communities that all passenger cars should be considered "like products" to the Timor, we consider that such a broad approach is not appropriate in this case. While it is true that all passenger cars "share the same basic physical characteristics and share an identical end-use", we agree with Indonesia that passenger cars are highly differentiated products. Although the European Communities have not provided the Panel with information regarding the range of physical characteristics of passenger cars, all drivers know that passenger cars may differ greatly in terms of size, weight, engine power, technology, and features. The significance of these extensive physical differences, both in terms of the cost of producing the cars and in consumer perceptions regarding them, is manifested in huge differences in price between brands and models. It is evident that the differences, both physical and non-physical, between a Rolls Royce and a Timor are enormous, and that the degree of substitutability between them is very low. Viewed from the perspective of the SCM Agreement, it is almost inconceivable that a subsidy for Timors could displace or impede imports of Rolls Royces, or that any meaningful analysis of price undercutting could be performed between these two models. In short, we do not consider that a Rolls Royce can reasonably be considered to have "characteristics closely resembling" those of the Timor.

14.176 The European Communities contend that we must consider all passenger cars to be "like" because any effort to differentiate between passenger cars with a multitude of differing characteristics would inevitably result in arbitrary divisions. We are aware that there are innumerable differences among passenger cars and that the identification of appropriate dividing lines between them may not be a simple task. However, this does not in our view justify lumping all such products together where the differences among the products are so dramatic. The parties to this dispute have submitted substantial data regarding a wide range of characteristics of the Timor and of the models that the European Communities and United States consider to be like products to the Timor. We must endeavour to find some reasonable way to assess the relative importance of the various differences in the minds of consumers and to devise some sensible means to categorize passenger cars.

14.177 One reasonable way for this panel to approach the "like product" issue is to look at the manner in which the automotive industry itself has analyzed market segmentation. The United States and the European Communities have submitted information regarding the market segmentation approach taken by DRI's Global Automotive Group, a company whose clients include all major auto manufacturers, including KIA, PT TPN's national car partner. DRI's Asian Automotive Industry Forecast Report of June 1997 ("Asian Forecast") divides passenger cars into five segments: Small Cars Segment (Segment A), Supercompact Segment (Segment B), Lower Medium Segment (Segment C), Upper Medium Segment (Segment D) and Executive Segment (Segment E). Three of those segments (C to E) are divided into subsegments 1 (lower end) and 2 (upper end). While the Asian Forecast does not identify the Timor, it does indicate that the Kia Sephia -- Timor's parent car -- falls within the C1 Segment (lower end of Lower Medium Segment). ⁷³² The Asian Forecast also places the Optima and the 306 in the C1 Segment, while the Vectra is placed in the D1 (lower end of Upper Medium Segment). The Escort and Neon are not identified in the portion of the Asian Forecast (Table 6 relating to sales in Indonesia) before the Panel, presumably because they are not sold in Indonesia. However, market segmentation data from DRI's World Car Industry Forecast Report of February 1997 ("World Forecast"), which appears to follows the same market segmentation approach as the Asian Forecast, places the Escort in the C1 Segment, while it places the Neon in the C2 Segment (upper end of Lower Medium Segment).

14.178 The question remains whether the criteria used by DRI in arriving at its market segmentation are appropriate to a "like product" analysis under the SCM Agreement. We consider that they are. The Asian Forecast indicates that:

The World Forecast provides some further explanation regarding DRI's market segmentation analysis. It states that:

Thus, DRI has in its analysis considered the physical characteristics of the cars in question when designing its segmentation. It has used as an initial filter the size of the vehicle, but it has then divided cars of a given size into upper and lower end categories, and has moved luxury cars, regardless of size, from lower segments to the E segment. We consider such an approach, which segments the market based on a combination of size and price/market position, to be a sensible one which is consistent with the criteria relevant to "like product" analysis under the SCM Agreement.

14.179 Indonesia notes that "like product" analysis may depend upon the market in question, and that DRI's segmentation analysis was developed for the European market. Indonesia further argues that the Timor sold in Indonesia is not in fact identical to the Kia Sephia, but is rather an older model of the Sephia with a smaller engine and fewer features than the new Sephia. In support of this proposition, it points to a US exhibit indicating that Sephia models sold in the United States beginning in 1996 had a larger 1.8 litre engine and new features such as an improved suspension and airbags. Indonesia also submitted data showing that the Sephia sold in the United States weighs from 68 to 103 kg. more than the Timor S-515. Indonesia has not presented any evidence regarding the size of the engine for Sephias sold in Europe, so the implications of this evidence for the validity of DRI's market segmentation analysis are unclear. ⁷³³ It does not appear, however, that DRI considered these differences sufficiently important to merit classifying the Timor sold in the Indonesian market differently from the Sephia sold in other markets. Further, DRI apparently considers its segmentation analysis to be applicable to the Asian market generally and to the Indonesian market in particular, as it applies that analysis in the Asian Forecast.

14.180 Indonesia further objects that the complainants have not submitted the full text of the Asian Forecast, and that the Panel should draw adverse inferences from this fact. However, the materials provided by the complainants appear to include that portion of the Asian Report relating to market segmentation, and Indonesia has not given the panel any reason to believe that relevant portions of the Asian Forecast have been excluded. Further, because the Asian Forecast is not a document to which the complainants can control access, resort to adverse inferences would not appear appropriate. Finally, Indonesia objects that the DRI approach is limited solely to physical dimensions, and fails to consider other physical characteristics much less "actual market forces or competition." As indicated by our discussion, however, Indonesia's argument on this score appears to be factually incorrect.

14.181 In our view, the DRI market segmentation analysis presented by the European Communities and the United States supports the view that all vehicles in the C1 Segment -- including the 306, Optima and Escort -- are "like products" to the Timor within the meaning of the SCM Agreement. By contrast, the DRI analysis places the Vectra and the Neon in different market segments (D1 and C2 respectively), and this in our view weakens the complainants' view that these products should be considered to be "like" the Timor. ⁷³⁴ The complainants have not provided any explanation as to why the Panel should consider DRI's market segmentation relevant yet overlook that segmentation where it is unfavourable to them. To the contrary, the United States' argument regarding the Vectra in particular damns with faint praise. The United States concedes that the Vectra "is positioned slightly higher in the market than the Timor", that it is longer and has a more powerful engine. The United States maintains only that the Vectra "is not all that dissimilar" to the Timor, a test which would appear to fall short of the SCM Agreement's definition.

14.182 While the DRI market segmentation analysis is useful, we are not prepared to rely on it without confirming, through an independent comparison of their physical characteristics, that the EC and US models in question are "like" the Timor. The European Communities, the United States and Indonesia have all provided the Panel with detailed comparative information regarding the physical characteristics and performance of the models at issue. This information varies in some details, and it is not always clear which versions of particular models are being compared with which. ⁷³⁵ Nevertheless, the information before us generally provides a good overview of the similarities and differences between the models in question. However, the parties have sometimes chosen to emphasize different physical characteristics and features or to characterize differently the significance of the differences that exist. In testing the DRI market segmentation approach, therefore, we will outline the Indonesian view of market segmentation in Indonesia and then test that view against the physical characteristics and other criteria identified by Indonesia as significant.

14.183 Indonesia's argument is that the Timor is a "low-technology no-frills budget car" which fills a unique niche at the bottom of the Indonesian market. Indonesia contends that a proper market segmentation for passenger cars would divide those cars into three classifications -- small, medium and large -- and that each of the classifications in turn should be divided into a budget, normal/regular and luxury categories. Indonesia contends that the Timor and one other model (the Bimantara Cakra) are alone in the budget small car category. ⁷³⁶ It considers that all of the specific EC and US models sold in Indonesia which are alleged to be "like" the Timor belong in a different category, that of small normal/regular cars.

14.184 Indonesia's analysis begins with a comparison of various physical features of the Timor with those of other models -- Escort, 306, Optima, Vectra, Neon -- specifically alleged by complainants to be "like products" to the Timor. Indonesia contends that there are four basic physical characteristics or specification groupings that differentiate passenger cars: passenger compartment, power plant, steering and suspension and safety features. Indonesia considers that an examination of these four specification groupings demonstrates that the five models identified above are not like products to the Timor. We will examine each set of specification groupings in turn.

14.185 With respect to passenger compartment, Indonesia considers that the relevant aspects are interior dimensions and number of passengers. Indonesia acknowledges that the Timor, Escort, 306, Optima and Neon are comparable in this regard. However, Indonesia contends (and the complainants have not contested) that the Vectra has a substantially larger interior and accommodates five passengers as opposed to four passengers for the other models.

14.186 With respect to power plant, the engine capacity of the Timor is 1498 cc. That of the other five models, in ascending order, are the Escort (1597 cc), 306 (1761 cc), Optima (1796 cc), Neon (1996 cc) and Vectra (1998 cc). Thus, the Timor does in fact have the smallest engine of the group. On the other hand, the record indicates the Timor has a sixteen-valve engine, and at least one of the Timor models sold in Indonesia (the Timor S-515i) ⁷³⁷ has a dual overhead cam engine. The other models all have single overhead cam engines, and the 306 and Optima have only an eight-valve engine. According to data submitted by the parties, while the power output of the Timor S-515 is 58kw/5500rpm, that of the Timor S-515i is 77kw/5500rpm, as opposed to 66 kw/5400 rpm for the Optima, 70kw/6000 rpm for the 306, 75kw/5500 rpm for the Escort, 85 kw/5400 rpm for the Vectra, and 97kw/6000rpm for the Neon. ⁷³⁸ Accordingly, it would appear that the engine of the Timor S-515i is actually more powerful than those of the Optima, 306 and Escort, but somewhat less powerful than those of the Vectra and Neon.

14.187 With regard to steering and suspension, Indonesia differentiates between the Timor and the Escort, Vectra and Optima on the grounds that the latter three have power steering. However, sales brochures for the Timor indicate that it has power steering as a standard feature, while the United States contends that the Escort has power steering as an option only. Thus, it would appear that with respect to this feature the Timor may actually be more advanced than the 306 and Neon. Regarding suspension, although Indonesia's summary table suggests that there are differences between the Timor and all of the models compared to it, the detailed information supplied by Indonesia shows that all of the models have identical suspension: in the front "Independent MacPherson strut", and in the rear "fully independent multi-linked".⁷³⁹

14.188 With respect to "safety features", Indonesia identifies the braking system, fuel tank capacity and mileage, curb weight and passive restraints as relevant features. With respect to the braking system, all models would appear to have front disc and rear drum brakes. The Vectra has ABS, and it appears that some models of the Optima and 306 may also have ABS, while the Timor and Escort do not (the Neon offers ABS as an option). The fuel tank capacity, in ascending order, is: Timor (50 litres), Optima (52 litres), Escort (55 litres), Neon (57 litres), 306 (60 litres), and Vectra (61 litres). Data regarding fuel efficiency were unavailable for most models. The models weigh in at the following: Optima (980 kg), Timor (1055 kg), 306 (1100 kg), Neon (1102 kg), Escort (1110 kg) and Vectra (1150 kg). Documentation submitted by Indonesia shows that all models, except the Timor and 306, have airbags; however, the United States indicated that the Indonesian version of the Escort would not have been equipped with airbags, while documentation from the European Communities indicates that one Optima model has a single airbag while another does not.

14.189 In our view, the data provided by the parties tend to confirm the market segmentation analysis of DRI. Even limiting ourselves to those features on which Indonesia has chosen to focus, the Timor does not appear to be notably inferior to the 306, Optima or Escort, and in fact in some respects may be superior to those models. With respect to the Vectra, on the other hand, the superiority seems relatively clear -- the Vectra is the heaviest passenger car in the group (95 kgs more than the Timor), shares honours with the Neon for the largest engine (roughly 500 cc larger than the Timor's), is the only one of the vehicles identified as appropriate for 5 passengers, and has all the features identified by Indonesia (power steering, ABS, airbags). The case of the Neon is intermediate: it is heavier and has a larger engine than the Timor and most of the other models in question, but it is not notably better equipped than the Timor in other respects. Thus, the data fit well with DRI's assessment that the Vectra is in a distinct market segment from the Timor while the Neon is in the upper end of the same market segment as the Timor.

14.190 A number of other factors combine to suggest that the Timor in fact belongs in the C Segment. First, according to DRI's World Forecast, Kia -- producer of the Sephia, parent car to the Timor --produces and sells in the Korean market a range of passenger cars, including entrants in Segments A (Morning) and B (Avella and Pride). Further, contrary to Indonesia's contention, DRI data show that several passenger cars classified by DRI in the B Segment (Suzuki Baleno, Toyota Starlet) are currently sold in the Indonesian market, and several others (Daihatsu Charade, Maleo Maleo) were sold in the Indonesian market as recently as 1996. We note that these B Segment cars would appear to be significantly smaller and, on the whole, less powerful than the Timor. ⁷⁴⁰ This undercuts Indonesia's arguments that the Timor and the Bimantara Cakra are in a class by themselves at the bottom of the Indonesian market.

14.191 In addition to comparing physical characteristics, Indonesia argues that the Timor differs from the other models alleged to be "like products" in terms of a variety of non-physical characteristics. Although Indonesia does not clearly identify which characteristics it considers to be non-physical, among the factors listed by Indonesia which arguably are not "physical characteristics" are brand loyalty, brand image/reputation, status, after-sales service and resale value. However, the only evidence cited by Indonesia in support of its view that the Timor should be distinguished from the other models in question on these bases is a single sentence in a newspaper article submitted as an exhibit by the United States, stating that "dealers say consumers ask a lot of questions about the quality of the Timor and the after-sales service." ⁷⁴¹ Further, it is by no means clear that the models alleged to be "like" the Timor would necessarily fare any better than the Timor with respect to these factors. For example, Ford and Chrysler do not have a well-established brand in Indonesia, while sales of Peugeots and Opels have also been relatively small. In short, Indonesia has provided almost no evidentiary support for its view that there are differences in non-physical characteristics which are important to this analysis.

14.192 Finally, Indonesia seems to suggest that the Timor's low price places it into a special market niche and thus renders it "unlike" the more expensive US and EC models. We do not preclude that price might be a relevant consideration in performing "like product" analysis, particularly where differences in price represent one way to assess the relative importance of differing physical characteristics to consumers. In this case, however, the complainants allege that the Timor is being sold at undercutting prices as a result of subsidization. If we were to conclude that the low price of the Timor in the Indonesian market were to render the Timor "unlike" other models which are similar in physical characteristics to the Timor but priced higher, the result would be that, in cases where the subsidization and resulting price undercutting were sufficiently high, price undercutting claims under Article 6 could never prevail. Thus, we do not consider that the Timor's lower price is a basis to conclude that it is unlike the models alleged by the complainants to be "like" the Timor.

14.193 In conclusion, we consider that the Optima, Escort and 306 are "like products" to the Timor within the meaning of the SCM Agreement. The Vectra, in contrast, is not a like product to the Timor within the meaning of the SCM Agreement. With respect to the Neon, the issue is more difficult. That model is not categorized in a different segment by DRI but is in the higher end of the C segment, and although it is larger and more powerful than the Timor, does not have the clear superiority in its features demonstrated by the Vectra. Thus, for the purposes of our further analysis, we will assume arguendo that the Neon is a like product to the Timor.

To continue with Treatment of Cars Imported Unassembled

⁷²⁵ Asian Automotive Industry Forecast Report of June 1997; World Car Industry Forecast Report of February 1997.

⁷²⁶ Shirts and Blouses, op. cit., Appellate Body Report, p. 13.

⁷²⁷ For example, Articles I and III of GATT. Article I:1: " . . . any advantage, favour, privilege or immunity granted by any contracting party to any product originating in or destined for any other country shall be accorded immediately and unconditionally to the like product originating in or destined for the territories of all other contracting parties." Article III:2: "The products of the territory of any other contracting party shall not be subject, directly or indirectly, to internal taxes or other internal charges of any kind in excess of those applied, directly or indirectly, to like domestic products." Article III:4: "The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin . . . "

⁷²⁸ Kennedy Round Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, Article I.A(b).

⁷²⁹ Tokyo Round Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, Article 2.2; Agreement on Interpretation and Application of Articles VI, XVI and XXIII of the General Agreement on Tariffs and Trade, note 18 to Article 6.

⁷³⁰ This interpretation is confirmed by the negotiating history of this definition. As noted above, this definition of "like product" is virtually unchanged from that which first appeared in the Kennedy Round Anti-Dumping Code. Thus, the penultimate draft of that Code defined the term "like product" to mean a product which "has physical characteristics close to those of the exported product." T.64/NAB/W/16, dated 3 March 1967. In the revised draft of 28 March 1967, the word "physical" had been deleted from the text, which was revised to the formulation ("characteristics closely resembling") that exists today. T.64/NAB/W/17.

⁷³¹ Op. cit., pp. 19-23.

⁷³² DRI identifies models sold in the Indonesian market by the more common model names used in other markets. However, it is clear from the sales data provided by DRI that the "Sephias" referred to in the Asian Forecast are in fact Timors imported from Korea or assembled in Indonesia.

⁷³³ The United States placed into the record a number of articles from US publications (the Washington Post, New York Times and Car and Driver magazine) suggesting that the Neon and the Sephia were competing in the same market segment in the United States, while Indonesia identified a publication (Automotive News) placing the Sephia in a separate class (budget) from that of the Neon and Escort (small). None of these articles provided any explanation for the market segmentation selected. In this context, we take note of evidence submitted by Indonesia regarding differences between the Sephia sold in the United States and the Timor.

⁷³⁴ Neither the European Communities nor the United States sell any B Segment cars in the Indonesian market, and we have not been asked to decide whether such cars are like products to the Timor for the purposes of the SCM Agreement.

⁷³⁵ The submissions of the parties refer to at least two versions of the Timor (S-515 and S-515i), three versions of the Optima (GLS, CDX and CDX A-BAG) and two versions of the 306 (M/T and A/T).

⁷³⁶ According to the European Communities, the Cakra is a rebadged Hyundai Accent.

⁷³⁷ Indonesia states in the Annex V process that the authorization to import the Timor from Korea applied both to the Timor S-515 and the S-515i and that the S-515i was the only national car produced in Indonesia. Thus, it is clear that at least some -- and perhaps most -- of the Timors to be sold in Indonesia are in fact the S-515i model.

⁷³⁸ While in its rebuttal submission, Indonesia indicated that the Optima's engine generated 66kw/5400rpm, at another point it indicated that the Optima had a power of 87.3/5400rpm. It would appear to the Panel, however, that the 66kw/5400rpm is the correct figure.

⁷³⁹ Indonesia's summary table mentions differences in tyre size in the context of "steering and suspension". It is not clear that tyre size is related to suspension, or is otherwise an important factor in defining the like product. Moreover, Indonesia's detailed table treats tyre size as a separate feature, unrelated to suspension.

⁷⁴⁰ It appears that the car identified by Indonesia as the Daihatsu G102 series is a rebadged Charade. According to figures supplied by Indonesia, the Starlet, Charade and Baleno weigh 725 kg, 845 kg and 965 kg (as opposed to 1055 kg for the Timor), and have engines generating 53, 56 and 87 kw respectively. The Timor S-515i generates 77 kw.

⁷⁴¹ Indonesia's national-car plan sputters under gaze of WTO, in Nikkei Weekly, 17 November 1997.