EC, December 17, 1998, No 98-C 394/04

THE COMMISSION OF THE EUROPEAN COMMUNITIES

A. INTRODUCTION

Articles 5, 6 and 7 of Council Regulation (EC) No 2026-97 of 6 October 1997 on protection against subsidised imports from countries not members of the European Community (Î) ("Regulation 2026-97') contain provisions on the calculation of the amount of subsidy. The purpose of this communication is to explain the application of these provisions in more detail, using examples where necessary, in order to clarify the methodology which will normally be used by the Commission in calculating the amount of subsidy in countervailing duty cases, unless special circumstances justify a departure from this methodology. In this way it is intended to render the process of calculation more open and to introduce greater certainty for economic operators and foreign governments.

This communication does not bind the Community institutions in any way, but rather provides guidelines solely for the purpose of conducting countervailing duty investigations under Regulation 2026-97.

The Commission intends to update this communication as necessary, in the light of experience in the administration of the Regulation and taking into account the practice of our major trading partners in this area.

B. BACKGROUND TO CALCULATION

(a)The Community and its Member States are members of the World Trade Organisation (WTO) and as such are bound by the provisions of the WTO agreement on subsidies and countervailing measures ("the Subsidies Agreement'), when carrying out countervailing duty investigations. Regulation 2026-97 implements the relevant provisions of this Agreement and lays down rules on the basis for the imposition of countervailing duty measures on injurious subsidised imports (Ï).

(b) Pursuant to this Regulation, such imports can be subject to measures if it is determined, following an investigation, that:

(i) they benefit from a countervailable subsidy;

(ii) they cause or threaten material injury to the Community industry producing the like product;

(iii) countervailing measures are in the Community interest.

(c) If measures are found to be warranted, they would normally take the form of a countervailing duty or of an undertaking from an exporter or the government of the subsidising country. The purpose of either type of measure is to offset the effect of injurious subsidisation. Therefore, as is explained in Article 15(1) of Regulation 2026-97:

(i) the amount of countervailing duty shall not exceed the amount of countervailable subsidy found;

(ii) the amount of countervailing duty should be less than the amount of subsidy, if such a lesser duty would be adequate to remove injury to the Community industry.

In either case, it is clearly imperative to know the exact amount of subsidy. This requires a methodology for calculation of the amount of subsidy.

This communication sets out a workable calculation methodology. It is important to note that only measures that constitute a countervailable subsidy are covered by these recommendations.

C. BENEFIT TO THE RECIPIENT

Subsidies can take many forms. Article 2 of Regulation 2026-97 defines a subsidy as either

-a financial contribution by a government,

or

-any form of income or price support within the meaning of Article XVI of the GATT 1994, which confers a benefit.

Pursuant to the rule laid down in Article 5 of Regulation 2026-97, the calculation of the benefit shall reflect the amount of subsidy found to exist during the investigation period and not simply the face value of the amount at the time it is transferred to the recipient or foregone by the government.

Thus, the face value of the amount of the subsidy has to be transformed into the value prevailing during the investigation period through the application of the normal commercial interest rate.

D. CALCULATION OF SUBSIDY PER UNIT/AD VALOREM

Regulation 2026-97, based as it is on the WTO Subsidies Agreement, assumes that an important effect of a subsidy is always to reduce a firm's costs and the methodology adopted to calculate CVD's therefore reflects this. The objective of the calculation is to arrive at the amount of subsidy per unit of production (Article 7(1) of Regulation 2026-97) during the investigation period (see section F). In the case of consumer products, such as television sets, the appropriate unit would be each individual item. If bulk products, such as fertilisers or chemicals, are involved, it would be appropriate to calculate the subsidy, say, per tonne, or other appropriate unit of measurement. Therefore, the simplest type of subsidy to calculate is that granted on a per unit basis (example 1).

The per unit subsidy can be converted into an ad valorem rate at the Community frontier by expressing the per unit subsidy as a percentage of the average cif. (duty unpaid) unit import price.

In this way it can be established whether the subsidy amount is de minimis, since this is expressed ad valorem in Article 14(5) of Regulation 2026-97 (1 % for imports from developed countries; 2 to 3 % for developing countries). In certain circumstances, it may also be considered to be appropriate to express the countervailing duty on an ad valorem basis.

E. CALCULATION OF CERTAIN TYPES OF SUBSIDY

(a) Grant

Introduction

In the case of a grant (or equivalent) where none of the money is repaid, the value of the subsidy is the amount of the grant corrected for any differences between the point in time of its receipt and the investigation period, i.e. the period in which the production or sales are allocated.

Therefore if the grant is expensed during the investigation period, (that is, its amount is entirely allocated to production or sales during this period), the interest that would have accrued during that period will normally be added. If however the grant is allocated over a longer period than the investigation period, the interest will be added as described in section F (a)(ii).

Any lump sum of revenue transferred or foregone (e.g. income tax or duty exemption, rebates, money saved from preferential provision of goods and services or gained from excessive prices for the purchase of goods) is considered as being equivalent to a grant (see examples 1, 1(i), 3, 4, 5).

Specific examples of grants or equivalent

In order to establish the full amount of subsidy, all of the amounts in specific examples below should be increased by interest as described in the introduction; the total amount of subsidy also depends on whether the subsidy is allocated or expensed.

(i) Direct transfer of funds

The simplest case. The amount of subsidy is the amount received by the company concerned (a subsidy to cover operating losses would fall into this category).

(ii) Tax exemptions The amount of subsidy is the amount of tax that would have been payable by the recipient company at the standard applicable tax rate during the investigation period.

(iii) Tax reductions The amount of subsidy is the difference between the amount of tax actually paid by the recipient 17.12.98 EN Official Journal of the European Communities C 394-7 company during the investigation period and the amount that would have been paid at the normal rate of tax.

(The same method should be applied to all other exemptions and reduction of obligation, e.g. import duties, social security contributions, redundancy payments)

(iv) Accelerated depreciation

Accelerated depreciation of assets under a government agreed programme should be considered as a tax reduction. The amount of subsidy is the difference between the amount of tax that would have been paid during the investigation period under the normal depreciation schedule for the assets concerned, and the amount actually paid under accelerated depreciation. To the extent that the accelerated depreciation results in a tax saving for the company concerned during the investigation period, there is a benefit.

(v) Interest rate subsidies In the case of an interest rate subsidy, the amount of subsidy is the amount of interest saved by the recipient company during the investigation period.

(b) Loans

Basic methodology

(i) In the case of a loan from the government (where repayment does take place) the subsidy is the difference between the amount of interest paid on the government loan and the interest normally payable on a comparable commercial loan during the investigation period (see example 2(i)).

(ii) A comparable commercial loan would normally be a loan of a similar amount with a similar repayment period obtainable by the recipient from a representative private bank operating on the domestic market.

(iii) In this regard, the commercial interest rate should preferably be established on the basis of the rate actually paid by the company concerned on comparable loans from private banks. If this is not possible, the investigation should consider the interest paid on comparable private loans to companies in a similar financial situation in the same sector of the economy, or, if information on such loans is not available, to any comparable private loan made to companies in a similar financial situation in any sector of the economy.

(iv) If there are no comparable commercial lending practices on the domestic market of the exporting country, the interest rate on a commercial loan may be estimated with reference to indicators of the economic situation prevailing at the time, (notably the inflation rate) and the situation of the company concerned.

(v) If all or part of a loan is forgiven or defaulted on, the amount not repaid will be treated as a grant depending on whether there was a guarantee. Specific cases

(vi) It should be noted that tax deferrals, or the deferral of any other financial obligation, should be considered as interest free loans and the amount of subsidy calculated as above.

(vii) In the case of reimbursable grants, these should also be considered as interest free loans until they are reimbursed. If they are not reimbursed, in whole or in part, they will be considered as grants rather than interest free loans from the date on which non reimbursement is established. From this date, the normal grant methodology will apply. In particular, if the grant is to be allocated over time, such allocation would start on the established date of non reimbursement. The amount of subsidy would be the amount of the grant, minus any repayments.

(viii) The same approach would apply to contingent liability loans. To the extent that such loans are given at a preferential rate of interest, the subsidy would be calculated as in paragraph (i). However, if it were to be determined that the loan would not be repaid, it would be treated as a grant from the date on which non repayment was established. The amount of subsidy would be the amount of the loan, less any repayments.

(c) Loan guarantees

(i) In general, a loan guarantee, by eliminating to some extent the risk of default by the borrower to the lender, will normally enable a firm to borrow more cheaply than would otherwise be the case. If the government provides the guarantee, the fact that loans are obtained at a lower interest rate than would otherwise be the case does not mean there is a subsidy, provided that the guarantee is financed on a commercial basis, since the financing of such a viable guarantee by the company would be assumed to offset any benefit of a preferential interest rate.

(ii) In this situation, it is considered that there is no benefit to the recipient if the fee which it pays to the guarantee programme is sufficient to enable the programme to operate on a commercial basis, i.e. to cover all its costs and to earn a reasonable profit margin. In such a situation, it is presumed that the fee covers the risk element involved in obtaining a lower interest rate. If the guarantee programme is viable during the investigation period as a whole and the recipient has paid the appropriate fee, there is no financial contribution from the government and therefore no subsidy, even if the recipient involved were to default on its loans during the period.

If the scheme is not viable, the benefit to the recipient is the difference between the fees actually paid and the fees which should have been paid to make the programme viable, or the difference between the amount the firm pays on the guaranteed loan and the amount that it would pay for a comparable commercial loan in the absence of the government guarantee, whichever is the lower.

(iii) In the case of ad hoc guarantees (i.e. not port of a programme), it should first be ascertained whether the fees paid correspond to those charged to other companies in a similar position which benefit from viable loan guarantee programmes. If so, there would normally be no subsidy; if not, the method explained in (ii) above would apply.

(iv) If no fees are paid by the recipient, the amount of subsidy is the difference between the amount the firm pays on the guaranteed loan and the amount that it would pay for a comparable commercial loan in the absence of the government guarantee.

(v) The same calculation principles would apply to credit guarantees, i.e. where the recipient is guaranteed against credit defaults by its customers.

(vi) In the particular area of export credits and guarantees, the provisions of the OECD arrangement in this area would of course guide our approach.

(d) Provision of goods and services by the government

Principle

(i) The amount of subsidy as regards the provision of goods or services by the government is the difference between the price paid by firms for the goods or service, and adequate remuneration for the product or service in relation to prevailing market conditions, if the price paid to the government is less than this amount. Adequate remuneration has normally to be determined in the light of prevailing market conditions on the domestic market of the exporting country, and the calculation of the subsidy amount must reflect only that part of the purchases of goods or services which are used directly in the production or sale of the like product during the investigation period. Comparison with private suppliers

(ii) As a first step, it must be established whether the same goods or services involved are provided both by the government and by private operators. If this is the case, the price charged by the government body would normally constitute a benefit to the extent that it is below the lowest price available from one of the private operators to the company involved for a comparable purchase. The amount of subsidy would be the difference between these two prices. If the company involved has not made comparable purchases from private operators, details should be obtained of the price paid by comparable companies in the same sector of the economy or, if such data is not availabe, in the economy as a whole. The amount of subsidy would be calculated as above.

Government monopoly suppliers

(iii) If, however, the government is the monopoly supplier of the goods or services involved, they are considered to be provided for less than adequate remuneration if certain enterprises or sectors benefit from preferential prices. The amount of subsidy will be the difference between the preferential price and the normal price.

If the goods and services in question are widely used in the economy, a subsidy will only be specific if there is evidence of preferential pricing to a particular firm or sector.

It may be that per unit prices charged vary according to neutral and objective criteria, for example large consumers pay less per unit than small ones, as sometimes happens in the provision of gas and electricity. In such situations, the fact that certain enterprises benefit from more favourable prices than others would not mean that the provision in this case was necessarily made for less than adequate remuneration, provided that the pricing structure in question was generally applied throughout the whole economy, without any preferential prices being given to specific sectors or firms. The amount of subsidy is in principle the difference between the preferential price and the normal price charged to an equivalent company, according to the normal structure.

(iv) However, if the normal price is insufficient to cover the supplier's average total costs plus a reasonable profit margin (based on sector averages), the amount of subsidy is the difference between the preferential price and the price which would be required to cover the above costs and profit.

(v) If the government is the monopoly supplier of a good or service with a specific use, e.g. television tubes, the question of preferential pricing does not arise, and the amount of subsidy will be the difference between the price paid by the firm involved and the price required to cover the supplier's costs and profit margin.

(e) Purchases of goods by government

(i) In a situation where private operators purchase the kind of goods in question as well as the government body, the amount of subsidy is the extent to which the price paid for the like product by the government exceeds the highest price offered for a comparable purchase of the same goods by the private sector.

(ii) If the company involved has not made comparable sales to private operators, details should be obtained of the price paid by private operators to comparable companies in the same sector of the economy, or, if such data is not available, in the economy as a whole. In such a case, the amount of subsidy should be calculated as above.

(iii) If the government has a monopoly for the purchase of the goods in question, the amount of subsidy as regards the purchase of goods by the government is the extent to which the price paid for the goods exceeds adequate remuneration. Adequate remuneration in this situation is the average costs incurred by the firm selling the product during the investigation period, plus a reasonable amount of profit, which will have to be determined on a case by case basis.

The amount of subsidy is the difference between the price paid by the government and adequate remuneration as defined above.

(f) Government provision of equity capital

(i) Government provision of equity capital is not considered as conferring a benefit, unless the investment decision can be regarded as inconsistent with the usual investment practice (including for the provision of risk capital) of private investors in the exporting country concerned.

(ii) Therefore, the provision of equity capital does not of itself confer a benefit. The criterion is whether a private investor would have put money into the company in the same situation in which the government provided equity. On the basis of this principle, the matter has to be dealt with on a case by case basis, taking account of the Commission's practice as regards State aid policy in this area and the practice of the Community's main trading partners.

(iii) Clearly, if the government buys shares in a company and pays above the normal market price for these shares (taking account of any other factors which may have influenced a private investor), the amount of subsidy is the difference between the two prices.

(iv) As a general rule, in cases where there is no market in freely traded shares, the government's realistic expectation of a return on the price paid for equity should be considered. In this regard, the existence of an independent study demonstrating that the firm involved is a reasonable investment is the best evidence; if this is not present, the onus is on the government to demonstrate on what basis it can justify its expectation of a reasonable return on investment.

(v) If there is no market price and the equity injection is made as part of an ongoing programme of such investments by the government, close attention should be paid not just to the analysis of the firm in question, but to the overall record of the programme over the last few years. If the records show that the programme has earned a reasonable rate of return for the government, there should be a presumption that the government is acting according to the usual investment practice of private investors with regard to the case in question. If the programme has not generated a reasonable return, the onus is put on the government to demonstrate on what basis it can justify its expectation of a reasonable return on investment.

(vi) The existence of a subsidy is determined by the information available to the parties at the time the equity injection is made. Thus, if an investigation considers an equity injection that was made several years before, the fact that the company has performed less well than expected does not mean that a subsidy exists, provided that the expectation of a reasonable return was justified in the light of the facts know at the time of the provision of equity.

On the other hand, a subsidy might exist even if a reasonable return has been achieved, if at the equity injection the prospect of such a return was so uncertain that no private party would have made the investment.

(vii) In cases where there is no market price for the equity and there is a subsidy and a benefit, i.e. the government has not acted according to the usual investment practice of private investors, all or part of the equity provided must be considered as a grant. A decision to consider all of the equity a grant would be made only in extreme cases where it is determined that the government had no intention of receiving any return on its investment and was in effect giving a disguised grant to the firm in question.

A decision on what portion of the equity to treat as a grant would depend on how near the government has come to meeting the private investor standard. This determination will have to be made on a case by case basis.

(g) Forgiveness of government held debt

Forgiveness of debt held by government or government owned banks relieves a company of its repayment obligations and should therefore be treated as a grant. If the subsidy is to be allocated, the allocation period should begin at the time of the forgiveness of the debt. The amount of subsidy will be the outstanding amount of the debt forgiveness (including any interest accrued).

F. INVESTIGATION PERIOD FOR SUBSIDY - CALCUÜ LATION EXPENSE VERSUS ALLOCATION

The amount of subsidy should be established during an investigation period, which should normally be the most recent financial year of the beneficiary enterprise (Article 4(1) of Regulation 2026-97). Although any other period of six months prior to initiation may be used, it is preferable to use the most recent financial year, since this will enable all appropriate data to be verified on the basis of audited accounts.

As many subsidies have effects for a number of years, subsidies granted before the investigation period should also be investigated in order to determine what portion of such subsidy is attributable to the investigation period (see below).

(i) If the subsidy is granted on a per unit basis, for example, an export rebate granted per unit of product, the per unit calculation normally consists of taking the weighted average value of the rebate over the investigation period (example 1).

(ii) Other kinds of subsidy are not readily expressed on a per unit basis, but involve a global sum of money which has to be allocated to each unit of product as appropriate.

Two exercises may have to be carried out, in this respect:

-Attribution to the investigation period of a portion of those subsidies granted before the investigation period but whose effects extend over a number of years (Article 7(3) of Regulation 2026-97).

-Allocation of the subsidy amount attributed to the investigation period per unit of the like product. In this case, the appropriate denominator for such allocation has to be selected (Article 7(1) and (2) of Regulation 2026-97).

(a) Attribution of a subsidy amount to the investigation Period

(i) Many types of subsidy, e.g. tax incentives and preferential loans are recurring and the effect is felt immediately after granting. Thus, the amount granted to the beneficiary can be expensed in the investigation period. The expensed amount should normally be increased by the annual commercial interest rate, to reflect the full benefit to the recipient, on the assumption that the beneficiary would have had to borrow the money at the beginning of the period and repay it at the end.

(ii) For non recurring subsidies, which can be linked to the acquisition of fixed assets, the total value of the subsidy has to be spread over the normal life of the assets (Article 7(3) of Regulation 2026-97). Therefore the amount of subsidy from, for example, a grant (for which it is assumed that it is used by the beneficiary to improve its competitiveness in the long term, and thus to purchase product assets of one kind or another), can be spread over the normal period used in the industry involved for the depreciation of assets. This will normally be done using the straight line method. For example, if the normal depreciation period was five years, 20 % of the value of the grant would be allocated to the investigation period (see example 6).

The approach of allocating over time means that non recurring subsidies granted several years before the investigation period can still be countervailed provided that they still have an effect during the investigation period.

Conceptually, this kind of allocation is equivalent to a series of annual grants, each having en equal amount. In order to determine the benefit to the recipient, the appropriate annual commercial interest rate should be added to each grant, to reflect the benefit of not having to borrow the money on the open market. In addition, in order to reflect the full benefit to the recipient of having a lump sum of money at its disposal from the beginning of the allocation period, the amount of subsidy should be increased by the average amount of interest which the recipient would expect to earn on the non depreciated amount of total grant over the whole period of allocation.

(iii) As an exception to (ii), non recurring subsidies which amount to less than 1 % ad valorem will normally be expensed, even if they are linked to the purchase of fixed assets.

(iv) In the case of recurring subsidies linked to the acquisition of fixed assets, e.g. import duty exemptions on machinery, which date back to before the investigation period, the benefits accruing from previous years within the depreciation period should be taken into account and the appropriate amount attributed to the investigation period (see example 7).

(v) In addition, recurring subsidies granted in large, concentrated amounts prior to the investigation period, can in certain circumstances be allocated over time if it is determined that they are likely to be linked to the purchase of fixed assets and still confer a benefit during the investigation period.

(vi) Consequently, in the case of subsidies expensed as in paragraphs (i) and (iii) no subsidies granted before the investigation period should be taken into account. For subsidies allocated over time, as in (ii), (iv), and (v), subsidies granted prior to the investigation period must be considered.

(vii) A more detailed illustrative table of subsidies to be allocated or expensed is included as Annex 1. This table is for illustrative purposes only; certain types of subsidy may be subject to a case by case analysis when deciding whether to expense or allocate.

(b) Appropriate denominator for allocation of subsidy amount

Once the subsidy amount to be attributed to the investigation period has been established, the per unit amount is arrived at by allocating it over the appropriate denominator, consisting of the volume of sales or exports of a product concerned.

(i) As regards export subsidies (Article 3(4)(a) of Regulation 2026-97) the appropriate denominator for allocation is the export volume during the investigation period, since such subsidies benefit only exports (see examples 2 and 3).

(ii)For non export subsidies the total sales (domestic plus export) should normally be used as the denominator, since such subsidies benefit both domestic and export sales (see example 4).

(iii) If the benefit of a subsidy is limited to a particular product, the denominator should reflect only sales of that product. If this is not the case, the denominator should be the recipient's total sales.

G. DEDUCTION FROM AMOUNT OF SUBSIDY

1. Article 7(1) of Regulation 2026-97 provides that only the following may be deducted from the amount of subsidy:

(i) Any application fee, or other costs necessarily incurred in order to qualify for, or to obtain, the subsidy

It is important to note that it is up to the exporter in the country concerned to claim a deduction; in the absence of such a claim accompanied by verifiable proof, no deduction will be granted. The only fees or costs that may normally be deducted are those paid directly to the government in the investigation period. It must be shown that such payment is compulsory in order to receive the subsidy. Thus payments to private parties, e.g. lawyers, accountants, incurred in applying for subsidies, are not deductible. Neither are voluntary contributions to governments, for example donations.

(ii) Export taxes, duties or other charges levied on the export of a product to the Community specifically intended to offset the subsidy

Such claims for deductions should only be accepted if the charges involved were levied during the investigation period, and it is established that they continue to be levied at the time when definitive measures are recommended.

2. No other deductions can normally be made from the amount of subsidy. No allowance can be made for any tax effects of subsidies or for any other economic or time value effect beyond that which is specified in this communication.

EXAMPLES OF SUBSIDY CALCULATION

For the purpose of these examples, the following assumptions are made:

1. The investigation period is the year 1996

2. The interest rate, with an annual repayment period, is 25 %

3. The depreciation period for machinery is five years

4. The import duty on machinery is 50 %

5. The like product is called product X, measured in tonnes

6. There are three producers in the exporting country; each exporting 100 000 tonnes per year. They are referred to as companies A, B and C

7. In all the examples, the amount of money is expressed in current ecus. In practice, the amounts would be denominated in the currency of the exporting country and the duty expressed in ecus at the Community frontier.

(a) Subsidies granted per unit of product

Example 1

(i) Exporters of product X receive a rebate from the government on each tonne exported. This rebate changes according to season. In the first half of 1996, when 200 000 tonnes were exported, the rebate amounted to ECU 5 per tonne; in the second half of the year, when 100 000 tonnes were exported, it rose to ECU 20 per tonne.

On a weighted average basis, the rebate was worth ECU 10 per tonne during the investigation period. Since this is equivalent to a grant, the amount of subsidy is calculated by adding the 25 % interest, which gives a figure of ECU 12,5 per tonne. In practice, since the rebates are granted regularly throughout the year, the company will not have benefited from the interest for the whole year on most transactions. If they can provide the appropriate evidence, the interest amount may be reduced in proportion to the weighted average period when they had access to the full amount granted. Companies A, B and C all benefit from this subsidy.

(ii)Where rebates vary on a seasonal basis, and the investigation establishes that this is a regular procedure, the weighted average amount should be used. If, however, the investigation showed that the ECU 20 per tonne rebate was to be maintained on a permanent basis, this amount, and not the weighted average, should from the basis of the subsidy amount, because this constitutes the actual benefit to the exporter on current shipments.

(b) Subsidies expensed in investigation period

Example 2

Company A - Allocation of export subsidies - company obliged to export all production

(i)Company A receives further benefits if it exports all the product X it produces. This is clearly an export subsidy as benefits are contingent upon export performance. Among other things, it receives an income tax exemption. During the investigation period, it produced and exported 100 000 tonnes. Assuming that it makes ECU 4 million profit on product X during the investigation period and the normal tax rate is 25 %, the value of the exemption in total would be ECU 1 million, which equates to a per tonne value of ECU 10. Since for calculation purposes the tax exemption has an equivalent C 394-14 EN Official Journal of the European Communities 17.12.98 effect to a cash grant, it is appropriate to add the 25 % interest, which makes the value of the subsidy ECU 12,50 per tonne.

(ii)Company A also receives a government loan of ECU 5 million at 5 % interest for the same project, and the normal commercial interest rate is 25 %. Since the loan is repayable, the amount of subsidy would be the difference between the interest paid and the normal amount payable at commercial rates (i.e. ECU 5 million at 20 %) = ECU 1 million for the investigation period. The subsidy per tonne would therefore be ECU 10. This subsidy will recur in future periods if the repayment period for the loan is more than one year; its exact amount will depend upon how much of the capital amount of the loan has been repaid.

Example 3

Company B - Allocation of export subsidy - incremental increase in exports

Company B, rather than being obliged to export all its production, benefits from being able to deduct from its taxable income 20 % of profits made on its increase in export value. This is clearly an export subsidy, since it is contingent upon export performance, and should be allocated only over the export sales of the company. It is equivalent to a cash grant for the purpose of calculation.

Company B produced 200 000 tonnes and increased its export sales from 50 000 to 100 000 tonnes in the investigation period, and the average export price per tonne was ECU 160. This equates to an increase in export value of 50 000  ECU 160 = ECU 8 million, of which ECU 2 million was profit.

20 % of ECU 2 million is ECU 400 000. If the tax rate is 25 %, the saving for the manuÜ facturer is ECU 100 000.

Expressed on a per tonne basis over the 100 000 units exported, the tax saving is ECU 1. By adding 25 % interest, we arrive at the value of the subsidy of ECU 1,25 per tonne.

Example 4

Non export - Allocation over total sales

Instead of receiving export subsidies, company C receives a production subsidy that is specific to the industry concerned and therefore countervailable. The subsidy is allocated over the total sales of the company, not just the exports.

(i) Company C produced and sold 200 000 tonnes of product X, 100 000 on the domestic market and 100 000 for export, in the investigation period. In calculating the amount of subsidy in the case of domestic subsidies, only the total sales volume, not the exports, is relevant.

(ii) Therefore, if the company in question obtained an income tax exemption and made ECU 4 million profit in the investigation period, the normal tax rate being 25 %, the tax saving of ECU 1 million would be allocated over 200 000 units, giving a subsidy of ECU 5 plus 25 % interest = ECU 6,25 per tonne.

Example 5

Preferential purchase of goods or services

For the purpose of this example, it is assumed that one tonne of gas is consumed to produce one tonne of producte X. The gas is supplied by a government owned distribution company, which is the single supplier in the country concerned.

In the absence of any private sector supplies, it is necessary to assess benefit in terms of whether the government supplier's price covers cost and whether there is a preferential price.

ECU 70 per tonne reflects the normal price of gas to large sized industrial consumers as part of a price structure and enables the government supplier to cover its overall costs.

However, in 1996 it was decided, in order to boost the competitiveness of the industry concerned, to reduce the price of gas to this industry only to ECU 50 per tonne. This is a preferential price, since other large industrial users continue to pay ECU 70 per tonne.

All three companies benefit from this subsidy to the extent of the difference between ECU 70 (the normal price) and ECU 50 (the preferential price) = ECU 20 plus the normal 25 % interest = ECU 25 per tonne.

(c) Subsidies allocated over time

Example 6

Grant for purchase of machinery (fixed assets)

Company A is given a non recurring grant of ECU 5 million to purchase machinery used to produce the product for export. The benefit of such a subsidy is spread over the normal depreciation period for such fixed assets. Assuming this to be five years for the industry involved, the annual amount attributed to one year (and thus the investigation period) would be ECU 1 million, which equates to ECU 10 per tonne.

Of course, in reality, a lump sum of ECU 5 million is of far greater value to a company than five annual grants of ECU 1 million each. In order to reflect the full benefit to the recipient, the face amount of ECU 10 has to be increased by the average annual amount of interest that the firm could earn from the non depreciated amount of the grant over the five year allocation. For example, in year 1, this amount would be ECU 5 million  25 % = ECU 1,25 million; in year 5, it would be ECU 1 million  25 % = 0,25 million. Over the period as a whole, the average annual amount of interest would be ECU 0,75 million, equivalent to ECU 7,5 per tonne.

Therefore the amount of subsidy for each of the five years is:

ECU 10 (face amount) plus ECU 7,5 (accumulated interest) = ECU 17,5 per tonne.

This amount can continue to be countervailed for a further four years.

Example 7

Import duty exemption on machinery (fixed Assets)

If the subsidy dates back before the investigation period, the annual value of benefits accruing from previous years within the depreciation period can be added. In the example below, company A has been granted import duty exemption on machinery since 1990. Since machinery is in no way consumed in the production process of the final product, the subsidy is countervailable. For calculation purposes, the amount of duty foregone is equivalent to a grant. Assuming that the investigation period is 1996 and that it imports ECU 1 million of machinery each year, the amount allocated to the investigation period will be ECU 500 000, equivalent to ECU 5 per tonne. The amount of the subsidy, including 25 % interest, is ECU 6,25 per tonne. (See Table 1 for a more detailed explanation).

(The same allocation methodology can be used for other types of subsidy, for example loans).

(d) Total amount of subsidy (ecu per tonne)

<table>

(e) Calculation of countervailing duty

(a) If the amount of subsidy is less that the injury margin, this amount should normally form the basis of the countervailing duty. The countervailing duty is collected at the time of the entry into free circulation of the goods and can be either:

(i) a specific duty, i.e. an amount per tonne, which in the above case would be:

Company A: ECU 83,75, company B: ECU 38,75 and company C: ECU 43,75;

(ii) An ad valorem duty, expressed as a percentage of the cif import price before duty. If, in the above case, the cif unit price was ECU 180 per tonne, the ad valorem duty would be:

Company A: 46,5 %, company B: 21,5 % and company C: 24,3 %.

The amount is rounded down to one decimal place.

(b) Of course, the duty will be limited by the injury margin, if the latter is lower than the subsidy amount.

ANNEX

Table 1

Import duty exemption on machinery - subsidy granted since 1990

Investigation period 1996 - depreciation period five years

Assuming ECU 1 million of machinery imported per year - duty rate 50 %

<Table>

The subsidy on machinery acquired in 1990 and 1991 has dropped out of the calculation for the investigation period (1996). The subsidy on machinery acquired in 1993 or after can continue to be countervailed until the allocation period expires.

Table 2

Illustrative table of subsidies to be expensed or allocated over time

<Table>

(Î)OJ L 288, 21.10.1997, p. 1.

(Ï)These guidelines also apply to investigations concerning

ECSC products conducted pursuant to Commission

(OJ L 245, 4.9.1998, p. 3).