14.212 Having determined that the EC and US models in the C1 Segment (and arguably those in the C2 Segment) are "like" the subsidized Timor, we consider it appropriate to analyze market shares for the C Segment. ⁷⁴⁸ A review of the data provided by Indonesia under the Annex V procedure demonstrates that the Timor quickly gained a very substantial share of the Indonesian C Segment passenger car market upon its introduction. As shown in Table 1, the Timor was not sold in 1995 and thus had a zero share of the C Segment market. In 1996, the year of introduction, the Timor captured a 16.9 per cent share of the Indonesian C Segment, while during the period January - May 1997 (the latest period for which we have been provided with data), that market share had climbed to 42.4 percent. Table 2 breaks this market share data down on a quarterly basis. These data indicate that Timor had no sales until the fourth quarter of 1996. In that quarter, its market share in the C Segment reached 40.9 per cent. It dropped to 38.8 per cent for the first quarter of 1997 but during the partial second quarter for which we have data (April-May) that share had climbed to 47.7 per cent.

14.213 In assessing whether this change in market share in fact amounted to a displacement or impedance of imports of EC and US origin products into Indonesia, our starting point is actual market shares for the three EC models we have found to be like products, and for the one US model which we assume arguendo to be a like product, to the Timor. The Neon was never introduced into the Indonesia market (allegedly because of the National Car programme), and the market share of US-origin passenger cars in the C Segment of the Indonesian passenger car market was therefore zero. Because the Escort also was never introduced into the Indonesian market, the EC market share data are based solely on sales of the 306 and the Optima. As shown in Table 1, infra at p.385, the European Communities in 1995 had an Indonesian market share in the C Segment of 2.4 per cent. The EC share climbed to 5.7 per cent in 1996, but dropped to 3.7 per cent for the period for which we have data (January - May 1997). Table 2, infra at p. 386, breaks down these data on a quarterly basis. This analysis shows that in the first three quarters of 1996, EC market share in the C Segment ranged from 6.9 to 7.8 per cent. In the fourth quarter of 1996 (the quarter in which the Timor first entered the market), EC market share dropped to 3.7 per cent. In the first quarter of 1997, EC market share fell even further to 3.3 per cent, and remained at a relatively low 4.2 per cent in the partial second quarter for which we have data (April-May 1997).

14.214 Focusing on market shares alone, the data before us show a potentially significant correlation between the introduction of the subsidized Timor and the decline in EC market share in the C Segment of the Indonesian market. The quarterly data show that the EC market share for C Segment cars in Indonesia increased substantially during the first three quarters of 1996, but that, in the fourth quarter of 1996, coincident with the introduction of the Timor, the European models' market share dropped to 3.7 per cent, where it remained on average during the first five months of 1997. Thus, there seems to be little question that the EC market share in the C Segment dropped substantially relative to that of the subsidized Timor, and the close correlation in time between the introduction of the Timor and the drop in EC market share suggests a causal link between the two.

14.215 If Article 6.4 of the SCM Agreement applied in this dispute, this showing of a change in relative market shares to the disadvantage of the non-subsidized like product might well have been sufficient to establish the European Communities' prima facie case of displacement or impedance. ⁷⁴⁹ In the absence of that article, however, it is not enough for the European Communities to demonstrate a decline in relative market share; rather, the European Communities must demonstrate that "the effect of the subsidy is to displace or impede the imports" of an EC-origin "like product" into the Indonesian market, i.e., that some imports that would have occurred did not occur as a result of the subsidies. While declining market share may be relevant to establishing such a situation, we consider that we must proceed further with the analysis and look at actual sales figures for the products in question.

14.216 In spite of their declines in market share, the absolute volume of sales of the relevant EC models did not significantly decline after the introduction of the Timor. Rather, sales of C Segment vehicles from the EC were 419 units in 1995 and 1,445 units in 1996. For 1997 we have full data from the Annex V process on the C Segment only for January - May, and this shows EC sales of 611 units, which amounts to 1,466 units on an annualized basis. We also have sales figures for the Optima (257 units) and 306 (656 units) for the period January - August 1997 from the European Communities. These figures, totalled and annualized, show sales of 1,370 units for 1997. On a quarterly basis, sales of the Optima and 306 remained relatively constant during the period from the fourth quarter of 1996 through May 1997 at between 300 and 400 units per quarter⁷⁵⁰.

14.217 The explanation for the loss of market share with no decline in absolute sales volume is that the size of the Indonesian market expanded after the introduction of the Timor. What is particularly relevant here is that the increase in the size of the market was largely attributable to sales of the Timor. In particular, between the third and fourth quarters of 1996, the market for C Segment cars increased by 6,326 units, of which the Timor accounted for 4,278 units (68 %). A similar pattern was evident during the first five months of 1997. Thus, relatively stable EC sales volumes in a rapidly expanding market resulted in market share declines but not in declines in absolute volumes.

14.218 We agree with the European Communities that a complainant need not demonstrate a decline in sales in order to demonstrate displacement or impedance. This is inherent in the ordinary meaning of those terms. Thus, displacement relates to a situation where sales volume has declined, while impedance relates to a situation where sales which otherwise would have occurred were impeded. The question before us is therefore whether the market share and sales data above would support a view that, but for the introduction of the subsidized Timor, sales of EC C Segment passenger cars would have been greater than they were.

14.219 In a usual case, a decline in market share in a stable or growing market, corresponding in time with the introduction of a subsidized product, might suggest that sales would have been higher but for the introduction of the subsidized product. This would be particularly the case where, in the period prior to the introduction of the subsidized product, the market share of the non-subsidized product had been rising. In this case, however, Indonesia contends that the introduction of the subsidized Timor was itself responsible for the rapid expansion of the market through the introduction of a new, highly affordable passenger car within the reach of first-time buyers. While the European Communities dismisses this argument as "purely speculative," the sales data in terms of volume of sales discussed above provide some support for this view. Thus, one possible interpretation of the data is that, if the subsidized Timor had not been introduced, the Indonesian C Segment market would have remained relatively stable or in any event would have posted during the last quarter of 1996 and the first eight months of 1997 more gradual increases, comparable to those experienced during the period between the first quarter of 1995 and the second quarter of 1996.

14.220 Assuming that, had the subsidized Timor not been introduced, the Indonesian C Segment market would have remained stable or grown at a more moderate rate during the period for which we have data, the question is whether sales of EC C Segment models in absolute terms would have been higher than those actually achieved. Here, the data are inconclusive. While the European Communities contend that its market share had been steadily increasing and that this trend would have continued but for the introduction of the subsidized Timor, actual sales consisted of only two models, the Optima and 306. ⁷⁵¹ As Table 2 shows, quarterly sales data for the Optima do not demonstrate any clear upward trend in the six quarters prior to the introduction of the subsidized Timor. ⁷⁵² Rather, the increase in EC sales was a result of the introduction of the 306 in the first quarter of 1996. While the European Communities state that 1996 sales of the 306 (1,086 units) were 400 units lower than planned, we have no knowledge of the basis for those sales forecasts. Sales of the 306 for partial year 1997 were at an annualized rate of between 984 units (January - August data ) and 1,070 units (January to May data), down slightly from the annualized rate in the first half of 1996 of 1,214 units. ⁷⁵³ Thus, if we assume that in the absence of the Timor the market would have remained stable or continued a more gradual increase through August 1997, ⁷⁵⁴ and that the 306 would have maintained the market share it had achieved in the first half of 1996, these sales might have been expected to increase at most slightly. Such a conclusion is, in any event, highly speculative based on the facts available. ⁷⁵⁵

14.221 Finally, we note the limitations inherent in the data before us. Full information about sales in the C Segment of the Indonesian market is only available through May 1997, i.e., for less than three quarters after the introduction of the Timor. Additional information regarding sales of the Optima and 306 through August 1997 cannot be placed in the context of the overall market situation. Further, the volume of sales of the EC models in question is small, to the point where the statistical significance of the changes in volume may be questioned. We note, for instance, that the sales volume for the Optima in 1995 (a year not affected by the subsidized Timor) jumped from 36 units in the first quarter to 186 units in the second quarter before dropping to 98 and 99 units in the third and fourth quarters espectively.

14.222 In short, the dramatic fall in EC market share in the C Segment is not in this case decisive evidence of displacement or impedance, as the data lend some credence to the Indonesian view that the introduction of the subsidized Timor actually created much of the market growth. Thus, we are required to speculate as to how the market would have performed in the absence of the introduction of the Timor, and as to the share of the market which EC models could have been expected to obtain in that hypothetical situation. It is quite possible that the Indonesian market would have remained stable or increased somewhat in late 1996 and early 1997, even without the introduction of the subsidized Timor, and that EC models would have at least maintained their market share, such that EC sales would have increased slightly. This conclusion is however highly tentative, and does not in our view satisfy the requirement, in the present case, that serious prejudice be demonstrated by positive evidence.

N/A Not available
*The DRI sales figures, on the basis of which the "C" class models were identified, refer to the Bimantara Cakra as the Hyundai Accent, to the Daewoo Nexia as the Daewoo Cielo, and to the Timor as the Kia Sephia.
Source: AV/3, attachment A-39/1-B, except January-August data, which were provided by the European Communities

N/A - Not available.
*The DRI sales figures, on the basis of which the "C" class models were identified, refer to the Bimantara Cakra as the Hyundai Accent, to the Daewoo Nexia as the Daewoo Cielo, and to the Timor as the Kia Sephia.
Source: AV/3, attachment A-39/1-B, except June-August 1997 data, which were provided by the European Communities.

14.225 The complainants have argued that Ford would have introduced the EC-origin Escort in the Indonesian market but for the subsidies provided under the National Car programme to the Timor. Specifically the European Communities have alleged that, "[a]t the time the National Car programme was adopted, Ford was about to start importing CKD Escorts made at its plant in Saarlouis (Germany)." Both the European Communities and the United States have stated that Ford had already committed US$1 million to the Escort programme. The source for these statements would appear to be a joint letter from GM and Ford to the US Trade Representative, dated 27 November 1996. ⁷⁵⁶ This letter indicates that the US$1 million commitment "included production and assembly equipment, tooling, component parts and engineering, all of which were in Indonesia prior to the National Vehicle programme." It further indicates that Ford had "an approved investment plan of $US56.0 million with the feasibility of future investment in assembly to be determined based upon the needs of the market and manufacturing requirements," and that "Ford's total planned investment was committed to the establishment of Escort in the market with both local assembly and increasing local content, ultimately reaching 40% by the fourth year." The European Communities indicated that, under a 1995 business plan, Ford would have exported substantial quantities of CKD Escorts to Indonesia, ⁷⁵⁷ while the joint letter indicated projected sales of 15,100 Escorts in the first four years. ⁷⁵⁸

14.226 The complainants have also alleged the existence of plans to replace the existing Optima with a new model. The European Communities state that Opel had "advanced plans" to expand and upgrade its assembly facilities in Indonesia, including the introduction of a new model of the Optima, while the United States states that GM "was considering investing in plant expansion, and had approval to bring in new models of the Opel Optima and Opel Vectra." The United States asserts that, while the precise figures are confidential, GM's business plan called for sales of Optimas and Vectras "in excess of 1,000 cars in 1996 and around 3,000 cars in 1997, with progressive increases in subsequent years." The joint letter discussed above contains similar information. The parties have not informed the Panel what portion of these planned sales would have been of the Optima.

14.227 In the context of a claim of serious prejudice arising from displacement or impedance, we must review the above information with an eye to whether it demonstrates that (i) there were concrete plans to increase sales of EC-origin passenger cars to the Indonesian market through the introduction of new models; and that (ii) the new models were not introduced because of the subsidies pursuant to the National Car programme.

14.228 Several factual elements have been alleged which could support the European Communities' assertions regarding the existence of concrete plans to introduce the Escort into Indonesia. We note the European Communities' statement that there was an "approved plan" to invest $US56 million for assembly of the Escort in Indonesia. Clearly, such information could be highly relevant to our analysis. The factual information before us with respect to such a plan is, however, extremely limited, and supporting documentation non-existent. Thus, we do not know, for instance, by whom and at what level this plan was "approved." ⁷⁵⁹ Similarly, the fact that Ford projected substantial sales of the Escort in Indonesia could be highly relevant, but we know nothing about the bases for the projections. Under these circumstances, it is difficult for the Panel to judge the weight to be given to these factors. The concreteness of Ford's plans is to some extent confirmed by the statement of the European Communities that Ford had already spent US$1 million to acquire production equipment which was already in Indonesia; such a commitment of resources suggests the Ford had moved beyond mere consideration of the Escort project and was in the process of implementation. ⁷⁶⁰ With respect to the reasons why Ford abandoned its plans to introduce the Escort into Indonesia, however, no direct evidence has been presented by the complainants. The United States cites Ford estimates that, if it had introduced the Escort in the Indonesian market, the Timor would have undercut the price of the least expensive version by at least US$5000. We have no basis to judge, however, whether this assessment had been made at the time Ford abandoned its project nor whether it represented the basis for the abandonment of those plans.

14.229 With respect to the Optima, there is even less factual evidence to support the European Communities' assertions than with respect to the Escort. Basically, the complainants are asking the Panel to conclude that the new model of the Optima would have been introduced but for the National Car programme on the basis of a statement regarding the existence of "advanced plans" and a statement that GM had approval to bring in the new model of the Optima. While this type of information could be highly relevant to the issue, the Panel again lacks any detailed evidence. We have been presented with virtually no information about these "advanced plans," e.g., whether a final decision had been taken within GM management and whether funds had been committed. With respect to the question of approval to bring in the Optima, we assume that this means approval by the Government of Indonesia, but we have no further information. Regarding the reasons why these plans were abandoned, again we have no evidence beyond the statements of company representatives in the joint letter and in newspaper articles. ⁷⁶¹

14.231 The United States stated in its first submission that, prior to the introduction of the National Car programme, Chrysler assembled Jeeps in Indonesia, and that Chrysler "had plans to introduce the Neon but, like GM and Ford, it had to abandon these plans once the National Motor Vehicle Programme was introduced." Its cited sources for this statement were articles in two publications, ⁷⁶² which contain no further information beyond that found in the US submission. In its second submission, the United States repeated that "Chrysler was planning to introduce the Neon to the Indonesian passenger car market, but that Chrysler, too, had to cancel its plans following the introduction of the National Car programme." In its response to a question from the Panel, the United States stated that "the Chrysler Neons would have been sourced from the United States; specifically, from Chrysler's plant in the state of Illinois." The United States indicated in its oral statement at the Panel's second meeting that "Chrysler's plan called for 1,000 - 2,000 Neons in 1997", and in response to a question from the Panel indicated that, "[a]ccording to Chrysler officials, the launch date for the Neon project would have been mid-1997 had the project not been cancelled due to the National Car Programme." Finally, the United States at the second meeting submitted as an exhibit a letter from Chrysler dated 19 December 1997 stating that, "[a]s noted in the attachment, announcement of the National Motor Vehicle Programme rendered non-viable Chrysler plans for additional investment in Indonesia." The full text of that attachment is as follows:

14.232 The evidence before the Panel regarding the intention to introduce the Neon to the Indonesian market is in our view extremely limited. We are told that Chrysler was "studying" an assembly joint venture to assemble the Neon in Indonesia, but we have no basis to judge how concrete those plans were. ⁷⁶³ With respect to the reason why those plans were cancelled, we are presented with no supporting argumentation or evidence beyond the bald statements of the United States and of Chrysler.

To continue with Assessment

⁷⁴⁸ According to data supplied during the Annex V process, only one passenger car model in the C2 Segment currently is sold in Indonesia (the Mitsubishi Lancer). Thus, excluding C2 Segment passenger cars when calculating market shares would increase the absolute market shares of the Timor and of EC C1 models somewhat, but would not affect the relative shares.

⁷⁴⁹ Assuming that the eight months of post-Timor data were considered to be "an appropriately representative period sufficient to demonstrate clear trends in the development of the market for the product concerned" within the meaning of Article 6.4.

⁷⁵⁰ In fact, the quarterly figure for the partial second quarter, if it continued for the rest of the quarter, would actually be 423 units, the highest quarter in the period 1995-second quarter 1997. Treating the period June-August as a quarter, EC C Segment sales amounted to 302 units.

⁷⁵¹ The European Communities pointed in its submissions to the rapid growth in its overall share of the Indonesian passenger car market from 1993 (9.9 per cent) to 1995 (23.72 per cent). However, this rapid growth in market share, which reflects EC gains in high-end vehicles such as Mercedes and BMW, is not relevant to displacement or impedance of "like products" to the Timor.

⁷⁵² It is possible that Optima sales would have increased upon the introduction of a new model. We discuss Opel's plans to introduce a new model of the Optima in paragraphs 14.228 to 14.232 infra.

⁷⁵³ The European Communities have stated that Opel has not ordered any new CKD Optimas since June 1996, although assembly of Optimas (presumably from inventory) continues.

⁷⁵⁴ We note that the Indonesian economy entered a period of serious difficulties beginning in the late summer of 1997. Thus, we must treat the European Communities' figures for January-August 1997 cautiously, particularly in the absence of any overall figures for the Indonesian market after May 1997.

⁷⁵⁵ Sales of the 306 actually climbed slightly from the third quarter (229 units) to the fourth quarter (250 units) of 1996, after the introduction of the Timor. It seems likely, however, that -- as argued by the United States -- the market as a whole in the third quarter was depressed in anticipation of the introduction of the Timor. Thus, third quarter 1996 data may not be reliable.

⁷⁵⁶ See paragraph 8.298 of the Descriptive Part.

⁷⁵⁷ The European Communities contend that, according to a business plan adopted in 1995, Ford would have exported to Indonesia the following quantities of CKD Escorts:

⁷⁵⁸ The joint letter also makes a variety of broader statements regarding Ford's commitment to the Indonesian market. However, in light of evidence in the record suggesting that Ford sources its Indonesian products in a number of countries, including Japan, these more general claims are of limited value to a specific assessment of whether EC-origin like products to the Timor would have been sold in the Indonesian market.

⁷⁵⁹ The plan apparently was not "approved" by the Government of Indonesia, as Indonesia has stated that Ford "did not submit a single application for approval of the development of passenger car production or assembly facilities in Indonesia during the period 1993 to the present." See paragraph 8.357 of the Descriptive Part.

⁷⁶⁰ Indonesia does not challenge that this is a factually correct statement. It does, however, seek to discount the significance of the amount involved in light of Ford's size and huge global sales volume.

⁷⁶¹ The European Communities attaches as an exhibit in support of this statement an article in the Financial Times, GM Halts Indonesia Move Over National Car Policy (date illegible). That article quotes a GM official extensively regarding plans to halt new investments as a result of the National Car programme; however, it states that the official "did not indicate the scale or nature of the expansion being put on hold."

⁷⁶² See paragraph 8.300 of the Descriptive Part.

⁷⁶³ As with respect to Ford, Indonesia states that Chrysler had not applied for approval to develop passenger car production or manufacturing facilities since 1993.