Special Report No 6/2000

concerning the granting by the Community of interest subsidies on loans by the European Investment Bank to small and medium-sized enterprises, through its temporary lending facility, together with the Commission's replies

(pursuant to Article 248(4), second subparagraph, EC)

(2000/C 152/01)

CONTENTS

>TABLE>

INTRODUCTION

Context and objectives

1. At its meeting in Copenhagen in June 1993, the European Council asked the EIB, in conjunction with the Commission, to add ECU 3000 million to the temporary lending facility of ECU 5000 million which had been adopted in Edinburgh in December 1992(1). Of this additional amount, ECU 1000 million was to be devoted to strengthening the competitiveness of small and medium-sized European enterprises (SMEs) by providing interest subsidies of up to a maximum of 3 % per year for five years, the granting of which would be linked to job creation. This measure is known as the temporary facility for SMEs, but is also referred to as the Copenhagen facility. It was created by a Council decision, adopted in April 1994(2) on the basis of Article 235 of the EC Treaty. In the recitals of the Decision, the Council recalled the urgent nature of the Edinburgh facility and stressed that this new facility should be introduced as quickly as possible.

2. The Economic and Social Committee issued a favourable opinion on the proposal for the measure, and considered, in particular, that it was "essential that the Commission and the EIB provide the Member States and the SME umbrella organisations in the Community with a detailed briefing, in good time, on the implementation of the facility and also formally involve the financial institutions and their umbrella groups in this briefing"(3).

Legal framework

3. The Council's Decision of April 1994 lays down a regulatory framework with the following main elements:

(a) the management of the measure by the EIB on behalf of the Community in accordance with a cooperation agreement(4) between the Commission and the EIB;

(b) the granting of the loans by the EIB on the basis of its usual methods;

(c) a total ceiling on subsidised loans of ECU 1000 million;

(d) that decisions to extend subsidised loans were to be taken by the EIB within 12 months of the date of entry into force of the Council Decision, namely before 28 April 1995;

(e) interest subsidies of 2 % per year for a period of five years;

(f) investment projects were to involve job creation;

(g) the drafting of an annual evaluation report by the Commission, giving details of the overall utilisation rate of the loans and their impact on job creation and interest rates(5).

The Court's audit

4. Following the Court's report concerning the financial year 1995, in its resolution of April 1997 giving discharge to the Commission in respect of the implementation of the general budget for the financial year 1995(6) the European Parliament asked the Court of Auditors and the Commission to issue opinions at the earliest opportunity "on the management and effectiveness of the Copenhagen facility, with particular reference to the functioning of the job creation criteria applied to interest rate subsidies".

5. The European Parliament's interest in this measure had been prompted by the Court of Auditors' observations on the legality, regularity and effectiveness of ECSC conversion loans(7). In fact, these loans are similar in many ways to the loans granted under the SME measure, especially as regards the clause linking the interest subsidy to job creation.

6. The audit's aims were, on the one hand, to check that the applicable regulations had been observed and, on the other hand, to evaluate the measure's effectiveness in meeting its specific objectives, which were to promote investment in the SMEs and to create new jobs.

7. All the Member States of the Union were visited in order to examine the implementation arrangements and the impact of the measure in the national context. Audit missions were carried out on the premises of 114 recipients and their financial intermediaries in six Member States.

SUMMARY OF THE MAIN OBSERVATIONS

8. Transfers made by the Commission to the EIB were treated as payments on account, and thus were recorded as final expenditure in the Commission's budgetary accounts. In reality, these sums were only paid over to the agents for distribution to the recipients after a long interval, with the result that at the end of 1994, 1995 and 1996, the real cumulative utilisation rate of the available budgetary funds amounted to 0 %, 11 % and 52 % respectively, whereas the budgetary accounts for these financial years showed utilisation rates of 100 % (paragraph 26).

9. The balance of the account held by the EIB on behalf of the Community does not appear on the Community's balance sheet. The interest revenue credited to this account, totalling ECU 9,6 million, was not established as revenue over the whole period from 1994 to 1998 when it was generated, until it was transferred to the Commission in September 1998. There is a general problem here concerning the rules governing the recording in the accounts of this type of payment on account and the auditing of it (paragraph 27).

10. The Court's examination of the measure's management and control system allowed some of the system's inherent weaknesses to be identified: the payment of aid on submission of a declaration by interested parties; entrusting the management to banks; criteria that were inadequately presented and explained; a lack of real penalties for making a false declaration; a lack of checks on recipients. The Court carried out audits on the premises of a selection of 114 undertakings. Such a selection cannot be used for a mere mathematical extrapolation. However, these audits revealed numerous errors which confirm the reality of the weaknesses (paragraphs 28 to 43).

11. The arrangements put in place restricted equality of access for all potential recipients in certain areas, despite the provisions of the Council Decision and the cooperation agreement (paragraph 45).

12. Contrary to the provisions of Article 4 of the cooperation agreement prohibiting the combination of an SME subsidised loan with other loans subsidised by the EIB under other agreements, a derogation was agreed between the EIB and the Commission in the case of ERDF aid which also takes the form of interest subsidies on the same EIB loans. Since this derogation was not agreed at the same hierarchical level as the cooperation agreement and was not incorporated into the contracts with the intermediaries, it has no legal value. The amounts paid in combination with other aids must therefore be considered to be in breach of the regulations. Furthermore, this exception does not seem to have been applied in all the countries (paragraphs 31 and 32).

13. The Court noted cases where, at the time of the application, the investment was already at an advanced stage of completion and the subsidised loan had actually replaced an existing financial arrangement (paragraph 53).

14. The undertakings which benefited from this measure were certainly creating jobs, but it is impossible to establish whether it was these subsidies which enabled the creation of the jobs in question. Moreover, the arrangements for implementing the measure do not include the information necessary to establish which jobs would not have been created without the subsidised loan (paragraph 57).

15. Coordination with the other aid on offer at regional, national or even Community level was not taken into sufficient consideration (paragraphs 62 and 63).

16. The banks which distributed the subsidies often integrated them so well into the financial packages they offered their customers that the origin of the subsidy disappeared from sight (paragraph 69).

IMPLEMENTATION OF THE MEASURE

Main characteristics

17. Detailed procedures for implementing the measure were set out in a cooperation agreement signed by the Commission and the EIB on 14 June 1994, and also in the addendum to this agreement, which was signed on 26 and 27 July 1995. Table 1 sets out the main data on implementation. The total amount available for the loans, ECU 1000 million, was divided among the Member States primarily on the basis of their populations, their gross national products and their unemployment rates. The sum of ECU 923,2 million was actually granted, resulting in the payment of ECU 92,3 million in interest subsidies(8). There were 4336 successful loan applications(9), and 53788 jobs were declared as satisfying the job creation criteria(10). At the time this report was drafted, 32 subsidy payments, representing a total of ECU 578600, had been cancelled or reduced.

Table 1

Implementation of the measure, by Member State

>TABLE>

18. The loans were granted according to the EIB's global lending mechanism(11). This consists in opening lines of credit in favour of financial intermediaries which then use them in order to finance individual loans made by them to their customers. The intermediaries manage the loans and take on the associated risk. For operations of this kind, the EIB has laid down conditions, of which the main ones are as follows: the customer must be an SME, its sector of activity must be included in a list of eligible sectors and the loan must be to finance an investment project. According to the EIB's criteria, an SME is defined as an undertaking with net fixed assets not exceeding ECU 75 million, fewer than 500 employees and with not more than one third of its equity owned by a company which is not an SME (in other words, not satisfying these criteria). Of these, the EIB favours SMEs with a staff of fewer than 50.

19. The subsidy, which is paid in a single instalment, is equivalent to 10 % of the ecu value of the loan, with a ceiling of ECU 3000 per job created. For this reason, the amount of the loan may not exceed 75 % of the investment, so that the subsidy may not exceed 7,5 % of the investment(12). It should be emphasised that the amount of the subsidy bears no relation to either the nominal loan interest rate or the loan repayment schedule.

Procedure

20. The payment of interest subsidies was carried out by national agents acting under a mandate from the EIB(13). The Community paid them a percentage calculated on the basis of the subsidies paid to the recipients(14). Some agents also acted as financial intermediaries and extended subsidised loans directly.

21. The undertakings involved sent the national agent, via the financial intermediary, one or more applications with details of the loan and investment amounts and the number of jobs they planned to create. Applications had to reach the national agent no later than 31 July 1995, although for certain countries the deadline was postponed to 15 December 1995 (see paragraph 47). Applications were added in order of receipt to a list held by the agent, and this list determined priority in the granting of subsidies on a first come, first served basis, as provided in Article 13 of the abovementioned cooperation agreement.

22. Subsequently, between 1 August 1995 and 30 June 1997, the applicant undertakings sent the national agent sworn statements giving:

(a) staff numbers on 28 April 1994 (publication date of the Decision establishing the measure);

(b) staff numbers on the date of the statement;

(c) a declaration to the effect that there had been an actual and uninterrupted increase in the number of jobs between these two dates during the six months preceding the date of the statement.

On the basis of this document, each agent calculated the subsidy amount and paid it from the funds placed at its disposal by the EIB until these were exhausted.

23. The measure was primarily publicised by the financial intermediaries, in the context of their normal marketing activities. In addition, they distributed a brochure issued by the EIB and aimed at businesses. Table 2 gives a summary of the implementation timetable.

Table 2

Summary of the timetable for implementing the measure

>TABLE>

MANAGEMENT OF BUDGET APPROPRIATIONS AND PRESENTATION OF THE ACCOUNTS

Changes in the appropriations

24. The appropriations to finance the measure were entered in the 1994 to 1996 budgets, under budget heading B5 - 322, "Employment and growth for Europe". The initial total of the appropriations, spread over three financial years, was ECU 106 million. This figure was increased in 1994 by a transfer of ECU 17 million and reduced in 1996 by a transfer of ECU 7,5 million to budget heading B5 - 323 for the pilot project "Growth and environment" managed by the European Investment Fund (EIF). Finally, ECU 25 million was cancelled in respect of the 1996 financial year (Table 3).

25. As a result, the final amount of appropriations used was ECU 93,5 million. The funds were first transferred to an account held by the EIB on behalf of the Community. From this account the EIB paid the interest subsidies claimed by the national agents for recipient undertakings which had fulfilled the conditions. Table 4 shows a summary of the movements on the account.

Table 3

Implementation of appropriations entered under budget heading B5 - 322, "Growth for Europe"

>TABLE>

Table 4

Summary of movements on the account held by the EIB on behalf of the Community

>TABLE>

Management of budget appropriations

26. Initially, the sums paid over by the Commission to the EIB were kept in an account yielding interest at a rate set out in the cooperation agreement(15). These payments were treated as payments on account, and as such were recorded as final expenditure in the Commission's budget accounts. In actual fact, these sums were only paid over to the agents for distribution to the recipients after a considerable interval. For example, whereas 73 % of the total amount of the subsidies was transferred to the EIB in 1994 and 1995, 92 % of the payments from these funds were made in 1996 and 1997. Thus, at the end of 1994, 1995 and 1996, the actual cumulative utilisation rates for the available budgetary funds were 0 %, 11 % and 52 % respectively, while the budget accounts showed utilisation rates of 100 % for these years (Table 5).

Table 5

Utilisation of available budgetary funds

>TABLE>

Presentation of the accounts

27. If the funds available in the EIB account are compared with the payments made (Table 6 and Figure 1), it is clear that there was a time lag between charging to the budget and making the payment. The size of this time lag calls into question the presentation of the transfers as final expenditure, which is in breach of the principles of annuality and transparency of the budget accounts(16). In addition, although the balance of the account held by the EIB on behalf of the Community represents a claim for the Community and a liability for the EIB, it does not appear on the Community's balance sheet. Furthermore, the interest revenue credited to this account, totalling ECU 9,6 million, was not established as revenue over the whole period from 1994 to 1998 when it was generated, until it was transferred to the Commission in September 1998(17). Since the temporary facility for SMEs is just one of several similar mechanisms, there is a general problem here concerning the rules governing the recording in the accounts of this type of payment on account and the auditing of it.

Table 6

Comparison between the amounts paid to agents and the budgetary appropriations available on the account held by the EIB on behalf of the Community

>TABLE>

Figure 1

Comparison between the amounts credited to the account and their utilisation

>PIC FILE= "C_2000152EN.001001.EPS">

(million ECU)

(1) Appropriations in the account at the end of the quarter (in combination)

(2) Payments of subsidies (in combination)

AUDIT OF REGULARITY AND LEGALITY

Results of the on-the-spot checks

Overall findings

28. The Court carried out on-the-spot checks on 144 recipients in six Member States, namely Germany, the United Kingdom, Italy, France, Spain and Austria. These recipients represent 123 payment files, as a number of them received several (two, three or four) of the facility's subsidised loans. During its checks, the Court found 61 errors in the files of 51 recipients, who had been granted a total amount of ECU 4,17 million of subsidies. After eliminating errors counted more than once, the total amount was ECU 1,26 million. The results of the audit findings are summarised in Table 7.

Table 7

Summary of errors found during the on-the-spot checks

>TABLE>

Number of eligible jobs

29. The amount of the subsidies was linked to the number of jobs created (and maintained for at least six months) by the recipient undertaking, with a maximum of 27 for second-category undertakings (see paragraph 58). However, in 23 of the 114 cases audited, the Court found that the number of jobs that were truly eligible was lower than the number being subsidised. For all the recipients audited, the total number of jobs subsidised was 1389, representing a sum of ECU 4,17 million. Two hundred and fifty-six of these posts, equal to a sum of ECU 768000, were with recipients where a discrepancy was found. The total number of discrepancies concerned 124 jobs. This means that ECU 372000 too much was paid out. Moreover, the figures relating to jobs created, whether subsidised or not, as declared by the undertakings, were discovered to be very unreliable: they were incorrect in 84 of the 114 cases examined.

30. Of the 124 posts for which, despite their ineligibility, subsidies were paid, 91 had in fact simply not been created and 33 concerned an undertaking in Spain in category two (see paragraph 58). This means that the ceiling of 27 posts which could be subsidised was obligatory. However, the subsidised loan was sub divided into four loans which were made by four different intermediary banks. Fewer than 27 posts were declared for each of these partial loans, so that the question of the category only arose at the overall level of the 60 jobs declared and subsidised. This fact was completely overlooked by the national agent and the EIB, even though two of the intermediaries declared that the undertaking belonged to the first category and the other two intermediaries declared that it belonged to the second.

Combination with other aid from Community sources prohibited

31. Article 4 of the cooperation agreement stipulated that, for each recipient, the SME subsidised loan could not be combined with other subsidised loans granted by the EIB under other agreements, or with ECSC-subsidised loans. Moreover, there was a declaration to this effect in the receipt which the recipient had to sign on collection of the subsidy. However, in July 1994 the EIB department managing the facility suggested to the Commission service responsible that it should not apply this provision to the combination of SME subsidies with other aid distributed in the form of interest subsidies as part of an ERDF measure(18). The Commission service concerned was supposed to have approved this derogation because they did not actively oppose it.

32. In fact, the Court's audit work detected numerous cases of combination with other aid, but only in Spain. This was due to the fact that intermediaries in other Member States do not seem to have known of the existence of the derogation. In Spain, of the 11 recipient undertakings audited (which received ECU 653000), 10 received a combination of SME subsidies (ECU 581000) and ERDF aid (ECU 370000). The latter was also in the form of an interest subsidy and was based on the same loan from the EIB, which was thereby subsidised twice. The rule prohibiting the combination of aid applies a fortiori to subsidies granted on the same loan, and regardless of which intermediary has granted the subsidy, provided that it comes from the Community budget. Therefore, since the combination of aid was prohibited under Article 4 of the cooperation agreement, and the value dates of the subsidies granted to the SMEs were later than those of the ERDF subsidies, the SME subsidies under discussion should be deemed as not in conformity with the regulations. A derogation agreed at interdepartmental level has no legal value. Only a document agreed at the same level as the cooperation agreement could have constituted a valid exception. Moreover, the contracts concluded with the agents and financial intermediaries contained the text of the agreement without amendment in the annex.

33. With regard to combination with ECSC-subsidised loans, the Court has already pointed out the risk of non-detection in its Annual Report concerning the ECSC (for the financial year 1996)(19) and the Commission has undertaken to carry out a check, once the list of recipients of the SME facility has been published.

34. VAT paid on the purchase of capital goods does not constitute a cost for undertakings since it is deductible(20). In the United Kingdom, however, in 10 of the 28 recipients audited, it had been included in the calculation of the subsidised loan, and thus resulted in too much subsidy being paid. This is one more example of the lack of clarity in the presentation and explanation of the rules to be observed. The total error is about ECU 40000, out of a total of ECU 253000 of subsidies paid to the 10 undertakings concerned and ECU 789000 for all the undertakings audited in the United Kingdom. This type of error was not found in other countries.

Ineligibility of the recipient or the loan

35. Of the 114 undertakings audited, the Court found four which had received a total of ECU 219000 of subsidies and which did not meet the criteria defining an SME selected for this measure or listing the eligible sectors of activity (see paragraph 18). As for the loan contracts, it was stipulated that they should be for a period of at least five years, which corresponds to the period for which the amount of subsidy is calculated. However, the Court's checks found three cases of contracts lasting less than five years and one case of early repayment before five years. The total of improperly paid subsidies was ECU 83000.

Non-compliance with the eligibility period

36. The sworn statement concerning the number of jobs created was supposed to be submitted to the agents before 30 June 1997, even though at this date there was no guarantee of the availability of the necessary funds for the payment of each subsidy. During the following three months, that is, before 30 September 1997, the agent was supposed to draw up the definitive list of recipients to be sent to the EIB.

37. In the United Kingdom, however, declarations made between these two dates were accepted. In fact, the financial intermediary, which was also the agent (see paragraph 68), asked some of its customers (for whom the availability of funds depended on the withdrawal of recipients with priority in the list) to produce this document after it had made sure that funds were available, and thus after 30 June 1997. As a result, the recipients had a longer period in which to create jobs than in other countries. Of the 28 recipients audited in the United Kingdom, the Court found six cases where the eligibility period extended beyond 30 June 1997. These cases concerned a total amount of ECU 149000 of subsidies paid. A similar case of creating and maintaining jobs outside the period was found in Spain (ECU 60000).

The system's weaknesses

38. The numerous errors detected reflect the inherent weaknesses in the management and control system that were identified by the Court during its analysis of the system. First of all, the fact that the subsidies were paid on the basis of declarations submitted by the recipients themselves did not offer an optimum guarantee of accuracy, given that the only risk in the case of an incorrect declaration was the repayment of the excess received. In addition, the lay-out of the declaration form and the instructions for completing it were not very precise, particularly with regard to what was meant by maintaining the increase in jobs for a period of six months prior to the date of the declaration. Most of the problems of interpretation encountered could certainly have been avoided with better drafted instructions and a more clearly designed form. This would have reduced the number of errors detected.

39. Also, the fact that at the operational level the measure was administered by banks does not in itself provide an adequate assurance of the regularity of the declarations. In general, the banks have only very occasionally introduced systematic arrangements for checking the eligibility criteria and in particular the additional jobs that have actually been created. Finally, no penalty was envisaged for intermediaries who sent in inaccurate declarations.

40. The arrangements did not include adequate compensating checks, as the EIB only verified the agents' procedures and that they had the documentation required by the facility and checked the consistency of the documents received from its agents. The Commission, for its part, did not carry out any regularity checks at all in connection with this measure.

41. Supplying proof of the creation of jobs is a problem with which the Court is familiar, as are the Commission departments which have been applying and checking measures of this type for many years (for example, the ECSC-subsidised conversion loans). There are variations from one country to another as to which documentation (the Court prefers official documents) is subject to checks by the national tax or social security authorities. Although the cooperation agreement lays down that "the beneficiaries shall be contractually required to make available the necessary evidence for the verification of the number of jobs created...", the Commission made no use of its previous experience and the Court only found a very few cases where an attempt had been made to carry out checks on this basis prior to the Court's own examination. In the only case where the national agent had made serious and systematic checks of the declarations on the basis of official documents, the reference documents used proved to be unsuitable because their figures were too general and the reference periods too long.

42. The cooperation agreement stipulated that subsidised loans had to be for a minimum period of five years and that there could be no early repayment before the end of this period without incurring the cancellation of the subsidy. On this point, the financial intermediary was responsible for informing the EIB and, at its request, obtaining repayment of the subsidy.

43. However, depending on the banks alone to identify and secure the repayment of subsidies which have become ineligible after being disbursed does not ensure the separation of duties which is essential in order for the control to be reliable. The observance of this rule cannot be adequately ensured without additional checks at another level, and in fact one case of undeclared early repayment was found during on-the-spot checks.

Individual aspects

Order of dealing with applications

44. Applications for interest subsidies were supposed to be dealt with in order of receipt, as shown in the list held by the national agents. In certain Member States, this order was not observed. In Belgium, the six financial intermediaries (including the agent) made an agreement to share the available subsidies between them(21), a practice which was contrary to the rules and rendered the chronological order rule inapplicable. In Ireland, too, since the system was based on quotas, it was impossible to use chronological order, and this was also true in countries where the agents took it upon themselves to choose the recipients (see paragraph 45(c)). In Finland, the agent did not draw up a chronological list; moreover, it accepted oral applications, making any check on the chronological order impossible.

Equality of access

45. The Council decision stipulated that the facility should be available throughout the Community(22) and the cooperation agreement provided for an "adequate promotion of the SME facility and a wide dissemination of information to SMEs on the availability of subsidised SME loans". However, the arrangements put in place restricted equality of access for all potential recipients in certain areas, for the following reasons:

(a) the measure was administered by banks which had global loan contracts with the EIB, thus excluding banks without such contracts, and as a result their usual customers (for example in the Netherlands, where the bank most heavily involved in the financing of SMEs was excluded for this reason);

(b) not all the banks which had signed global loan contracts with the EIB were chosen as intermediaries for the implementation of this measure, even if they wanted to be (for example, in France and also, at first, in Belgium);

(c) some agents or intermediaries chose the applicants themselves rather than publicising the facility widely (for example in Denmark, Greece, Ireland, Luxembourg and Austria).

Implementation deadline

46. Article 1 of the Council Decision establishing the measure required the EIB to take the decisions to make subsidised loans within 12 months of the date of publication of the Decision, and therefore before 28 April 1995. In effect, this date was the deadline not only for the granting of the loan to the recipient but also for the receipt by the national agent of the application for an interest subsidy. This application was in fact the element which ensured that the underlying loan was a subsidised loan made by the financial intermediary under the conditions in force. However, this deadline was not correctly transposed into the cooperation agreement between the Commission and the EIB, for the following two reasons.

47. First, this agreement set 28 April 1995 as the deadline for global contracts to be concluded with the financial intermediaries and not for the loan contracts to be concluded with the recipients: these latter could be signed up to 12 months after the signing of the global loan contracts, and thus after the date of 28 April 1995 set in the Council Decision. Secondly, the agreement set 31 July and 15 December 1995 as the deadlines for the receipt of applications for interest subsidies in the Member States, whether or not they had exhausted the funds allocated as at 31 July 1995; in other words, three and seven and a half months respectively after the deadline laid down in the Council Decision.

48. In addition, the speed of implementation of this measure was mentioned in the recitals of the Council Decision and also in the Economic and Social Committee's opinion.

EVALUATION OF EFFECTIVENESS

Impact on investment and job creation

49. The measure was designed to encourage and accelerate investment by the SMEs in job creation by making available a loan package of ECU 1000 million exclusively to SMEs, together with a reduction in financing costs of 2 % per year for five years. The assistance was thus supposed to be a response to the main concern of many SMEs, particularly at the creation and expansion stages, namely, in the absence of their own resources or adequate guarantees, to obtain funds by borrowing while reducing the costs involved.

50. During its audit, the Court systematically asked the recipients and the financial intermediaries whether the availability of the subsidised loan had been a decisive factor in making the investment and whether it had had any influence on its amount, nature or timetable. Whilst making the subsidy available led, in 8 % of the cases audited, representing 6 % of the amount of subsidy, undertakings to bring forward their investments, in none of the cases examined was the availability of the subsidised loan the decisive factor in deciding to invest.

51. The main reason for this lack of influence on the entrepreneurs' strategic decisions is the low level of the subsidy. In fact, its average impact on the cost of the investments financed (excluding financing charges) is scarcely 2,4 % (Table 8), which is too low a percentage to be considered a determining factor in deciding on an investment project. When compared with the level of investment needed to create a new job, ECU 3000 is not a sufficient incentive. To have a real influence on decisions of this kind, the money should be allocated in a more concentrated and targeted fashion.

Table 8

Impact of interest subsidies on investments and loans disbursed

>TABLE>

52. The fact that the banking risk for individual loans is taken on by the intermediary banks leads them to apply their own criteria for accepting requests for finance. Thus, only projects that would have been financed anyway were accepted for financing on the back of subsidised loans. The existence of subsidised loans did not therefore create any additional financing capacity. During its visits to the Member States, the Court examined considerable evidence which indicated that the main problem for SMEs in finding financial backing is the insufficiency of their own funds and the guarantees they can offer to external lenders. This being so, more recent initiatives focusing on financing guarantees or consisting in taking out temporary participations in the capital of small undertakings at the creation or expansion stages will do more to encourage the financing of projects which are not attracting backers(23).

53. The use of the "first come, first served" principle at the national level meant that the projects which were submitted by applicants for subsidies and processed by the banks were mainly at an advanced stage of planning. Because of the risk that other applicants might obtain the subsidies before them, business owners did not have the time to draw up new projects, but generally submitted projects where the investment and its financing had already been decided on and the forecast impact on the number of jobs was already known. In fact, the Court noted cases where, at the time of the application, the investment was already at an advanced stage of completion and the subsidised loan had actually replaced an existing financial arrangement.

54. Despite their name, the interest subsidies were not perceived as such because their payment was not linked to the interest payments and they were not made at regular intervals. The amount was received in a one-off transfer unconnected with the loan's banking operations, following the declaration that the jobs had been created. Moreover, according to the statements made by entrepreneurs and bankers, a reduction in financing charges only seems to be considered as urgent when the interest rate becomes too high, or at any rate higher than it was during this measure's period of application.

55. Under the rules laid down for implementing this measure, undertakings which had applied for and obtained a bank loan in order to finance an investment and which had recorded an increase in jobs were eligible for a subsidised loan. There was therefore no requirement for a causal link between the creation of the jobs and the investment in question. Sometimes the investment had no effect on employment or was even made to fund a rationalisation entailing a reduction in jobs in a specific area, while the undertaking increased the overall number of jobs. The loans served primarily to channel subsidies to expanding undertakings and the mechanism of decentralised distribution of the subsidies in small individual amounts by the banking network took precedence over the aim of stimulating job creation.

56. The regulations in question did not exclude take-overs, where the investment consisted in financing the purchase of an existing undertaking; in this case job creation comes down to maintaining existing jobs. For ECSC-subsidised loans, this type of project is strictly excluded, even for take-overs of companies in difficulty, where jobs can be saved that would otherwise possibly be lost. During its audit of the Copenhagen facility, the Court found numerous cases where the financed and subsidised investment consisted in taking over another company along with its staff. On some occasions, this was a rescue operation which allowed some of the jobs to be retained; on others, neither the undertaking nor the continued existence of its jobs were at risk.

57. Although subsidies were indeed given to job-creating undertakings, it is impossible to prove that the subsidies enabled the creation of the jobs in question. Moreover, the arrangements for implementing the measure do not include the information necessary to establish which jobs would not have been created without the subsidised loan. The figures diclosed by the Commission(24) are a sum of the increases in staff members declared by the undertakings between the reference date and the declaration date. They include all increases, whether or not they were linked to the investment financed by the subsidised loan. Furthermore, these data were not checked by either the Commission or the EIB, and during its on-the-spot checks the Court uncovered many inaccuracies in the figures (see paragraph 29).

Size of the recipients

58. The implementation arrangements defined eligible SMEs according to the definition usually used by the EIB within the context of its own policies for this category of undertaking (see paragraph 18). Since this definition includes SMEs of quite a considerable size (net fixed assets of up to ECU 75 million and up to 499 employees), the arrangements distinguish between two categories of SME. The first includes undertakings with a turnover of less than ECU 20 million or total assets of under ECU 10 million and not more than 250 employees and the second all other SMEs which meet the EIB's criteria. Then it was stipulated that there would be a ceiling of ECU 810000 on loans for the second category (and thus a total subsidy ceiling of ECU 81000) and that at least 60 % of the recipients had to fall within the first category.

59. These two rules were observed, and the percentage of recipients from the first category was actually considerably in excess of the 60 % mentioned above(25). In this regard, many intermediary banks played quite an active role in encouraging potential recipients from this category, and some of them made more efforts to find small undertakings. On the other hand, others preferred larger customers, which also had the effect of reducing the number of files to be managed. In fact, 81 undertakings were granted subsidies of over ECU 100000 even though they belonged to the first category. Thus, 2 % of the recipients received 14 % of the total amount of subsidies (ECU 13 million out of ECU 92 million).

Geographical breakdown of the subsidies

60. The Member States were allocated the sum of ECU 1000 million in subsidised loans in four stages. First, in 1994, the first tranche of ECU 650 million was allocated to the then 12 Member States according to their population (weighted at 50 %) and their gross national product (20 %) in 1993, together with the average of their unemployment rates in 1992 and 1993, weighted according to the population (30 %). An additional allocation was then carried out for the three new Member States according to the same criteria, except that the result of the calculation was halved, to give ECU 27,8 million. Subsequently, the remaining balance of ECU 322,2 million was shared out between the 12 original Member States on the basis of their rates of utilisation of the first tranche as at 28 February 1995. Finally, the sums unused as at 15 December 1995 were redistributed to those countries which had, as at 31 July 1995, a list of applications for a total amount greater than the available funds.

61. This procedure was designed to make maximum use of the available budgetary funds, but in practice two aspects were not entirely satisfactory. On the one hand, the new Member States were treated differently, in that they received a lower amount in comparison with the other countries and were not able to take part in the subsequent redistributions. However, the actual absorption capacity in these three countries proved to exceed by far the amounts allotted. On the other hand, the data on which the calculations of the second tranches and the final redistribution were based did not seem sufficiently reliable. The main reason for this was that at the time of the redistributions, there was no checking procedure to eliminate from the lists applications which had lapsed in the interim.

62. Finally, ECU 92,3 million was paid in interest subsidies. The utilisation level is shown by country in Table 1. The utilisation rate in Germany is low in comparison with the initial applications, including both the main lists and the reserve lists. The main rea son for the underutilisation is early repayment caused by a reduction in interest rates. Because the minimum period of five years had not been observed, these early repayments made the loans ineligible, so that many files were withdrawn before reaching the declaration stage. The utilisation rates in Belgium, Greece and Finland may be explained by the management methods used by the national agents. The very low rates in Ireland and Portugal reflect the availability of other more advantageous support programmes offered by the national authorities (in Ireland) or the Community authorities (in Portugal).

63. Coordination with the other aid on offer at the regional, national or even Community levels was not taken into sufficient consideration. During its missions in the Member States, the Court found that sometimes a measure of this type was the only one of its kind, sometimes there were much more advantageous alternatives (see paragraph 61) and sometimes it was possible to combine it with regional or national aid, which was sometimes of highly significant proportions (for example in Germany).

Competition aspects within the banking market

64. Under the provisions of Article 267 of the EC Treaty, the EIB finances investment projects in less-developed regions or where the means available in the individual Member States are not sufficient. It is therefore called upon to support the Community's policies while observing the principle of subsidiarity; in other words, intervening only where the resources of the market or the Member State are not sufficient.

65. On this basis, in 1968 the EIB established a global loan facility for SMEs operating in a limited number of productive sectors and located in the regional development areas. In doing so, while it avoided taking on any risks connected with the undertakings, it developed a method of financing investment projects through a network of existing financial intermediaries.

66. Because of their success, these loans were subsequently granted to finance the projects of SMEs located throughout the Community and operating in an increasing number of sectors (for example, since 1995 the retail trade has been eligible). Thus, between 1993 and 1997, the EIB signed global loans for SMEs for a total of ECU 24362 million, accounting for approximately 25 % of all its loan activities(26).

67. In recent years, however, this situation has changed. The banking market has become so competitive that increasingly similar terms are on offer, particularly for interest rates. In this context, the financial intermediary's priority is no longer the availability of funds, which it can obtain from the market, but a borrowing cost that will give it better operating margins than the competition. Including a bonus such as the Copenhagen facility subsidy in a financial package proposed to a potential customer constitutes an asset in comparison with the other banks. The financial intermediaries with exclusive rights to distribute these subsidies had an advantage over other lenders who were not included in the network of banking relations with the EIB (see also paragraph 45(a)).

68. In addition, the role played by some agents contributed to creating circumstances under which distortions of competition between the banks distributing the subsidies could also arise. In the United Kingdom, although five financial intermediaries were authorised to distribute the subsidies, 83 % of the available subsidised loans were extended by one intermediary which also acted as an agent. This does not correspond at all to its actual commercial penetration of this financial market (a market share of approximately a third). In Belgium, five financial intermediaries experienced losses of customers to one financial intermediary, which was the only one initially able to extend subsidised loans. In France, only eight financial intermediaries were authorised to grant subsidised loans, and they were not among the largest in the national market. Moreover, some of them only learnt of the measure indirectly and after a certain delay (see also paragraph 45(b)).

69. The banks which distributed the subsidies often integrated them so well into the financial packages they offered their customers that the origin of the subsidy disappeared from view. One bank went so far as to present the facility as "Bank X's temporary facility for SMEs". Of those recipients that were aware that the subsidy came from elsewhere, the majority thought it was an EIB subsidy. Only in a few exceptional cases did the recipient know that the subsidy was financed by the Community budget.

CONCLUSION

70. In practice, the temporary facility for SMEs has consisted above all in distributing premiums, through the banking network, to job-creating undertakings which have invested by borrowing mainly against the background of a stage of development which was already well advanced. Since it is impossible to show a direct link between the subsidy in question and the creation of jobs, it is also impossible to affirm that the facility contributed to these jobs being created. Since the banks distributing the subsidies took on the banking risk, they chose projects according to their own criteria and which they would therefore have financed in any case without the subsidy. The application of the "first come, first served" principle led to numerous projects being put forward which were at an advanced stage of planning and whose effect on employment was in fact already guaranteed (see paragraphs 49 to 57).

71. The reduction in financing costs was not perceived as much of an incentive, given the levels of interest rates recorded over recent years. More recent initiatives offering guarantee facilities or consisting of taking shares in small undertakings with financing difficulties at the creation or expansion stages seem to be better suited to their real needs (see paragraph 54).

72. The checks carried out by the Court on the premises of 114 undertakings, which cannot be used for a mere mathematical extrapolation, revealed numerous errors (see Table 7). A number of the irregularities concern jobs declared to have been created and maintained for at least six months, when in fact this was not the case (91 jobs out of 1389; see paragraphs 28 to 37).

73. The cases where the rules were not observed are in large part due to the temporary nature of the facility, which is based on a brief regulation and an ad hoc mechanism, administered by banks and relying on sworn statements made by the recipients. Neither the past experience of the Commission departments nor the Court's previous recommendations(27) were turned to account to offset the system's deficiencies. In fact, neither the Commission nor the EIB carried out checks to compensate for these weaknesses (see paragraphs 38 to 43).

74. The banks chosen to distribute the subsidies took the opportunity to keep their customers loyal, or even to attract new clients. In some cases this created situations which may lead to distortions of competition on the banking market (see paragraphs 67 and 68).

75. The facility was financed by transfers of significant sums from the Commission to the EIB a long time in advance of the actual date of use. During the five-year period during which the funds were in the EIB's cash resources, they were not monitored within the framework of the Community's accounting system (see paragraphs 26 and 27).

This report was adopted by the Court of Auditors in Luxembourg at its meeting of 23 and 24 February 2000.

For the Court of Auditors

Jan O. Karlsson

President

(1) In the conclusions reached at its meeting on 11 and 12 December 1992 in Edinburgh, the European Council requested the Council and the EIB to consider, together with the Commission, establishing a new temporary lending facility of ECU 5000 million, which would be designed to accelerate the financing of infrastructure projects, especially in connection with the trans-European networks.

(2) Council Decision 94/217/EC of 19 April 1994 (OJ L 107, 28.4.1994, p. 57).

(3) Opinion of the Economic and Social Committee, 20 October 1993 (OJ C 352, 30.12.1993, paragraph 3.3).

(4) The cooperation agreement was signed on 14 June 1994.

(5) Four reports were submitted between October 1994 and December 1997.

(6) European Parliament Decision of 10 April 1997 (OJ L 162, 19.6.1997, paragraph 57).

(7) Annual Report of the Court of Auditors concerning the ECSC (for the financial year 1995) (OJ C 377, 13.12.1996, paragraphs 2.14 to 2.20).

(8) The subsidies paid amount to 10 % of the loans granted (see paragraph 19).

(9) Since a given undertaking could make more than one application, the precise number of recipients is not known, but in any case it is lower than 4336. The same applies to the number of jobs created, as where one recipient made several applications, the same number was given each time in the application.

(10) Sum of the figures appearing in the recipients' declarations. Source: EIB.

(11) There is a description of the EIB's global loans in "EIB INFORMATION" No 86, November 1995 (pp. 7 and 8).

(12) The Commission decided to apply the same 7,5 % ceiling to State investment aid for SMEs outside the assisted regions; Commission Communication on the Community guidelines on aid to SMEs (OJ C 213, 19.8.1992, paragraph 4.1).

(13) In Ireland, the EIB itself paid the subsidies.

(14) In Greece, the Ministry of Finance disbursed the subsidies and declined payment.

(15) Interest was credited at the end of the month at a rate equivalent to the monthly average of the daily interest rates set by the ecu clearing system.

(16) Opinion No 4/97 on the proposal for a Council Regulation (Euratom, ECSC, EC) amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities (OJ C 57, 23.2.1998, paragraphs 3.7 and 3.8).

(17) The Court dealt with the issues of the accounting treatment of advances and instalments, as well as interest, in its Opinion No 4 of 10 July 1997 on the proposal for a Council Regulation (Euratom, ECSC, EC) amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities.

(18) Commission Decision of 25 November 1994, C(94) 3044, establishing a global subsidy for Objective 1 regions in Spain in the framework of the ERDF operations during the 1994 to 1999 period.

(19) OJ C 380, 15.12.1997, paragraphs 2.28 to 2.33.

(20) The VAT paid when purchasing goods is, according to the very principles governing this tax, a financial claim which the undertaking has on the tax authorities and by this token is entered under the assets on the undertaking's balance sheet.

(21) In Belgium, initially, there was only one financial intermediary (also an agent) able to extend subsidised loans. Subsequently, five other financial intermediaries, which were, in addition, already customers of the EIB, became aware of the loss of their customers to the first intermediary, displayed their dissatisfaction with this situation and requested from the EIB the opportunity to extend subsidised loans to their customers too. This right was granted to them several months after the launch of the measure by the first financial intermediary. In order to make up for the delay, the five other financial intermediaries demanded an agreement setting quotas for the subsidies which each of the six could distribute to its own customers.

(22) Recital 8 of Council Decision 94/217/EC of 19 April 1994 (OJ L 107, 28.4.1994, p. 57).

(23) In particular the "European technology facility - start-up facility" and the "SME guarantee facility" initiatives, which come within the "SME technology facility" framework, which was the subject of the conclusions of the Extraordinary Council meeting of November 1997 in Luxembourg and of the Council decision of 19 May 1998.

(24) First, second, third and fourth reports of the Commission relative to the implementation of the measure, respectively COM(94) 434 final of 19 October 1994, paragraph 4.3; COM(95) 485 final of 30 October 1995, paragraph 4.1; COM(96) 522 final of 25 October 1996, paragraph 4.1 and COM(97) 645 final of 2 December 1997, paragraph 4.1.

(25) Fourth report of the Commission relative to the implementation of the measure, COM(97) 645 final of 2 December 1997, p. 10, paragraph 4.3. The Commission mentions the figure of 80 % without giving its source whereas the data obtained from the EIB give a figure of over 90 %, although these data contain numerous classification errors.

(26) See the EIB publication "40 years of activity", p. 10, table headed "Loans signed in the European Union".

(27) The Court's Annual Report concerning the financial year 1995 (OJ C 340, 12.11.1996, paragraphs 8.16 and 8.17).

The Commission's replies

SUMMARY OF THE MAIN OBSERVATIONS

8 and 9. It is true that the way in which the interest subsidies for small and medium-sized enterprises were recorded in the budgetary accounts gives the impression that the utilisation rate was 100 %, whereas in fact the subsidies were not simultaneously paid to the final beneficiaries by the EIB. During the time that they were held by the EIB they should have appeared on the assets side of the balance sheet as Commission claims.

10. The system was based on a sworn declaration by the final beneficiary. The Bank checked that the agents had instituted a proper system for verifying receipt of these declarations.

Article 16 of the cooperation agreement provides for wrongly paid subsidies to be paid back.

11. The Council provided for the scheme to be implemented via EIB global loans, a well-known means of Bank funding. To extend the scheme further would have been unmanageable and economically unjustified, given the small amounts involved.

12. The scheme could not have reached the Objective 1 regions without this possibility of combining Community grants.

13. It is true that at the beginning of the programme some on-going projects were funded. This was difficult to avoid, given the urgency with which the scheme had to be set up. However, the impact has diminished over time.

14. The difficulties of quantifying the employment effects of interest subsidies at microeconomic level are well known. However, the scheme did deliver a message of support and encouragement for job creation from the public authorities to the business leaders at European level. It was one of a number of measures, and even if it was necessarily limited it did fulfil a positive function by helping to create an environment conducive to promoting growth and job creation by small businesses, in accordance with the broad objectives set out in the preamble to the Council Decision.

15. Given that the scheme had to be applied uniformly throughout Europe it was quite simply impossible to ensure coordination with all regional and national aid schemes. The resulting combination was not contrary to the subsidiarity principle.

16. Given the number of financial intermediaries that took part in the system it is hardly surprising that the Court found a number of differences in presentation. It is also quite likely that the origin of the subsidy "disappeared from view" for perfectly legitimate reasons, given the scale and complexity of the banking systems involved and the variety of national and regional aid schemes managed by certain financial intermediaries.

IMPLEMENTATION OF THE MEASURE

Procedure

20. The decision to allow agents to combine this role with that of financial intermediary was dictated by considerations of efficiency, which was the Bank's paramount concern in the short start-up stage of the scheme. The operational guidelines drawn up by the Bank contained specific clauses to this effect. Several agents did indeed act as financial intermediaries, directly disbursing the subsidised loans. This was because of their special experience in managing financial schemes of this type, including other Community schemes (e.g. ECSC loans).

23. In conjunction with the Commission, the EIB issued 90000 multilingual brochures outlining the scheme, which were distributed to all of the participants. Numerous presentation meetings were also held in the beneficiary countries to bring together the Commission, the Bank and organisations representing the target group.

MANAGEMENT OF BUDGET APPROPRIATIONS AND PRESENTATION OF THE ACCOUNTS

Management of budget appropriations

26. In future the Commission will ensure that sums held by the EIB pending transfer to the final beneficiaries of the subsidies appear on the balance sheet.

Presentation of the accounts

27. As regards the management of the Commission's account by the Bank under the scheme, the Bank and the Commission agreed in an Exchange of Letters of 6 August 1998 and 7 September 1998 that amounts above ECU 250000 would be returned to the Commission on 31 May and 30 November each year, so that budget funds would not be kept in this account in the post-disbursement phase. These amounts currently consist of interest and early repayments.

The proposal for recasting the Financial Regulation should include provisions requiring all public and private financial intermediaries to provide information about the state of the funds they hold on behalf of the Commission, and the interest paid and bank charges incurred.

The Commission is in favour of treating such payments as budgetary expenditure, regarding them as advances (payments without transfer of ownership) to ensure greater transparency in budget implementation.

As regards the issuing of recovery orders for the interest generated, the Commission shares the Court of Auditors' view that interest should have been established annually as an entitlement. The proposals for recasting the Financial Regulation should also contain provisions on this point to tighten up the obligations on authorising officers.

AUDIT OF REGULARITY AND LEGALITY

Results of on-the-spot checks

Overall findings

28. Cases where aid under the SME scheme is combined with ERDF aid account for 38 % of the total amount of errors.

Combination with other aid from Community sources prohibited

31 and 32. The Commission is of the opinion that the scheme could not have reached the Objective 1 regions without combining Community subsidies.

33. The Commission and the EIB are carefully monitoring the question of combination with ECSC loans. The Commission has carried out checks to identify duplication of funding between this scheme and the ECSC. These revealed a tiny number of potential cases of duplication. In all, 317 cases were identified with the same or similar name and were analysed. 61 of these were considered potential double financing and are being analysed individually. Cases are being cleared as further information is received, but at the present time it would appear that about 25 cases will require either a visit or audited statements before a final decision can be taken.

Depending on the further information received, joint Commission/EIB inspections could be arranged.

Ineligibility of the recipient or the loan

35 to 37. As part of its monitoring of other schemes involving delegation to financial intermediaries the Commission will tighten checks on eligibility to avoid such cases.

The system's weaknesses

38 to 41. The scheme relied on delegating responsibility to agents and intermediaries, in the interests of efficiency and in keeping with the nature of the global loan facility itself. Given the importance of the group targeted by the scheme, the number and nature of the beneficiaries (several thousand SMEs) and the average size of the subsidies (ECU 20000), it is hard to imagine another system that would not have buckled under the weight of the procedure.

It should be noted that the sworn declarations completed by the recipients and submitted by the financial intermediaries stressed that recipients laid themselves open to civil and criminal prosecution if they made false declarations.

The incorrect declarations detected by the Court have been investigated by the EIB. In cases where there is no legitimate defence the EIB has initiated the recovery procedure via the financial intermediaries.

42 and 43. The banks are required to inform the EIB of any early repayment of the global loan and to repay the subsidies, in line with the philosophy of the scheme which relies on delegating responsibility to the agents and intermediaries. The EIB checks these repayments against the list of recipients.

Individual aspects

Order of dealing with applications

44. The exceptions to the "first come, first served" rule were authorised by the Bank in certain Member States in the light of circumstances connected with the economic and/or administrative situation.

Equality of access

45. (a) The Council decision specified that the loans available under the scheme, worth ECU 1000 million, had to be granted in accordance with the EIB's global loan mechanism. Given the urgency with which the scheme had to be put into effect throughout the Community, the Bank restricted itself almost exclusively to its existing network of financial intermediaries. These intermediaries are selected on the basis of operational criteria related particularly to their financial capacity, their type of activity and, of course, their willingness to fulfil the conditions laid down by the Bank regarding project management and follow-up. One might legitimately expect the Council to realise that it would have been impossible to meet the tight deadline had a scheme been adopted whose implementation extended beyond the network of banks involved in the EIB's global loans. Moreover, given the limited volume of the scheme such an attempt would have been impossible to justify on economic grounds.

(b) Some type of selection of participating banks was inevitable, if only to meet the tight deadline imposed by the Council. Moreover, some banks and financial institutions with global loan contracts with the EIB simply did not wish to participate in the system because of the charges the scheme would inevitably entail.

(c) The small sums available in the countries mentioned more than justify the lack of wide publicity. Such publicity would have aroused false expectations and over-burdened the banks with applications that could not be satisfied. In the countries in question the banks elicited projects from their clients which fulfilled the conditions laid down.

In general, the fact that 92 % of the scheme funds were used, despite a general fall in interest rates which led to a series of repayments because of failure to comply with the required duration of the loan, demonstrates the satisfactory level of promotion and equality of access to the scheme. The size of the reserve list in certain countries also reflects the genuine equality of access.

Implementation deadline

46 to 48. The "decisions to extend loans benefiting from the [...] subsidy", which Article 1 of the Council Decision required the Bank to take "within 12 months of the date of entry into force of this Decision" were the decisions taken by the EIB's Board of Directors on extending global loans to the financial intermediaries. The text of the Decision did not refer to the allocation of the global loans. It would have been physically impossible to allocate all of the funds under the scheme to final beneficiaries within a period of 12 months. The Commission and the Bank would have raised this question back in 1994 had they thought there was any ambiguity on this point, whereas, on the contrary, the EIB complied with the Council's request and set up the scheme with the financial intermediaries (concluding new contracts and amending existing ones) within the deadline specified. The disbursement phase then extended beyond this period, which is perfectly normal in the context of global loans.

EVALUATION OF EFFECTIVENESS

Impact on investment and job creation

49 to 52. The European Council has taken a political decision to foster the creation of jobs and has fixed certain parameters within the framework of its Decision. The Commission and EIB have respected the Council's decision.

Client creditworthiness and project viability are essential elements of banking practice. It is only natural that banks should apply their own objective criteria in order to select the projects they consider most promising and viable from the list of applicants. On the other hand, the Commission believes that the fact that a loan looks sound in terms of the financial criterion of the creditworthiness of the firm does not necessarily mean that the project will receive a loan. Another decisive factor in the Bank's decision as to whether or not to extend the loan will be the ratio between the firm's financial obligations and its operational results, i.e. its ability to honour its commitments. The reduction in company financial charges thanks to the Community contribution is therefore also a factor influencing the extension of loans to the firms in question.

53. It is true that at the start of the programme ongoing projects were funded. However, it is difficult to avoid such a situation in the start-up phase and the impact has declined over time.

54 to 57. The payment of the interest rate subsidy in a lump sum has had two advantages:

- it is the simplest and cheapest solution from an administrative point of view,

- it increases the visibility of the scheme.

In addition, reducing the expenses of the companies concerned improved the general level of their financial charges, which is an important factor in the decision to extend a loan. When the scheme was launched this was a persistent concern of the directors of small businesses.

The Commission would not claim that the interest rebate is the single determining factor in the beneficiaries' decision to undertake job-generating investment projects. Rather, interest subsidies form part of a package which also includes measures on a local, regional or national level which together create a favourable context for the achievement of priority Community objectives.

The important thing about the subsidy scheme, and this was undoubtedly the political intention of its supporters, is that it sent a Europe-wide message of support and encouragement from the public authorities to entrepreneurs to promote job creation. It was one of a number of measures, and even though it had to be limited in scope it fulfilled a positive function by helping to create an environment conducive to promoting growth and job creation by SMEs, in line with the broad objectives formulated in the preamble to the Council Decision. The scheme undoubtedly helped to speed up the implementation of the investment decisions taken by entrepreneurs.

Small businesses try to improve their competitive position by investing in rationalisation, modernisation and increased capacity. Job creation results from success in the market and the need to fill new orders, regardless of whether the business is in an expanding or a declining sector.

At a macroeconomic level, the real rate of interest is known to be a factor influencing the level of investment.

Size of recipients

58. The Court's finding confirms the banks' commitment to complying with the rules on allocating subsidies under the scheme (and even applying more stringent criteria).

Geographical breakdown of the subsidies

60 and 61. Provision was made from the start for the accession of new Member States. The calculation took account of the amount of time remaining for allocating these sums.

63. It was impossible to ensure far-reaching coordination with all of the other aid schemes available at regional, national or even Community level, because the scheme had to be available throughout the Community and be applied uniformly in all Member States. It was inevitably superimposed on all other national and Community measures (particularly in regions receiving special aid), and this was not considered to be contrary to the principle of subsidiarity.

Competition aspects within the banking market

67. The EIB cooperates with a vast network of banks and financial institutions in all of the countries in which it operates (over 130 in the EU), in accordance with Article 16 of its Statute. The Council takes this feature into account when entrusting the Bank with tasks, including, where necessary, the granting of interest-rate subsidies inside and outside the European Union. The EIB selects its financial intermediaries by applying objective criteria of creditworthiness and quality. It would have been impossible for an unlimited number of financial intermediaries in Europe to administer Community interest-rate subsidies for SMEs available in limited amounts and for a limited period. Moreover, for a number of reasons, not all of the banks satisfying the EIB's criteria agree to manage global loans or interest-rate subsidies. Finally, some form of selection of the participating banks was inevitable, if only to meet the tight deadline imposed by the Council.

68. It was unlikely that the implementation of the scheme would reflect the market shares of the participating banks. As in any scheme of this kind, providing quotas are not applied to the banks taking part and the principle of "first come, first served" is adopted, there will inevitably be differences in the level of commercial motivation depending on the institution.

The comparatively large share accounted for by one intermediary in the subsidies granted reflects its traditional importance as an intermediary for EIB global loans in the United Kingdom. The scheme was the subject of commercial procedures and was made available to other banks in the United Kingdom. This particular intermediary also acted as agent because it combined the experience (including experience with other Community schemes, like the ECSC), the human resources and the willingness to take on the role.

CONCLUSION

70. It is a well-known fact that the causal link between interest-rate subsidies and investment is difficult to prove at microeconomic level. But at macroeconomic level the real rate of interest is generally recognised as a factor affecting the level of investment. Moreover, the scheme sent a Europe-wide message of support and encouragement from the public authorities to entrepreneurs to promote job creation. It was one of a number of measures, and even though it had to be limited in scope it fulfilled a positive function by helping to create an environment conducive to promoting growth and job creation by SMEs, in line with the broad objectives formulated in the preamble to the Council Decision.

It is true that at the beginning of the programme a number of ongoing projects received funding. Given the speed with which the scheme had to be implemented it was difficult to avoid such a situation in the start-up phase. However, the impact declined over time. In addition, the reduction in the financial charges of the companies concerned favourably influenced the banks' decisions on extending credit in so far as it improved the ratio of companies' financial charges to their operational results. This ratio is an important factor in the decision to grant a loan.

71. The Commission welcomes the Court's favourable response to its more recent initiatives.

72. The Commission will carefully investigate the cases referred to by the Court.

73. In view of the economic climate at the time, the Council Decision stressed the need for the scheme to be implemented rapidly, as a matter of urgency. It would have been impossible to reconcile developing complex regulations with this need for urgency.

Given the size of the group targeted by the scheme, the number and nature of the beneficiaries (several thousand SMEs) and the average size of the subsidies (ECU 20000), it is hard to imagine another system that would not have buckled under the weight of the procedure.

74. Because the scheme involved limited funds and a limited time-span it could not have been implemented by an unlimited number of financial intermediaries in Europe. Some sort of selection was inevitable. Moreover, some banks were unwilling to administer global loans and interest rate subsidies.