What's New?
 - Sitemap - Calendar
Trade Agreements - FTAA Process - Trade Issues 

espa�ol - fran�ais - portugu�s
Search

World Trade Organization

WT/DS54/R
WT/DS55/R
WT/DS59/R
WT/DS64/R


2 July 1998
(98-2505)
Original: English

Indonesia - Certain Measures Affecting the Automovile Industry

Report of the Panel

(Continued)


(7) Calculation of subsidization from the government-directed $690 million loan

8.104 With respect to the government-directed $690 million loan to TPN, the United States estimates the amount of the subsidy to be 7.1 per cent in 1998, 28.95 per cent in 1998, and 5.3 per cent in 2000. The United States calculated these percentages in the following manner.

8.105 As noted above, newspaper reports indicated that the terms of the loan were 3 per cent over the 3-6 month deposit rate, a maturity of 10 years, and a grace period of 3 years. Thus, the first task is to determine the interest rate to be paid by TPN, which is based on the 3-6 month deposit rate. Because Indonesia did not provide this rate in its Annex V responses, the United States was forced to rely on other sources. The most recent information available to the United States is from The Economist Intelligence Unit Limited (April 1997) (EIU) 457,which indicates that as of the end of February 1997, the average rate on Bank Indonesia certificates was 8.75 per cent. 458 Adding 3 per cent to this figure results in an interest rate on the $690 million loan of 11.75 per cent.

8.106 The next task is to determine the "benchmark" interest rate against which TPN�s interest rate should be judged. Because the Government of Indonesia did not provide the loan to TPN directly, but instead directed banks to provide the loan, it would be inappropriate to use the Government�s cost as the benchmark. Instead, one must use the cost of the actual lender; in this case, the banks. The cost to a bank is reflected in the interest rate that it would charge a comparable commercial borrower. Again, because the Government of Indonesia refused to provide relevant information in its Annex V responses, the United States has been forced to estimate the benchmark interest rate based on the best information otherwise available to it.

8.107 The United States began with the rate on a 10-year "Yankee Bond" that the Indonesian Government began to issue in July 1996. 459 According to EIU, the rate on this US$ denominated bond was set at 7.825 per cent, 1 percentage point above the 10-year Treasury bond rate. 460 The use of a US$ denominated benchmark is appropriate, because in AV/16, p. 4, Answer 13, Indonesia referred to a US$ denominated rate.

8.108 The United States added a spread of seven percent to the Yankee Bond rate of 7.825 per cent. According to EIU, p. 28 (Exhibit 16), Indonesian bank spreads range from three to seven percentage points, depending on the reputation of the client. As discussed above, TPN was not a reasonable credit risk for a loan of the magnitude of $690 million. 461 Therefore, a bank would have charged the maximum spread to a commercial borrower comparable to TPN. Adding 7 per cent to the Yankee Bond rate results in an interest rate of 14.825 per cent.

8.109 Finally, the United States added on a "risk premium" to more accurately reflect the fact that TPN was an unsound credit risk. In other words, a bank lending to a borrower comparable to TPN would have charged more than 14.825 per cent in order to cover its (the bank�s) costs. In the absence of any other information, the United States relied on the methodology used by the US Department of Commerce to calculate a risk premium under the US countervailing duty law. Under this methodology, in the case of a company considered to be "uncreditworthy," the Department of Commerce adds to the "benchmark" interest rate an amount equal to 12 per cent of the prime rate in the country in question. Here, the United States used 10.825 per cent as the prime rate. This was based on the Yankee bond rate (7.825) plus 3 per cent, 3 per cent being the low spread for bank lending. 462 12 per cent of 10.825 equals 1.29 per cent. Adding 1.29 per cent to the benchmark rate of 14.825 results in a final benchmark rate of 16.124.

8.110 Having determined the interest rate paid by TPN (11.75 per cent) and the cost to the banks (16.124 per cent), we next must calculate a "grant equivalent" of the benefit received by TPN from the loan. The first step in this process is to calculate the net present value of payments under the loan to TPN and the benchmark loan. The difference in payments between the two loans constitutes the subsidy.

8.111 Assuming that there will be interest-only payments during the 3-year grace period, and assuming an 11.75 per cent interest rate, the payment schedule for the loan to TPN is as follows:

Table 18

Payment Schedule for Loan to TPN

Year

Beginning

Balance

Add:

Interest

Less:

Payment

Ending

Balance

0

0

0

(690,000,000)

690,000,000

1

690,000,000

81,075,000

81,075,000

690,000,000

2

690,000,000

81,075,000

81,075,000

690,000,000

3

690,000,000

81,075,000

81,075,000

690,000,000

4

690,000,000

81,075,000

149,994,645

621,080,355

5

621,080,355

72,976,942

149,984,645

544,062,652

6

544,062,652

63,927,362

149,994,645

457,995,368

7

457,995,368

53,814,456

149,994,645

361,815,179

8

361,815,179

42,513,284

149,994,645

254,333,817

9

254,333,817

29,884,224

149,994,645

134,223,396

10

134,223,396

15,771,249

149,994,645

(0)

8.112 Using 16.124 per cent as the interest rate, the payment schedule for the benchmark loan is as follows:

Table 19A

Payment Schedule for Benchmark Loan

Year

Beginning

Balance

Add:

Interest

Less:

Payment

Ending

Balance

0

0

0

(690,000,000)

690,000,000

1

690,000,000

111,255,600

111,255,600

690,000,000

2

690,000,000

111,255,600

111,255,600

690,000,000

3

690,000,000

111,255,600

111,255,600

690,000,000

4

690,000,000

111,255,600

171,477,239

629,778,361

5

629,778,361

101,545,463

171,477,239

559,846,584

6

559,846,584

90,269,663

171,477,239

478,639,008

7

478,639,008

77,175,754

171,477,239

384,337,523

8

384,337,523

61,970,582

171,477,239

274,830,866

9

274,830,666

44,313,729

171,477,239

147,667,355

10

147,667,355

23,809,884

171,447,239

0

8.113 Having calculated payment schedules for both the loan to TPN and the benchmark loan, we now must calculate the net present value of the payment differentials. This results in the following schedule, using a standard net present value calculation and a discount rate of 16.124 per cent:

Table 19B

Payment Differential

Year

TPN Loan

Payment

Benchmark Loan

Payment

Payment

Differential

NPV of

Pmt. Diff.

0

0

0

0

0

1

81,075,000

111,255,600

30,180,600

25,989,976

2

81,075,000

111,255,600

30,180,600

22,381,227

3

81,075,000

111,255,600

30,180,600

19,273,559

4

149,994,645

171,477,239

21,482,594

11,614,049

5

149,994,645

171,477,239

21,482,594

10,173,650

6

149,994,645

171,477,239

21,482,594

8,761,023

7

149,994,645

171,477,239

21,482,594

7,544,541

8

149,994,645

171,477,239

21,482,594

6,496,970

9

149,994,645

171,477,239

21,482,594

5,594,855

10

149,994,645

171,477,239

21,482,594

4,818,001

Total

240,919,960

122,847,850

8.114 Having calculated a "grant equivalent" of the loan payment differential of US$122,847,850, the next step is to translate this lump sum into annual benefits. For this purpose, the United States has prorated the amount of the "grant equivalent" over a 10-year period based on the life of the loan to TPN. An amount is added to each annual allocation to account for the time value of money of the remaining unallocated portion, using a discount rate of 16.124 per cent. Using a "declining balance" formula, these two amounts are added together and divided by one plus the discount rate to obtain the amount of the "grant equivalent" allocable to any one year. This results in the following schedule of annual benefits:

Table 20

Annual Benefit from Loan

Year

Benefit to Year

0

0

1

27,636,640

2

25,930,879

3

24,225,117

4

22,519,355

5

20,813,594

6

19,107,832

7

17,402,070

8

15,696,308

9

13,990,547

10

12,284,785

8.115 The final calculation step is to translate the annual subsidy amount into an ad valorem percentage:

Table 21

Annual Subsidy Rate for Loan (1998-2000)

1998

1999

2000

Benefits in US$

27,636,640

25,930,879

24,225,117

Sales in US$

387,569,0434463

89,560,000

453,180,000

Subsidy

7.1%

28.95%

5.3%

(c) Arguments of Indonesia regarding serious prejudice as a cause of action in this dispute

8.116 The subsidies at issue technically fall within the scope of Article 3.1(b) as "subsidies contingent (whether solely or as one of several other conditions) upon the use of domestic over imported goods". As Indonesia is a developing country, it is within the ambit of Article 27.3 of the SCM Agreement, which provides that "[t]he prohibition of paragraph 1(b) of Article 3 shall not apply to developing country Members for a period of five years...from the date of entry into force of the WTO Agreement (i.e., until 1 January 2000)". Instead, the provisions of Articles 5 to 7 regarding "actionable" subsidies apply.

8.117 Subsidies which are not "prohibited" may be either "actionable" or "non-actionable." The subsidies involved in this dispute do not meet the criteria of Article 8 of the Agreement, and so must fall into the residual category of "actionable" subsidies. This rationale for application of Articles 5 to 7 must be used because the language of Article 27.7 is ambiguous. It states that:

The [remedial] provisions of Article 4 shall not apply to a developing country Member in the case of export subsidies which are in conformity with the provisions of paragraphs 2 through 5. The relevant [remedial] provisions in such a case shall be those of Article 7. (Emphasis added.)

8.118 "Export" subsidies and "domestic content" subsidies are not synonymous. The former are covered by Article 3.1(a) and Annex I of the Agreement, while the latter are covered by Article 3.1(b). Given the context of Article 27.7 (including its citation to Article 27.3), the drafters of the Agreement clearly meant Article 27.7 to cover domestic content subsidies, as well as export subsidies. This is confirmed by the analysis above. Under Article 32 of the Vienna Convention on the Law of Treaties, recourse to supplemental means of interpretation is permissible in this instance due to the ambiguity of the text of Article 27.7.

8.119 The European Communities and the United States assert that the actual and alleged subsidies of the National Car programme bestowed by the June 1996 programme, the February 1996 programme and the $US690 million loan exceed the 5/15 per cent ad valorem thresholds established by Article 6.1(a) of the Subsidies Agreement. They properly recognize, however, that Article 27.8 stipulates that serious prejudice in terms of Article 6.1(a) shall not be presumed where, as here, the subsidy is granted by a developing country Member. Rather, in such a case, a complainant must demonstrate serious prejudice by positive evidence in accordance with the provisions of paragraphs 3 through 8 of Article 6.

8.120 Thus, whether proceeding on the basis of Article 6.1(a) or independently on the bases of Article 6.3, complainants must demonstrate serious prejudice by positive evidence in accordance with the provisions of paragraphs 4 through 8 of Article 6 before the remedial powers of Article 7.8 may be applied.

8.121 In response to a question from the Panel, Indonesia indicated that the approximate ad valorem amount of subsidization conferred on the Timor by the exemption from the luxury tax was as follows:

CBUs imported from Korea

29.54%

Timors assembled at Tambun

26.20%

Timors produced at Karawang

18.68%

To continue with Like product


457 US Exhibit 16.

458 According to EIU, these certificates are the primary tools used by Bank Indonesia to control interest rates and money supply.

459 See EIU, p. 3 (US Exhibit 16).

460 Because it is highly unlikely that anyone but the US Government is in the business of issuing US$ denominated Treasury bonds, the United States assumed that the reference in EIU to the 10-year Treasury bond rate is to the 10-year US Treasury bond rate.

461 In this regard, the United States emphasizes that in the Annex V process, Indonesia refused to answer questions regarding TPN�s financial situation. See AV/15, Question 12/29(d) and AV/16, Question 12/29(d).

462 EIU, p. 28 (US Exhibit 16).

463 The United States has assumed that payment on the TPN loan will not begin until 1998, thus rendering 1998 "Year 1" of the loan. In addition, pursuant to paragraphs 2 and 3 of Annex IV of the SCM Agreement, the denominator in the ad valorem subsidy calculation is based on sales in the preceding year. Accordingly, the denominator for the 1998 subsidy is based on 1997 sales, the subsidy for the 1999 subsidy on 1998 sales, and the subsidy for the 2000 subsidy on 1999 sales.

The figures for 1998 and 1999 sales (Column 1999 and Column 2000) were taken from Attachment A-28, AV/14. The figure for 1997 sales (Column 1998) was estimated based on data in Attachment U-12, AV/14. The United States calculated the total sales value of cars imported during a particular year by calculating an average sales price.