REPORT FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT CONCERNING INTEREST-RATE SUBSIDIES

TABLE OF CONTENTS

INTEREST-RATE SUBSIDIES

1. INTRODUCTION: COUNCIL REQUEST TO THE COMMISSION

2. PART I: PROBLEMS ASSOCIATED WITH INTEREST-RATE SUBSIDIES; POSITIVE AND NEGATIVE EFFECTS

2.1. Reasons for public intervention

2.2. Interest-rate subsidies: positive effects

2.3. Interest-rate subsidies: negative effects

3. PART II: THE LESSONS OF EXPERIENCE

3.1. Action at international level

3.1.1. Loans subsidised by the World Bank

3.1.2. Experience at the Council of Europe

3.2. Member States: measures involving interest-rate subsidies

3.3. Community measures involving interest-rate subsidies

3.3.1. A tentative classification

3.3.2. Community programmes

4. CONCLUSIONS AND RECOMMENDATIONS

ANNEXES: COMMISSION WORKING PAPER ON INTEREST-RATE SUBSIDIES

5. Annex I: INTEREST- RATE SUBSIDIES: COMMUNITY Programmes

5.1. Summary table of programmes

5.2. ECSC subsidised loans (Article 56: industrial conversion)

5.3. Trans-European Networks (TENs)

5.4. Interest-rate subsidies to help certain Member States to participate in the European Monetary System

5.5. Edinburgh facility (SMEs)

5.6. Subsidised loans in the ACP countries

5.7. Interest-rate subsidies linked to EIB loans in the Mediterranean

5.8. ERDF Programme

5.9. EAGGF Programme

5.10. Interest-rate subsidies in the central and eastern European countries - Phare budget

5.11. Exceptional aid for the regions of Italy devastated by the earthquake of 1980

5.12. Exceptional aid to the regions of Greece devastated by the earthquake of 1981

5.13. Exceptional aid for the reconstruction of Kalamata (Greece)

5.14. Aid for the reconstruction of Madeira after the cyclone in 1993

5.15. The Special Support Programme for Peace and Reconciliation in Northern Ireland and the Border Counties

5.16. Interest-rate subsidies under the Financial Mechanism of the European Economic Area

5.17. Interest-rate subsidies for projects in the western Balkans

6. ANNEX II: ANALYSIS OF OPERATIONAL ASPECTS

6.1. Management and control

6.2. Visibility

6.3. Evaluation of Community programmes

6.3.1. Reconstruction programmes

6.3.2. Other programmes

6.4. Compliance with the competition rules

6.5. Final remarks

7. Annex III: INTEREST-RATE SUBSIDIES: ExpErience IN MEMBER STATES

7.1. Italy

7.2. France

7.3. Germany

7.4. Spain

7.5. Sweden

7.6. United Kingdom

8. ANNEX IV: The use of subsidies in the context of EU policies with special reference to soft loans: theoretical considerations

8.1. Introduction

8.2. On the use of subsidies in general

8.3. On the equivalence of different forms of subsidy

8.4. On the suitability of different instruments for one and the same objective

8.5. On the implementation of soft loans

8.6. Conclusion

9. ANNEX V: Macroeconomic Effects of an Interest-Rate Subsidy fOr Firms in the EU

9.1. Introduction

9.2. Assumptions

9.3. Simulation results

INTEREST-RATE SUBSIDIES

1. INTRODUCTION: COUNCIL REQUEST TO THE COMMISSION

In response to the Court of Auditors' Special Report No 3/99 on the management and control of interest-rate subsidies by the Commission (OJ No C 217 of 29 July 1999), the Council "questioned more generally whether interest-rate subsidies are still justified [...] taking into account in particular the cost of effective control measures and the availability of other [...] financial instruments, such as loan guarantees" (doc. 6224/100 rev. 1). Accordingly, it asked the Commission to present as soon as possible "a comprehensive report on the general usefulness of interest-rate subsidies in terms of cost and economic efficiency, in the light of all relevant factors".

This report, which has been drawn up following numerous contacts with the EIB, the national authorities and the different Commission departments responsible for Community spheres of activity where interest-rate subsidies are used, is in two parts: (i) a Commission communication, and (ii) a Commission working paper attached as an annex.

The communication examines the question in its entirety and sets out some general guidelines for future use of interest-rate subsidies.

The Commission working paper is more empirical in nature, describing the operational aspects of the instrument and experience of interest-rate subsidy programmes within the European Union, making a tentative assessment of those programmes and, lastly, examining which financial instrument is more adapted to the objective being pursued.

2. PART I: PROBLEMS ASSOCIATED WITH INTEREST-RATE SUBSIDIES; POSITIVE AND NEGATIVE EFFECTS

2.1. Reasons for public intervention

Interest-rate subsidies are above all transfers of resources to natural or legal persons. Like any transfer, they are generally justified by the public interest, either as a sign of solidarity (e.g. in support of housing policy) or as an instrument of industrial or sectoral policy (e.g. to promote investment in infrastructure or R&D, to promote exports, etc.) or both (e.g. as support for regional policy). It is to be noted that, where the beneficiaries are enterprises, interest-rate subsidies fall within the definition of state aid, itself subject to Article 87(1) of the Treaty.

The point must be made that distortions caused by any form of public intervention may be justified (positive distortions).

Subject to the rules on monitoring aid, intervention by the public authorities may be considered justified where:

(a) the market is failing to provide the requisite financing for non-profitable activities or projects regarded by the public authorities as being necessary to the welfare of the population at large, e.g. TENs, R&D, agriculture and the environment. Also included here are measures to assist less-favoured regions and firms in difficulty.

(b) certain external shocks have to be dealt with, e.g. by way of measures to assist countries/regions affected by natural disasters, etc. or negative externalities.

While these principles justify intervention by the public authorities, they are silent as to the form such intervention should take: interest-rate subsidies, subsidies, guarantees, loans, equity participations, or a combination of all of them- Now, these instruments differ not only in nature and as regards their effects but also in terms of their costs, whether budgetary management or control costs. In view of these particular characteristics, the choice of instrument or form of intervention by the authorities must be a function of the objectives pursued and actual circumstances, even if recourse is had very often to two or more instruments at the same time.

SpECIAL CHARACTERISTICS OF THE INSTRUMENTS FOR PROMOTING INVESTMENT

* Direct grants: // Objective: public-interest projects in general

* Interest-rate subsidies: // Objective: to improve project profitability by reducing financing costs

Example: infrastructures

* Loan guarantees: // Objective: access to credit for high-risk projects

Example: SMEs at the start-up or development stage

* Equity participations: // Objective: risk-sharing and flexibility of return on capital

Example: innovative SMEs

* Tax breaks: // Objective: to improve the profit outlook

Examples: craft sector, SMEs, regions lagging behind

In all cases of such intervention, it is essential that certain conditions regarding control, transparency and effectiveness in relation to the objectives being pursued should be met. This imposes a range of obligations on the public authorities entailing additional administrative costs that must be taken into consideration when the decision to intervene is taken. It must be borne in mind that the follow-up of such operations extends throughout the life of the loan.

2.2. Interest-rate subsidies: positive effects

Ø Interest-rate subsidies have the direct effect of reducing financing costs; indirectly, they have a beneficial effect on the demand for credit, investment and, ultimately, employment. Given their specific characteristics, they are the ad hoc instrument for action by the public authorities to reduce costs, to ease the financing burden and thus to rectify any lack of profitability. They have only an indirect bearing on project risk, for which there are other, more appropriate instruments such as guarantees or risk capital.

Ø Since interest-rate subsidies are tied to loans, they have the advantage over other forms of intervention of imposing the management stringency or discipline that any bank loan requires. Viewed from this angle, there is, in principle, less risk of abuse or misuse of public money than in the case of grants or other forms of aid.

Ø Interest-rate subsidies have a significant financial leverage effect; for example, the EUR92.3 million of subsidies under the SME facility generated just under EUR 1 billion of investment credits. From a budgetary angle, they are therefore a relatively economic tool not simply because limited financial involvement allows substantial volumes of finance to be raised but also because the budgetary commitment can be spread over a fairly long period.

Ø One of the features of the instrument is that, from the outset, the body granting the subsidies knows for certain what its intervention will cost, something which is not the case with other instruments such as guarantees or equity participations.

Ø Although interest-rate subsidies are an instrument of micro-economic policy very different from macro-economic policy instruments such as the interest rate (monetary policy), the effects work through in the same way in both cases: the reduction in the cost of credit has a beneficial effect on economic activity, and in particular on investment and then employment.

Ø Using the Quest model, we have been able to estimate the impact on investment and employment of an interest-rate subsidy policy covering all enterprises (we were unable to use the model to carry out an econometric estimation for SMEs alone). The result (see Annex V: Tentative econometric simulation) is clear: an interest-rate subsidy of 0.5% in real terms reduces the cost of capital by 7% at the end of the period and increases industrial investment by 5%.

Ø These estimations are sufficiently robust to answer the criticism levelled in certain quarters ("windfall") to the effect that the investor receives the subsidies but his decision to invest is taken regardless of the measure in question.

Ø The instrument of interest-rate subsidies is used for various purposes: environmental projects, infrastructure projects, social policy projects, etc. We have not been able to undertake any quantitative estimations of the effectiveness of such intervention but few doubts remain as to the beneficial effect of interest-rate subsidies on such forms of investment: witness the level of financial charges in the business plans for such projects.

2.3. Interest-rate subsidies: negative effects

The undesirable effects of this instrument are well known.

Ø Like any public instrument, it may lead, macro-economically speaking, to an inefficient allocation of resources, with financial efficiency being defined as an allocation of resources to investment with the highest anticipated rate of return. This may in turn result in less profitable projects being undertaken or give rise to activities that the market would not have financed, etc.

Ø Other, more specific distortions of competition may, however, occur: on the demand side either because certain measures are not sufficiently publicised or because the complexity of the conditions imposed means that the scheme is not readily understandable to those eligible; on the supply side either because certain financial institutions reserve the instrument for their best clients or because the public authorities entrust management of the instrument to a preferred intermediary and exclude competitors.

Ø From an operational viewpoint, the instrument also has a number of pitfalls that are associated with similar instruments, in some cases to a greater degree:

(a) the passing on in full of the interest rebate to the final beneficiary sometimes lacks visibility and transparency and is not even always guaranteed;

(b) "public" control of interest-rate subsidies, which is essential in the case of such expenditure since it involves public monies, is not always easy. It is even difficult sometimes to ensure that all the conditions for granting subsidies have been met; abuses may occur;

(c) moreover, such controls and the management of the programme add significantly to the administrative costs. It is estimated, for example, that a posteriori documentary checks on the regularity of the operations of financial intermediaries cannot process more than 25-30 projects a day. For some programmes, the total management cost has been put at some 7-8% of the appropriations earmarked for the subsidies.

Ø It should be borne in mind that interest-rate subsidies have become much less attractive for promoters since interest rates in Europe have fallen. None the less, a given percentage subsidy still represents a large proportion of the interest burden.

Ø Lastly, it must be added that these positive and negative effects of interest-rate subsidies are not an exclusive feature of this instrument but are common to other instruments too.

3. PART II: THE LESSONS OF EXPERIENCE

The interest-rate subsidy instrument is commonly used at international level (by multilateral financial institutions), at national level (by the Member States) and at Community level (by the European Union). For the purposes of this report, use at international and national level is examined briefly and only by way of example.

3.1. Action at international level

We will confine our analysis to two cases: the initiatives undertaken by the World Bank and those undertaken by the Council of Europe.

3.1.1. Loans subsidised by the World Bank

The bulk of World Bank loans to its member countries are provided by the IBRD (International Bank for Reconstruction and Development) on market terms, i.e. without any interest-rate subsidy. A second group of World Bank loans is designed for direct project financing via the International Finance Corporation (IFC), which offers credit lines (without interest-rate subsidies) to local financial intermediaries for five priority purposes: infrastructure, SMEs, strengthening of financial structures, information technologies and social needs (education and health).

Subsidised loans for the poorest countries are disbursed under the IDA programme (International Development Association), which is administered by the IBRD. Finance is provided in the form of outright grants from the richest countries in line with a programme that is updated every three years. These resources and the redemption of earlier loans permit the disbursement of loans to the most heavily indebted countries at a rate of 0% (plus 0.75% loading) for a period of 30-40 years and with a grace period of ten years. [1]

[1] At the time of the first oil shock in 1975, the World Bank inaugurated a Third Window to act as a half-way house between the two previous facilities and financed by donations from the OECD and OPEC countries. The intention was to grant subsidised loans to Member States experiencing difficulties in the wake of the increase in interest rates. The experiment was short-lived (1976-77) because donors were reluctant to provide top-up financing.

Assessment

During the tax year 2000 (which ended on 30 June), the World Bank lent USD 15.3 billion, of which USD 10.9 billion in the form of commercial loans and USD 4.4 billion (29% of the total) in the form of subsidised loans. It had been much more active the previous year (USD 29 billion in total) in the wake of the Asian crisis. The assessment made by the World Bank of its assistance in the poorest countries is a function of how stringent financial management in the beneficiary country has been. According to the World Bank, in those countries that respect the fundamental equilibria and free trade, USD 1 in aid can be tripled thanks to the confidence it imparts to private investors. On average, however, USD 1 in aid is reckoned to generate USD 1.5 in investment. These estimates relate to aid as a whole (all types taken together) granted by the institution, and not merely to interest-rate subsidies.

3.1.2. Experience at the Council of Europe

Lending by the Council of Europe takes place through its Social Development Fund, which, under its statute, intervenes in two clearly defined situations: (i) aid to migrant workers (housing, healthcare, education, training, etc.) and (ii) reconstruction of areas affected by natural disasters. In the latter case, the Fund has in the past launched initiatives similar to those of the Commission in disaster areas in Greece and Italy. Its involvement is on a smaller scale than that of the World Bank, with funding of EUR2 billion a year for between 30 and 40 projects.

Concessionary assistance from the Fund is intended for initiatives concentrated in very needy areas (currently the CEECs) via financing obtained on the foreign exchange market and on the best terms available at the time (triple A). This results in a substantial interest-rate differential relative to the local market to which can be added an interest-rate subsidy of between 1.5% and 3% according to circumstances.

Assessment

The effectiveness of such lending must be assessed in the light of the social mandate entrusted to the Social Development Fund. The Charter's signatory states attach importance to such lending; clear and easily verifiable objectives ensure that the loans enjoy political visibility and are attractive. Intervention by the Fund signals support for the policy being implemented by the recipient country, thereby facilitating the inflow of private capital from abroad.

3.2. Member States: measures involving interest-rate subsidies

Ø Many Member States grant interest-rate subsidies for a variety of objectives: general aid schemes for productive investment by SMEs, the craft sector, etc. and specific measures (export promotion, fostering of innovation, help for young entrepreneurs, etc.) or social measures (housing policy).

These interest-rate subsidy schemes are sometimes part-financed by Community programmes under the Structural Funds (ERDF, EAGGF) with a view to promoting economic and social cohesion. The public authorities responsible for carrying out multiannual development programmes in eligible regions administer themselves the interest-rate subsidies for final beneficiaries or entrust their administration to financial institutions on a contractual basis.

Annex III reviews government measures in Italy, France, the United Kingdom, Spain, Germany and Sweden (for operational reasons, we have been obliged to limit the number of cases studied). Recourse to interest-rate subsidies is fairly widespread in the other Member States too (Finland, Portugal, etc.).

A number of examples will serve to illustrate the orders of magnitude involved in Member States:

- in Italy in the period 1995-99 subsidised loans totalled some EUR0.7 billion a year;

- in Germany in 1999 the Special ERP Fund, which centralises the management of interest-rate subsidies on behalf of the State, granted loans amounting to EUR6 billion for investment projects by SMEs totalling EUR15 billion;

- in France the interest-rate subsidy scheme administers loans amounting on average to EUR1.5 billion a year and costs the State some EUR250 million in subsidies;

- in Spain interest-rate subsidies administered by the ICO under the ERDF programmes for SMEs during the period 1994-99 totalled EUR194.6 million in respect of loans totalling EUR3.3 billion for investment projects costing some EUR6.1 billion;

- in the United Kingdom the Small Business Loan Scheme provided subsidies for loans totalling EUR70 million;

- in Sweden, where interest-rate subsidies are used largely for housing policy, state assistance amounted to only EUR80 million in 1999, sharply down following a reform undertaken in 1993 and the fall in interest rates.

Ø The main findings of our investigations, which were conducted primarily among enterprises, are the following:

(a) The objectives pursued by a policy on interest-rate subsidies are, generally speaking, fairly well targeted (purchase of equipment, business start-ups, etc.). The reduction in the cost of financing has fostered productive investment and thus strengthened productive capacity and competitiveness, notably in the regions assisted. The effectiveness of national measures and of Community measures administered by Member States is generally deemed to be satisfactory.

The Member States have also set in place an administrative and operational framework designed to guarantee the transparency of the operation. Announcements of these measures are made in the media in order to inform all potential beneficiaries. A particular point to be noted is that the monopoly of administration and indeed of allocation of interest-rate subsidies by preferred financial intermediaries is gradually being replaced by a system of competitive tendering.

(b) The policy of the Member States concerning interest-rate subsidies varies since they are attempting to find the most economic way of achieving their policy goals. Some are restructuring their interest-rate subsidy products to suit the changing enterprise landscape (France), some apply packages of loan conditions (Sweden), some make little, if any, use of interest-rate subsidies (United Kingdom), and some continue to use them to pursue their policy aims (Germany, Finland, Spain, etc.).

3.3. Community measures involving interest-rate subsidies

Annex I gives a concise description of EU measures and programmes providing interest-rate subsidies and of the resources devoted to them.

Unlike Member States' measures part-financed by the ERDF and EAGGF programmes, Community measures are implemented directly by the Commission, which in many cases entrusts their management to the EIB on a contractual basis.

3.3.1. A tentative classification

For the purposes of the analysis, these measures can be classified into two groups (see also table below):

(a) Measures designed to combat market failures, with the practical objective being:

(i) macro-economic, such as the Edinburgh facility, designed to create jobs in SMEs without distinction;

(ii) meso-economic: this includes measures to promote sectors such as the environment (MEDA) or infrastructures (Trans-European Networks), where projects are characterised by a clear public interest and poor profitability prospects. It also includes measures to assist sectors undergoing conversion (ECSC) and regions lagging behind, where there is a mix of social and development objectives. Subsidies tied to soft loans for developing countries also fall into this subgroup;

(iii) micro-economic: this includes Community measures in support of innovative firms, business start-ups, etc. where the entrepreneurial risk is considered by investors to be unduly high.

(b) Measures to deal with fairly unpredictable external shocks; this group includes:

(i) measures with a social or solidarity objective, such as the operations to assist regions devastated by natural disasters and the measures in favour of Madeira, Kalamata and southern Italy that were authorised by the Council and amounted respectively to EUR15.9 million, EUR100 million and EUR1 billion in loans, together with a 3% interest subsidy that was paid to the European Investment Bank or direct to the regional authorities as in the case of Madeira;

(ii) measures with a macro-economic objective, such as the former European Monetary System facility for countries facing external difficulties threatening their continued participation in the EMS as a result, for example, of a sharp increase in oil prices or an attempt by foreign-exchange markets to destabilise the currency. This facility was applied only once (for Italy and Ireland jointly) and has since been discontinued.

In aggregate, the EU's budgetary resources devoted to this instrument have been substantial, approaching EUR3 billion. With a subsidy rate of around 3%, this figure gives some idea of the volume of loans involved (leverage effect). Annex I provides a breakdown of the subsidies by instrument. It should be borne in mind here that the bulk of the subsidies have been granted and administered by the EIB, which also disbursed the corresponding loans.

Community programmes involving interest-rate subsidies

>TABLE POSITION>

>TABLE POSITION>

3.3.2. Community programmes

The task of assessing the effectiveness of this instrument in the various programmes mentioned above (described in Annex I) is not easy; the small amounts involved, the difficulty of calculating accurately management and control costs, the lack in some cases of quantifiable objectives, the induced effects of the measures, etc. make any quantitative assessment of effectiveness unreliable.

Generally speaking, however, the analytical criteria described in Part I suggest that interest-rate subsidies are an appropriate intervention instrument particularly as:

- they are designed primarily to make up for poor project profitability and not to offset risk;

- they are properly targeted, thereby reducing the negative effects described above, with a resulting increase in effectiveness;

- ex post control and evaluation of the programmes are possible and reasonably guaranteed, as is transparency (the subsidies are passed on in full to the final beneficiaries).

By following this line of reasoning, the following observations can be made:

(a) As regards the Community measures described at point 3.3.1 ("solidarity measures"), interest-rate subsidies appear appropriate since, on the one hand, these measures are closely targeted and involve investments - notably in infrastructures - with low profitability and, on the other, the negative effects (discrimination, distortion of competition, etc.) are insignificant. [2] In some cases, however, there is still a problem of control because of the increase in the number of bodies intervening and the large number of recipients.

[2] It should though be pointed out that, since they are geared to enhancing solidarity, these Community measures must be evaluated by applying more than a simple economic calculation.

The same can be said of concessionary loans to public and private operators in the ACP countries, especially where tangible investment projects are being financed. Concessionality here often takes the form either of a separate financial allocation earmarked for specific technical assistance linked to the investment in question or of a reduction in the rate with a view to facilitating debt servicing. It is thus more than a simple matter of an automatic reduction in the rate and contributes to attainment of the development objectives.

(b) Interest-rate subsidies for transport infrastructures (excluding the telecommunications sector and certain energy sectors) and for environmental projects are especially appropriate, notably in the case of projects where the Community interest is in addition to the national interest (missing links). These are tangible investment projects that have poor profitability but are of considerable public utility, and the control of the public monies involved is relatively easy. What is more, not only are the negative effects of any distortion of competition very limited but, when completed, these major works are quite often an important factor enabling economic operators to participate fully in the internal market.

(c) In the case of competing enterprises, the way in which interest-rate subsidies are administered differs according to whether or not they are located in regions receiving assistance from the Structural Funds. Subsidies granted to enterprises in such regions are included in point 3.2. under subsidies administered by Member States.

In the other regions, a distinction must be made between newly established SMEs and SMEs requiring new investment in order to expand. In the former case, and notably where the enterprises are located in growth sectors, the problem is less one of poor profitability and more one of the high risk of failure associated with early-stage development; in many cases they do not have access to credit, irrespective of the interest rate they are prepared to pay. [3] By contrast, the rates of return of those enterprises which move beyond early-stage development and are located in growth sectors are often very high. Under the circumstances, therefore, it is evident that subsidised loans are not the appropriate tool here since, in addition to the rigid nature of loan redemption, it is highly probable that the subsidies will be redundant. In the event, other instruments such as guarantees or risk capital as well as horizontal measures such as tax reform and simplification of administrative procedures may possibly be more appropriate (although each instrument has its own limitations).

[3] Newly established SMEs have to invest heavily in production, distribution and marketing and do not make profits during the first few years. The breakeven point could be in five years, or even later. Banks are reluctant to extend financing under such circumstances.

(d) The granting of interest-rate subsidies to SMEs generally, e.g. under the Edinburgh facility, poses certain questions. Leaving aside the fact that the programme objective (job creation) was achieved at a very low budgetary cost, [4] the first remark concerns the size of the programme. Since they have a very limited budget, these measures benefit only a small number of enterprises, and the macro-economic objective (reduction in unemployment) cannot be achieved using the instrument. There is, therefore, something of a mismatch between the objective and the instrument.

[4] The average subsidy for each new job was only EUR1 716 and represented extremely good value for money (Commission report COM(2000) 376).

From an operational viewpoint, since the financial circuit through which the public aid is channelled is, of necessity, very long, transparency is not fully ensured, but this is a feature common to other programmes too.

4. CONCLUSIONS AND RECOMMENDATIONS

Ø Interest-rate subsidies are an instrument used widely in Community programmes in pursuit of various objectives:

- political objectives: natural disasters in the countries of the Union as a symbol of solidarity; subsidised loans in the CEECs (preparation for enlargement); the Balkan countries (regional stability) or the ACP countries (EU presence), etc.;

- economic objectives: structural problems, regional disparities, market failure, support for SMEs, launch of infrastructure projects, anti-cyclical instruments: unemployment, etc.

Ø The main conclusion of this report is that it is not possible to state definitively that interest-rate subsidies are an effective instrument; this depends not only on the nature of the objectives of a particular measure and the manner in which the subsidies are administered or utilised [5] but also on the scope for exercising strict control and, lastly, on the distribution chain for the subsidies and the quality of intermediaries.

[5] Interest-rate subsidy schemes differ greatly, e.g. between TENs and MEDA or other RELEX programmes; subsidies are paid and administered in the former case by the Commission according to its own control criteria and in the latter by the EIB according to a different arrangement.

The analysis made in this report has nevertheless brought to light a number of underlying trends regarding the merits or limitations of the instrument. With all the reservations that any attempt at simplification entails, the situation can be summarised as follows:

Strong points:

- Low budgetary cost; [6]

[6] However, where the choice of financial instrument for certain Community programmes (e.g. EAGGF) is left to the Member States, experience has shown that the choice made is not neutral from a financial viewpoint and in particular that it is unlikely to minimise the budgetary costs to the Community.

- Leverage effect (multiplier);

- Additionality;

- Linkage to a well-known form of general support (bank loan);

- Symbol of Community solidarity.

Weak points:

- Risk of inefficient allocation of resources;

- Cumbersome administration;

- Not-inconsiderable administration costs;

- Control and transparency not always ensured;

- Risk of discrimination;

- Low visibility.

Ø Management and control are of major importance. Since interest-rate subsidies are tied to loans, overall management is, in most cases, entrusted to the EIB under an agreement signed with the Commission. Given the number and type of beneficiaries (SMEs, etc.), other national financial institutions intervene in the distribution chain, and this necessarily complicates programme monitoring and control. It would seem that this constraint is, however, difficult to avoid.

The issue of administration costs should also be commented on. These are, it is true, significant in the case of interest-rate subsidy programmes or any other programmes that cover thousands and even tens of thousands of small beneficiaries. That said, the commercial nature of the intermediaries and competition between banking institutions help to keep down costs. Moreover, in some cases, the amount received by intermediaries has to be considered inadequate in relation to the costs incurred, especially in the case of intermediaries who set up procedures to handle small subsidies. [7]

[7] For instance, in the implementation of the SME facility, the EIB, the national agents and the intermediaries shouldered a not inconsiderable administrative burden. In many cases, the intermediaries created new internal procedures to accommodate the scheme and produced publicity material at their own expense. The intermediaries were remunerated and received fees amounting to 1% of subsidies disbursed (SME Facility Fifth Report).

Lastly, the constraint imposed by administrative costs must be set against the advantage of having access to a vast banking network allowing Community initiatives to be implemented within a reasonable time frame.

Ø The question as to which financial instrument is better adapted to the objective is discussed in detail in the report. [8] To sum up, this analysis suggests that interest-rate subsidies are appropriate for remedying serious deficiencies in the financial sector and for supporting projects with low profitability in special areas requiring assistance (developing countries, assisted regions, etc.). They are also appropriate when the project generates indirect positive effects (externalities) that cannot be reflected in the user price of the service (environment, transport links, etc.) as well as for dealing with external shocks (natural disasters, etc.).

[8] See Commission working paper, Annex IV "Theoretical considerations".

Ø The comparison between instruments does not permit any general clear-cut conclusions either.

Interest-rate subsidies and direct donations (lump-sum) are virtually identical. [9] Donations (discounted subsidy paid as a single amount anticipated at the time of the loan) also aim to reduce the cost of borrowed capital. But, unlike a donation, the beneficiary of a subsidy is still subject to the monitoring conditions imposed by banks throughout the loan redemption period. Such monitoring comes on top of that exercised by the public body granting the subsidy and thus complements it. However, donations are simpler to administer and more transparent [10] but they have a more adverse impact on the Community budget (concentrated in the first year).

[9] However, the implications for the beneficiary's cash flow differ.

[10] TENs financing is one example even though this project category also contains exceptions for interest-rate subsidies (see point 5.3 in Annex I).

Recourse to guarantee-type instruments seems more justified than recourse to interest-rate subsidies in cases where the financial system is reluctant to intervene because of the high risk of the project (risk aversion), where information is asymmetrical or where the small scale of the project does not justify the risk-assessment costs. In these cases, the guarantee allows SMEs access to credit by diversifying the risk among a large number of lenders. In some cases, however, guarantees may be prejudicial to the principles of sound administration (so much so that this possibility can be minimised by close project monitoring and control).

Risk capital is a more flexible instrument than bank loans (interest payments are not rigidly fixed) and very appropriate when it comes to risk sharing, notably in the earlier phases (start-up, early-stage, etc.) of firms with high growth potential. The clear advantages of the instrument are offset by the very small number of potential users and by the fact that the institutional investor (venture capitalist) does not have a guaranteed exit (small, fragmented stock market and regulatory barriers [11]).

[11] See Risk Capital Action Plan (COM (2000) 658 final).

It goes without saying that some of the advantages and disadvantages of interest-rate subsidies are also common to other financial instruments.

Ø A number of conditions have to be met if interest-rate subsidies are to remain effective:

- Strict compliance with state aid legislation in order to limit any undesirable side effects.

- Genuine competitive bidding between financial intermediaries can help to reduce administrative costs. Moreover, the lack of any real competition between intermediaries can create distortions on financial markets and jeopardise the transfer of the subsidy to the final beneficiary. In such cases, interest-rate subsidies must be avoided.

- Improved targeting of objectives/beneficiaries must be accorded priority (the allocation of interest-rate subsidies to enterprises in general can be criticised except in the case of specific regions, sectors or tangible investment projects). Such assistance must be for a limited period.

- The programmes should have a budget that will trigger a visible impact and amortise implementation costs.

- Public controls must be a requirement that is explicitly spelt out in contracts but any duplication should be avoided. The cost of controls must not outweigh their benefits. In the case of financial intermediaries, there is no manageable way to ensure to what extent an interest-rate subsidy is actually transferred to the intended final beneficiary and to what extent it benefits the intermediary. This is not a question of abuse that could be resolved by stepping up control operations; it is inherent to the nature of interest-rate subsidies and is due to the fact that the conditions for individual loans cannot be regulated by the Commission.

- Interest-rate subsidies must be accompanied by simple and clear rules. Complicated implementation and quantification conditions must be avoided.

- Programme implementation must generate value added for national measures (subsidiarity).

- The monitoring and assessment aspects must be reinforced. Any programme must be subject to an ex ante evaluation, a mid-term evaluation and an ex post evaluation.

Ø The fall in interest rates [12] and the availability of other financial engineering instruments have now reduced the relative importance of interest-rate subsidies. [13] Encouragement should, therefore, be given to the use of complementary or alternative instruments. What is more, a combination of two or more instruments may prove more beneficial than reliance on a single and exclusive financial instrument. Partnership (co-financing, etc.) between the public and the private sector should also be stepped up. It should, however, be pointed out that interest-rate subsidies are tied to bank loans, which were, and are still, the main source of financing for productive investments (it is a well-known fact that most SMEs are undercapitalised and depend to a large extent on bank finance).

[12] However, a fall in nominal rates does not necessarily mean a corresponding fall in real rates.

[13] It is important to recognise that interest-rate subsidies are becoming a less important part of overall loan conditions. For entrepreneurs and start-ups, it is much more relevant to obtain long-term finance (up to twenty years) and a grace period, which cuts capital costs in the early phase (up to three years). For commercial banks channelling promotional loans and quasi-equity to entrepreneurs, the built-in risk-sharing elements of such loan schemes are a relevant factor reducing their exposure. Currently, access to long-term finance at reasonable cost, and not merely the interest-rate subsidy, makes schemes attractive for entrepreneurs and start-ups in all sectors of the economy.

Ø For a full appraisal of the instrument, account also needs to be taken of the views of all those involved, particularly national intermediaries and final beneficiaries.

Financial intermediaries, which play the key role in the distribution chain, mostly take a positive view of interest-rate subsidies for the following reasons:

* The direct effect and simplicity of the instrument: banks provide interest-rate subsidies to firms to which they have granted credit. This is a more effective and more direct mechanism than a guarantee line, which requires a more complex and more uncertain procedure.

* The temporary and revocable delegation of responsibilities, with a posteriori control, allows banking networks to assume part of the distribution cost under optimum conditions. Value added stems from the dilution and transformation of the cost within an overall financial and commercial relationship.

Where financial beneficiaries are concerned, it is well known that bank loans (and hence interest-rate subsidies) are preferred to equity participations by outside shareholders on account of the partial loss of management control entailed by risk capital. However, for some enterprises representing a very small proportion of the total number of enterprises in the European Union, other forms of financial engineering have their attractions. Since enterprises learn as they go along, SMEs could abandon loan financing in favour of more sophisticated options.

Ø For the reasons given above, interest-rate subsidies have played an important role in the past. Generally speaking, given the historical context, their positive aspects can be regarded as having outweighed the constraints they impose. The question of abandoning them must not, therefore, be based on results (overall, the objectives have been achieved) but rather must be viewed in the context of the emergence of a new financial environment.

Under the circumstances, it would seem expedient for innovative firms facing high risks to opt, among the financial instruments available, for guarantee or risk capital arrangements when having to contend with market failure, notably in the start-up phase.

Elsewhere, the question of abandoning interest-rate subsidies in favour of other instruments should be examined closely. [14] This observation applies above all to programmes for areas where financial markets are not yet highly developed (developing countries, CEECs), for certain less-favoured regions in the European Union or for certain project categories (environment, TENs, etc.) that give rise to externalities.

[14] For example, risk capital schemes cover a very small number of enterprises. In 1999 only 3000 start-ups were financed by risk capital in the EU whereas some 1. 5 million SMEs had recourse to bank lending.

ANNEXES: COMMISSION WORKING PAPER ON INTEREST-RATE SUBSIDIES

5. Annex I: INTEREST- RATE SUBSIDIES: COMMUNITY Programmes

5.1. Summary table of programmes

Programme // Objective

Market failures: //

ECSC conversion (Art. 56) // Employment and access to credit for SMEs

Trans-European Networks (TENs) // Transport infrastructure

Edinburgh facility (SMEs) // Employment and access to credit

Development cooperation (ACP) // Infrastructures and access to credit for SMEs

Economic development (MEDA and Slovenia) // Environment and transport infrastructure

Economic and social cohesion (ERDF, EAGGF) // Access to credit for SMEs

Phare programme (CEECs) // Access to credit and privatisation

External shocks: //

Natural disasters (Italy, Greece, Kalamata, Madeira) // Reconstruction and restoration of productive capacity

European Monetary System // Infrastructures

Peace and reconciliation in Ireland // Access to credit for SMEs and infrastructures

Reduction in disparities (EEA) // Access to credit for SMEs and infrastructure

Reconstruction in the Balkans (FYROM and Bosnia) // Restoration of productive activity

5.2. ECSC subsidised loans (Article 56: industrial conversion)

A. Description

The purpose of subsidised conversion loans (Article 56 of the ECSC Treaty) is to assist in the economic conversion of regions affected by the decline of the coal and steel industries, by making a financial contribution to job-creating investment. The first loans were disbursed in 1961.

Amount of subsidies

Initial subsidies granted between 1993 and 1997

1993 EUR37 693 920 for 20 911 jobs created

1994 EUR26 275 492 for 13 213 jobs created

1995 EUR21 614 248 for 11 138 jobs created

1996 EUR13 732 288 for 7 249 jobs created

1997 EUR22 948 045 for 11 482 jobs created

Total : EUR122 263 993 for 63 993 jobs created

Since the ECSC Treaty will expire in July 2002, the last loan instalments were disbursed in May 1997, so that they can be repaid in May 2002, before the expiry of the Treaty.

Beneficiary industries/firms

Any firms located in an ECSC area, but particularly SMEs.

All industries are eligible, except those on a negative list, such as:

- the retail trade where the vendor does not process goods;

- hospitals;

- educational establishments;

- the steel industry, etc.

Criteria for granting interest-rate subsidies

- Investment

The amount invested must be at least twice the ECSC loan.

- Jobs

Jobs must be created not later than three years after disbursement of the loan (assessment date).

They must be maintained for at least the following two years.

B. Operational aspects

Management

The initial amount of the interest-rate subsidy is calculated on the basis of the number of jobs to be created and the loan amount requested.

The subsidy is paid in three equal instalments, on the following due dates:

1. First instalment: one year after disbursement of the loan.

2. Second instalment: three years and six months after disbursement of the loan, following verification of the business report demonstrating that the objectives of carrying through the investment and creating jobs (social clause) have been achieved.

3. Third instalment: five years after disbursement of the loan, following verification of the business report demonstrating that the jobs created have been maintained.

Control and procedures

Project control is decided on as part of the annual missions schedule laid down by DG ECFIN's Financial Operations Unit 01 (Budget and Financial Aspects).

Depending on the case, the unit member is accompanied by members of other units or institutions, such as:

- the European Court of Auditors;

- the European Commission's Financial Control Directorate-General (DG Audit);

- DG ECFIN's Evaluation and Internal Audit Unit;

- DG ECFIN's Project Management Unit.

Visits are made in the presence of an agent of the financial intermediary concerned by the projects to be checked.

Control is carried out on the basis of official documents or declarations, such as:

- monthly returns of social security contributions;

- wages book;

- annual balance sheets confirmed by the auditors' report, etc.

C. Evaluation

A report exists for the period 1981-90, and a new study is planned for the period 1991-2000. In answer to questions put to it on several occasions by the Court of Auditors and the European Parliament, the Commission has provided general replies on the effectiveness of the actions carried out.

The real impact on job creation of interest-rate subsidies paid by the Commission under Article 56 of the ECSC Treaty can be measured quantitatively as follows:

- the ECSC disbursed 15 000 loans and sub-loans in the period between 1982 and 1997, most of which were granted to small and medium-sized enterprises;

- these loans helped to create about 285 000 new jobs at a budgetary cost of ECU 490 million.

The ECSC has never claimed that these loans and interest-rate subsidies alone were responsible for creating these 285 000 new jobs; rather, they formed part of a package of measures at European, national and local level which stimulated job-creating investments.

According to information received from banks acting as the Commission's financial intermediaries and from beneficiary enterprises, the loans and interest-rate subsidies received under Article 56 of the ECSC Treaty for job creation have, in fact, played an important role in the investment decisions of the enterprises concerned.

5.3. Trans-European Networks (TENs)

A. Description

From 1995-99, legal basis: Council Regulation (EC) No 2236/95 of 18 September 1995 laying down general rules for the granting of Community financial aid in the field of trans-European networks.

From 2000, legal basis: European Parliament and Council Regulation (EC) No 1655/1999 of 19 July 1999 amending Regulation (EC) No 2236/95 laying down general rules for the granting of Community financial aid in the field of trans-European networks.

Amount of interest-rate subsidies committed in the last five budget years (EURm, approximate):

1995: 25

1996: 18

1997: 37

1998: 20

1999: 71.

Ø Objectives

Interest-rate subsidies are intended to assist in the launch of transport infrastructure projects, mainly by reducing the financial burden on implementing authorities in the pre-operational phase. They reduce the interest costs, or financing costs in general, of a loan. Some large TEN projects have been funded primarily by loans from public financing bodies (e.g. EIB, National Debt Offices, etc.) or from the international banking system.

Ø Criteria for granting interest-rate subsidies

To ensure agreement on the administrative procedures to be followed in the use of interest-rate subsidies, criteria were laid down by the Commission departments responsible for transport, subjected to inter-departmental consultation and agreed by the TEN Financial Assistance Committee in 1997.

Regulation (EC) No 2236/95 stipulated in Article 4(1)(b) that Community aid for projects may take the form of subsidies of the interest on loans granted by the European Investment Bank or other public or private financial bodies. In addition, a declaration attached to the minutes of the Council meeting at which this Regulation was adopted gave some further indication of the proposed use of this form of support; the Commission declared that the amount of interest-rate subsidy from the Community budget allocated to a project in a given year should always represent a fraction of the volume of interest payments actually borne by that project during the year in question.

The rules for interest-rate support for TENs proposed by the Commission in 1997 made a distinction between different types of promoter; in particular, between single-purpose authorities, on the one hand, and multi-purpose authorities, on the other.

Single-purpose authorities:

In the case of single-purpose authorities, all expenditure is effectively aimed at meeting the objectives of the project which the Commission seeks to support. It therefore seemed unnecessary and unjustifiable to demand that each individual loan taken out by the authority should be assigned to a physical objective. In such cases, loans are generally raised to meet overall liquidity needs and not to pay for individual actions at any event. Loans are swapped in some cases according to the conditions on international currency markets.

In such cases, the following rules apply:

- interest-rate subsidies not to exceed five years, this being the general rule laid down in Article 4(1)(b) of Regulation (EC) No 2236/95 (although loan repayment itself may need to be monitored over a longer period, depending on circumstances), and to be below actual accumulated interest expenditure in each eligible period;

- interest-rate subsidies only for amounts of over ECU 1 million per year (for cost-effectiveness reasons);

- support can be granted to either newly negotiated or existing loans;

- interest-rate subsidies would be paid after receipt of invoices showing that interest had been paid, and progress reports showing the achievement of physical milestones, and related payments, as set out in the relevant Commission decision and application form;

- interest-rate subsidies will be paid direct to the beneficiary, where possible;

- overall support under the Regulation will be below 10% of the total expenditure on the project during the entire period of support (i.e. within the spirit of the TENs Financial Regulation limits on the financing of construction). (This may give the Commission the flexibility to pay interest-rate subsidies of over 10% of expenditure for specific years, e.g. during the launch period, when financial costs may be especially high. This form of "accelerated" payment will in a way mirror the normal advance payment of 40% for direct grants.)

Multi-purpose authorities:

In the case of multi-purpose authorities, the above individual points still apply but it cannot be said in these cases that any loan is eligible for interest-rate support. Clearly, the loan or loans supported by the Commission must be earmarked for the specific projects (i.e. track sections, equipment installations, etc.) which it wishes to support. In practice, this specific link is more likely to exist when the loan has been taken out from an institution like the EIB, which itself demands regular reports on the achievement of physical objectives.

B. Operational aspects

Ø Control

Essentially, the two key principles of control are the same as those applying for traditional direct grant projects, i.e. (i) measuring physical progress and (ii) checking the eligibility of expenditure and verifying the correctness of accounts.

For both types of project authority (see above), the Commission needs to ensure as a condition of payment that proof of loan agreements and obligations, and the actual repayment of financial costs, is accompanied by satisfactory physical and expenditure progress on launching or continuing the project (according to the stage in its evolution). Payment of initial tranches of interest-rate support and of the final tranche is contingent on progress towards and achievement of the objectives respectively. Payments to contractors should be verified and audited.

Lack of physical progress, or failure to launch, will mean that interest-rate subsidy payments should be withheld by the Commission, as would be the case with payment tranches in a direct grant scenario.

Finally, the amount paid should be flexible, although linked to physical progress, and will not exceed 95% of the amount of the interest-rate subsidy specified in the decision so as to allow for an amount to be held back pending accounting controls and assessment of the achievement of physical objectives set out in the decision.

Ø Visibility

The exact nature of this concept is hard to define. TEN projects supported by interest-rate subsidies are essentially subject to the same general rules regarding physical progress as direct grant projects and hence to associated requirements (e.g. for publicity to be given through signboards, etc. showing the role of the European Union). Part of the purpose of control is to ensure visible progress in this sense.

C. Evaluation

As part of the mid-term evaluation of the TEN-T budget interventions, which the Commission departments responsible for transport commissioned and which was delivered by outside experts in 1999, the efficiency of interest-rate subsidies ("interest rebates") as a financial instrument was discussed. An extract from the evaluator's report follows:

"The instrument of interest rebates is seen as the second best solution and generally of more value with high and volatile interest rates. Nowadays nominal interest rates are lower and in some Member States some transport operators (such as railways) can receive high grants from their governments and interest-free loans making the instrument of interest rebates even less attractive.

In addition the duration of the interest rate subsidy is not always compatible with the loans and project duration. The present limit of five years is considered as a serious constraint for the development of this instrument.

However Member States do not reject the possibility of applying for interest rate rebates in the future in the light of PPP approaches. Therefore, the Commission should better explain and spread more information on the reasons and advantages for project promoters to apply for this instrument. More information on the existing TEN-T projects benefiting from interest rebates should be provided to other Member States.

Three projects have applied for this instrument to date : the Øresund fixed link, the Belgian HST, and Malpensa Airport. In these three cases, the choice of this form of intervention was suggested by the Commission and the three projects received loans from the European Investment Bank.

* Interest rebates function well for the Øresund connection but have otherwise not been applied in Sweden. This type of support has been discussed as a possibility for the third runway at Arlanda airport. The civil aviation administration did, however, not consider them as a feasible solution as loans for the runway will have to be raised by the national civil aviation administration proper, and not by the airport, which organisationally is only a division within the administration.

* Interest rebates helped to complete the financial plan of the Belgian HST project and launch the construction phase. As the Belgian Railways had to borrow the funds needed to build the HST, the financial return of the project was highly influenced by the interest charges. This instrument appeared to be the most adequate one in this particular case and at that particular moment in time. It allowed Belgian Railways to be provided with EU aid that was not directly linked to the stretches of routes for which the Belgian requests were originally made.

* The Malpensa 2000 project is an example of a project for which this kind of support can be well suited. It was the Commission itself who suggested this solution, believing that it was the most appropriate one for this project.

Some Member States suggested that the interest rate form of aid should be allocated to the promoter as a lump sum rather than be related to a specific loan. Railway companies do not always borrow money for a specific project, but loans are arranged to finance their global investment programmes. If the interest rate subsidy form of aid is granted provided that it is linked to a specific and well-defined loan scheme, then it can perturb its financial management or impose increased auditing procedures."

There are also advantages to interest-rate subsidies for TENs : they are good for private or semi-private project authorities which have to face high financial costs in the years before their project starts to yield revenue from tolls, etc. (this was certainly the case for the Øresund fixed-link project).

5.4. Interest-rate subsidies to help certain Member States to participate in the European Monetary System

A. Description

At the end of December 1978 the Council adopted Regulation (EEC) No 3181/78 relating to the European Monetary System (EMS). Previously, the Council had recognised the need to support the less prosperous Member States in their efforts to participate in the EMS. Support was to take the form of assistance in carrying out infrastructure programmes and projects financed by the EIB and by the recently established NCI (Decision 78/870/EEC of October 1978).

The Regulation implementing this support was adopted in August 1979 (Regulation (EEC) No 1736/79); it granted a 3% interest subsidy for loans relating to Commission-approved programmes of up to one billion units of account a year for a period of five years. It was also agreed that the discounted value of the subsidies to be entered in the general budget of the Community should not exceed 200 million units of account a year for a period of five years. At the same time, the Council designated Ireland and Italy as the Member States which were eligible for these subsidies. The implicit subsidy for them was: 200 million units of account in 1979, 197 million in 1980, 193 million in 1981, 210 million in 1982 and 200 million in 1983.

In accordance with the above-mentioned Regulations, eligibility depends on the execution of major infrastructure projects in Italy and Ireland that meet EIB and NCI operational criteria and are capable of anchoring these countries' economies more firmly in the European Monetary System. By financing individual projects of undisputed public utility, the aim was therefore to pursue an overall (macroeconomic) objective in parallel with other instruments with the same purpose. It will be remembered that at the time the oil crisis had destabilised the industrialised countries' balances of payments and exchange rates and Community solidarity had been mobilised on behalf of the weakest countries, i.e. Italy and Ireland, in the form of Community borrowings (Article 108 of the Treaty). The interest-rate subsidies for infrastructure projects thus reinforced measures to restore external equilibrium.

B. Operational aspects

The subsidies were managed by the EIB, which both selected the projects and disbursed the aid.

In view of the relatively large subsidy amounts decided by the Council, the EIB kept very large-scale projects separate from those it usually managed so that they could be monitored more easily. In some cases it selected tranches of programmes which had been started before or were to be completed after the eligible period (1979-83) because of the scale of the projects and the economic importance of the tranches in question.

The subsidies were paid to the project promoters only for the five-year period laid down by the Council. Each half year, the EIB presented the Commission with a refund application calculated on the basis of the subsidised amounts, updated at the beginning of each half year. Italy thus received aid totalling ECU 669 million in respect of a loan volume of ECU 3 336 million; the subsidies for Ireland totalled ECU 331 million in respect of a loan volume of ECU 1 470 million. Control was exercised by the EIB and the Commission, in cooperation with the central banks of the two countries concerned.

C. Evaluation

During its investigations, the Court of Auditors noted delays in the execution of the projects and increases in costs as compared with the original estimates. These overspends were covered by other sources of finance such as the ERDF and state aid.

These criticisms, relatively minor by comparison with the objectives, did not prevent the Court from acknowledging that the objectives had been attained. It stated that these projects had contributed to the development of the water supply and drainage infrastructure in Ireland in the context of the expansion of industrial, agricultural, tourist and urban activity. The Court's assessment of the extension of an aqueduct in Italy was more qualified because of the delays caused by the 1980 earthquake but it did not call into question the key importance of the work for the economy of the Mezzogiorno.

5.5. Edinburgh facility (SMEs)

Council Decision 94/217/EC of 19 April 1994

A. Description

Ø Description

The purpose of this programme was to help enterprises, in particular SMEs, to create jobs. It was decided on in the early 1990s, a period of severe recession, fairly high unemployment and interest rates, and deep pessimism on the part of various economic actors.

The programme is managed by the EIB on behalf of the Commission. In each Member State the EIB dealt with a national agent that coordinated the network of financial intermediaries which selected the individual projects.

The interest subsidy related to interest on a five-year loan financing a job-creating investment project. The financial intermediaries carried out the financial analysis of the credit provided by the EIB. The subsidy was based on the number of jobs created.

The SME facility is a temporary "one shot" facility which, at the request of the Council, had to be introduced very rapidly. The operational part, which began in 1995, was completed in 1997.

Ø Amount of subsidies

- Total amount of interest subsidies granted by the Council: ECU 100 million

The subsidy was tied to EIB loans of ECU 1 000 million to SMEs in the Community. On 8 June 1998 the amount distributed in interest subsidies totalled ECU 92 321 473.

At the end of 1997 subsidies paid totalled ECU 92.3 million. This amount corresponds to a 2% per year interest-rate reduction for a maximum of five years, which is equivalent overall to 10% of the loan amount, or an average of ECU 3 000 per job created.

Ø Beneficiaries

- SMEs, in particular those with no more than 250 employees and a turnover not exceeding ECU 20 million or assets not exceeding ECU 10 million;

- virtually all sectors of activity (some countries introduced exceptions);

- all Member States;

- The loans could not be combined with other subsidised loans granted by the EIB under other agreements or with ECSC-subsidised loans.

Ø Criteria for granting interest-rate subsidies

In its application for subsidies, the SME had to state the amount of financing required from the EIB, the capital cost of the investment and the number of jobs it planned to create; only jobs created after 28 April 1994 were eligible.

In order to qualify for a definite subsidy payment, the beneficiary had to submit a sworn statement concerning the actual net increase in the number of jobs and declaring that the additional jobs had been in existence for more than six consecutive months prior to the date of the declaration. If the number of jobs actually created was smaller, the proposed subsidies were reduced accordingly.

B. Operational aspects

Ø Management

The EIB shared out the facility among the national agents and its network of intermediary banks.

- The interest subsidy is generally paid in a single instalment (lump sum). In Germany it was used to reduce the last repayment of the borrowing.

- In its report the Court of Auditors criticised the fact that the rules were less clear in some countries than in others.

Ø Control

The national agents verified each beneficiary's entitlement to its respective subsidy on the basis of the declarations made by the intermediaries and beneficiaries. As well as the usual monitoring of its global loans, the EIB also made spot checks on the documentary evidence presented in support of the requests for subsidy payments. In 1996, 1997 and 1998 the EIB monitored the procedures applied by agents in eleven countries

In its Report No 6/2000 the Court of Auditors criticised the fact that the Commission did not monitor this facility in detail, despite the cooperation agreement.

Ø Community visibility

- The financial intermediaries' documentation generally referred to Community aid, at least that of the EIB; this aspect was criticised by the Court of Auditors in its report.

Ø Accounting

- The accounting and presentation of the accounts were criticised by the Court of Auditors and the Commission has undertaken to improve them.

C. Evaluation of the action

* Four reports have already been presented to the European Parliament, the last on 2 December 1997 (COM (1997) 645 final).

* The fifth report was being drafted at the same time as this report. It contains statistical tables and is more complete than the previous ones.

5.6. Subsidised loans in the ACP countries

A. Description

Successive European Development Funds have supported the action of the EIB and of local banks by granting general interest-rate subsidies in order to reduce the very high level of interest rates and to provide access to credit for small commercial and craft firms in the countries which are signatories to the Lomé Convention. Since the pressing need for finance is, however, very extensive, subsidies have also been allocated for infrastructure and sewerage works.

During the life of the most recent Conventions, the WTO expressed reservations about this general practice and about the risk of distorting competition in relation to neighbouring countries not eligible for EDF loans. It has so far been possible to circumvent this obstacle by granting loans to public enterprises at market rates, with the State receiving the interest-rate subsidy under a contract signed with the enterprise concerned.

Nevertheless, in the spirit of openness which will guide future international conventions, it has appeared essential to review the system as a whole in order to boost direct aid while reducing indirect aid tied to a loan spread over time. The new ACP-EC Partnership Agreement signed at Cotonou on 23 June 2000, which will replace the Lomé Convention, contains a new financial mechanism - the Investment Facility (IF) - that will make it possible to finance essentially private-sector investment. Like the risk capital funds under the old Convention, this Facility will be managed by the EIB. The main changes introduced by this Facility as compared with the former risk capital funds can be summarised as follows: unlike the risk capital funds, the Facility will function as a revolving fund, thus offering ACP countries a degree of security as to the finance available for private-sector development in the short, medium and long term. In addition, the operations financed by the Facility (essentially in the form of risk capital and loans) will be financed on conditions based on those of the local financial market. However, the agreement provides that interest-rate subsidies could be granted for certain specific operations, namely for:

(1) infrastructure projects in the least-advanced countries or in countries in a post-conflict situation that are essential for private-sector development. In these cases, the interest rate will be reduced by 3 percentage points, and

(2) projects involving restructuring operations in connection with privatisation or projects entailing substantial and clearly demonstrable social or environmental advantages. In these cases, loans may carry interest-rate subsidies the amount and form of which will be decided on the basis of the special features of the project and thus on a case-by-case basis, in line with predefined criteria. For this type of operation, the interest-rate reduction will not exceed 3 percentage points and will therefore be calculated on a case-by-case basis in line with the criteria jointly laid down by the EIB, the Commission and the Member States.

In all cases, the final interest rate on the loan will be no less than 50% of the reference rate, in order to avoid situations in which the final rate would be close to 0% or even negative.

The specific criteria for granting interest-rate subsidies will form part of the Facility's policy guidelines to be laid down jointly by the EIB and the Commission and submitted to the Investment Facility Committee, which will include representatives of the Member States.

The interest-rate subsidy will be financed by the EDF's long-term development resources. The funds necessary will not exceed 5% of the total amount allocated to finance investment by the Investment Facility and by the EIB out of its own resources, i.e. some EUR200 million.

B. Operational aspects

The EIB is currently managing the equivalent of 20.7% of the budget allocations of the 8th EDF, comprising EUR1 658 million in own resources (11.34% of the resources of the Convention) and EUR1 370 million in risk capital and interest-rate subsidies (19.36% of the Fund). As regards operations, the EIB has so far signed over 660 operations in the ACP countries, 190 out of its own resources and 470 out of EDF funds.

The activity of the new Investment Facility will be more targeted than in the past as far as interest-rate subsidies are concerned, but without the automatic aspect which existed in the past; it will still be possible for concessionary loans to be granted in some cases and on the basis of predefined criteria (see above).

The financial allocation for the Facility will be EUR2.2 billion for the First Financial Protocol (2000-05), which will be used to finance investment. To this will be added a further EUR200 million or so to finance interest-rate subsidies for projects financed both by the Facility and by the EIB out of its own resources.

C. Evaluation

It is fairly difficult to make a general evaluation of the impact of subsidies in the context of operations with 72 different ACP countries. Nevertheless, the purpose of the major changes concerning the Investment Facility is to correct the shortcomings of the past and to take into account the more open direction of international trade (WTO rules).

* The new Facility will have to take care not to generate distortions on local capital markets or to undermine local financial institutions. Instead, the financial resources provided should enable these institutions to mobilise other sources of finance.

* The Facility must not disrupt or conflict with national and regional financial and monetary policies or with policies advocated by international institutions.

* Unlike the public sector in the heavily indebted ACP countries, the private sector does not seem to consider the subsidy to be a determining factor in investment decisions, unlike other criteria, viz. provision of medium- and long-term financial resources, rapid disbursement, loans in local currency, etc.

* Concessionary loans may, however, be beneficial in the cases mentioned above since they encourage economic development and support investors in situations where the political and economic risks are particularly high.

5.7. Interest-rate subsidies linked to EIB loans in the Mediterranean

A. Description

Ø Protocols

From 1978 to 1992 EIB loans formed part of the cycle of five-year financial protocols that constituted an overall legal framework for financial and technical cooperation between the European Community and each Mediterranean country and territory; each protocol fixed a financial allocation indicating the origin of credits and the nature of the finance.

The beneficiary countries were then able to use the budget funds to subsidise all EIB loans by up to two percentage points, whatever the sector (infrastructure, water, industry, etc. ). The total amount of budget commitments for these interest-rate subsidies was EUR214 million

Ø Environmental protection: from 1992

In 1992, under the Renewed Mediterranean Policy, the EIB was instructed to implement a specific financing programme for projects to protect the environment; the corresponding Financial Regulation (outside the protocol) provided for an interest subsidy of three percentage points on these projects.

Subsequently, the Barcelona Declaration, adopted on 28 November 1995 at the Euro-Mediterranean Conference of Foreign Ministers, established the Euro-Mediterranean partnership, a joint initiative of the 15 members of the EU and 12 Mediterranean partners. The Council Regulation of 23 July 1996 laid down the financial and technical measures (MEDA) to accompany the reform of economic and social structures of the Mediterranean partners. It provides for a 3% interest-rate subsidy on EIB loans in the field of the environment.

In accordance with these provisions, interest-rate subsidies totalling EUR255 million were granted from 1992 to 1999 to 11 Mediterranean countries and territories and mobilised for 37 different projects. These projects concerned the treatment of waste water in order to reduce pollution in coastal rivers and in the Mediterranean, the reduction of industrial pollution and the treatment of solid waste.

Ø Financial and technical cooperation with the Palestinian territories

The July 1994 Regulation laying down the provisions for this cooperation was amended in December 1998, by providing in particular for the possibility of subsidising the loans granted by the EIB to these territories at the rate of three percentage points. Two subsidies totalling EUR8.6 million were decided in 1999. These subsidies are not restricted to the environment but cover the priority economic and social areas specified in the 1994 Regulation.

B. Operational aspects

Ø Area of intervention

The areas of intervention, which were extended under the protocol system, were restricted to the environment from 1992; the environment has therefore become the main eligibility criterion for the budget grant in the form of interest-rate subsidy, with the exception of the EIB loans granted to the Palestinian territories.

Ø Management

The EIB selects and prepares projects, draws up the financing proposal and then seeks both the opinion of the "Article 14 Committee" (MEDA), consisting of the representatives of the Member States, and the Commission's opinion. On the basis of this consultation, the EIB asks the Commission to take a decision to fund the granting of an interest-rate subsidy. Following an internal consultation process, the Commission submits to the MED Committee a draft decision authorising or, where appropriate, refusing the funding. The Commission decision is forwarded to the EIB; if the decision is favourable, the EIB may grant the subsidised loan.

The EIB and the Commission jointly see to the financial management of interest-rate subsidies on the basis of a management agreement signed between them.

No problem of repayment of EIB loans by the Mediterranean partners has been indicated.

In its Report 1/98 on the management of the protocols, the Court of Auditors referred to dormant commitments or commitments in excess of the needs of the operations concerned; the Commission replied that it had introduced a regular information system and that it had reexamined the dormant commitments as part of a periodic review.

C. Evaluation

Ø MEDA evaluations carried out in 1998 and 1999

In November 1998 the evaluation by external consultants of certain aspects of EU development aid for the Mediterranean region showed that coordination between the EIB and the Commission was working well and was the subject of regular consultations but that it could be improved by incorporating the EIB activities linked to the granting of budget funds into the national indicative programmes at an earlier stage.

In February 1999 the evaluation of the MEDA provisions carried out under Article 15(6) of the MEDA Regulation recommended that the procedure for deciding interest-rate subsidies should be simplified, alone the lines of the other operations managed by the Commission.

Ø Proposal for amending the MEDA Regulation

In its recent proposal to amend the MEDA Regulation (COM (1999) 494 final), the Commission stressed the need to streamline coordination with the EIB, i.e. in the preparation of strategies and at the programming stage; as far as interest-rate subsidies specifically are concerned, the Commission has followed the recommendation made in the evaluation and has proposed lightening the decision-making procedure by providing that, as interest-rate subsidies would be covered by approved national financing plans, they would not need to be submitted separately to the MED Committee; the "Article 14 Committee" approval procedure would require no modification.

The subsidy rate applied to each payment would be equal to half the interest rate relating to the payment in question, but in no case higher than the nominal rate of 3%.

The Commission proposal maintains the area in which interest-rate subsidies would apply, i.e. the environment; given the scarcity of water resources, the growing degradation of natural resources and the dense urbanisation of the region, the environment will become an even more important issue for all the countries bordering on the Mediterranean. Interest-rate subsidies constitute an incentive mechanism which has effectively enabled the Mediterranean partners to embark on specific environmental action programmes; the exception for the Palestinian territories is justified by the very special situation there in the context of the peace process.

Ø Evaluation of EIB activities in connection with the financing of the Community budget

An evaluation of the financial assistance to the Mediterranean countries managed by the EIB on behalf of the EC was launched at the beginning of this year; it covers the interest-rate subsidy and risk capital mechanisms. It will review the period 1979-99 and contain an analysis of the performance and impact of interest-rate subsidies and of the mechanisms and procedures; it will conclude by addressing recommendations to the Commission and the EIB. The results of this evaluation are expected shortly.

5.8. ERDF Programme

A. Description

Interest-rate subsidies part-financed by the Structural Funds, essentially the European Regional Development Fund, are managed not by the Commission but by the national and regional authorities in Member States charged with implementing multiannual regional development programmes within specific eligible areas.

B. Operational aspects

- SMEs in the less-developed areas of the EU face particular problems in terms of the cost of credit: although base interest rates charged by financial institutions are uniform and set nationally, a margin of additional interest is charged by banks to reflect the risks of a particular lending operation, these being higher for SMEs than for larger enterprises and often higher still for SMEs located in the less-developed regions.

- Applications for loans are vetted and granted in the normal manner by lending institutions, which agree the terms of the financing with the SME. The interest-rate subsidy is covered by a separate agreement and is paid to the SME during the term of the loan by the lending institution. The total financial cost to the SME is therefore the commercial interest rate paid to the lending institution, minus the interest-rate subsidy received from the managing authority under a regional development programme measure (such as a global grant) part-financed by the Commission.

C. Evaluation

In regional development terms, the interest cost of the finance required for an investment is reduced, enabling SMEs to invest and thereby to increase their production capacity and competitiveness. The subsidy is covered by a separate agreement from the loan agreement, thus ensuring a swift decision by the lending institution and reducing the administrative burden on the SME.

Interest-rate subsidies available under the Structural Funds are considered to be on an equal footing in terms of effectiveness with the several other forms of grant applied by the Funds. A number of lessons arising out of experience with interest-rate subsidies during the previous programming period are now being put into practice in the Commission's guidelines for the period 2000-06, in particular the need for an initial diagnosis of SME requirements, the need for prior consultation with intermediaries on the design of such schemes so as to avoid an unnecessary administrative burden on the banks and their information systems, and the fact that the targeting of interest-rate subsidies on particular SME sectors has generally enhanced their impact.

Article 29(4) [15] of the new Council Regulation laying down general provisions on the Structural Funds [16] explicitly encourages managing authorities to use forms of finance other than direct assistance, including interest-rate subsidies, by permitting an additional 10% for the maximum rate of contribution, in accordance with Article 28(3).

[15] Article 29 (4) (a) and (b) (iii).

[16] Council Regulation (EC) No 1260/1999 of 21 June 1999.

5.9. EAGGF Programme

A. Description of the action

- Amount of subsidies (statistics for the last five years):

Since the Agriculture Directorate-General does not grant interest-rate subsidies directly but only refunds the subsidies granted by the Member States for agricultural structures (see also the answers to point B below) and since Member States often combine different forms of investment aid (capital grants, interest-rate subsidies, guarantees for loans contracted, etc.) for the same aid scheme, the EAGGF Guidance Section does not possess any analytical statistics relating to the form of aid which it has helped to finance.

Nevertheless, by way of an example to indicate an order of magnitude, France declared for 1999 alone an amount of EUR244 million in interest-rate subsidies incurred in 1998, while each year the EAGGF Guidance Section refunds Member States' expenditure at variable rates set by the rules (25% to 60%) on the basis of various parameters such as the type of aid scheme, the status of the beneficiary or its location.

- Objectives

Interest-rate subsidies have so far been granted mainly in connection with the implementation of Objective 5a of the reform of the Structural Funds, namely the granting of official aid for investments made in agricultural holdings, and setting-up aid to young farmers (Articles 7, 10 and 11 of Regulation (EC) No 950/97).

The system of interest-rate subsidies has been extended for the new programming period 2000-06 (see Article 28(3) of Regulation (EC) No 1260/99 or Article 8(2) of Regulation (EC) No 1257/99).

- Criteria for granting subsidies

Numerous criteria and restrictive conditions have been laid down by Council Regulation (EC) No 950/97 in connection, for instance, with the economic conditions of the agricultural holding, the status of the farmer (practising farming as his main occupation, occupational skill, etc.) or with the type of production (no aid for surplus production).

B. Operational aspects

- Management: procedures for payment of subsidies

Within the fields of intervention of the EAGGF Guidance Section, it is up to the Member States to opt for the financial instrument they deem appropriate (interest-rate subsidies, grants, etc. or a combination of these instruments) in order to implement the aid scheme provided for in a multiannual programme of assistance decided on by the Commission.

Official aid is co-financed by the EAGGF Guidance Section and the Member States. The Community rules or the programme fix the rate of Community financial intervention. Payment by the EAGGF Guidance Section is made on the basis of a system of advances, followed by refund of the annual expenditure which each Member State declares to the EAGGF and which it has already financed in advance (system in force until 1999) or in the form of down payments/intermediate payments/payment of programme balance (system in force from 2000).

When Member States are responsible for choosing the financial instrument for implementing certain Community programmes, experience has shown that this choice has not been neutral from a financial point of view and in particular that it has been unlikely to minimise budget costs for the Community.

If there was genuine competitive bidding between financial bodies, the cost of managing these aid schemes could be partially reduced, but it should be remembered that the Commission disagrees with most of the Member States on the question of applying the public procurement Directives to financial services.

- Control: who is responsible and what are the procedures-

Under Council Regulation (EC) No 1260/99 (for instance, Article 38(1)), Member States take responsibility in the first instance for the financial control of programmes. The Commission, as part of its responsibility for the implementation of the general budget of the European Communities, ensures that Member States have smoothly functioning management and control systems so that Community funds are efficiently and correctly used (Article 38(2)).

The staff of the EAGGF Guidance Section regularly carry out on-the-spot checks, including sample checks, in the Member States.

- Visibility

As stated above, it is not up to the Commission to opt for any particular financial instrument to implement its programmes within the spheres of intervention of the EAGGF Guidance Section since this is a matter for the Member States (subsidiarity principle). Consequently, the question of desirability of choosing the system of interest-rate subsidies as such does not arise when it is a matter of implementing rules laid down by the Council.

C. Evaluation

- Do evaluation reports exist-

Evaluations are obligatory under the rules, but they chiefly cover the operational content of the programmes. However, there is as yet no properly targeted evaluation in terms of the type of financial instrument (e.g. the system of interest-rate subsidies).

Nevertheless, this lack of evaluation does not mean that there is no visibility as regards the system of interest-rate subsidies in this area. On the contrary, the on-the-spot checks we have carried out in the Member States have sometimes prompted hefty criticism of the economy of such a system for the Community budget, and in particular its cost effectiveness. The risk is that the Member States, in their subsidy statement, charge part of the cost of administering the scheme to the EAGGF Guidance Section by taking advantage of the reference rates, which diverge significantly from market interest rates.

5.10. Interest-rate subsidies in the central and eastern European countries - Phare budget

A. Description

Ø Historically, Community funds have provided two types of interest-rate subsidies for the central and eastern European countries: Slovenia benefited from the continuation of EIB subsidised loans to the former Yugoslavia, and the Phare programme has provided interest-rate subsidies for loans to SMEs in the applicant countries.

Ø From September 1993 to 1997 Slovenia benefited from a financial protocol with an EIB allocation of EUR150 million accompanied by interest-rate subsidies from the Community budget of EUR20 million for transport projects in the public interest.

An amount of EUR47 million was spent on repairing the main rail network in 1993; two loans totalling EUR41 million were signed in 1994.

A loan of EUR32 million was granted in 1995 in support of the existing motorway project, with a further EUR30 million being released in 1996 for the continuation of the project.

Subsequently, Slovenia reverted to the ordinary law applicable to the applicant countries, with pre-accession EIB loans which no longer included interest-rate subsidies but some of which enjoyed the Community budget guarantee. To a very large extent, these loans were made in the telecommunications sector.

B. Management

Ø Under financing provided by the Phare programme, interest-rate subsidies have been granted for loans to SMEs. These operations were carried out mainly in the early- or mid-1990s. Since inflation was high at the time, the beneficiary countries co-financed these operations (at very low rates) with banks which were to lend at the market rate. Of the total amount of the loan, the final beneficiary thus received a loan at a very advantageous rate. Once the operation had been carried out, it repaid both its bank and the State which had obtained the Phare funds and thus had funds available for allocation elsewhere.

Micro-loans were also granted at preferential rates to small firms under the Phare programme. They were often distributed by non-bank institutions.

C. Evaluation

According to various sources, these operations do not seem to have worked to the total satisfaction of the various players involved. There were repayment defaults. This can perhaps be explained in part by the difficulties that persisted while these operations were being carried out. It was also often difficult to find partners to disburse the loans. These operations were frequently organised in parallel with grants of guarantees, e.g. in the Czech Republic and Poland; the least encouraging experience was in Bulgaria.

The conclusions which can today be drawn from the provision of interest-rate subsidies for loans to SMEs are the following:

Ø The banking environment, coupled with the level of inflation, the weak involvement of the beneficiary States and the absence of specific policy objectives, adversely affected policy implementation, with it being impossible to obtain convincing results in several countries.

Ø However, relatively satisfactory results were obtained in Hungary because of the greater maturity of the banking system, the attitude of SMEs, and support of the Finance Ministry (possibility of co-financing, existence of an agency for SMEs) and of SME associations (which were members of the agency's board of management).

Ø A satisfactory rate of disbursement through the banking system depends less on the level of subsidy to the SME and more on other factors. The most important factors would appear to be the margin that can be earned by the bank compared to alternatives (e.g. using deposits), whether the risk is shared with or borne entirely by the bank, and who is responsible for exchange-risk losses. [17]

[17] Final synthesis report of the evaluation of Phare-financed programmes in support of SMEs (February 2000; pp. 46-48)

Ø See point 5.16 above for the FYROM and for Bosnia and Herzegovina.

5.11. Exceptional aid for the regions of Italy devastated by the earthquake of 1980

A. Description (Council Decision 81/19 of 20 January 1981)

The purpose of this action was to finance investments for the reconstruction of the means of production and the economic and social infrastructure in the areas affected by the earthquake of November 1980 which hit the Italian regions of Campania and Basilicata.

The Commission was empowered to contract loans in order to provide finance under the New Community Instrument (NCI). The amount of such loans could not exceed a principal equivalent to ECU 1 billion, subject to deduction of any operations of the same nature carried out by the EIB from its own resources.

Amount of the interest-rate subsidies

The budget of ECU 1 billion in loans authorised by the Council was covered up to an amount of ECU 613 million by NCI loans and up to an amount of ECU 353 million out of EIB own resources. In both cases, the loans carried an interest-rate subsidy of 3% per annum for a maximum of twelve years. The total amount in subsidies charged to the Community budget in respect of the NCI and EIB loans was ECU 240 million.

Objectives

To remedy the damage caused by the earthquake quickly and effectively by financing the investments required for the reconstruction of the means of production and the economic and social infrastructure.

Beneficiaries

Contractors of loans granted under the Council Decision on the basis of projects presented by the Italian authorities.

B. Operational aspects

Management

The Council had decided that the programme would be managed by the EIB on behalf of the Commission. An agreement was signed between the Italian Treasury, the Commission and the EIB on the procedures to be followed. Under this agreement, the loans were granted to state bodies, regional authorities, credit institutions and other bodies designated by the Italian Treasury.

Control

The agreement between the Treasury Minister, the Commission and the EIB specified that applications for subsidised loans were to be forwarded by the Italian Treasury with its favourable opinion; the loans were to be used exclusively to finance investment projects in accordance with the Council Decision.

The Court of Auditors pointed out that EIB checks were inadequate and that the Commission carried out no checks at all. In its replies, the Commission stated that the projects were eligible in accordance with the criteria laid down in the Decision and had been examined individually by the Commission in accordance with the procedure under Article 21 of the Bank's Statute.

C. Evaluation

Usefulness

The Italian authorities reported the estimated cost of the damage to be ECU 13 596 million. The Community contribution of ECU 1 billion accounted for just over 7% of that amount.

The breakdown of the loans per sector is as follows (in ECU million):

>TABLE POSITION>

Reconstruction of the damaged infrastructure and means of production was achieved by projects presented by the Italian authorities and declared eligible by the Commission in accordance with the Council Decision and the tripartite agreement.

However, in its reports, the Court of Auditors criticised:

- the time taken to complete the work, and

- the lower-than forecast number of jobs created.

Regarding the time taken to complete the work, there were various reasons for the delays: changes to the original projects, difficulties linked to expropriations, legal and administrative problems that arose in connection with the implementation of complex projects in difficult circumstances. Regarding job creation, the Commission indicated in its replies that this criterion was secondary to the main objective (reconstruction).

Cost-effectiveness analysis

No cost-effectiveness analysis was carried out. The reason for choosing interest-rate subsidies is the leverage effect: at a modest cost to the budget, they help to generate a high level of investment.

Did the beneficiaries gain from the financial advantages conferred by the interest-rate subsidies-

Almost all the projects were public-sector projects (water, railways, roads, public buildings, etc.). The effect of the subsidies was therefore to reduce the cost to the public authorities' budget.

Evaluation reports

The Bank produced progress reports on the work and summary reports by project and by group or economic sector (see above table). A general evaluation of the NCI was done internally by the Commission and sent to the Court of Auditors in 1994. Representatives of the Commission and the Bank were present at several on-site inspections by the Court of Auditors.

5.12. Exceptional aid to the regions of Greece devastated by the earthquake of 1981

A. Description (Council Decision 81/1013 of 14 December 1981)

The interest-rate subsidies concerned were intended for the reconstruction of the means of production and the economic and social infrastructure, including housing, in the affected regions.

Under the New Community Instrument (NCI), the Commission was empowered to contract borrowings of ECU 80 million on the capital market for financing loans in the affected regions. The loans carried an interest-rate subsidy of 3% per annum for a maximum of twelve years. The same subsidy was granted to recipients of EIB loans made for the same purposes.

The subsidies were granted in respect of projects in the western parts of Athens and Piraeus, in the towns of Corinth, Loutrakio, Megara, Thive, Kiato and Preveza and in 140 other localities.

The projects were presented by the Greek authorities. The EIB participated in the selection procedure.

B. Operational aspects

Borrowings were contracted by the Commission (NCI) and the EIB. The latter managed the loans according to the requirements of a cooperation agreement with the Commission and its usual loan criteria.

C. Evaluation

The subsidies, covering more than a third of the amount of the loans, considerably reduced the financing costs for the State and/or the local authorities and final beneficiaries at a time when budgetary circumstances were difficult.

Some ECU 27 million in subsidies helped the Greek authorities to finance the individual projects concerned.

No cost-effectiveness analysis was carried out. The reason for choosing interest-rate subsidies is the leverage effect: at a modest cost to the budget, they help to generate a high level of investment.

The EIB is monitoring implementation of the projects; if there are delays in implementation and/or technical changes to the projects, the EIB is informed and participates in checks.

5.13. Exceptional aid for the reconstruction of Kalamata (Greece)

A. Description (Council Decision of 7 November 1988 (OJ of 15 November 1988))

On a proposal from the Commission based on Article 235 of the Treaty, the Council decided to grant exceptional aid to the population affected by the earthquake of September 1986 in the Kalamata region of Greece.

The loans were granted by the EIB out of its own resources within the authorised limit, equivalent to a capital value of ECU 100 million.

The rate of the subsidy was set at 3% per annum for a maximum of twelve years. The objective was to finance the reconstruction of infrastructure, buildings and homes damaged by the earthquake and to help re-create the economic conditions that prevailed in the region, of which Kalamata is the main town, prior to the earthquake.

B. Operational aspects

The ECU 100 million in EIB loans was divided into two instalments:

- ECU 80 million paid to the Ministry for National Economic Affairs (MEN), and

- ECU 20 million allocated to the Deposits and Loans Fund (DLF) for global loans to be on-lent to the organisations directly responsible for carrying out the work. [18]

[18] Part of the ECU 20 million was transferred from the DLF tranche to the MEN tranche because of the latter's higher absorption rate.

In the first case, the subsidies were paid direct by the Commission to the EIB, enabling it to reduce its interest rate. In the second, the Commission paid the subsidy direct to the DLF, which then passed it on to the final beneficiaries.

The practical arrangements were set out in the EIB-Commission agreement of 30 March 1990 and in the Commission-DLF agreement. A financing contract was also signed between the EIB and the DLF on 24 October 1989 for an amount of ECU 7 250 000. The Commission paid to the EIB on each interest payment date the amount in each currency equivalent to the difference between the interest to be paid in that currency by the borrower and the interest which should have been paid on that date if the normal interest rate being subsidised had been increased by 3 percentage points a year.

The EIB monitored project implementation and administered the loans. Monitoring was undertaken in accordance with its usual procedures.

C. Evaluation

Did the beneficiaries gain from the financial advantages conferred by the subsidies-

This was clearly the case as regards the MEN tranche since the State was the only final beneficiary of the subsidy. A cost-effectiveness analysis was not carried out. The reason for choosing an interest subsidy is the leverage effect: at a modest cost to the budget, it helps to generate a high level of investment.

The EIB is monitoring project implementation. If there are delays in implementation and/or technical changes to the projects, the EIB is informed and takes part in checks.

5.14. Aid for the reconstruction of Madeira after the cyclone in 1993

A. Description (Council Decision 95/250/EC of 29 June 1995)

Subsidies of three percentage points for a maximum period of twelve years were disbursed on loans granted by the EIB out of its own resources up to a total capital value equivalent to ECU 15.85 million. They total ECU 4 294 309 over the life of the loans.

Objectives: loans for the reconstruction of areas devastated by the cyclone which hit Madeira in October 1993, in order to mitigate its impact on the economic and social well-being of the population affected, etc.

Sectors: from 1993 onwards, road infrastructure, improvement and repair of riverbanks, water distribution, hydroelectric plant and social housing.

Criteria for granting the subsidies: projects proposed by the Autonomous Region of Madeira.

B. Operational aspects

The subsidies were paid on each interest due date.

For control purposes, the Autonomous Region of Madeira is required to present twice yearly, before 30 April and 30 October, a report to the EIB on the progress of the projects until their completion. The most recent report confirms that the regional authorities have met all their contractual obligations. The authorities had to report to the Commission on project implementation before 31 December 1997 and 31 December 2000.

Visibility is guaranteed at project sites by means of signboards indicating clearly that Community aid has been provided.

C. Evaluation

Representatives of the Commission visited Madeira at the end of October 1998 and produced an evaluation report (COM (1999) 286 final of 11 June 1999).

5.15. The Special Support Programme for Peace and Reconciliation in Northern Ireland and the Border Counties

A. Description

This is part of sub-programme 5 "Productive investment and industrial development" under the Special Support Programme for Peace and Reconciliation in Northern Ireland and the Border Counties of the Republic of Ireland (Community Initiative 1995-99).

The interest-rate subsidies during the last five years totalled:

- EUR29 000 000 for Northern Ireland;

- EUR5 667 000 for the border counties of Ireland.

Objectives of the initiative: "to create a positive investment climate throughout the area by making available an interest-rate subsidy for capital investment projects which will contribute to economic development, growth and employment (especially in the most deprived areas) and thereby foster reconciliation".

Ø Which are the sectors/firms/projects that benefit-

Project investment in industry and related services, in tourism, in large and small urban centres and in rural areas.

The scheme prioritises smaller companies (particularly those with fewer than fifty employees) and projects that help to create/safeguard employment, promote industrial competitiveness and innovation, develop new markets, make a significant contribution to tourism development, demonstrate a particular linkage to the peace environment, benefit deprived areas (in Northern Ireland) or involve the development of non-displacing or innovative indigenous start-ups in services related to industry (in the border counties).

Ø Criteria for granting interest-rate subsidies

The criteria for granting the subsidies are the same as conditions under the EIB global loan scheme. The same exclusions/restrictions apply to specified sectors of industry.

Given the large number of applications, there has been a focus on manufacturing, processing and accommodation-related tourism projects, with priority given to smaller companies (in Northern Ireland).

Eligible project costs comprise new investment in fixed assets and stocks (maintenance expenditure and normal vehicle replacement are specifically excluded). The cost range for projects is between EUR40 000 and EUR12 million; qualifying loans are limited to EUR3 million in Northern Ireland and EUR1 million in the border counties.

Loans were approved by intermediary banks up to 31 March 1997 in the form of term loans, leasing agreements or hire-purchase agreements. The term must not be less than five years and can be up to fifteen years, with repayment of principal in equal instalments commencing not later than two years after the first disbursement.

B. Operational aspects

Applications are submitted by intermediary banks to the scheme's appointed agents. Recommendations are made to the EIB by the agents, with the consent of the lead departments (DED in Northern Ireland; Department of Finance in the border counties). Decisions by the EIB are communicated to the intermediary banks, which inform the borrowers.

The subsidy available under the scheme is calculated at 4% per annum on qualifying loans over a five-year period.

Two database-management systems were set up by the agents for monitoring purposes.

Regular progress reports are sent to the Programme Monitoring Committee.

C. Evaluation

The review carried out by the EIB in 1997 concluded that "the general reaction to the scheme in both Northern Ireland and the border counties has been very positive".

Ø Can the rate of achievement of the objectives be quantified-

There were no specified targets attached to the scheme.

Ø Cost-effectiveness analysis

Eligible capital investment projects for which applications were made to the scheme far exceeded the amount of subsidies available, necessitating early closure of the scheme. The intermediary banks have indicated that their lending to the eligible sectors in Northern Ireland has increased by between 30% and 40% in relation to what was expected to be advanced during 1995/96.

The estimated increase in employment attributable to the approved projects is over 7 700 new jobs in Northern Ireland and 1 170 in the border counties.

Ø Are there any evaluation reports-

The EIB reviewed the scheme with Wilson Gillespie and Company.

The ECA audited the scheme in 1997. [19]

[19] European Court of Auditors, Special Report No 7/2000.

5.16. Interest-rate subsidies under the Financial Mechanism of the European Economic Area

A. Description

A Financial Mechanism to assist development projects in Greece, Ireland, Northern Ireland, Portugal and certain regions of Spain (Objective 1 regions, as defined in 1988) was set up on 1 January 1994 (Articles 115 and 116 of the EEA Agreement, and Protocol 38).

To start with, the Mechanism was financed only by the EFTA member countries. From 1 January 1995, when Austria, Finland and Sweden joined the EU, the contributions of these three countries to the Financial Mechanism were paid out of the EU budget.

Up to 31 December 1998 the Financial Mechanism provided funds to the recipient regions primarily in the form of interest-rate subsidies of 2% per annum on loans of a total volume of ECU 1.5 billion granted by the EIB.

Since then, the EIB has continued to manage the project portfolios and will do so until the loans are repaid.

Ø Annual subsidy amounts

1999: EUR 9 398 033

1998: ECU 30 260 769

1997: ECU 60 964 034

1996: ECU 23 226 896

1995: ECU 7 314 073

1994: ECU 3 005 709

Ø Objectives

To strengthen economic cohesion and promote the development of the regions of the European Economic Area (EEA).

Ø Which sectors benefit-

Priority was given to projects that placed emphasis on the environment (including urban development), transport (including infrastructure), education or training.

B. Operational aspects

Ø Management

The subsidies are linked to loans or tranches of loans from the EIB, which administers the entire mechanism. They are paid at the same time as the associated EIB loan or tranche.

Ø Control and procedures

The subsidy applications were evaluated by the EIB. These evaluations were sent to the Commission for its opinion and to the Financial Mechanism Committee (appointed by the EFTA member countries and the Commission) for a decision.

Under Article 8 of the Cooperation Agreement of 30 June 1992 concluded between the EFTA member countries and the EIB, financial control of the operations is carried out in accordance with the audit and discharge procedures set out in the Bank's Statute for all its operations.

The European Court of Auditors is preparing an audit of projects part-financed by the Financial Mechanism.

Ø Visibility

Under the operational guidelines of 10 June 1997 approved by the Financial Mechanism Committee and the EIB, promoters are asked to give appropriate publicity to projects receiving loans subsidised by the Financial Mechanism in such a way as to make recipients and the general public aware of its assistance with project implementation (commemorative plaques in the case of infrastructure works).

C. Evaluation

Ø Usefulness

In so far as the borrower benefits fully from the financial advantage conferred by the interest-rate subsidies paid by it, the Mechanism acts as an incentive for promoters to carry out projects in the sectors covered.

Ø Do beneficiaries gain fully from the financial advantages conferred by the subsidies-

- In the case of subsidies linked to a direct EIB loan, the only intermediary between the Mechanism and the borrower is the EIB. The subsidies are paid direct by the EIB to the borrower, after deduction of the margins laid down in the EFTA-EIB cooperation agreement.

- In the case of subsidies linked to a global EIB loan, it is more difficult to verify the financial advantage conferred on the beneficiaries. The matter is among those being addressed as part of an audit currently being carried out by the Financial Mechanism Committee.

Ø Can the rate of achievement of the objectives be quantified-

Every year the EIB produces a general progress report on the implementation of the projects receiving assistance under the Mechanism.

Ø Cost-effectiveness analysis

The EIB carried out an overall ex ante evaluation of each project proposed prior to the Financial Mechanism Committee's taking a decision.

Ø Are there any evaluation reports-

A Commission report on the implementation of operations financed by the Mechanism is sent each year to the Council, Parliament, the Economic and Social Committee and the Committee of the Regions.

5.17. Interest-rate subsidies for projects in the western Balkans

A. Description

For the countries of the western Balkans, interest-rate subsidies are granted for EIB loans in respect of projects in two countries: FYROM, and Bosnia and Herzegovina.

To date, the Commission has granted subsidies amounting to EUR20 million in respect of two EIB loans to FYROM totalling EUR150 million.

In the case of Bosnia and Herzegovina, the Commission has decided to grant a EUR7 million interest-rate subsidy on a first EIB loan to the country. This represents a partial interest-rate subsidy in respect of this first loan, which totals EUR60 million. A further subsidy of EUR5.27 million will be required to cover the entire amount. This will probably be granted in 2001. In addition, if the EIB grants further subsidised loans to the country of EUR40 million (bringing the total of such loans to EUR100 million), the Commission will be required to grant a further interest-rate subsidy. On past experience this will require a subsidy of the order of EUR8 million that will need to be disbursed in 2001.

The legal bases for subsidies for projects in FYROM are the cooperation agreement and the transport agreement, which mentions infrastructure projects of mutual interest as being eligible for Community funding. For Bosnia and Herzegovina, the legal basis is the Council decision granting the Community guarantee to the EIB for loans to that country.

B. Management

The interest-rate subsidy is disbursed after the decision by the Board of Directors to grant the loan.

It is committed when the loan is signed. It is paid as a lump sum at the time of payment of each instalment of the loan, on the basis of a calculation of its present value. The visibility of the Community contribution is very poor. It is difficult to ensure consistency between the project cycle, managed by the EIB, and the requirements of managing the Community budget.

C. Evaluation

The instrument's usefulness is relative in that the capacity of the countries concerned to borrow on commercial terms is very limited. This limitation aside, the interest-rate subsidies on EIB loans for transport infrastructure projects in the Balkans, and especially in the republics set up following the break-up of Yugoslavia, have proved very useful to the development of projects of mutual interest to the Community and the beneficiary countries.

The Community's particular interest derives from the international transit role these countries have and its importance for the smooth functioning of the single market and for connecting the Community to the applicant countries of southern and eastern Europe. The transport agreements in force with Slovenia and FYROM form a specific legal basis for Community financial assistance, including in the form of interest-rate subsidies.

However, the pattern of cooperation followed in Albania, i.e. parallel financing of (infrastructure) subprogrammes, produces the same financial result without the drawbacks mentioned above.

The granting of an interest-rate subsidy or direct co-financing may, however, be envisaged on a case-by-case basis where the EIB has a comparative advantage as implementing agent. In such a case, it is important, notably in order to ensure satisfactory visibility of Community assistance, that the two institutions cooperate closely in vetting projects, which should also correspond to Community priorities.

Evaluation is carried out by the EIB in its regular activity reports.

6. ANNEX II: ANALYSIS OF OPERATIONAL ASPECTS

For each of the programmes examined, specific attention has been paid to the operational aspects of interest-rate subsidies. A number of general observations now follow.

6.1. Management and control

Interest-rate subsidies have always been linked to mechanisms introduced in very specific circumstances, such as the deep recession in the early 1990s to support job creation, structural adjustment and the launch of transport projects. In most cases, the facilities have been administered by the EIB on behalf of the Commission, while the subsidy mechanisms have been laid down in a cooperation agreement between the Commission and the EIB.

For smaller loans, and SME loans in particular, financial intermediaries have carried out the project appraisal on the basis of criteria previously agreed upon with the EIB, which has subsequently given its approval.

The agents have ensured that approved subsidy applications are registered and processed on a "first come, first serve" basis. They have also been responsible for processing and checking the evidence which beneficiaries have been asked to submit in relation to the criteria set out in the Council decisions. They have therefore played an important role in ensuring that potential beneficiaries are treated in a uniform manner and that all administrative procedures linked to verification of applications and payment of the subsidies have been handled in a consistent fashion.

In most cases, in order to qualify for a subsidy, the beneficiary has had to declare in a sworn statement that the conditions laid down in the decision (job creation, etc.) have been met.

Where the nature of the programme so permits, projects are selected from approved national programmes so as to reinforce the monitoring and control obligations of financial intermediaries and local authorities. Member States, the EIB or financial intermediaries are responsible first and foremost for the proper functioning of the management and control systems.

Adding to the different levels of intervention complicates monitoring and control. The reconstruction aid programmes for earthquake damage (where even the regional and local authorities are involved) are one example of a complex monitoring and control instrument.

In most programmes, given the number and nature of beneficiaries (SMEs, etc.) and the low level of subsidy, it is difficult to image a system other than that based primarily on the delegation of powers to, and the responsibility of, the EIB, the agents and the intermediaries. The Council has indicated that further control by other Community institutions would be superfluous.

6.2. Visibility

Projects financed by soft loans are subject, in general, to the rules regarding technical progress and thus to associated requirements (e.g. signboard publicity) highlighting the role of the European Union.

As for the documentary aspects, the Community origin of interest-rate subsidies is clearly spelt out in the financing contracts signed between financial intermediaries and beneficiaries. However, in a number of programmes (e.g. SME facility), the origin of the subsidies may be lost sight of because of the degree of dispersion among the banking networks concerned and the proliferation of national and regional aid managed by certain financial intermediaries.

Visibility can also be tested, to the extent that the borrower benefits fully from the financial advantage conferred by interest-rate subsidies. Subsidies linked to a direct EIB loan are paid direct by the EIB to the borrower. In the case of subsidies linked to a global loan or forming part of an overall grant, it is more difficult to verify that the financial advantage has been passed on. Nevertheless, competition between banks should ensure that this advantage is passed on in full to final beneficiaries.

6.3. Evaluation of Community programmes

6.3.1. Reconstruction programmes

In the case of exceptional aid granted in the wake of natural disasters, the prime objective of restoring the means of production and permitting a return to normal living conditions has been achieved within satisfactory periods and on satisfactory terms, bearing in mind the difficulties caused by the earthquakes (Italy and Greece) and by the cyclone that devastated Madeira.

Mention should be made of cases where fewer jobs than expected have been created, where additional costs have been incurred and where there have been delays in carrying out the work (e.g. reconstruction following the earthquake in Italy). Assessing the impact on employment calls for very detailed analyses that take account of how jobs created and/or safeguarded are measured and of the effects of technological change.

NCI/EIB financing operations were not the only ones to encounter additional costs and delays, and the lack of additional resources and delays in resolving administrative or legal problems are, in a way, attributable to the national, regional or local authorities (in some cases, the national legislation governing public works contracts laid down special conditions that differed from those contained in EIB loan contracts). It is evident that adding to the different tiers of decision-making gives rise to delays and to monitoring, management and control difficulties.

Below are a number of observations regarding the effectiveness of programmes:

- As regards budgetary costs, none of the guarantees entered in the budget have as yet been called. The budgetary impact has, therefore, been confined to the amount of the interest-rate subsidies themselves.

- The basic rate charged by the EIB was invariably below the rates applied by commercial banks in the regions affected. The life of EIB loans was also longer and the repayment conditions more favourable (grace period). These advantageous terms relative to the conditions generally imposed by commercial banks for similar operations helped in the long term to diversify the resources generally available to the region (and to develop capital markets).

- Community assistance for public infrastructures has had a positive effect on the volume of production in the private sector (the GDP multiplier is particularly high in assisted regions) as a result of the overall demand triggered by the projects and the increase in productivity resulting from lower costs (e.g. the installation of more modern equipment than that which had been destroyed).

- Although difficult to quantify, the impact of the reconstruction work following natural disasters has, in all cases, been a factor in the recovery of the devastated regions (roads, communications, energy, water supplies, environment, etc.). The work has also led to job creation (direct and indirect).

- Alongside infrastructures, the loans have also served to finance housing reconstruction (e.g. Kalamata and Madeira), also necessary in order to restore normal living conditions. It must be mentioned here that, in the normal course of events, the EIB does not finance housing construction. In these cases, however, the Council decisions made express reference to the reconstruction of housing exclusively for those whose homes had been seriously damaged.

- Together with the favourable conditions for loans granted by the EIB, interest-rate subsidies have helped speed up the reconstruction work without imposing an excessive burden on national budgets.

6.3.2. Other programmes

Irrespective of the aid schemes introduced by Member States, greater recourse has been had to these programmes in structural action areas and this has boosted job creation. Below are a number of points for consideration:

- Subsidised-loan programmes have exploited the synergies between bank loans and budgetary resources. They have in some cases led to, and in others reinforced, close cooperation between the EIB and the banks or financial intermediaries, imparting a significant multiplier effect to Community assistance.

- For the most part, SMEs (and hence productive investment) have benefited from these programmes. It goes without saying that SMEs account for the bulk of jobs created in the European Union.

- In all cases, the amounts available have been used up quite quickly. Since the budgetary funds earmarked for the programmes are not renewable (understandably so since the measures had a limited objective), it has not been possible to improve cost effectiveness. Therefore, an insufficient rate of return was obtained as regards efforts to implement an administrative management package (as well as the learning curve of intermediaries and clients).

- Most interest-rate subsidies are tied to EIB loans, which were subject to a ceiling equivalent to 70% or 75% of the amount of investment. Taking all programmes into account, the share financed by the EIB was lower on average. Since the remainder was financed by intermediary banks or other financial institutions, this shows that the Community contribution had an additionality effect.

Impact of interest-rate subsidies within Community and national programmes: examples (in EURm)

>TABLE POSITION>

To sum up, the economic advantages of these programmes have been:

1. For the beneficiary:

Direct effects

- Reduction in the interest-rate burden: It is evident that the most immediate and visible effect has been a fall in the price of money, this being the benchmark (other things being equal) for the different alternatives. A lower price makes the entrepreneur's decision easier:

- reducing uncertainties and bringing forward the project's payback period;

- making it possible to increase available cash flow.

Indirect effects

Since interest-rate subsidies are tied to longer-term projects and are conditional on a prior examination of project eligibility and monitoring, they have made it possible:

- to improve the financial structure of liabilities and to transform short-term debts into longer-term debts;

- to create a new banking relationship culture (particularly in the less-developed regions).

Such monitoring obliges the enterprise to improve the programming and follow-up of the more demanding projects, all to the advantage of the project itself.

2. For the intermediary:

- Since these loans cost less, intermediaries are able to expand operations and gain access to new clients.

- If the subsidy applies to the reduction in the principal, this allows banks to reduce their risks but also the need for own funds (Cooke ratio).

- The projects are better studied and programmed by the entrepreneur; this generally results in a much lower failure rate.

Most intermediary banks take a positive view of interest-rate subsidies. Below is a summary of the arguments:

* Competition between the banking networks involved, lending know-how and capacity, prudential solvency and responsibility on the part of the financial institutions ensure that the system functions properly.

* Direct effect and simplicity

The intermediary allocates the interest-rate subsidy direct to its client, passing on the credit and cash-flow risk. This is more effective and more direct than a guarantee, which does not entail credit and requires a more complicated and more unreliable procedure.

* Delegation with a posteriori control creates value added

It is the temporary and revocable delegation of responsibility that allows competing banking networks to assume part of the cost of allocating interest-rate subsidies under optimum conditions by incorporating it into existing, tried-and-tested systems and into the overall service provided to the enterprise. The value added stems from the dilution and transformation of the cost within an overall financial and commercial relationship.

6.4. Compliance with the competition rules

Like measures taking other forms, interest-rate subsidies caught by Article 87(1) are, in principle, incompatible with the common market. This provision states that, save as otherwise provided for in the Treaty, any aid granted by a Member State or through state resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods is, in so far as it affects trade between Member States, incompatible with the common market. However, the principle of incompatibility does not amount to a full-scale prohibition. Articles 87(2) and 87(3) of the Treaty specify a number of cases in which state aid could be considered acceptable ("exemptions").

The existence of these exemptions also justifies the vetting of planned state aid measures by the Commission, as provided for in Article 88 of the Treaty. This article stipulates that Member States must notify to the Commission any plan to grant state aid before putting such plan into effect. It also gives the Commission discretionary power to decide whether the proposed aid measure qualifies for exemption or whether the "State concerned shall abolish or alter such aid".

The most important exemption clauses are those of Article 87(3)(a) and (c):

- Article 87(3)(a) covers "aid to promote the economic development of areas where the standard of living is abnormally low or where there is serious underemployment";

- Article 87(3)(c) refers to "aid to facilitate the development of certain economic activities or certain economic areas, where such aid does not adversely affect trading conditions contrary to the common interest".

In exercising its discretionary powers, the Commission has developed specific approaches depending on the size of the firm, its location, the industry concerned, the purpose of the aid, etc.

As its power is discretionary, the Commission has sought to publicise its approach in order to ensure that its discretion is exercised with proper openness and that authorities and businesses are clear about their legal position. To do so, it has published the criteria it applies when deciding whether aid measures notified to it qualify for exemption. These publications have taken the form of regulations, communications, notices, frameworks, guidelines and letters to Member States. In many cases they define "eligible costs" (i.e. types of expenditure which can attract aid) and "permissible intensities" (i.e. the proportion of these costs which can be met by aid).

It is a principle of, for example, Structural Fund operations that the use of Community resources must also be consistent with Community competition rules.

In general, though with some exceptions, the state aid rules do not specify the form of an aid measure. Different forms of aid (grant, interest-rate subsidy, guarantee) are analysed for their "net grant equivalent" for the application of the state aid rules. Considerations of state aid policy are therefore not a critical factor in the comparison of interest-rate subsidies with other forms of aid.

As a rule, the Commission considers only the borrower to be the beneficiary of state aid in the form of an interest-rate subsidy, provided that mechanisms are in place to ensure that the subsidy is fully passed on in reduced interest payments. This is the case even though (as shown in Section 6.3 above) there may also be advantages for the lender.

6.5. Final remarks

- Interest-rate subsidies are not a determining factor in investors' decisions. However, they do provide an incentive in that they reduce enterprises' financial charges and thus promote access to credit.

- Interest-rate subsidies are an effective mechanism for mobilising additional funds (additionality). At a modest cost to the budget, they have helped to generate a high volume of investments (leverage effect). They are tied to, and accompany, long-term financing, thereby fostering investments which create stable jobs. In some cases, there are synergies with national programmes and/or other financial engineering instruments.

- A European framework can provide significant efficiency gains as compared with a national approach. The participation of private banks in programmes involving interest-rate subsidies is one way of publicising Community initiatives.

- When interest rates are channelled via financial intermediaries which operate in a competitive market, problems of market distortion could arise if full and free access by these intermediaries to management of the instrument cannot be guaranteed. In cases where financial intermediaries operate in a non-market environment, fully transparent terms and conditions (e.g. interest rate to be paid by the final beneficiary) help to ensure that the interest subsidy is transferred to the final beneficiary.

- An alternative to interest subsidies would be direct grants (with the grant being equal to the net present value of an interest subsidy). The advantage of using a direct grant lies in increased transparency and administrative simplicity. However, there is less risk of abuse when interest-rate subsidies are tied to bank loans.

- There are shortcomings in programme implementation, particularly as regards control. Nevertheless, the burden of control must not outweigh its benefits.

- Interest-rate subsidies must be accompanied by simple, clear provisions regarding implementing conditions and procedures. Complicated conditions regarding outturn and quantification should be avoided.

7. Annex III: INTEREST-RATE SUBSIDIES: ExpErience IN MEMBER STATES

7.1. Italy

A. Description

In Italy there are currently over 80 laws designed to support business investment (capital grants and interest-rate subsidies, often combined), only half of which were activated in 1999. Of these, only half a dozen may be considered to be significant because of their widespread application.

The system is based on highly targeted programmes with the following main objectives:

- Law No 1329 of 1965, better known as the Sabatini Law, aimed at encouraging SMEs to acquire productive plant and machinery with no sectoral bias (industry, distributive trades, services, agriculture, craft industry). At first, the purchase of machinery was encouraged by a financing scheme based on bills of exchange. This scheme, advantageous for all parties, was later improved by the granting of an interest-rate subsidy at the time the debt was paid off, calculated on the basis of a reference rate fixed by the government (slightly below the market rate), together with a reduction ranging from 50% in non-assisted regions to 100% in Objective 1 regions, with all the arrangements being published periodically in the press to ensure complete transparency.

- Law No 598/94, aimed at encouraging investment by innovative SMEs or in the environment. In December 1999 the reduction in the reference rate fixed by the government (6.15%) ranged from 1.40 percentage points for medium-sized firms to 4.9 percentage points for small firms located in Objective 1 areas.

- Law No 317/81, aimed at encouraging the formation of producer groupings in industry, the distributive trades and the craft sector. The reduction in the reference rate ranges from 40% in non-assisted regions to 70% in Objective 1 and Objective 2 areas.

- Law No 46/82, aimed at supporting applied research and technological innovation, both in SMEs and in large firms, with a sliding scale of reductions.

- Law No 341/95, aimed at consolidating the debt of SMEs in Objective 1 regions and covering all sectors. The medium-term loan granted in this connection, with the support of a guarantee fund, is eligible for a maximum reduction of 4.5 percentage points, provided that the equivalent discounted grant equivalent does not exceed 40% of the loan.

- Law No 227/77, aimed at promoting export credit for industrial and commercial firms. The rate applicable in the country of exportation is reduced by 4.75 percentage points.

B. Operational aspects

The management of the credit lines relating to these facilities is usually entrusted to a lead bank (Mediocredito Centrale for SMEs, IMI for large firms, etc.) which allocates them among the other banks on an ongoing basis, ensuring as far as possible distribution in eligible areas of the country. The final recipient bank assumes all the risks of the operation, including the follow-up and recovery of any unpaid debts.

Generally speaking, checks are carried out at three levels: (i) by the local bank, according to its own procedures; (ii) by the lead bank on a more general basis in order to demonstrate to the authorities that the system is functioning properly and that the economic objectives are being met; (iii) by the competent Ministry (e.g. Industry, Treasury, Agriculture), on the basis of documents and on-the-spot surveys. Naturally, such checks, which are necessary throughout the period in which the loan is being paid off (around ten years), impose a heavy administrative burden, especially on the Industry Ministry, which is responsible for a larger number of smaller-scale (SMEs) and riskier operations.

The visibility of the instruments, conditions and objectives is strictly ensured, both in the wording of loan contracts and in the promotion of specific instruments, with the implementing arrangements being published regularly in the press. This allows each firm to calculate accurately the financial advantage it can expect; in the event of a dispute, the financial intermediary risks criminal proceedings

C. Evaluation

The usefulness of the instrument is evaluated at regular intervals on the basis of surveys and questionnaires sent to beneficiaries.

The surveys conducted by Mediocredito Centrale over the period 1992-94 among a sample of SMEs comprising firms which did and firms which did not benefit from the scheme described in point A produced the following results:

- 95% of firms eligible actually carried out new investment during the relevant period, while 80% of firms not eligible did so; the propensity to invest was therefore definitely boosted, albeit to a modest degree, by interest-rate subsidies.

- Comparison of the economic and financial situation of the two groups of firms shows a more marked improvement in the beneficiary companies as regards turnover, labour productivity, labour costs, net result and return on investments.

- In terms of job creation, it would appear that 87% of the SMEs which availed themselves of the scheme have created new jobs, compared with 77% of those which have not. Particularly as regards the Sabatini Law (purchase of productive plant and machinery) and Law No 317/81 (producer groupings) it was observed that over the three-year period the former led to an increase in recruitment of 0.8% (despite being "capital-intensive" in scope) and the latter to an increase of 5.5%.

- With regard to the medium-term consolidation of SME debt (Law No 341/95), the SanPaolo-IMI Institute found that, of ITL 3 000 billion of loans with a subsidy and a guarantee of 60% from the guarantee fund, only 1% were not repaid, resulting in an improvement in the quality of debt.

7.2. France

A. Description

Since 1946 interest-rate subsidies in France have been used almost exclusively in two sectors: craft industries and agriculture. In both cases, they pursue several objectives: for example, setting up in business of young farmers, development/modernisation of farms, and compliance by farms with environmental standards. In recent years, for strictly limited amounts and for a strictly limited period, the government has occasionally had recourse to interest-rate subsidies in order to restore forests destroyed by the winter storms in 1999.

Public assistance, which was available on a generous scale at first, has tended to decline over time, with the result that the craft sector was recently excluded. By 1999 the FRF 15 billion to 20 billion of subsidised loans granted each year in the 1970s had declined to between only FRF 10 billion and FRF 12 billion, and this exclusively for the agricultural sector, together with a public grant (subsidy) of the order of FRF 2 billion per year, including FRF 700 million from the EAGGF. This trend is attributable to the fall in interest rates, but also to the narrower targeting of farms. The criteria for granting an interest-rate subsidy are laid down in great detail by law, by the Rural Code and by ministerial decrees (age of the farmer, size of the enterprise, type of production, quantities produced, etc.) For most loans, the beneficiary must carry out a multiannual investment programme approved by the administration; the applicant must, therefore, be assisted by a management consultant or by the administration in drawing up and implementing the programme.

B. Operational aspects

The system of interest-rate subsidies in France developed in three main phases. From 1946 to 1978 (credit controls), the State provided both the capital and the subsidy. In practice, it made available to financing bodies specialised in specific sectors (i.e. Crédit Agricole for the agricultural sector, the "banques populaires" and Crédit Coopératif for the other sectors) credit raised on the market (at an interest rate of 10% at the time). These bodies were charged with vetting the projects and financing them with an interest-rate subsidy of around 4-6% . With growing demands on the budget, the State decided in the 1980s to redirect its efforts towards the loan categories with the highest rate of subsidy, in particular establishment and modernisation. In the early 1990s these two categories accounted for around 75 % of annual subsidised loans disbursed. However, the arrangements for granting subsidies remained more or less unchanged compared with the previous period.

Since 1983 the budget allocation for the agricultural subsidy has fallen sharply, in part because of the increase in EAGGF refunds with respect to establishment and modernisation aid, but mostly because of the increase in the cost of bank financing since 1986. In 1990 the distribution of subsidised loans in the agricultural sector, previously entrusted to Crédit Agricole, was opened up to various banking networks, once accredited by the State. Since 1995 successive improvements aimed at opening up access (for farms run as joint-stock companies, for example) were made to the system of subsidised loans at the same time as the subsidy rates were lowered. As for the financial management arrangements, subsidies are, in general, paid out quarterly by the Ministry of Agriculture and Fisheries to a public body (CNASEA) which each month reimburses the financial intermediaries for expenditure calculated on the basis of loan repayment requirements. Once the files have been vetted by decentralised departments, each bank is responsible for financial accounting on the basis of a central computer system which filters files vetted by the various bank branches throughout the country, according to the contractual conditions laid down.

Monitoring and checks are carried out at three levels: by the public authorities (Ministry of Finance or Agriculture) on a representative sample selected each year and via ex post and on-the-spot documentary checks, by the bank's head office on supporting documents and by subsidiaries. Visibility of the measure and provision of information to potential beneficiaries are ensured by administrative departments, by bank branch officials and by trade associations that advise applicants, help them present their projects and filter or redirect applications to the most appropriate source of financing.

C. Evaluation

During the period of credit controls, when there was therefore a shortage of fresh capital, the scheme had a major impact on investment by small agricultural and craft enterprises and, conversely, imposed a very heavy burden on the state budget, given the very advantageous conditions offered.

During the phase of liberalisation and simultaneous tightening of the allocation criteria, the scheme remained highly effective, especially in the agricultural sector, when it came to increasing the capitalisation of small farms and boosting their productivity. Its success was due in part to the back-up and information provided by the trade organisations to their members, which made it possible to optimise the scheme's impact. The scheme is judged to have been very effective in the sector in promoting unprecedented modernisation and giving an important boost to the establishment of young farmers.

According to some financial intermediaries, the tightening of the allocation criteria, on the one hand, and the fall in interest rates, the cost of the computer changeover to the year 2000 and the introduction of the euro, on the other, have seriously complicated the management of interest-rate subsidies and made the authorisation procedures a less attractive proposition for lead banks.

In the craft sector, the State is moving towards abolishing the system of subsidies to craftworkers and replacing it with a system of public guarantees. This trend is dictated by the constraints imposed by the EU on Member States' budgetary discipline. By means of the guarantee, the Treasury therefore aims to reduce the rationing of credit to small craft businesses and thus allow an increase in more risky financing that would not have been possible on market conditions. Given the present cost of money, the reduction in the cost of credit brought about by the subsidy no longer appears to be a priority objective in this sector.

Official opinions appear to diverge considerably. Some point to the bureaucracy of the system, in particular as regards checks, the outdated nature of the instrument given the current level of interest rates, and the need to improve computer tools in order to reduce management costs. Others, however, stress the proven effectiveness of interest-rate subsidies as regards agricultural investment, with subsidies being considered to be an instrument well suited to the capitalisation levels in the sector (average investment of FRF 1 million, i.e. around EUR 150 000), and the rate of success of projects, which is regarded as satisfactory.

More generally, interest-rate subsidies are considered a highly appropriate instrument for carrying out investments which have low profitability but are necessary to meet statutory standards. This is the case, for example, when SMEs have to meet pollution standards or when an unforeseeable event, such as a natural disaster, disrupts normal economic activity or when, even in normal circumstances, the return on investment in forestry is satisfactory but very long-term.

7.3. Germany

A. Description

In Germany the measures taken by the federal administration to encourage economic development (in the case of SMEs) traditionally take the form of loans managed primarily through the special European Recovery Programme Fund (ERP). However, subsidised loans are also granted by the Länder.

In 1999 the special ERP Fund helped to strengthen SMEs through loans totalling some DEM 11.5 billion to investment projects with a total value of DEM 30 billion. In the main, this assistance has been in the form of repayable loans, with facilities such as:

* an extended repayment period (up to twenty years);

* a grace period (up to five years);

* a preferential rate;

* possibility of early repayment at any time (without penalty).

The loans are granted mainly through the enterprise's own bank, with the result that all credit institutions are involved in these operations. As they are responsible for repaying the loans to the Fund, the borrower must provide the usual guarantees required for any bank loan (e.g. claim on property, collateral or surety).

The ERP equity loan scheme ("Eigenkapitalhilfe" or EKH) is an exception. A subsidy under this scheme does not consist in granting a lower fixed rate valid for the life of the loan, but in fixing a progressive rate from the first to the sixth year, followed by a stable rate up to the tenth year. In addition, in the case of a twenty-year loan, the grace period is ten years. For this type of assistance, a traditional bank guarantee is not required, which means that the credit is granted in the form of a personal loan. Here too, however, the profitability of investment projects is monitored by the credit institutions participating in the overall financing.

German credit institutions are involved in the overall financing of SME investment projects via all the forms of credit available under the ERP because, on the one hand, subsidiarity is verified (self-financing capacity) and, on the other, investment projects can be financed only in part by this type of credit. Thus, from the outset, they are directly or indirectly associated with the financing of businesses. It is then easier for them to grant "growth" financing directly.

Interest-rate subsidies are not calculated in the form of a fixed spread relative to the market rate in Germany. The rates granted under the programmes vary in line with market developments. Thus, the normal rate for a ten-year ERP loan with a two-year grace period is currently 5.88% in western Germany. Since the reference rate (fixed by the Commission) for calculating the amount of subsidy is currently 5.70%, the subsidy in this case is nil. By comparison, in the eastern Länder an ERP loan with the same terms is currently granted at an effective rate of 5.36%.

The beneficiaries of the ERP programmes are mainly young entrepreneurs running SMEs; there are no regional limits or sectoral preferences (only sensitive sectors in the EU are excluded).

Financing under the special ERP Fund may therefore be granted without discrimination to a wide range of firms. Under the circumstances, promotion of economic development involves support for:

- business start-ups via equity assistance and launch funding;

- the expansion and modernisation of existing firms in assisted regions;

- R&D in respect of new products and processes up to the marketing stage;

- the creation of more apprenticeship places;

- investments linked to environmental protection (waste water treatment, recycling of waste, air quality protection, energy saving and use of renewable sources of energy).

B. Operational aspects

The procedures are quite straightforward. The special ERP Fund sets the criteria for allocating funds for the various programmes and makes repayable funds available to its two institutions, the Kreditanstalt für Wiederaufbau (KfW) and the Deutsche Ausgleichsbank (DtA). Applications for subsidised loans under the various support programmes are submitted by firms to their own bank, which examines them and forwards them to the two institutions.

These ensure that the funds are transferred and used for the stated objectives. To do this, they rely on the usual criteria and principles of sound banking management. The special ERP Fund is also subject to the general budgetary rules in Germany, which means that the overall use of resources is regularly monitored by the Federal Court of Auditors.

The special ERP Fund promotes economic development under transparent conditions, its procedures being published for the information of the businesses concerned. Credit institutions are kept informed by the two institutions via bank circulars and regular conferences reviewing practical matters and planned changes in procedure.

C. Evaluation

The granting in Germany of loans subsidised to a greater or lesser extent has proved its worth in recent years. This form of financing has witnessed a revival of a sort since unification in 1990. At the time, the main aim was to create in the new Länder a hitherto virtually non-existent network of small and medium-sized businesses. Thanks to the ERP programmes, the authorities have managed to provide these new small businesses that had little equity and a limited self-financing capacity but also offered viable projects with the means to finance those projects.

The granting of subsidised but repayable loans is an "economic" way for the State to promote small businesses, enabling them to overcome the difficulties of gaining access to capital markets and at the same time allowing them to help themselves.

The interest-rate subsidies granted by the special ERP Fund are of direct benefit to businesses. The rate offered to the final borrower by his own bank is imposed by the Fund, and the bank receives the corresponding funds for on-lending.

The special ERP Find and the two institutions associated with its programmes carry out regular (internal) cost-effectiveness analyses. External expert reports are also commissioned at less regular intervals. For instance, a check on the effectiveness of state measures to stimulate the economy via the ERP programmes was carried out in 1995. At the moment, an independent economic research institute is also conducting a detailed evaluation of the ERP programmes aimed at stimulating the economy.

7.4. Spain

A. Description

Interest-rate subsidies and repayable advances without interest (i.e. fully subsidised) are economic policy instruments for promoting productive investment that are widely used in Spain by the State and its agencies, both financial (Instituto de Credito Oficial - ICO) and technological (CDTI), and by the Comunidades Autonomas (the regions have complete budgetary autonomy and some fiscal autonomy) at regional level.

There is no general law governing the use and allocation of subsidies. However, subsidies are granted under specific schemes or programmes for a limited period.

* Programmes implemented by the State

There is a wide variety of programmes, and we will therefore refer only to a few by way of example. The most recent is the programme for expanding technological research and innovation (PROFIT). Assistance is in the form of :

- repayable aid of up to 75% of the cost of the project, with zero interest for projects with a low technological risk; repayment periods vary according to the characteristics of the projects but do not exceed fifteen years.

- subsidies for projects with a high technological risk; the rate of aid varies between 25% and 60% of the cost of the project, depending on its characteristics.

SMEs and projects in assisted regions qualify for the most advantageous rates of aid.

Under no circumstances can assistance exceed the limits laid down by the Community rules on the cumulation of aid (state aid).

Aid may be distributed through partner institutions (e.g. the Centro de Desarrollo Tecnológico Industrial), which undertake (i) to check that the beneficiaries comply with the conditions laid down, (ii) to transfer the funds, (iii) to monitor the projects and (iv) to check that the funds are used for the stated objectives.

* Programmes implemented by the "Comunidades Autonomas"

All the "Comunidades Autonomas" have set up programmes to support small and medium-sized enterprises.

For example, the "Comunidad de Andalucía", through the Instituto de Fomento, grants subsidies for long-term loans [20] signed by private banks or by the EIB under a cooperation agreement. The subsidies come out of their own budget but also partly out of the ERDF budget. The terms of the loans (interest rate, repayment period) applied by the banks are those applied to their best customers. The rate of subsidy is 3.5% (on average). The beneficiaries are small and medium-sized businesses. An additional subsidy of 2% is granted to loans guaranteed by a mutual guarantee company.

[20] Loans are for seven years with a two-year grace period. Interest may be at a fixed or variable rate.

* Programmes implemented by other public institutions.

Instituto de Crédito Oficial (ICO)

The ICO is a financial institution whose assets are held by the State. It is therefore a publicly owned bank and a financial agent of the State.

The ICO has implemented interest-rate subsidy programmes using subsidies from Spanish State funds and Community funds. The programmes are specific in nature and of limited duration. These are some examples:

- "PYME 2000": Interest-rate subsidies on loans to finance investment by SMEs. The loans may cover up to 70% of the project, and the repayment period is seven years. The subsidy is provided by the ERDF and by the ICO.

- "Creación de empleo"(Copenhagen facility): With funding from the EIB, subsidies per job created come out of the Community budget.

- "Internacionalización de la empresa española": This programme aims to promote projects abroad. The subsidies come out of the State budget.

- EEA Financial Mechanism: Interest-rate subsidies for investment in the environmental and renewable energy sectors in particular. The ICO receives a global loan from the EIB and uses this to grant subsidised loans. The ICO funds may also be added to the EIB loans (in this case, the ICO loans are not subsidised).

* Centro para el Desarrollo Tecnológico Industrial (CDTI)

The CDTI is a public institution responsible for promoting investment and technological development and innovation. It finances projects either from its own resources or jointly with other bodies (e.g. ICO, ERDF) at low or zero interest rates. In some cases, it also shares the risk of the operation (i.e. it does not ask for guarantees); 45% of projects financed are SMEs with fewer than fifty employees. The CDTI is the Spanish representative for European Union programmes (e.g. Eureka). Its most important activities are:

- Financing of R&D projects: Five-year loans from own resources and from the ICO; interest-rate subsidies of 1.25%

- Eureka projects

* 60% of CDTI financing over eight years at zero interest

* 25% of subsidies financed by the State budget or the "Comunidades Autonomas"

- Projects abroad: CDTI loans over eight years at zero interest

B. Operational aspects

Comunidad de Andalucía

The application procedures are straightforward (form and a brief file on the firm).

Where the funds are not used for the stated objectives, the firm must repay the loan and the proportional share of the subsidy plus statutory interest.

For operations of a certain size, on-the-spot inspections are carried out. For the rest, spot checks are made. In all cases, SMEs must submit documents justifying the investment.

As regards the rules on aid cumulation, the application form states that potential beneficiaries must declare any existing or future aid.

In all cases, loans are limited to 70% of the investment. The rate of subsidy varies between 2.15 and 3 percentage points. All economic sectors are eligible (except agriculture, fisheries and the wholesale trade).

ICO

The ICO ensures that the conditions laid down by the EU (de minimis rules) are met.

Payment procedures differ from one programme to another. In some cases, the subsidy serves to reduce the loan principal. In others, it is paid to the beneficiary, who must then pay the full interest rate without deduction.

The intermediary banks monitor compliance with eligibility conditions (almost all sectors are eligible). The beneficiary submits a sworn statement on the use of the financing, which is verified by documents or by spot checks carried out by the intermediary bank. Given the large number of operations and the small size of the loans (around EUR 70 000), stricter controls would be difficult to implement. In any case, the ICO has carried out inspections for operations amounting to more than EUR 600 000. For the EEA Financial Mechanism, controls are also carried out by the EEA Committee.

Subsidised loans are allocated on a "first come, first served" basis. All available funds have been used.

Visibility is ensured as the origin of the funds is stated in the loan contract.

CDTI

Financing applications are examined and a decision taken within two or three months. Payments are made as the project progresses. There is therefore a prior check on the proper use of the funds.

C. Evaluation

ICO

The aim of most of the programmes (PYME 2000, etc.) was to lower interest rates in order to encourage investment. Clearly, this has been achieved. The ICO has made certain that the intermediary banks do not levy commission or extra charges on the final beneficiaries, who have therefore received the full amount of the subsidy. Payment to these banks for their services was, in fact, clearly established separately. Job creation, which is the criterion for eligibility, has also been ascertained.

The theoretical maximum ceiling for loans (70% of the investment) has not been used in most cases, which suggests that the subsidised loans have had a multiplier effect as they have attracted additional funds.

According to the ICO, the failure rate for subsidised loan operations is much lower than the average for other operations. This may be due to better preparation and follow-up.

CDTI

According to the CDTI, the programmes are considered to be very successful in terms of cost effectiveness. Long-term financing and subsidies reduce the uncertainty attaching to innovative and technological projects and can improve the structure of the firm's liabilities (long-term financing). Moreover, project monitoring by the CDTI necessitates better planning by the entrepreneur, and this is beneficial for the project.

7.5. Sweden

A. Description

In Sweden interest-rate subsidies are currently used as a policy instrument within the SME and housing sectors. [21] In terms of budget expenditure, housing subsidies are by far the most important policy instrument and this chapter will therefore deal only with the subsidy programmes for that sector.

[21] Interest-rate subsidies for export credits have effectively been phased out as the consensus rates agreed within the OECD for export credits now equal market rates.

By the budget year 1992/93, the annual cost of interest-rate subsidies had escalated to more than SEK 30 billion (around ECU 3.5 billion). The increased cost of the programme, due in part to a large increase in interest rates, and the resulting strain on the overall budget led to a reform of the programme in 1993. The reform meant that the subsidy element was reduced significantly during a transitional period. Previous generous programmes which targeted both single- and multiple-family dwellings and which included unlimited interest subsidies above a set interest rate were replaced by a programme targeting multiple-family housing only and with a reduced subsidy.

The reduction of government expenditure on the reformed subsidy programme can be seen from the table below. It should be noted that declining interest rates in the late 1990s also contributed to the sharp reduction in government expenditure on the programme.

Government expenditure - interest-rate subsidies for housing

>TABLE POSITION>

Source: Ministry of Finance.

The current programme offers a government contribution of 30% towards the "estimated" interest-rate costs for the construction or conversion and modernisation of multiple-family housing. The recipient of the subsidy is the registered owner of the property and may be a private individual, a company, a housing association or some other legal entity.

Subsidy payments according to the above formula are made in perpetuity and regardless of how the project has been financed. [22]

[22] Certain exceptions to this rule exist for housing associations.

B. Operational aspects

The Swedish Board of Housing, Building and Planning (Boverket), a central government agency, has overall responsibility for implementing the programme, which is administered mainly by the regional authorities (länsstyrelserna). Boverket issues regulations concerning the implementation of the programme, provides information and offers training on how to implement the programme to the regional authorities. [23] It also serves as the highest instance for appeals.

[23] With the exception of payment execution, which is handled by Boverket.

In its supervisory role, Boverket ensures on an ongoing basis that the regional authorities implement the programme according to the relevant legislation and according to its instructions. For internal control purposes, both Boverket and the regional authorities conduct their own internal audit units and, as government bodies, they are also subject to regular audits of their activities by the National Audit Office. Boverket's internal auditors also regularly visit the regional authorities in order to conduct detailed checks on the administration of the programme.

Prime responsibility for ensuring that the rules are being applied falls to the regional authorities, which vet and approve applications and authorise subsidy payments. The risks of irregularities in connection with the programme are limited thanks to a few basic controls built into the programme:

- The basis on which subsidies are calculated is determined by reference to standardised construction costs specified in the regulations governing the programme. [24]

[24] In the case of conversions/modernisations, where the subsidies are calculated on the basis of actual costs, the beneficiary has to select the lowest of three independent estimates. This is also verified by the regional authorities' experts.

- Subsidy applications have to include a construction plan and a building permit. This ensures that approval is given only to projects that comply with the rules of the programme.

- A check by independent inspectors approved and authorised by the regional authorities that the building work has been carried out according to plan is undertaken before the subsidy is paid out.

An indication of the cost of administering the programme is given in a study of policies in the housing sector [25] undertaken in 1996. During 1994/95 Boverket's administration of the interest-rate subsidies cost some SEK 56 million, of which SEK 30 million concerned EDP systems and postage. At regional level it was estimated that the regional authorities spent a further SEK 58 million. During 1994/95 the number of subsidy recipients totalled 273 000 and the number of payments exceeded 1 000 000. Together, the administrative costs amounted to around 0.35% of total subsidies paid in 1994/95.

[25] SOU 1996:156.

C. Evaluation

There is no mechanism in place for evaluating the programme. As part of its ongoing review of overall policies in the housing sector, the government may, of course, initiate an evaluation or study of the programme as it sees fit.

The National Board of Auditors, which is the highest audit body in Sweden, may, in addition to its annual audits of government activities, undertake "specific performance audits". To date, no such audit has been undertaken with regard to the interest-rate subsidy programme in the housing sector.

7.6. United Kingdom

A. Description

No interest-rate subsidy programmes are currently being operated by the UK government. The last such programme was the Small Business Loan Scheme (SBLS) in Scotland. It was launched in November 1994 as a two-year pilot project but was subsequently extended for a third year.

The programme was developed by Scottish Enterprise (a non-departmental government agency responsible for economic development in Scotland), Highlands and Islands Enterprises (which subsequently withdrew) and four Scottish clearing banks with a view to providing a source of stable, long-term finance for new and expanding businesses. The scheme was approved by the Board of Scottish Enterprise and the Treasury.

Under the scheme SMEs could obtain loans with a capped interest rate through the participating banks. The interest-rate cap was set in relation to the Bank of England base rate plus one percentage point at the time the loan was made. Each bank would add its own credit margin as normal, in order to maintain competitiveness between banks in lending to customers. SMEs would therefore have access to medium- and long-term loans and interest rates could rise by only one percentage point above the rate applicable at the outset. Should interest rates rise any further, Scottish Enterprise rather than the SMEs themselves would cover the additional loan interest payable to the banks. The scheme was therefore intended to give companies an element of certainty in terms of knowing the maximum payment required, while at the same time allowing them to benefit from any reductions in interest rates over the loan period.

The programme applied only to SMEs and for loans of between UKL 20,000 and UKL 500,000000 with terms of up to seven years. Certain eligibility restrictions with regard to the sector in which the applicant SME operated and to the use of loan proceeds were built into the programme. In volume terms, the SBLS was a relatively small programme. Towards the end of the programme [26] 264 loans had been taken out. The total value of those loans was around UKL 44 million, with the cost of the caps purchased (i.e. the subsidy element) totalling some UKL 2.4 million.

[26] That is to say, November 1997, when a report "Evaluation of the Small Business Loan Scheme" was compiled by DTZ Pieda Consultants.

B. Operational aspects

While the overall management of the SBLS was undertaken centrally by Scottish Enterprise, its network of local enterprise companies and the participating banks played an important role in the operation and delivery of the scheme. All three parties were involved in vetting applications. Scottish Enterprise and the local enterprise companies focused on assessing eligibility, while the banks assessed the viability of proposals.

In general terms, companies entered the SBLS "system" through contact with their own banks. The banks appraised the project on a standard commercial basis. If SBLS seemed an appropriate funding route, then the bank would begin to process the company's application. The application would then be passed to the relevant local enterprise company for an assessment of eligibility. Once the application was "accredited" by Scottish Enterprise, the loan finance agreement was finalised and the company joined the programme.

Scottish Enterprise had access rights to the firms which joined the programme in order to collect information and access accounts and business plans relating to the SBLS loan. The banks monitored the ongoing progress of the loans, reporting as required to Scottish Enterprise, which undertook its own monitoring and analysis of the scheme.

C. Evaluation

In November 1997 an evaluation of the SBLS was undertaken by an outside consultancy. [27] With regard to management and delivery, the consultants reported that they were impressed with the day-to-day management and administration of the scheme by Scottish Enterprise. The scheme was said to be very tightly controlled and an excellent database monitoring cap activity had been developed as well as good communication channels and arrangements with the banks. However, some of the banks considered the level of administration required for cap "repayments" burdensome. Control aspects of the SBLS were not assessed in the study.

[27] "Evaluation of the Small Business Loan Scheme", a report by DTZ Pieda Consulting, November 1997.

The study's overall conclusions concerning the scheme were as follows:

* The SBLS had attempted to address a genuine problem of medium/long-term finance for SMEs. However the main issues of market failure did not lay with the banks but were related to the perceptions, attitudes and knowledge of small firms.

* The SBLS was too blunt an instrument to tackle that problem effectively. A high proportion of assisted cases were not additional and there was a need to minimise deadweight assistance through improved targeting, marketing and screening. In addition, there may have been room for closer integration with the business development activities of the local enterprise companies - it was a question not just of selling the scheme but also of working with companies to improve their financial and business performance.

* It is sensible and practical that efforts to encourage small firms to make more effective use of finance should operate through the commercial lenders. However, it is essential that future initiatives are targeted effectively on small firms with little experience of finance. Some form of incentive to use loan finance may be necessary and the cap may be an effective instrument in this regard. However, this is essentially a matter of expediency and there is no specific market failure which calls for the subsidisation of capped loans per se.

The Small Business Loan Scheme in Scotland was not extended after 1997 and no other scheme is currently planned to replace it.

8. ANNEX IV: The use of subsidies in the context of EU policies with special reference to soft loans: theoretical considerations

8.1. Introduction

At first sight, the issue of recourse to soft loans instead of other forms of subsidy gives rise to two contradictory interpretations. The first maintains that different forms of subsidy are essentially equivalent in their effect as regards a similar objective, while the second maintains that each instrument should have a single objective because an instrument's characteristics, and hence its suitability for a given objective, vary. This annex attempts to demonstrate that this contradiction is apparent rather than real.

When examining the suitability of recourse to soft loans compared with other types of subsidy, three different levels of analysis need to be borne in mind. At the most general level, the issue of recourse to subsidies as an instrument of economic policy has to be addressed. In particular, the different types of justification for providing subsidies may have an impact on the choice of instrument. At an intermediate level, the appropriateness of a given instrument can be evaluated only in the context of the specific individual project. In turn, this requires that the type of expenditure eligible for subsidy and the different possibilities for financing such expenditure by the recipient be known. Finally, each instrument will have its own advantages and drawbacks at the implementation stage when viewed from the point of view of the public authorities, any intermediary responsible for administering the scheme and the recipient.

The conclusion reached is that it is not possible on theoretical grounds either to condone or to reject the use of soft loans as an instrument for subsidising economic activity. In particular there is no single instrument that can replace soft loans in all circumstances. If certain developments point to less use of this instrument in the future, this should not be taken to mean that it would ever be inappropriate to have recourse to soft loans.

8.2. On the use of subsidies in general

Soft loans (more properly, interest-rate subsidies) are a form of subsidy and must be treated as such. It follows that they are subject to the same general considerations as other subsidies. The subsidy element on the loan is the difference in price that the borrower would pay to a commercial lender in the absence of subsidy and the price actually paid. This will usually be expressed as a reduction in the rate of interest paid.

From a macro-economic point of view, public subsidies to enterprises should be treated as transfers. Accounting for over 2% of GDP in the EU as whole, they are far from negligible although they represent only a tenth of public transfers to households. Certain characteristics of subsidies make them particularly harmful from the point of view of economic efficiency. The most harmful effects concern their impact on resource allocation and on competition between enterprises.

At the highest level of abstraction, it can be argued that all taxes are levied directly or indirectly on the market economy in the terminology of European national income accounts. The market economy is composed essentially of enterprises, public or private, whether for profit or not for profit (such as producer cooperatives and mutuals). Since enterprises are both the source of the means to provide subsidies and the recipients of subsidies themselves, transfers to enterprises should properly be seen therefore as transfers between enterprises via the tax system. Because this exercise is essentially non-transparent, the potential for misallocation of resources and unforeseen side effects is rather high. Little information on the whole cycle of taxation and subsidisation is available but it can be conjectured that the direction of transfer is most likely to be from small and medium-sized enterprises to large ones by virtue of the superior organisational possibilities and greater access to decision-makers enjoyed by large enterprises.

A number of other objections to recourse to subsidies may be mentioned. They raise public expenditure and therefore either the level of taxation or public deficits, both of which can have a negative effect on growth. They distort market signals particularly in periods of high inflation, when restrictive monetary policies are often required in order to achieve macro-economic stability. As such, they may undermine the effectiveness of general instruments of economic policy.

From the point of view of EU integration, the principal concern with regard to the harmful effect of subsidies for enterprises has always been a concern with competition, particularly with regard to their effect on trade between Member States. For this reason, such subsidies are in principle prohibited where they are likely to affect intra-EU trade. Since the state aid provisions of the Treaty apply to subsidies provided by Member States, it may be objected that the subsidies provided by the EU, which are the subject of this report, are not affected by competition considerations. However, this would ignore the need for coherence between the actions undertaken by the EU and those undertaken by Member States.

A number of texts indeed make this point, e.g. the Commission communication to the Member States on the links between regional and competition policy "Reinforcing concentration and mutual consistency". [28] The linkage between EU and national subsidies is explicitly made in other areas, such as the Community framework for state aid for research and development (point 1.3). [29] The guidelines on aid for SMEs go further; see point 4.2.1, where they state that "The aid ceiling will apply regardless of whether the aid is provided entirely from domestic sources or is part-financed by the Community from the Structural Funds, and more especially the European Regional Development Fund (ERDF)". [30] Indirectly, this places a limit on the aid intensity also of EU subsidies. The basic Structural Funds Regulation 1260/1999 requires that "operations financed by the Funds or receiving assistance from the EIB or from another financial instrument shall be in conformity with the provisions of the Treaty, with instruments adopted under it and with Community policies and actions, including the rules on competition, on the award of public contracts, on environmental protection and improvement and on the elimination of inequalities and the promotion of equality between men and women."

[28] OJ C 90, 26.3.1998, p. 3.

[29] OJ C 45, 17.2.1996, p. 5.

[30] OJ C 213, 25.7.1996, p. 4.

In practice, the corpus of legislation, decisions and practice of EU policy with regard to state aid provides a sound basis also for evaluating the pertinence of recourse to subsidies at EU level. [31] Indeed, to ignore this aspect by applying different criteria or giving differing justifications for similar types of subsidy at EU level to those applied by EU competition policy at national level might justifiably be criticised on the grounds of consistency and equality of treatment between the actions undertaken by differing levels of authority.

[31] For the purposes of this annex, the terms "state aid" and "subsidies" will be used interchangeably, even though certain subsidies which do not affect intra-EU trade, such as subsidies for local services, are not considered as state aid under the EU's competition policy.

As a general rule, subsidies find their justification as a means of compensating for an externality that market forces are unable to integrate into the final end-user price for a product in the normal operation of competition or for resolving another market failure. Such general considerations must be applied on a category-specific basis if they are to be operational. For the purposes of this annex, a number of the EU provisions governing state aid are relevant, particularly the horizontal frameworks or guidelines for research and development, environmental protection, small and medium-sized enterprises, employment and training as well as the notice applying to state aid in the form of guarantees, the guidelines on national regional aid and the multisectoral framework on regional aid for large investment projects. [32] They illustrate why, although in principle not an efficient instrument of economic policy, subsidies to enterprises may be justified in particular instances.

[32] The guidelines on state aid for environmental protection are particularly useful for the analysis of externalities, and the framework on training aid for the analysis of other market failures.

A second, more practical consideration with regard to subsidies relates to the distinction between subsidies and public investment, which is not always easy to make. This concerns certain forms of investment in infrastructure (which may be financed, built or operated on either a public or a private basis) but more particularly intangible investment. Here, certain expenditures may be denominated as public investment or even consumption, but an equivalent result may be obtained through providing subsidies to enterprises. A typical example concerns training where vocational training may take place in publicly financed and operated establishments or in enterprises with public financing. In the first case, the expenditure would not normally be considered as a subsidy, but in the second it would fall within the remit of the appropriate Commission framework on state aids for training. In such cases, the decision by public authorities to provide a service directly or through subsidies to enterprises may be based essentially on criteria of efficiency of supply

Of course, the fact that Member States are authorised under certain conditions to provide subsidies to enterprises does not oblige the EU to follow them. Indeed, in view of the persistently high level of such subsidies and the desire to reduce them, it would be preferable for the Commission to set an example by having only sparing recourse to subsidies in its own policies. Many areas remain outside the EU's remit or fall within it only to a limited extent. In such cases, it would be quite normal for the Commission to refrain from proposing subsidies even though Member States are providing them. It would also be legitimate for the EU to refrain from subsidisation where the necessary results were already being achieved by action by or within Member States.

Nevertheless, there are certain cases in which the Commission could be led to provide subsidies, even where Member States either do not or cannot provide them. One of these cases concerns infrastructure where the major beneficiary or beneficiaries were situated mainly or partly in other Member States. The Member State in whose territory the infrastructure was located might decide that the use made of the infrastructure by its own citizens or enterprises did not warrant the public expenditure involved. In this case, an EU subsidy could be justified in order for the wider benefit to be felt. It represents in a sense a particular form of externality based on the fact that users and administrative territories do not coincide. Article 87(3)(b) EEC (previously Article 92) implicitly provides for this case where it permits aid for promoting the implementation of an important project of European interest. If subsidies provided by Member States for such a project are admissible, it should follow a fortiori that EU subsidies for such projects also represent a valid case for intervention. Trans-European networks constitute a typical application. [33]

[33] Another example of trans-national externalities is provided for by the guidelines on environmental protection, which state in point 3.7 that "Aid to promote the execution of important projects of common European interest which are an environmental priority and will often have beneficial effects beyond the frontiers of the Member State or States concerned can be authorized under the exemption provided for in Article 92 (3)(b) of the EC Treaty."

The Treaties have also placed on the EU specific responsibilities in certain areas, such as economic and social cohesion, which may require recourse to subsidies. These may complement the efforts of Member States or may, in certain circumstances, be required independently of them, particularly where the application of the additionality principle would place an undue burden on the public finances of a Member State. Similar considerations apply to extra-EU interventions, e.g. in the field of development policy with the European Development Fund or for specific interventions linked to foreign and security policy, as in the Balkans. To the extent that it is more efficient for the EU to act with one voice and to avoid dispersal of resources ("saupoudrage"), recourse to subsidies at EU level without recourse to complementary sources from Member States may well be justified in certain cases.

Even where the major responsibility in a policy area lies with Member States, there can be aspects which either are not covered by national schemes or are better dealt with by EU subsidies. For instance, aid for the internationalisation of small enterprises, particularly through industrial cooperation, is better organised at EU level because, by definition, the matching of partners requires the participation of more than one country. Using a multilateral framework in this context provides significant efficiency gains over a bilateral approach between pairs of Member States, although there is still a question about the efficiency of any aid for such activities.

Just as there are cases for which subsidies to enterprises are justified and permissible under EU state aid rules, so there are cases where the EU needs to have recourse to such subsidies. The same theoretical considerations that are applied to state aids can by and large also be applied to the EU's own measures.

8.3. On the equivalence of different forms of subsidy

Probably the most common confusion that arises with regard to soft loans concerns their particular usefulness, or lack of usefulness, compared with other form of subsidies. Sometimes soft loans are endowed with almost mystical capacities in the financial field, while at other times they are reviled. In practice, soft loans are a form of subsidy with similar effects to other forms.

The essential equivalence of soft loans and direct grants is the easiest to demonstrate. Loan guarantees and tax incentives are more difficult to compare directly, but it is always possible to reduce a subsidy system to a single figure by replying to the question "what is it worth to the recipient-". [34]

[34] In any substantial company this is the question that the finance director will pose before deciding whether to take up the possibility offered. In a smaller company it will be the CEO, or more often the company's accountant, that will reply to the question.

Take a simple example, that of a fixed investment in plant and equipment. An enterprise may finance this from its own resources as reinvested earnings, may take a loan or may raise new equity either on external markets or in the form of private equity. Since it is relatively rare to raise new equity for the purpose of financing fixed investment, the main choice will be between taking a loan or financing from own resources. Where a subsidy is provided in the form of a direct grant, the need for external finance is diminished by an equivalent amount. Any loan taken out will bear a full rate of interest on a diminished principal. In the case of a soft loan, the full principal is charged, but the interest repayments are reduced. Any difference between the forms of subsidy therefore concerns the timing of the cash flows associated with the repayment of the loan. Since the value of money diminishes over time, a subsidy in the form of interest paid in the future will be less valuable than a cash sum allotted at the outset. In order to calculate the worth of the two possibilities, a finance director will discount any interest-rate subsidies to find their net present value. Any soft loan can therefore be compared to a direct grant. It also follows that, for any specified loan, there will be some interest-rate subsidy that is equivalent to a given direct grant.

It will be immediately apparent that the rationale and steps indicated above are those applied over many years by the Commission in the context of its evaluation of state aid. Any given state aid is reduced through appropriate calculations to either a gross (before tax) or net (after tax) grant equivalent. Similar levels of subsidy are deemed to have similar effects on competition and therefore to be equivalent for the purposes of competition policy. [35] Certain countries, in particular the UK, have traditionally used direct grants as the main form of subsidy, while others have relied mainly on soft loans. However, Member States typically use a combination of different instruments, albeit in different proportions.

[35] The steps taken by the Commission to calculate net grant equivalents for a capital grant or a subsidised loan are set out in Annex 1: Net grant equivalent of investment aid to the "Guidelines on national regional aid".

In passing, it may be mentioned that, as with any net present value calculation, one of the most difficult choices to be made is that of the discount rate. Choice of discount rate is particularly important because it significantly affects the outcome of the calculation. The Commission usually, but not always, sets discount rates to be applied in state aid cases as the five-year interbank swap rate in the relevant currency plus a premium (which may vary between Member States). The use of a single rate has advantages in terms of transparency and ease of administration. Since projects and enterprises have different risk characteristics, it is not evident that a single rate can adequately capture the true net present value of a soft loan compared with a direct grant. It merely indicates that, although equivalence between the effect of different instruments may be true in theory, there are problems in applying the principle in practice. These problems have not prevented policies on state aid from developing and being applied on a daily basis.

Since the Council, in its report to Coreper on the special report of the Court of Auditors on interest-rate subsidies, specifically draws attention to the availability of "other, possibly more efficient, financial instruments, such as loan guarantees", Commission treatment of such guarantees under competition policy is also of interest. In particular, it can help elucidate major differences between instruments. The Commission notice on the application of Articles 87 and 88 of the EC Treaty to state aid in the form of guarantees sets out the procedure to be followed.

Not all guarantees constitute state aid. A number of conditions must be met for guarantees not to be considered as aid. A key provision requires that the market price for a guarantee be paid. In the case of guarantee schemes, the premiums must cover both the normal risks associated with granting the guarantee and the administrative costs of the scheme, including a return on any initial capital provided. Two main methods are used to calculate the amount of subsidy attached to a guarantee. The first covers the difference between the market rate on a loan and the rate obtained thanks to the guarantee after any premiums have been deducted. The guarantee can then be likened to a soft loan. The second covers the difference between the risk-adjusted sum guaranteed and the appropriate premium for such a sum. The second of the two methods already indicates that risk constitutes the problem that guarantees are intended to address. Although risk constitutes an element in any financing decision, soft loans and direct grants are more directed towards solving problems of cost or availability than risk. It suggests that for different types of problem there can be different instruments, even when the net grant equivalent of the subsidy is the same.

8.4. On the suitability of different instruments for one and the same objective

The doctrine according to which each financial instrument has its own characteristics and each objective to be met will have an optimum financial instrument is well anchored in Commission thinking. At an abstract level, it is necessary to define clearly the public policy objective to be met, especially whether an externality or market failure is the cause of the problem, the characteristics of the action to be undertaken and the type of recipient of the subsidy. For any given profile of problem-action-recipient, instruments that are close substitutes will usually give a range of acceptable outcomes. The decision to use one or other of them will then depend on other characteristics of the instruments such as their ease of deployment or the degree to which they can be adequately monitored for abuse. To the extent that instruments are not close substitutes, it can be expected that one or other instrument will be appropriate and the others inappropriate to the task in hand, unless several are required jointly in order to meet the policy objective.

In the case of a large infrastructure project, e.g. a transport link with pan-European implications, there can be several possible outcomes. Certain projects require very large outlays of capital for returns far in the future and of a very uncertain nature. In this case the problem to be addressed is risk which cannot be adequately evaluated by the market, either by the enterprise wishing to execute the project or by the financial intermediaries wishing to invest in it. [36] The Channel Tunnel was an example of how difficult it is to assess and price risk adequately in such cases. The difficulties it encountered and the losses borne by shareholders may make it difficult for similar projects in future to find financing on a purely commercial basis. Faced with this situation, the public authorities could either decide to build and finance the link themselves, thereby assuming all of the risk and the necessary finance, or provide a guarantee such that the risk became manageable for commercial operators and markets. In such a case, soft loans or direct grants would not be an appropriate instrument because the problem concerns incalculable risks rather than weak profitability.

[36] It is necessary to distinguish unpredictability from duration. Not all infrastructure projects, even those with very long pay-off periods, encounter problems of financing. Typically, power stations display very stable returns over a long period and do not encounter any problems of financing when operated on a commercial basis. However, the uncertainty surrounding decommissioning costs of nuclear power stations may inject a sufficient degree of unpredictability to jeopardise their commercial viability.

The case of a transport link that relieves congestion and/or reduces pollution constitutes an entirely different type of problem. Here, we are faced with one or more externalities rather than a market failure. In the case of congestion, the cost appears in terms of time taken to travel to destination, more accidents and higher pollution. This cost is not or only partially borne by the owners and operators of the current infrastructure. In the case of commercial exploitation (e.g. of a motorway or rail link), only part of the benefit to society can be recuperated by the end user. In this case, infrastructure provision may be suboptimal for society.

As with any externality, the provision of a subsidy may be justified when for any reason it is not possible to reflect adequately the value of the externality in the end-user price (e.g. by levying taxes on emissions). However, the level of subsidy should be calculated according to the value of the externality rather than in order to ensure that any particular project becomes commercially viable. In principle, the projects with the highest return in terms of increasing a positive externality or reducing a negative one should be subsidised. An infrastructure project to which externalities are attached should be undertaken if the commercial value of the project plus the subsidy equivalent to the value of the externality is sufficient to ensure commercial viability. The value of the subsidy should not be calculated as the difference between the total cost of a project and that which a commercial promoter would be willing to pay. In such a case, the cost in subsidy might well outweigh the value of the externality.

In terms of the financial instrument to be used to meet the problem of an externality, it is clear that it is the cost of the project rather than its risk characteristics which is being addressed. Both soft loans and direct grants reduce the cost to the promoter and, since both are close substitutes, either could be used. It would be unlikely for a loan guarantee to be an appropriate instrument in this case.

Aid to small and medium-sized enterprises (SMEs) represents a second typical case for which subsidies are often provided. Under point 1.2 of the Community guidelines:

"SMEs suffer from a number of handicaps that can slow down their development. One of the main handicaps is the difficulty in obtaining capital and credit, the chief causes of which are imperfect information, the risky nature of financial markets and the limited guarantees that SMEs are in a position to offer; SMEs' limited resources also restrict their access to information, notably regarding new technology and potential markets. The introduction of new regulatory arrangements often entails higher costs for SMEs. The imperfections in the market which limit the socially desirable development of SMEs justify the favourable consideration which the Commission has traditionally been prepared to give to state aid to SMEs, provided that such aid does not affect trade to a disproportionate extent relative to the contribution it makes to the achievement of Community objectives allowed by Article 92(3). The Community has in fact itself been implementing an action programme for SMEs."

The objective identified here is to remedy a market failure or failures leading to insufficiencies in a number of areas, each of which could be the subject of one or more actions. When a decision is taken on the choice of financial instrument for the provision of subsidies, risk and limited resources will be key considerations. The recipients obviously are small and medium-sized enterprises and the Community dimension of the problem is acknowledged.

When it is the intention to provide subsidies to remedy a market failure with regard to the provision of financing for SMEs, it is necessary to establish properly the specific financing needs that are currently either not being provided or being provided only inadequately. This implies a good knowledge of SMEs' specific financial structures and needs. Since corporate finance texts and business school classes deal mainly with the financing of large firms, there is a dearth of easily accessible information on this. A study for DG ECFIN which makes a comparison between the financial structure of SMEs versus large enterprise using the BACH databank helps to fill a gap. [37] As is well known, SMEs have little access to external markets either for equity or for loans (corporate bonds). In spite of this, debt/equity ratios and the degree of leverage for SMEs are quite similar to those of large enterprises. Since the report does not distinguish between different sources of equity, the amount of self-financing through reinvested earnings compared with outside sources of equity is not available. However, as mentioned above, this distinction is not significant for most investment decisions. By placing SMEs in a more dependent relationship with banks, the lack of access to bond financing could be of greater significance in terms of the market failure to be remedied.

[37] I.D.H.E. Final Report by Dorothée Rivaud-Danset et al, June 1998.

One of the most important findings of the report concerns the much lower capital intensity of SMEs and their much higher share of working capital in total capital. SMEs rely on flexibility to respond to market signals quickly. Most subsidy schemes relate either to investment in fixed assets - whether tangible or intangible in the form of patent rights, licences, know-how or unpatented technical knowledge - or to specific aids for meeting a shortfall in specific areas such as recourse to consultancy schemes. Since working capital does not give rise to a fixed asset, no provision is made for it by the Community guidelines and operating aid is specifically excluded. In practice, a significant area for SME financing for which loans are the overall source, and hence the possibility of providing soft loans appear to fall outside the scope of existing Member State and indeed EU schemes. [38]

[38] There is an exception in the case of agriculture.

Concerning recourse to loan guarantees or soft loans for those categories of SME investment that are eligible for subsidy, the appropriateness of one or other instrument depends again on the specific market failure to be addressed. Assessing the degree of riskiness of a loan for an SME may be difficult even for an established financial intermediary such as a bank. High turnover among SMEs may mean that there is an insufficiently long period of operation to judge the financial health of an enterprise. The large number of firms in very specific market niches, particularly in service industries, means that specialist knowledge of the conditions for success on these markets will often be lacking within commercial lenders. Success in the market may, in any case, depend more on the individual capacities of the SME manager than on any objectively verifiable criteria on which, for instance, a credit rating can be made. The attendant difficulty of accurately pricing a loan can lead a bank to withhold credit, especially when the collateral offered is insufficient. In this instance, guarantee schemes which diversify risk among a large number of SMEs appear the most appropriate financial instrument. They address the failure to obtain credit at any (or almost any price) rather than the high cost of credit. In a sense, they can be seen as a response to a particular form of credit rationing. [39]

[39] There are other forms of credit rationing, e.g. those stemming from inadequate balance-sheet resources, by lending institutions that cannot be dealt with through guarantee schemes.

There are a wide range of loan guarantee schemes for SMEs currently in operation within Member States. For instance, the "banques populaires" in France operate a scheme whereby potential borrowers in the craft sector are interviewed by other craftsmen to determine whether their professional capacities make them a good or bad credit risk (peer review). Once their professional capacities have been assessed, the borrower has access to a guarantee scheme for which a premium must be paid. Such a system acts to solve the risk evaluation problem both by injecting specialist knowledge into the loan application system and by diversifying risk. In principle, the subsidy element is absent, so there is no market failure. However, since the French Government provides soft loans to the "banque populaires" for craftsmen and since the lender obliges the recipients of these loans to participate in the guarantee scheme, it can be said that there is an indirect subsidy element. The scheme also illustrates how soft loans and guarantee schemes can be complementary.

Other Member States, such as Germany, Italy and Spain, operate guarantee schemes where, in principle, the premiums cover the full cost of operation but where there is a public guarantee in the case of particularly distressed conditions either because of a crisis in an individual sector or because of a particularly severe business downturn which could generate large numbers of bankruptcies if borrowers were forced to default. Such schemes inevitably imply a subsidy element of imprecise nature. Nevertheless, they may be very efficient at remedying the market failure outlined above.

Soft loans to SMEs, whether or not in conjunction with loan guarantee schemes, can be valuable where the cost rather than the availability of capital is an issue. This applies where economically or socially desirable investment is forgone because of the incapacity of SMEs to obtain finance on sufficiently favourable terms. Weak bargaining power with regard to banks, the lack of a competitive alternative to banks in the form of the bond market and a higher risk of default means that loans to SMEs are generally more expensive than those for large enterprises, sometimes substantially so. Since SMEs are less capital-intensive and more labour-intensive than large enterprises, subsidies for their investment needs may be justified in periods of severe employment shortfall.

The value of subsidising investment changes over time along with the situation of financial markets. During a period of low interest rates and low inflation, the need for public subsidies for investment may decline, particularly if profits are growing steadily and the self-financing capacity of enterprises is improving. However, low nominal interest rates do not necessarily lead to low real interest rates. On the grounds that real interest rates should determine investment behaviour, there may still be some justification for maintaining investment subsidies such as soft loans even when nominal interest rates are low. Figs. 1, 2 and 3 illustrate movements in nominal long-term interest rates on government bonds and real interest rates (nominal rates adjusted ex post for inflation) for a sample of countries over the past decade. [40] The first impression is, to say the least, hazy. For instance, in Japan the relationship between nominal and real interest rates appears to be downwards sloping while in the UK it is upwards and then downwards sloping; in the US there is no discernible pattern in the relationship. On closer inspection, it would appear that in Europe there is some sort of pattern to the relationship but that this pattern varies in timing from one country to another. Generally speaking, at the beginning of the disinflationary cycle real interest rates fall but then rise as inflation falls faster than interest rates. They then resume their downward path until reaching a minimum around which there is some stability. In certain cases, such as Italy, real interest rates rose at the beginning rather than towards the middle of the disinflationary cycle.

[40] It is not suggested that this provides an adequate proxy for the actual financing conditions of SMEs. It is intended only as an indication of the overall relationship in an economy between nominal and real interest rates.

Certainly, it would be difficult to adjust public subsidy schemes in line with changes in real interest rates, particularly since such adjustments would have to be based on inflationary expectations rather than realised outcomes. The best that could be hoped for would be that schemes are able to move in line with a relatively steady-state situation. If, as now appears to be the case, there is some stability in low real interest rates, then it becomes possible to adjust aid schemes and intensities in line with changed economic situations. An era of abundant and cheap capital clearly does not call for the same level of subsidies as one of scarce and expensive capital. Therefore, investment subsidies for enterprises in the form of soft loans may have become a less appropriate instrument than previously. In specific instances, either where capital remains expensive or where there is a need for particularly significant incentives (usually associated with an important externality), soft loans may still be appropriate. Identical considerations apply to close substitutes for interest-rate subsidies, such as direct investment grants. They clearly apply only to the EU and cannot be extended, e.g. to developing countries and the European Development Fund.

Developments in the "new economy" and the greatly enhanced importance of intangibles also presage a change in the financial needs of enterprises and thus in the suitability of soft loans as an instrument of economic policy. Much of the expenditure on intangibles takes place on a regular rather than discrete basis and is treated as an operating cost rather than a balance-sheet asset. As such, traditional forms of subsidy as a percentage of a fixed investment may become less relevant in the future. Instruments such as soft loans either will have to be adjusted to take this development into consideration or are likely to decline in importance.

For the limited number of firms with high growth potential, often but not exclusively in high-technology industries, a lack of seed and start-up capital has been identified as a major problem. The Commission has already proposed action at both EU and Member State level to address this matter. The general approach is to improve the functioning of financial markets rather than subsidisation. To the extent that recourse to public subsidies might be justified, it would probably be required in the form of equity injections rather than loans. However, once such firms enter into the high-growth phase (so-called gazelles), the need for all kinds of finance grows very rapidly. Cash-flow problems are often encountered in this phase because revenues are growing more slowly than investment. In order to ease cash-flow constraints, soft loans concentrated in the early phase of the repayment period may have a role to play.

When reviewing the future of an instrument such as soft loans, possible developments in the financial environment and institutions for potential recipients must be acknowledged. SMEs face a number of potentially contradictory developments as regards their access to finance. On the one hand, they are becoming a much more significant part of the customer base for banks. Major companies are now able to raise finance more cheaply on bond markets than they can borrow from banks and in many cases more cheaply than banks themselves can borrow on the market. As a result, the borrowing and lending activity of banks has been heavily curtailed, leaving them with rather unprofitable business such as effecting transfers. At the same time, consumers are no longer willing to leave large sums unremunerated in their bank accounts. In order to attract such savings, banks are increasingly required to pay market rates of interest. SMEs represent the last major source of potentially profitable business from conventional lending activities. Banks themselves are also facing much greater competition, both from other banks and from non-banks. All these developments point to a lowering in the cost of finance for SMEs.

On the other hand, the cost of capital to banks is rising because of capital adequacy provisions and the desire of retail customers for remuneration for their savings. Very heavy investment in information technology and the infrastructure for processing financial flows must be repaid. If SMEs remain the principal source of profitable business for banks, a larger element of the cost base must fall on them. This points to higher financing costs for SMEs.

8.5. On the implementation of soft loans

A major part of the criticism of the Court of Auditors echoed by the Council concerns the implementation of soft loans by the Commission or by agents appointed by the Commission to manage them, such as the European Investment Bank. In its report to Coreper, the General Secretariat of the Council even goes so far as to state that:

"...several delegations questioned more generally whether interest-rate subsidies are still justified in the light of experience taking into account in particular the cost of effective control measures..."

Several types of criticism are levelled against soft loans including the cost of administration, the uncertainty as to whether the borrower receives the subsidy or whether the financial intermediary pockets all or part of it, a lack of transparency in the choice of intermediary and the possibility of distortion of competition between financial institutions and the assertion that the investment financed by a soft loan would have taken place even without the subsidy. Some of these criticisms are justified but apply to subsidy schemes in general, others are inherent in the soft loan instrument but are offset, wholly or partly, by other advantages while still others appear to have little merit or to be based on a misconception.

Soft loan schemes typically target a very wide audience, with many recipients for relatively small amounts. Such schemes are expensive to administer. The same is true of direct grants and other instruments with a comparable impact. Because soft loans are, as a general rule, administered by commercial firms, these require their costs to be covered in some form or other (either through a fee for administration or in the form of the difference between the interest subsidy received and that provided to the borrower). The cost of administration is reduced by the incentive that any commercial firm has to reduce costs in order to maximise profits and by the fact that the infrastructure for providing loans is already place and does not have to be remunerated by the soft loan scheme or at least not remunerated in full. In the case of an alternative to soft loans administered by the public authorities themselves such as direct grant schemes, the cost of administration is not transparent and rarely, if ever, calculated. Experience with the privatisation of public enterprises and the outsourcing of public services demonstrates that cost savings in the field of administration are huge. It defies belief that the cost of administering soft loan schemes is higher than, for instance, that associated with direct grants.

It is true that under soft loan schemes it is difficult to ensure that the borrower and ultimate recipient receives the full benefit of the subsidy. This is in part because of the administrative costs mentioned but mainly because of the oligopolistic structure of the banking industry and asymmetric information on the part of the borrower and the lender. This represents a very serious drawback because the high price of borrowing constitutes one of the essential reasons why public authorities provide soft loans. The only solution to the problem is to create a more competitive environment for lenders. Progress in implementing the single market and the developments in the banking industry outlined in the previous section should make banks more eager to compete for SME business. Under competitive conditions, it would be counterproductive for a lender to fail to pass on to the borrower the maximum benefit of a soft loan since the borrower would then have an incentive to go elsewhere. In some cases of state aid schemes in the form of soft loans, the Commission has insisted that the benefit be available through all lenders for precisely this reason.

A key advantage from the point of view of public authorities that stems from recourse to soft loans as compared with other forms of subsidy resides precisely in the expertise of banks in assessing the commercial viability of borrowers and their projects. As indicated in a previous section, this expertise may be imperfect and there are occasions where it is impossible or difficult to evaluate risk precisely. However, commercial operators whose success depends on this ability are more likely to make a thorough and proper evaluation of a request for a loan than public administrations are in the case of a grant. Where all risk is taken away from the lender, a problem of moral hazard arises. For this reason, it is legitimate to limit subsidies to a certain percentage of the total, even where there is no legal requirement to do so. In the case of loan guarantees for example, Member States usually limit the amount guaranteed to between 50% and 70% of the principal.

It has been the practice of the Commission, either directly or through the European Investment Bank, to restrict access to soft loans to a limited number of institutions. This can only diminish the amount of competition between such institutions. However, the practice can be justified on a number of grounds. The choice of institutions is usually made by virtue of their specific expertise in the field to be covered by the scheme (e.g. banks specialised in lending to SMEs). Enhanced efficiency in implementing the scheme and lower administration costs for the Commission or its agents should be treated as valid bases for restricting the number of eligible institutions.

A large number of intermediaries make administrative control of the execution of a scheme more difficult. However, faced with a general problem of large numbers of final beneficiaries, recourse to intermediaries may be the only cost-effective way of supplying a service. Particularly when speed of execution is called for, administrative control may have to be sacrificed in order for the initial objective to be met. Taking the Edinburgh facility for SMEs as an example, the European Council entrusted the Commission with the task of rapidly executing a scheme to promote investment by SMEs in order to help relaunch growth in the EU. Inevitably, only projects already in the pipeline could be financed and a considerable degree of delegation to intermediaries was required. Any aid scheme implies a trade-off between efficiency of execution and efficiency of control. The optimal trade-off cannot be established in abstract or without reference to the particular characteristics of the action involved. In particular, speed and control are likely to be contradictory.

The most difficult and controversial aspect of soft loan schemes is their degree of efficiency in meeting the results expected. Although soft loans may be provided for a variety of reasons and to finance a range of activities, the most common case is that of a loan to finance a fixed investment. The criticism levelled is that the investment would have been made even in the absence of the aid provided. Contrary to the opinion expressed by the Council, this constitutes a general problem with subsidy schemes and is by no means restricted to soft loans.

According to corporate finance literature, the returns on an investment project should be compared with a firm's cost of capital to ascertain whether or not an investment should be undertaken. All projects with a profitability greater than or equal to the cost of capital should be financed. In theory, therefore, it should be sufficient to establish whether or not the projected return on an investment covers the cost of capital in order to determine whether the aid provided has been decisive in the investment decision. In practice, things are not so simple. The cost of capital varies not only from firm to firm but also from project to project. A large investment financed by a loan will alter the degree of leverage, which may affect negatively a company's credit rating. The attendant increase in the cost of capital can then make other investments no longer worthwhile. Even on a firm-wide basis, cost-of-capital calculations are complicated to make, particularly if the most reliable method of evaluating equity, the capital asset pricing model (CAPM), is employed. Non-quoted firms and firms in countries with small or illiquid stock exchanges are precluded from applying CAPM at all. A study undertaken for the European Commission by Coopers and Lybrand at the beginning of the 1990s found that even large firms do not calculate their cost of capital for the purpose of making investment decisions.

An alternative approach which appears more in line with corporate practice assumes that an enterprise has a portfolio of possible investments at any moment in time. According to prevailing conditions, the most profitable or the least expensive or some combination of these two criteria will be applied and an investment carried out. The relevant variables are market conditions, which determine the extent to which investment possibilities are present and their profitability, the cost of finance, in particular the prevailing interest rate faced by a firm, and internal financing possibilities. According to the pecking order theory of corporate finance, external financial sources are a second best. In order to finance the growth of fixed assets, managers give priority to retained earnings, then to borrowing and finally to issuing shares. When retained earnings are high, many investments are made, even with low profitability and conceivably even below a firm's average cost of capital.

Under such circumstances, the question whether a specific project would have been financed in the absence of aid becomes irrelevant. If a supplementary project were to be financed from the portfolio of possible investments, the intention of the aid could be very well met. The difficulty then is to establish whether all of the portfolio realised would have been undertaken in the absence of the aid. Where there is a definable credit constraint either in total amounts available to the firm or in a sharply rising cost curve for borrowing, it may be possible to demonstrate that a subsidy has really led to additional investment. In many cases, this will not be possible. Applying the pecking order theory, an investment that could be financed by retained earnings would probably have been executed in any case, while an investment requiring an equity injection would not have been undertaken. This leaves a considerable grey area covered by investments for which a loan would in any case have been required.

Similar considerations apply to the lender. Although any individual loan may have been made in the absence of a subsidy, aid which expands the total amount of loans for a given category of investment that it is the intention of public authorities to assist can be deemed to meet the objectives of the scheme. Where the number of potential recipients is small or market conditions unfavourable, then the probability that the loan portfolio will have expanded is reduced. Where the lending institution faces a limitation on the quantity of loans it can provide, e.g. as a result of balance-sheet restrictions, or where the lender's cost of funds is high, then the probability that the loan portfolio has expanded as a result of the subsidy is increased.

In the final analysis, the impact of a subsidy scheme for a given policy objective should be seen as an empirical and not a theoretical question. It relates to the tools of evaluation that are available and the capacity of the Commission to apply them. Systems involving public subsidies are subject to a deadweight loss or "effet d'aubaine". Rigorous evaluation procedures require that such losses be calculated in a formal way [41]. In the United States, where evaluation methods are more developed, systematic recourse to control groups for public aid schemes is encouraged and often mandated. Control groups are a useful tool for evaluation, but aid recipients may not themselves be typical of firms in the economy as a whole. It follows that differences in outcomes between recipients and the control group do not necessarily stem from the programme of aid. Closer definition of control groups by matching on the basis of defined objective criteria (size of firm, sector, location, age of entrepreneur, etc.) can reduce but not eliminate this problem. Very rigorous selection procedures are required in order to take into account various forms of bias in making comparisons between recipients and non-recipients. In the end, it is the lack of a culture of evaluation in Europe that leads to a situation in which the effect of subsidy schemes is unknown and therefore open to criticism on efficiency grounds.

[41] "Six steps to heaven: evaluating the impact of public policies to support small businesses in developed economies", David Storey, Warwick Business School Working Paper No 59, September 1998, sets out the problems of evaluation in an area of significance for the Commission in a practical way.

8.6. Conclusion

Since capital in developed countries is no longer scarce and is increasingly cheap and because of the growing importance of intangibles in the economy, it is likely that the use of soft loans as an instrument of economic policy will decline both at EU and Member State level. The same does not apply to developing countries or therefore to the Commission's development policy.

In certain specific cases, in particularly where there are externalities which cannot be dealt with in other ways, soft loans may remain an appropriate instrument. Excluding any future recourse to this instrument on theoretical grounds would not appear to be justified. Only on a case-by-case evaluation of the advantages and disadvantages of soft loans compared with other instruments will it be possible to judge whether such loans are appropriate for a given objective. In any case, it will not be possible always to replace soft loans by loan guarantees since they address different kinds of problem and in some cases are complementary.

FINANCIAL INSTRUMENTS: SOME EXAMPLES

>TABLE POSITION>

>REFERENCE TO A GRAPHIC>

>REFERENCE TO A GRAPHIC>

>REFERENCE TO A GRAPHIC>

9. ANNEX V: Macroeconomic Effects of an Interest-Rate Subsidy fOr Firms in the EU

9.1. Introduction

This annex presents results concerning the effects of an EU-wide interest-rate subsidy for the corporate sector, using DG ECFIN's macroeconomic model QUEST II. An interest-rate subsidy is one way to promote private investment. The channel is via a reduction in capital costs for firms. The analysis presented here looks at the macroeconomic impact of such a policy, i. e. the implicit assumption is made that the policy is not targeted at specific forms of investment, specific sectors or specific types of firm but applies to the corporate sector in general. The simulations can therefore only give a broad picture and illustrate the various channels through which policy affects the main macroeconomic aggregates. In contrast to a purely microeconomic or sectoral analysis, a macroeconomic analysis will, however, consider many possible feedback effects. In particular, this simulation analysis will shed some light on the following issues:

* How responsive is investment to a change in interest rates-

* Are there offsetting effects, e.g. an increase in the interest rate when aggregate demand increases-

* What are the exchange-rate effects-

* What are the budgetary implications-

* How much of a GDP and employment effect can be expected from an increase in investment-

* What is the time horizon for the effects to materialise-

Though feedback mechanisms are important, the initial impact of interest rates on investment is, of course, crucial for the aggregate impact. The QUEST model allows for such a direct interest-rate effect since investment is specified using the standard neoclassical model of the firm with adjustment costs, i.e. investment is a function of the present value of profits. This model of investment is also known as Tobin's q model. This hypothesis postulates that firms equate the marginal product of capital to capital cost and marginal adjustment costs. The interest rate enters the capital-cost term together with the depreciation rate, the corporate tax rate and the purchase price of new investment goods. Policy has three instruments available to promote investment; it can either lower corporate taxes/increase depreciation allowances, subsidise investment or subsidise interest rates. The last of these policy alternatives is considered in this annex.

9.2. Assumptions

In the simulation presented below, a permanent EU-wide subsidy is assumed which reduces the real interest rate of the corporate sector by half a percentage point. This amounts to a reduction in capital costs of some 7%. Such a subsidy would represent a considerable fiscal stimulus amounting to more than 2% of EU GDP. [42] It is assumed that the government does not offset this subsidy with either a reduction in expenditures or an increase in taxes but instead runs a deficit/surplus. In this way one can see most clearly the budgetary implications of the policy measure. A debt rule is invoked after ten years which adjusts government transfers to households in order to restore the baseline debt-to-GDP ratio in the medium term.

[42] QUEST does not distinguish between different sources of financing and different forms of ownership. In the simulation full debt financing of investment by firms is assumed, with the result that the interest-rate subsidy is a subsidy to the total capital stock.

9.3. Simulation results

The results of this simulation for the EU economy are contained in Table 1. First, it is interesting to look at the effects on GDP, private consumption, investment, employment and wages. Starting with the direct impact on private investment, it will be seen that investment is sensitive to changes in capital cost. Investment increases by about 5% on average in the first ten years. The elasticity of investment with respect to the user cost is about 0.7. This is within the range of recently reported estimates for OECD countries. [43] But notice also that, despite a strong investment response, the GDP effect is rather small. There exists both a supply-side and a demand-side interpretation of this result. It takes time for investment to increase significantly the capital stock of the economy. As can be seen from the table, the capital stock increases only by 0.17% in the first year and it takes ten years for the capital stock to increase by 2%. With an output elasticity of capital below 0.5, it is obvious that the GDP response will be limited, even with such a boost to investment. An increase in employment would be another factor which could increase GDP. However, the employment effect is limited. There are two reasons for this: first, the policy lowers the relative price of capital and therefore induces substitution effects and, second, since wages are partially linked to an increase in labour productivity, the policy leads to an increase in real wages.

[43] There is considerable scepticism in the macroeconomics profession as to whether capital costs are a significant investment determinant. This scepticism is based mostly on time-series estimates of investment equations where generally a strong accelerator term is found, reflecting the stylised fact of a strong positive correlation between investment and output over the business cycle. Moreover, real interest rates themselves are partly endogenously determined and show a procyclical pattern. This makes it difficult to estimate the effect of exogenous interest-rate shocks on investment with time-series data. Recent empirical research into the effects of capital costs on investment has therefore moved away from time-series analysis to individual-firm data sets, which make it possible to look more closely at the investment response to "exogenous" changes in capital costs, e.g. by looking at the investment response of firms after tax reforms. An influential paper by J. G. Cummins, K. A. Hassett and R. G. Hubbard (1996), "Tax reforms and investment: A cross-country comparison" (Journal of Public Economics 62, pp. 237-273), arrives at the conclusion that for a set of seven OECD countries the investment elasticity with respect to the user cost of capital lies between 0.5 and 1.0. A recent study by R.S. Chirinko, S. M. Fazzari and A. Meyer (1999), "How responsive is business capital formation to its user cost- An exploration with micro data" (Journal of Public Economics 74, pp. 53-80), arrives at an estimate of 0.25.

There is also a demand-side aspect which should not be neglected. The consumption response is negative. This response is important for understanding the macroeconomic effect of the policy. On a macroeconomic level a policy of promoting investment will, as a result, increase the rate of savings, at least temporarily. Lowering interest costs for firms allows them to borrow at higher rates and thereby stimulates savings. [44] Higher interest rates also attract capital inflows and lead to an appreciation of the euro. While this is necessary to increase investment, it obviously has demand repercussions which limit the expansion of the economy as a whole. Nevertheless, the permanent rise in investment increases the capital stock and therefore permanently raises potential output. The adjustment process is, in fact, not completed after ten years but it takes about 25 to 30 years before 90% of the total adjustment is completed, with a permanent increase in the level of GDP of around 2.5%. The policy also generates employment effects of some 0.4% in the long run.

[44] As can be seen from the table, the increase in real interest rates partly offsets the initial interest-rate subsidy.

Although this policy seems attractive from a macroeconomic point of view, there are also considerable costs involved and the macroeconomic welfare effects do not seem to be overwhelmingly positive. First notice that, as can be seen from the rise in the government deficit, the interest-rate subsidy is far from self-financing. This becomes clear if one compares the trend of income (GDP) with the trend of consumption, which shows that the losses in VAT revenue outweigh the induced increase in income taxes. In order to stabilise the debt-to-GDP ratio, it is assumed in the model that transfers to households will be reduced. This is another reason why consumption is not picking up. Of course, this mode of financing is relatively non-distortionary; if other financing alternatives were chosen such as an increase in labour taxes, the GDP and employment effects would be likely to be smaller.

Table 1: Macroeconomic effects of an interest-rate subsidy (EU economy)

>TABLE POSITION>

PCER: % deviations from baseline

ER: % point deviations from baseline.