51997AA0004

OPINION No 4/97 on the proposal for a Council Regulation (Euratom, ECSC, EC) amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities

Official Journal C 057 , 23/02/1998 P. 0001


OPINION No 4/97 on the proposal for a Council Regulation (Euratom, ECSC, EC) amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities (98/C 57/01)

THE COURT OF AUDITORS OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Atomic Energy Community, and in particular Article 183 thereof,

Having regard to the Treaty establishing the European Coal and Steel Community, and in particular Article 78h thereof,

Having regard to the Treaty establishing the European Community, and in particular Article 209 thereof,

Having regard to the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities (1), as last amended by Council Regulation (EC, Euratom, ECSC) No 2335/95 of 18 September 1995 (2), and in particular Article 140 thereof,

Having regard to the Commission proposal of 26 July 1996 (3),

Having regard to the request for an opinion on this proposal sent by the Council to the Court on 19 September 1996,

Whereas the draft Commission Regulation (Euratom, ECSC, EC) of 5 September 1996 (4) amending Commission Regulation (Euratom, ECSC, EC) No 3418/93 of 9 December 1993 laying down implementing rules for certain provisions of the Financial Regulation (5) must also be taken into account,

HAS ADOPTED THE FOLLOWING OPINION:

1. Since the Court of Auditors came into being in 1977, it has been required to deliver about 20 opinions relating to proposals for amendments to the Financial Regulation applicable to the general budget, and in one year, 1987, it was called upon to produce four such opinions.

2. Indeed, right from the earliest years of the Community, even before the Court was set up, and subsequently as a result of its opinions, the Financial Regulation has been amended numerous times, at a rhythm which has tended to fluctuate a great deal. Instead of making a regular review of whether there is cause to revise certain provisions, the Commission, except for the general revision it made in 1990, has usually gone about settling particular problems on a case-by-case basis, without ensuring that the text as a whole is consistent.

3. As a consequence, the Financial Regulation, which, even before 1977, showed some degree of inconsistency owing to the co-existence of several different sets of rules, each with numerous derogations, has over the years become a heterogeneous document, to which, each time there is a revision, passages have been added that fit in less and less with the rest of the text.

4. The time has therefore come for the Commission to stop proposing piecemeal, stop-gap reforms and instead to embark on a general overhaul of all the Communities' financial regulations; the Commission needs to redefine clearly the fundamental principles on which these regulations must be based, so that it can then establish on this basis the whole set of financial regulations - the Financial Regulation, the Implementing Rules and also the internal rules for the implementation of the budget and all the other more specialised, subordinate regulations.

5. Until now, the Court of Auditors has endeavoured in its opinions to reply paragraph by paragraph to each of the amendments put forward by the Commission. However, in taking stock of these successive reforms, the Court has found that, although on the face of it rectifying certain specific shortcomings, they in fact gradually undermine the consistency of the text as a whole. In this Opinion, the Court makes a critical analysis in the light of the whole Financial Regulation as it would be if the Commission's proposal were adopted as it stands, and then makes some general recommendations. These recommendations, which are merely touched on in paragraph 16, are set out in greater detail in the Annex to this Opinion, which gives the broad outlines of what could form the basis for a reformed set of financial regulations.

6. This Opinion, submitted to the Council, is delivered pursuant to Article 209 of the EC Treaty and the corresponding articles of the other treaties, and therefore concerns only the proposal relating to the Financial Regulation. Nevertheless, in making its overall scrutiny, the Court has not been able, in the interest of consistency, to exclude the draft Regulation amending the Implementing Rules. The Court has had to take that into consideration, as the two documents are obviously interrelated.

7. Once the procedure currently under way has been completed and the Council has adopted a regulation amending the Financial Regulation, the Court will give its opinion on the draft Regulation amending the Implementing Rules, an opinion which it will deliver to the Commission pursuant to Article 139 of the Financial Regulation.

8. There is no shortage of reasons to support a general overhaul of the Financial Regulation. Some of the terms used have never been defined and some do not spell out the responsibilities which thus become incumbent upon the various financial officials involved. Many of the expressions it uses are vague or, by their wording, introduce 'grey` areas in legal terms. Expressions such as 'in particular`, 'where appropriate`, 'if need be`, 'as a rule`, 'in principle`, 'endeavour`, 'exceptionally` and others of the same ilk are to be found by the score. The main drawback of these expressions is that at the same time as laying down a rule, they also lay down the possibility of derogating from it, without specifying the cases where such a derogation would be permitted. As a result, many provisions have virtually no legal application.

9. Furthermore, the terminology is in some cases inconsistent. Close to 20 different expressions are used to denote commitments in the Financial Regulation, the Implementing Rules and the explanatory memorandum of the two documents submitted to the Court for its opinion, but the concepts that need to be distinguished do not warrant more than five or six terms (see paragraphs 2.21 and 2.22 of the Annex). Moreover, for a number of provisions in the Financial Regulation, the different language versions contain discrepancies in terms of meaning. The Court will send the Council and the Commission a list of the disparities it has found. The terminology must be scrupulously standardised.

10. For some of the Financial Regulation's exceptional arrangements, there is uncertainty as to the limits of the applicability of the provisions that have general validity (i.e. save where otherwise provided for). This is the case for Title IX because of the proposed wording of Article 112 (less clear than the present version in this respect), Title XI and Part II.

11. The interaction between the Financial Regulation and its Implementing Rules is not always consistent. For example, Articles 19 and 34 of the Implementing Rules needlessly repeat Articles 79 and 24 respectively of the Financial Regulation. Article 21 of the Implementing Rules is sufficiently important to warrant being incorporated into the Financial Regulation. The same is true for much of Articles 133 to 136, which distinguish between the budget accounts and the general accounts; the main gist of these Articles ought to be placed near to the amended version of Article 70 of the Financial Regulation, which introduces a new distinction between budgetary expenditure and revenue on the one hand and non-budgetary expenditure and revenue on the other.

12. Essential financial provisions relating to certain aspects of the implementation of the general budget are to be found in regulations governing certain policies or special measures (Structural Funds, new financing mechanisms relating to external aid, etc.). The basic regulations governing the Structural Funds contain budgetary provisions relating thereto which are incompatible with the Financial Regulation, but the latter makes no mention of the Structural Funds. Provisions of this type should be incorporated into the Financial Regulation, once it has been ensured that they are consistent with all the provisions of the latter. The primacy of the Financial Regulation in the hierarchy of the legal texts derived from the Treaties would thus be restored and the provisions in question would no longer, without good reason, fail to be subject to the procedure stipulated in Article 209 of the EC Treaty.

13. Certain matters, whilst not completely omitted, are dealt with in an altogether unsatisfactory and inadequate manner. This is true of the definition of and the detailed rules for the implementation of the liability to disciplinary action and payment of compensation of the prime movers in the financial procedure - the authorising officers, financial controllers, accounting officers (6*) and administrators of imprest accounts.

14. As a result of the successive proposals for revising the Financial Regulation, a good many 'facilities` have been arranged, or allowed to emerge; these discretionary arrangements are regarded as useful for managers - a view which still has to be proved on a case-by-case basis - but tend to run counter to a disciplined approach and hugely complicate the accounting and financial management, especially its computerised aspects. In the Annex to this Opinion, a large number of these 'facilities` are analysed and specific recommendations about them are made.

15. Of these 'facilities`, which the Court does not consider to be indispensable and even regards as irregular in some cases, special mention may be made of:

- the authorisation and existence of negative items of revenue, which are really items of expenditure; negative items of expenditure, which are really items of revenue; and negative reserves of appropriations, which enable the overall amount of the budget to be artificially kept at a lower level than the total sum of the allocations considered heading by heading; all of which are procedures that lack transparency and frequently infringe the principle of the universality of the budget (see paragraphs 1.11 to 1.14 of the Annex),

- the possible financing of the same measure or expenditure of the same kind from several budget headings, which infringes the principle of specification of the budget (see paragraph 1.17 of the Annex),

- the reduction in the number of budget headings reserved for areas which have, however, been allocated rapidly growing amounts of appropriations, especially in respect of the Structural Funds and the Cohesion Fund, to the extent that certain articles or items have several thousand million ecu in appropriations entered against them, a fact which once again serves to water down the principle of specification (see paragraph 1.20 of the Annex),

- the proliferation and complexity of the cases where carryovers are possible or even automatic, and where the principle of annuality is thus ignored, contrary to the Treaty itself, which explicitly states this principle (Article 199 of the EC Treaty), and also the proliferation and complexity of the reasons cited by way of justification for some of these carryovers, reasons which tend to elevate the Institution that decides on the carryover to the status of court of appeal for the budgetary authority (see paragraphs 1.25 and 1.26 of the Annex),

- the fact that the Financial Regulation contains no provisions relating to the accounting principles which apply to the preparation of the Community financial statements, to the extent that it is only in one of the very last paragraphs of the Implementing Rules that any provisions on this matter are to be found (see paragraphs 1.33 and 1.34 of the Annex),

- similarly, the lack of any precise definition of the objectives connected with the financial statements and of the accounting framework to be used to trace back in a transparent manner the implementation of the budget and the changes in the assets and liabilities making up the Community balance sheet (see paragraphs 1.39 and 1.40 of the Annex),

- the lack of accounting practices which are common to all the institutions, despite the provision expressly stipulating that they be introduced (Article 21 of the Implementing Rules) (see paragraph 1.41 of the Annex),

- the fact that mini-budgets, having undergone one metamorphosis after another, still persist; mini-budgets entail the use of operating appropriations to effect administrative expenditure that usually comes under Part A of the budget; this has the effect of blurring the distinction between the two parts of the budget (see paragraph 2.4 of the Annex),

- the system of appropriations for the Joint Research Centre; over 80 % of these appropriations are entered in the budget according to the type of expenditure, without any clear indication as to how they are allocated to the research objectives, i.e. to the programmes; this is because, in the case of the JRC, the principle of defining research appropriations according to their objective has been accompanied by exceptions which have proliferated from one reform of the Financial Regulation to the next (see paragraph 2.5 of the Annex),

- the confusion between the concepts of payments by instalment and advances, and the booking of advances as budgetary expenditure in cases where they cannot be regarded as budgetary expenditure since they are simple cash payments which are offset by the establishment of debts for the same amounts (see paragraphs 3.5 to 3.14 of the Annex),

- the existence of revenue generated by loan and borrowing operations which, instead of being entered in the accounts on one heading of the budget, goes - without any budgetary authorisation - into a fund, the use of which for expenditure is also not covered by the budgetary procedures (see paragraph 1.10 of the Annex),

- the awarding of contracts in respect of external aid without certain generally applicable provisions being observed, since the Commission mistakenly considers that they are not applicable in this instance (see paragraphs 4.3 to 4.7 of the Annex),

- the abuse of subcontracting, which in the most serious cases results in public authority being stripped of its powers and responsibilities (see paragraphs 4.8 to 4.17 of the Annex),

- the discontinuation of the prior approval of the Financial Controller, which has to a large extent already been embarked on, de facto, by the practice of making prior random-sample checks, and which, according to the Commission's proposal, would be extended to cover budgetary commitments relating to individual contracts ('specific` commitments) (see paragraph 5.5 of the Annex),

- the suggestion that the Financial Controller might be assigned the function of internal auditor; the Court would in principle support such an extension of his role but, since the organisational and operational consequences of this are not clearly defined in the Commission's proposal and since no clarifications or guarantees are made, it considers that this extension of the Financial Controller's role would not offer sufficient compensation to protect the financial system against any internal control defect that might result from the discontinuation of the prior approval proposed in other respects (see paragraphs 5.7 and 5.8 of the Annex).

16. The proposal for an amendment to the Financial Regulation presented by the Commission is not likely to solve the problems set out above. The main objectives of the whole set of financial regulations should be to help the Communities to achieve their political aims effectively by allocating their budgetary resources in the best possible way, to guarantee the legality and regularity of their acts, to protect their assets and to ensure the disclosure and the transparency of their financial operations. To this end, a new Financial Regulation should in particular contain the following principles and provisions:

(a) in respect of the budget:

- there should be a return to a strict application of the budgetary principles, in particular the distinction between differentiated and non-differentiated appropriations and the various mechanisms for carryovers and for offsetting (for example, negative amounts) should be abolished, and the way in which expenditure is classified in the budget should be simplified and rationalised,

- the concept of commitment should be defined in a sufficiently clear and general manner for it to be applicable to all aspects of the budget,

- all appropriations, even administrative ones, should be allocated to a set Community policy, a solution that would help to achieve optimum use of the available resources and precise knowledge of the total costs of each measure; or, alternatively, there should be a return to making a clear distinction between administrative appropriations and operating appropriations, and, in any case, hybrid solutions such as 'mini-budgets` should be eliminated,

- it should be pointed out forcefully that the budgetary procedure is the only source of authorisations for revenue and expenditure,

- advances should be strictly limited to real cash needs and it should be forbidden to charge them to the budget until such time as they are used to cover eligible expenditure,

- rules should be laid down that are common to all the institutions for transfers of administrative appropriations,

(b) in respect of the drawing-up of financial statements:

- the Commission should be required to prepare every year a consolidated revenue and expenditure account showing the annual implementation of the budget, a consolidated statement of the assets and liabilities and the notes attached to the financial statements, as it already does,

- however, the Financial Regulation should refer to the abovementioned notes, stipulate that the statement of assets and liabilities and these notes should be drawn up in accordance with the provisions of the Fourth Directive and with international accounting standards and make provision for an interim statement between the consolidated revenue and expenditure account and the statement of assets and liabilities, showing the impact on the latter of the income and expenditure (e.g. depreciation) that do not stem from the management of the budget,

- the Commission should be required to send the draft consolidated financial statements to the Court of Auditors, so that the latter can audit them before they are definitively adopted,

- provision should be made for the formal adoption of the definitive consolidated financial statements and this adoption should be made the task of the Members of the Commission,

- there should be a requirement that the audited consolidated financial statements must be published in the Official Journal of the European Communities every year,

(c) in respect of exercising the internal control function in the institution:

- the status and powers of the Financial Controller should be redefined, on the assumption that he would exercise the functions of internal auditor of the institution,

- a clearer indication should be given of the scope of the authorising officers' powers in respect of financial management and financial control. Since the prior approval procedure is likely to be used significantly less or even, in the end, to disappear completely, additional provisions need to be introduced to ensure the legality, regularity and sound financial management of Community measures and to protect the Communities' financial interests, in particular to prevent officials from having to face any situation where a conflict arises between the interests of the service and the interests of third parties,

- on the same assumption, the powers of the accounting officer need to be strengthened.

17. An overhaul of this kind could only be achieved if the Commission were to use an organisational approach and apply a working method which were very different from those to which it has had recourse hitherto. If it did not want to give up its prerogatives, it could, for example, set up an internal working party enhanced by the inclusion of high-ranking experts selected from circles in the institutions and the Member States with extensive practical experience and sound knowledge of the matters to be tackled.

18. According to the Commission, the purpose of its successive proposals for drawing up and then revising the Financial Regulation (and its Implementing Rules) is to update the Financial Regulation, to rationalise it and even to make it stricter. Thus the Commission now invokes the needs that arise because of 'SEM 2000`, the initiative for improving financial management.

19. The fact is that, even if the Financial Regulation were to be amended in line with the Commission's proposal, it would still contain many weaknesses and inconsistencies, and that is why a complete recasting is warranted.

20. These weaknesses and inconsistencies tend to make it easier for the number of cases of fraud and irregularities against the Community budget to grow. In the light of this increase, which is a source of concern for all the institutions, the Court would very much like to see a stricter and more consistent Financial Regulation.

This Opinion was adopted by the Court of Auditors at its meeting of 10 July 1997.

For the Court of Auditors

Bernhard FRIEDMANN

President

(1*) Translator's note: It would be preferable to translate the French term 'comptable` as 'accountant` in English rather than 'accounting officer`, which is misleading as the 'Accounting Officer` in the UK and Irish public systems has a completely different function to the 'comptable`. However, the term 'accounting officer` is retained here for the sake of consistency with the Financial Regulation.

(2) OJ L 356, 31.12.1977, p. 1.

(3) OJ L 240, 7.10.1995, p. 12.

(4) Document COM(96) 351 final and OJ C 296, 8.10.1996, p. 13.

(5) Document SEC(96) 1356 final (not published in the EC OJ).

(6) OJ L 315, 16.12.1993, p. 1.

ANNEX AN ANALYSIS OF HOW THE FINANCIAL REGULATION COULD BE REFORMED

TABLE OF CONTENTS

Paragraph reference Page

1. BUDGETARY AND ACCOUNTING PRINCIPLES 1.1 - 1.48 8

Introduction 1.1 - 1.2 8

The principles of budgetary law 1.3 - 1.32 8

Introduction 1.3 - 1.4 8

The principles of unity and universality 1.5 - 1.14 8

The principle of budgetary specification 1.15 - 1.21 9

The principle of annuality 1.22 - 1.28 10

The principles of disclosure and transparency 1.29 - 1.32 11

Accounting principles 1.33 - 1.48 11

The general framework 1.33 - 1.34 11

The various elements of this framework 1.35 - 1.41 12

Ways towards reshaping the Financial Regulation 1.42 - 1.48 12

2. THE SYSTEM OF APPROPRIATIONS 2.1 - 2.41 13

The distinction between administrative appropriations and operating appropriations 2.1 - 2.7 13

The principle underlying the distinction 2.1 - 2.2 13

The blurring of the distinction 2.3 - 2.5 13

Recommended approaches 2.6 - 2.7 14

Differentiated appropriations and non-differentiated appropriations 2.8 - 2.18 14

The basic concept underlying the system now in force 2.8 - 2.10 14

The system's malfunctions 2.11 - 2.17 14

The remedy: the differentiation of all appropriations 2.18 16

Towards two or more levels of commitment 2.19 - 2.41 16

The broad outline of the system envisaged 2.19 - 2.20 16

The shortcomings of the existing and proposed provisions 2.21 - 2.41 16

3. TOWARDS AN ACCRUALS-BASED ACCOUNTING SYSTEM 3.1 - 3.20 20

The recording of assets and the drawing-up of the balance sheet 3.1 - 3.2 20

The valuation of assets 3.3 - 3.4 20

The accounting treatment of advances and payments by instalment 3.5 - 3.14 20

The treatment of interest 3.15 - 3.20 21

4. AWARDING OF CONTRACTS 4.1 - 4.17 22

The role of the ACPC 4.1 - 4.2 22

The case of external aid 4.3 - 4.7 23

Misuse of subcontracting 4.8 - 4.17 23

5. THE RENEWAL OF THE CONTROL FUNCTION 5.1 - 5.27 24

The respective roles of the various officials responsible for control 5.1 - 5.17 24

Introduction 5.1 24

Departures from the present system 5.2 - 5.9 25

A new definition of the role of the financial officials 5.10 - 5.16 26

The Court of Auditors' rights of access 5.17 27

Compensation of financial damage suffered by the Community 5.18 - 5.19 27

The liability to disciplinary action and to payment of compensation of each institution's financial officials 5.20 - 5.23 27

Overcoming the weaknesses in adapting to computerisation 5.24 - 5.27 28

Introduction 5.24 28

Access to supporting documents 5.25 - 5.27 28

1. BUDGETARY AND ACCOUNTING PRINCIPLES

Introduction

1.1. The main objectives of the financial regulations, as the Court has pointed out in paragraph 16 of the Opinion, are to help the Communities to achieve their political aims effectively by allocating their budgetary resources in the best possible way, to guarantee the legality and regularity of their measures, to protect their assets and to ensure the disclosure and the transparency of their financial operations.

1.2. These objectives are reflected in the definition and implementation of, on the one hand, principles of budgetary law governing the collection of resources and the execution of expenditure, and, on the other hand, accounting principles relating to the recording of these operations and their presentation by means of financial statements.

The principles of budgetary law

Introduction

1.3. The Communities' budgetary law is traditionally governed by the following five fundamental principles:

- unity,

- universality,

- specification,

- annuality,

- equilibrium between revenue and expenditure.

In a democratic framework, it is essential to add to this list the principles of disclosure and transparency. The 'transparency of the decision-making process` has, moreover, been incorporated into Community law by Declaration No 17 on the right of access to information, which is annexed to the Maastricht Treaty.

1.4. The Court will discuss the principles of unity and universality together in the next section. Universality is a corollary of unity. Budgetary balance will not be discussed as it scarcely poses a problem in the Community context. The principles of specification and annuality will both be dealt with in detail, however, as they do indeed give rise to serious problems in the current situation. Lastly, there will be some comments on the principles of disclosure and transparency.

The principles of unity and universality

1.5. Certain exceptions to the principle of budgetary unity have arisen because of political choices that are beyond the scope of the Financial Regulation, which can do no more than take note of them. This applies to the existence of the ECSC operating budget, the Euratom Supply Agency, the European Development Funds and the various satellite bodies.

1.6. Other structures have also arisen recently as a result of policy decisions reflected in a legal basis consisting of a Council Regulation or Decision and, here too, the Financial Regulation can only take cognisance of them, but these cases involve a number of situations which are, to varying degrees, in conflict with the principle of unity. This is particularly the case for the Guarantee Fund for External Operations, the European Investment Fund and the European Guarantee Fund to encourage cinema and television productions which is to be created in the near future by a Council Decision based on an existing Commission proposal.

1.7. A special feature of these new structures is that the general budget is thereby financing, by means of appropriations that can be identified under one or more budget headings, the creation of 'authorised capital` or of a sort of working capital or 'reserve`. Payments from the Community budget are entered in the accounts as 'budgetary expenditure` although, viewed from a commercial accounting point of view, they would not be regarded as 'costs` as they are offset by the formation of a Community asset. Once these Funds have been constituted, they produce expenditure and revenue of their own which are not included in the Community's revenue and expenditure account. By creating a sort of reserve/safety buffer between the final expenditure, which is subject to erratic variations, and the budgetary expenditure, which is being regularised by the existence of this 'shock-absorber`, the Community is abandoning the direct recording of the expenditure and revenue ultimately incurred as a result of its budgetary interventions and, by virtue of this fact, runs the risk of losing both a clear perception and, therefore, control of them.

1.8. Loss of budgetary unity is, of course, bound to happen where the Community is merely one partner amongst others, but it should not be seen as so inevitable when the Fund is 100% financed by the Community (as is the case with the Guarantee Fund for External Operations).

1.9. The Community Institutions must become aware of the fact that these complex and variable financial mechanisms, which are tending to proliferate, are paving the way for an increasing fragmentation of the Community finances and a loss of financial transparency. It could thus become increasingly difficult to obtain an overall view of the Community's financial activities. This being the case, appropriate provisions concerning these mechanisms need to be incorporated into the Financial Regulation.

1.10. Equally counter to the principle of budgetary unity but lacking any higher legal justification is the fact that lending and borrowing activities generate revenue which is not recorded as such in the general budget. The practice of entering this revenue as extra-budgetary revenue, on the pretext that the loans and borrowings are themselves extra-budgetary, leads to the formation of a cash reserve which is used in part for expenditure without any legal framework and outside the scope of the budgetary procedure. The Financial Regulation must, therefore, as a matter of urgency, include a provision to incorporate this revenue and expenditure into the general budget.

1.11. The budget includes a number of negative amounts. These are primarily negative revenue, which is actually expenditure, and, secondly, negative expenditure, which is actually revenue and, lastly, negative reserves which, formally, represent a negative allocation of appropriations in the 'expenditure` section of the budget, an allocation which is not required to be 'used` by means of commitments and payments, but offset and cancelled out by transfers of positive, unused appropriations.

1.12. In some cases these negative amounts are authorised by the present Financial Regulation: this applies to the negative reserve (Article 19(5)) and to the clearance of the EAGGF Guarantee Section accounts (Article 102(3)). In other cases these negative amounts have been created within the framework of the budgetary procedure even though the Financial Regulation makes no provision for them: this applies to the co-responsibility levies which are imposed under certain common market organisations, or to the costs incurred by the Member States in collecting traditional own resources. Moreover, Article 19(5) of the Financial Regulation authorises only one negative reserve. However, in the budget for 1997, for the first time, two distinct, separate negative reserves have appeared.

1.13. Every negative amount introduces a lack of transparency and makes reading and understanding the budget more difficult. Furthermore, when these negative amounts are of a special type and are used to reduce amounts under positive headings, certain items of revenue are offset against certain items of expenditure, i.e. there is a breach of the principle of budgetary universality. A similar criticism must be made when certain special negative reserves are offset, by means of transfers, against specific positive allocations of appropriations.

1.14. The existence of any negative amount in the budget must therefore be forbidden. To achieve this it is not enough to repeal the provisions permitting certain negative amounts, since the Commission has taken the liberty of creating others for which there are no provisions in the Financial Regulation: there also needs to be a single, simple provision forbidding the entry of any negative amount in the budget.

The principle of budgetary specification

1.15. The purpose of budgetary specification is to make sure that the Institutions are entrusted with the management of budget resources only for the specific purposes set out in the budget and defined in a sufficiently detailed manner, heading by heading. This definition requires the observance of several rules that the Financial Regulation does not make sufficiently clear.

1.16. Only expenditure which is authorised under a budget heading may be effected. It follows that no activity or programme which is not explicitly mentioned, together with its legal base, in the remarks relating to a budget heading, may be financed by this heading. It has, however, happened, particularly in respect of the Structural Funds, e.g. the EAGGF Guidance Section, that the 'measures` entered under a budget heading have been so numerous that they were not all named individually, with the result that there could be some doubt as to whether the 'right` heading had been chosen. In this respect, Article 20(2)(bb) is not precise and rigorous enough in stating that the remarks on each budget heading must include 'all such explanations as may be necessary concerning the nature and purpose of the appropriations`. For example, when 'the basic legal instrument` does not exist, what document is supposed to take its place?

1.17. Expenditure may be authorised only by one budget heading. This second rule is far from being complied with in practice and the vagueness of the remarks gives the Commission a pseudo-justification for charging the same kind of expenditure either to one heading or to another, depending on the respective availability of appropriations. In this respect, Articles 5(1), 19(2) and 26(1) are superfluous and at the same time not sufficiently strict.

1.18. It needs to be clearly stated which parts of the budget may include expenditure classified according to type and which parts may include expenditure classified according to use. In principle, a given heading should not contain a mixture of expenditure classified according to type and expenditure classified according to use, as is the case, for example, for shared-cost research and, more generally, for all operating appropriations headings to which administrative expenditure is charged.

1.19. The degree of specification according to titles, chapters, articles and items should be defined in a uniform manner. There is total confusion in this respect. On the one hand, the degree of specification is not defined anywhere. On the other, the hierarchy of divisions and subdivisions differs from one text to another. Article 202 of the EC Treaty, Articles 5(1), 19(2) and 26(1) of the Financial Regulation each set out a list of subdivisions of the budget headings, but these lists are all different. Lastly, the scale of the subdivisions affected by transfers of research appropriations is out of line with that generally applicable. These areas of confusion and inconsistency must be eliminated.

1.20. It is all too easy to circumvent the principle of budgetary specification by reducing the number of headings even though the total volume of the budget is growing relentlessly from year to year, and to allocate to certain headings huge amounts of appropriations which in themselves constitute budgets within the budget. This is illustrated particularly well by the Structural Funds and the Cohesion Fund. The ERDF's most important budget heading is a simple item which, in 1997, was allocated ECU 8 800 million in payment appropriations (10,7 % of the total budget). In the face of budget headings containing such amounts, the very concept of budgetary specification loses all meaning. Paradoxically, there are also a large number of headings in the budget corresponding to needs defined in excessively narrow terms and therefore containing minimal amounts of no more than a few thousand ecu. There should be a provision in the Financial Regulation aimed at reducing these disparities.

1.21. When a transfer can be approved by the institution without the agreement of the budgetary authority, this is yet another case of derogation from the principle of budgetary specification, since the use of certain appropriations has been changed. More generally, transfer arrangements which are too broad may jeopardise transparency by reducing the normative value of the budget as it was voted. In this respect, the provisions of Article 26 could be improved. Whilst derogation in respect of operating appropriations is a serious matter, as it offers the possibility of altering the content and relative importance of the measures decided by the legislative and budgetary authorities, it is much less so in respect of administrative appropriations. It may even be considered favourable to efficiency to allow the institution greater freedom to make the best use of the means at its disposal. This is, moreover, the guiding philosophy behind the rights conferred upon the Commission regarding transfers by which it can make transfers within its administrative appropriations from chapter to chapter but only from article to article within its operating appropriations. In this respect, however, the argument is valid for all the institutions and there is no reason to restrict the liberty of the Court of Justice, the Court of Auditors, the Economic and Social Committee and the Committee of the Regions to make transfers from article to article, whilst the Commission, the Parliament and the Council may make transfers from chapter to chapter. For administrative appropriations, all the institutions should be subject to the same arrangements.

The principle of annuality

1.22. There is only one exception to the principle of annuality which is necessary for obvious reasons: the need to be able to commit Community finances to multiannual programmes or activities. As discussed later in this document with regard to the system of appropriations, this need has been adequately catered for by the introduction of differentiated appropriations and the possible multiannual character not, strictly speaking, of commitment appropriations which, if not used, lapse at the end of the financial year for which they were entered (Article 7(2)), but of the underlying legal commitments which are entered into on the basis of these appropriations.

1.23. The desire to shorten the execution periods for commitments is reflected in the Commission's proposal to make it compulsory to enter a time-limit for execution for every commitment proposal (Article 1(7), first subparagraph), and, no doubt to cover those cases where the authorising officer might omit to specify it, to introduce a generally applicable time-limit, i.e. the end of the financial year following the one in which the commitment was entered into (Article 36(2)). These provisions are welcome. It should, however, be noted that Article 36(2) only concerns commitments based on differentiated appropriations, which means that, if there are none, the new provision, instead of tightening the regulations, would only worsen the situation as far as non-differentiated appropriations are concerned in relation to what is stipulated in Article 7(1)(b).

1.24. The Commission has made the opportune suggestion that an obligation to cancel commitments when their period of validity expires should be included in the Financial Regulation. It would, in fact, be preferable for this essential obligation to be included in the Financial Regulation rather than in the Implementing Rules as is the case at the moment (Article 71 of the latter). But this same Article 71 lays down another obligation just as essential in itself for the prevention of dormant commitments and for encouraging the authorising officers to remain vigilant in this respect: the outstanding balances of commitments for which, once the final payment has been made, the total is known, must be cancelled. This provision should also be transferred to the Financial Regulation.

1.25. Allowing for the exceptional cases of certain commitments being multiannual in nature, the principle of annuality must be enforced with all its consequences. The complex 'array` of carryovers, whether automatic or simply a possibility for the Commission, pursuant to Article 7 of the Financial Regulation, should, in the main, be abolished. In particular, it is not acceptable to cite cases where 'the appropriations made in the headings in question in the budget of the next financial year are not sufficient to cover requirements` (Article 7(1)(a) and Article 7(2)(b)) in order to justify the usefulness or the necessity of the carryover (1*).

1.26. Undoubtedly, the Commission's concern is to guarantee the continuity of Community action. But either:

- the budgetary authority has made provision, for the next financial year, for appropriations with the same purpose as those which, in the current state of the regulations, have been or may be carried over. In this case, the institution has the powers, as from 1 January of the new financial year, to continue the measures which are already in progress,

- or the budgetary authority has not provided for appropriations with the same purpose for the next financial year, thus clearly demonstrating its determination to terminate the measures in question.

1.27. As regards cases where the Commission is faced with emergencies calling for very speedy financial intervention (humanitarian aid and emergency aid) which could not be foreseen, the possibility of coping with them must and can be provided for every year in the framework of the budgetary procedure. Alternative solutions to carryovers, such as that of the emergency aid reserve (Chapter B7-91 of the budget), already exist in this respect.

1.28. The old Article 99 of the Financial Regulation (which has become Article 102 in the meantime) states, in paragraph 3, that 'any differences which may exist between the expenditure charged to the accounts of a financial year . . . and that established by the Commission when clearing the accounts . . . shall be charged as under- or over-expenditure to the financial year during which the accounts are cleared` (2**). Regulation (ECSC, EEC, Euratom) No 2049/88 of 24 June 1988 deleted the final part of the sentence ('to the financial year during which the accounts are cleared`). As a result the Commission is now free to choose the financial year in the course of which this discrepancy will be entered in the budget. Thus the Commission has even been able (see Court Opinion No 6/95, paragraph 26 (3)) to spread recoveries from Member States over four financial years. This is why the drafting of Article 102 should be reviewed and the part of the sentence which was deleted in 1988 should be reinserted. However, it would be acceptable for there to be a provision allowing exceptions to this principle in cases where the amount to be recovered exceeded the budgetary capacities of any given Member State, and for the recovery to be spread over several financial years. This spread, however, should be accompanied by conditions defining amounts and duration, and these conditions should themselves be included in the Financial Regulation.

The principles of disclosure and transparency

1.29. At present, the Community has a long way to go before it can publish the accounts showing the implementation of the budget with the same detail and according to the same headings as the budget itself. There is therefore both a failure to disclose fully and a lack of clarity, as it is extremely difficult, if not impossible, for the public outside the Community bodies to make any comparison between the planned and the actual implementation.

1.30. What is more, a large part of the budgetary and accounting movements relating to the Joint Research Centre (JRC) are even beyond the comprehension of most of the Community departments themselves. This is because the JRC has set up, for internal use, an accounting system that is outside the budget; this system does indeed produce documents, which are circulated to a certain extent, but they are so complex as to discourage any non-experts from reading them.

1.31. Similar practices are to be found in many other budgetary areas, for example, that of the Structural Funds, where crucial documents concerning the distribution of financial allocations are drawn up outside the scope of any budgetary procedure.

1.32. The Financial Regulation should therefore, on the one hand, stipulate that procedures of a purely administrative nature may not form the basis for the detailed allocation of certain appropriations and, on the other, require that the implementation of the budget must give rise to a disclosure in the same media, giving the same degree of detail and broken down into the same headings as the budget itself.

Accounting principles

The general framework

1.33. Pursuant to Article 54(3)(g) of the EC Treaty, the European Commission is responsible for submitting proposals for the harmonisation of accounting and auditing practices in the Member States. Some provisions exist on this basis in the financial regulations but only at the level of the Implementing Rules, Article 136(10) of which states that 'except where otherwise provided for by regulation, all the financial statements shall be presented in accordance with generally accepted accounting principles, including the principles provided for in Council Directives` (4).

1.34. It is not normal that the provisions applicable to a subject of this importance are not to be found in the text of the Financial Regulation itself and, in addition, are treated in such a brief manner. The Financial Regulation must include precise details on this subject.

The various elements of this framework

1.35. Accounting principles form part of a whole made up of three elements: the basic accounting framework, the accounting principles themselves and, lastly, the accounting policies and practices.

1.36. The basic accounting framework (cash-based accounting, accruals-based accounting or a compromise between the two concepts) determines at what point in the implementation of a transaction the effects of the latter are entered in the accounting books and registers. The basis to be used must be chosen according to the environment in which the organisation concerned operates and according to the purpose of the financial statements: for the public sector, Intosai (5) has published standards which are applicable in the Community context.

1.37. The accounting principles determine the rules according to which the financial statements are to be drawn up. The basic accounting principles are set out in Article 31 of the Fourth Council Directive.

1.38. Lastly, the accounting policies and practices are defined within the organisation and specify the detailed rules according to which it records accounting and financial operations and produces financial statements respecting the basic accounting framework and accounting principles.

1.39. As far as the Community system of accounting and producing financial statements is concerned, the present Financial Regulation (and its Implementing Rules) fulfils three main roles. Firstly, it complements the definition of the budgetary principles laid down in the Treaty. Further, it defines the accounting policies and practices which must be used to ensure application of the budgetary and accounting principles. It also defines the structure of the accounting system to be used to produce the financial statements. Lastly, it introduces a number of exceptions in order to 'adapt` the accounts or the underlying accounting principles to certain specific requirements. In other words, it authorises divergences from the budgetary and accounting principles and, in so doing, allows a certain 'creativity` which is generally regrettable in accounting matters.

1.40. The Community system works on a mixture of cash-based and accruals-based accounting. A large part of the serious uncertainties affecting this system - such as the problem of the treatment of advances - arise because the Community's accounting and budgetary principles do not allow it to properly take into account activities or events of an accruals-based accounting nature.

1.41. As the DAS special report for 1995 underlines (6):

'In the context of the Community's accounts, the Commission has been empowered by Article 21 of the Implementing Rules to decide on the accounting methods and harmonise the presentation of the financial statements. Thus, in particular, after the Commission's accounting officer has consulted the accounting officers of the other institutions, the Commission must adopt the accounting methods that are to apply to all the institutions.

In spite of this the information included in the financial statements was not always uniform (for example, as regards the entry of computer equipment in the balance sheet in the case of the ESC and the COR or the conversion into ecu of the value of the tangible assets in the case of the Council).`

This situation is abnormal, and it is therefore necessary for the wording and scope of Article 21 of the Implementing Rules and the obligations which arise from it to be strengthened, if need be, on the basis of interinstitutional consultations.

Ways towards reshaping the Financial Regulation

1.42. One of the first stages in reshaping the Financial Regulation must be to redefine the basic accounting framework. This means adopting a series of budgetary and accounting principles which reflect the juxtaposition of activities requiring cash-based accounting and those activities which are better reflected in accruals-based accounting. In fact, the major effort should be directed at incorporating, more and more, accruals-based accounting principles (Intosai acknowledges the benefits of this type of accounting, which has the advantage of forming a coherent and recognised framework).

1.43. Secondly, the Financial Regulation will have to define the accounting policies and practices which are a direct result of these principles. Any exceptions deemed necessary would have to be justified both technically and politically. The Implementing Rules would supply the provisions for the application of these rules, listing the details, if necessary, in subordinate texts (e.g. the internal rules concerning management of inventories).

1.44. This approach would enhance transparency, obviate the risks of a drift towards accounting by derogation and towards questionable practices and would lead to a proper solution to present accounting and financial management problems, such as the treatment of advances.

1.45. In respect of the financial statements, the Commission, as it does at present, is required every year to draw up a consolidated revenue and expenditure account showing the implementation of the budget, a consolidated statement of the assets and liabilities plus the notes attached to the financial statements.

1.46. However, the provisions contained in the financial regulations need to be strengthened: they should not only explicitly provide for the inclusion of the aforementioned notes, but also stipulate that the statement of assets and liabilities and the explanatory notes should be drawn up in accordance with the provisions of the Fourth Directive and with the international accounting standards. Similarly, as the Commission suggests in its proposal for an amendment to Article 70a of the Financial Regulation, it would be useful to provide for an interim statement between the consolidated revenue and expenditure account and the statement of assets and liabilities, showing the impact on the latter of the income and expenditure (for example, depreciation) which do not stem from the management of the budget. By presenting financial statements in accordance with these principles, the Commission should thus be able to provide high-quality financial information equivalent to that of a company of similar size.

1.47. The finishing touch to these provisions would be to make it obligatory for the Commission to send the draft consolidated financial statements to the Court of Auditors so that it could audit them before they are finally adopted. The Financial Regulation ought to make provision for this formal adoption of the definitive consolidated financial statements, and the adoption should be the responsibility of the Members of the Commission. Lastly, the fact that public management calls for transparency should mean that the Financial Regulation stipulates, as the Court has pointed out above, that the audited consolidated financial statements are to be published every year in the Official Journal of the European Communities.

1.48. If the rational approach described above were adopted, this would make it possible to overcome the absurd situation in which the institution entrusted with drawing up the regulatory framework for the private accountancy profession in Europe is not itself in a position to abide by generally accepted accounting principles.

2. THE SYSTEM OF APPROPRIATIONS

The distinction between administrative appropriations and operating appropriations

The principle underlying the distinction

2.1. For some considerable time, the section of the budget devoted to the Commission has been divided into a Part A, for the entry of the institution's administrative appropriations, and a Part B, for the entry of 'operational appropriations`, i.e. appropriations specially allocated to Community policy areas and, within each of these, to programmes and measures.

2.2. Only the Commission has this distinction between the two sorts of appropriations because it is the only institution that implements Community policies. The sections of the budget devoted to the other institutions therefore only contain administrative appropriations comparable to those in Part A of the Commission's section.

The blurring of the distinction

2.3. Although its principle is clear, this distinction has in fact been eroded with time. On the one hand, appropriations are to be found in Part A of the Commission's section which are, in fact, of the type of operating appropriations. All Title A3 (expenditure resulting from special functions carried out by the Institution) falls within this category, as does part of Chapter A-10 0 (provisional appropriations). In addition, a whole series of budget headings in Part A give rise to a distribution, admittedly under a small number of items, across large fields of operational activity. This sui generis classification of certain administrative resources does not even have the merit of corresponding to the headings used in the financial perspective, of which there are approximately the same number.

2.4. Conversely, part of the amounts entered under a large number of budget headings in Part B, although theoretically of an operational nature, are used for expenditure which is, in fact, administrative expenditure of the same type as entered in Part A of the budget. These are 'mini-budgets` which, for two years - 1991 and 1992 -, were even distinctly identifiable as such in a special subsection (B8). This concern for transparency had made the management of the budget so complicated that this special subsection of the budget was abandoned in 1993 but, contrary to what was frequently and erroneously asserted at the time, it was not a case of 'abolishing` the mini-budgets but of merging them once again with the corresponding operating headings, a return to a solution which was simpler but, it must be said, less transparent. For the PHARE, TACIS and MEDA programmes, a supplementary and amending budget for 1996 made explicit use of this opportunity within the limits of stipulated percentages. This solution, which is undoubtedly a little more transparent, does not, however, solve the problem of the mini-budgets.

2.5. The special case of the Joint Research Centre (JRC) is even more serious. In spite of the principle, which is still to be found in the Financial Regulation (Article 93(1)), whereby the budgetary nomenclature of the research appropriations must be based 'on the purpose of the expenditure in achieving research and technological development objectives`, i.e. as a genuine nomenclature for operating appropriations, the Commission, from one reform of the Financial Regulation to another and in spite of the repeated Court opinions on this subject (7), has virtually rendered this principle meaningless and defines the greater part of the appropriations by the type of means used and not by the research objectives. Thus in the budget for 1997, only 16,8 % of the JRC's payment appropriations are still defined 'by purpose`.

Recommended approaches

2.6. In the final analysis, the crucial question which must be asked about the mixture of functions served by Parts A and B of the budget is one of budgetary philosophy. Either one takes the view that a distinction needs to be made between operating appropriations, which are allocated for the purpose of achieving special political objectives, and administrative appropriations, which are more or less permanent means of action and may be redeployed for various policies and activities at any time, or one takes the view that, in the end, all means must serve clearly stated purposes and, in that case, all appropriations must be considered to be operating appropriations.

2.7. Starting initially from the first approach, the Communities have taken some steps towards the second without ever, however, openly opting for it. This explains why we are now in the ambiguous situation described above. Unless the Communities unequivocally decide to opt for the second approach, a clear definition of Parts A and B must be achieved, excluding any mixing of the two. It is the task of the Financial Regulation to define precisely this distinction, which is at present completely blurred.

Differentiated appropriations and non-differentiated appropriations

The basic concept underlying the system now in force

2.8. Although the principle of annuality is a fundamental principle of the Community's budgetary law, it is impossible to overlook the multiannual aspects of budgetary management. The adoption of a multiannual programme necessitates the immediate entering of the total expenditure that its implementation will incur parallel to the entering of the payments as and when they are made. The creation of differentiated appropriations allows, for the commitment appropriations, the immediate entry of the total foreseeable expenditure, and with payment appropriations, the entering of the actual expenditure. This formula allows, in a certain sense, the safeguarding, even at the level of commitment appropriations, of the principle of annuality since, at the close of the financial year, appropriations which have not been used, whether commitment or payment appropriations, are cancelled except in cases which are narrowly defined by the Financial Regulation (Article 7(2)). Multiannuality does not affect the appropriations themselves but the commitments.

2.9. For those operations which can be committed and paid in the same financial year, the principle of annuality can, it seems, be strictly applied and thus the concept of non-differentiated appropriations was introduced for them, with the same appropriation serving to authorise the commitment and subsequently the corresponding payment or payments in the course of the financial year in which the appropriation is entered.

2.10. Whatever formula is adopted, whether it be the differentiation or the non-differentiation of appropriations, the above arguments highlight the mechanism of the budgetary commitment. In its most usual sense, this serves two main functions:

(a) to provide for the total amount of an item of expenditure, whether or not it is spread out over several financial years, and to set out the timetable for the corresponding payments;

(b) to allocate a certain amount of appropriations for a set purpose by means of freezing them, making these appropriations unavailable for any other use.

The system's malfunctions

Annuality is even less often observed in the case of non-differentiated appropriations than in the case of differentiated appropriations

2.11. Non-differentiated appropriations, which, in theory, are more likely to obey the principle of annuality than differentiated appropriations, are, in fact, perhaps even less likely to do so in view of the complex mass of exceptions represented by the carryovers which prolong the life of these appropriations by a second financial year. Differentiated commitment appropriations which are not used in the course of the financial year in which they are entered may not be carried over except in very special circumstances where the commitment can only be entered into just before the end of the financial year (Article 7(2)(a)). In contrast, where a non-differentiated appropriation has not been committed in the course of the financial year in which it was entered, it may be carried over for much broader reasons: when the appropriations provided for the same purpose in the following financial year 'are not sufficient to cover requirements` (Article 7(1)(a)).

2.12. A similar paradox applies to payments. Differentiated payment appropriations may be carried over when the appropriations provided for the same purpose in the subsequent financial year 'do not cover requirements` (Article 7(2)(b)). This is already a broad enough justification, similar to the last case cited above. The possibilities for carrying over non-differentiated appropriations are even broader: on condition that they have been the subject of a commitment before 15 December of the financial year in which they are entered, appropriations which are not provisional or do not relate to salaries 'shall be carried over automatically`, i.e. systematically (Article 7(1)(b)).

Non-differentiated appropriations used as differentiated appropriations

2.13. Part A of the Commission's section and the other Institutions' sections only include non-differentiated appropriations entailing, in principle, a commitment in the financial year in which they are entered and payment during the subsequent financial year at the latest. However, amongst the operations financed from Part A of the Commission's section and from the other institutions' sections there are multiannual operations, sometimes of long duration, and very similar to the operations financed in Part B by differentiated appropriations. This concerns, in particular, the construction or acquisition of buildings. How can such an operation be 'committed` on the basis of appropriations which are, by nature, annual? How can the financial controller, pursuant to Article 38(1), confirm the availability of appropriations before commitment when, in theory, the necessary (multiannual) appropriations do not exist? In practice, the solution, which is irregular in terms of the Financial Regulation, is that these acquisitions may be initiated without budgetary authorisation and without a commitment proposal, i.e. without approval.

Differentiated appropriations used as non-differentiated appropriations

2.14. Conversely, although the Financial Regulation does not mention the Structural Funds, a financial regulation exists for them in the basic regulations on this subject. However, Article 20(2) of Council Regulation (EEC) No 4253/88 (8), as amended by Council Regulation (EEC) No 2082/93 of 20 July 1993 (9), which constitutes the regulation containing the provisions for the application of the basic regulations for all Structural Funds (10), stipulates that 'commitments in respect of operations to be carried out over a period of two or more years shall, as a general rule, (. . .) be effected in annual instalments`. A single commitment is envisaged only for operations to be carried out over a period of less than two years or where the assistance granted by the Community does not exceed ECU 40 million (Article 20(3) of the same Regulation).

2.15. This commitment by annual instalments is an anomaly and an oversight of the fact that, for multiannual operations to which differentiated appropriations are allocated, it is the payment appropriations that must ensure conformity with budgetary annuality and the regularity of the amounts spent from one financial year to another.

2.16. This false principle, which consists of fragmenting, year by year, even commitments within the framework of allocations of differentiated appropriations, undoubtedly arises from the philosophy of the financial perspective, the purpose of which is to limit and to define, but also to regulate over a period of seven years (1993-99) not only the flow of expenditure (appropriations for payments) but also the flow of commitments (appropriations for commitments). Moreover, in the financial perspective, the annual amounts are much more restrictive for the commitments than for the payments as only amounts of appropriations for commitments are detailed according to categories of policy and expenditure. It is a contradiction to enter 'en bloc`, for 1993, the basic texts and the amounts of appropriations which will be allocated to the Structural Funds for the whole period 1994-99 (6 years) and, at the same time, to distribute the commitments themselves over the whole of the period, with their annual amount even being entered as increasing from year to year. It would be more sensible if the Member States presented their programmes at the beginning of the period, which would be committed in their entirety on adoption, and this would, by contrast, mean that the annual amounts would decrease markedly year after year, with only the payments being spread in an even manner over the whole of the period. Only the evenness of the flow of payments demands, in fact, a matching evenness in the flow of revenue and only this latter evenness must be monitored in order to regulate the burden imposed on the European taxpayer.

2.17. The Commission, which is a co-signatory of the successive interinstitutional agreements on budgetary discipline, should recommend an approach of this kind to the two other co-signatory institutions. In parallel to this, it should propose that Article 20 of Regulation (EEC) No 4253/88 be repealed or radically amended, in order to abolish the commitment of Structural Fund operations by annual instalment payments.

The remedy: the differentiation of all appropriations

2.18. In view of these inconsistencies in the system of appropriations and also in the manner of using them, the Court recommends the amalgamation of the systems of appropriations which should all be differentiated. This general differentiation should not lead to a multiplication of derogations from the principle of annuality. In fact, for every entry under a budget heading in the form of commitment appropriations, there should be, in the remarks, a corresponding provisional schedule of payment appropriations. This solution would allow:

- the strict limitation of the schedule to the foreseeable and reasonable duration of the implementation of the assistance and, thus, of the payments. In many cases, the schedule of payments would not exceed one or two financial years as is the normal case at present in Part A of the budget, but without the possibility of a carryover,

- limits to be set on the life of appropriations since, after the first year of the schedule, any eventual 'amounts still to be paid` would be cancelled. The final dates of implementation in the contracts should never exceed the budgetary limits,

- the Budgetary Authority to fix, easily and every year, the amount budgeted for payment appropriations,

- the standardisation and clarification of the accounting for commitments, for payments and for 'amounts still to be paid`, and avoidance of the anomalies mentioned above in the use of non-differentiated and differentiated appropriations, in breach of their legal definition, and, therefore, many irregularities,

- the re-establishment of the principle of annuality in all its force, while making possible the implementation of multiannual operations.

Towards two or more levels of commitment

The broad outline of the system envisaged

2.19. The Commission's proposal aims at establishing a system of accounting on two levels in order to allow strict monitoring of entries of 'primary` or 'global` commitments against the appropriations available, 'secondary` or 'specific` commitments against amounts committed at the 'primary` level, and payments against secondary commitments. At each level, it would therefore be possible to monitor the amount of appropriations used at the lower level and the amount which remained available. Once the time-limits for implementation had run out, at one level or another, the amounts remaining available would be cancelled and this cancellation would be passed on from level to level up to the highest, i. e. up to the level of appropriations available, which would be annulled up to the amounts not used within the time-limits. This system results from an analysis of the proposed amendments to Articles 1(7), 7(6), 36(1), (2) and (3) of the Financial Regulation and Articles 52 and 53, first, fourth and fifth subparagraphs, of the Implementing Rules, which are also to be modified, according to the Commission's views.

2.20. The Court favours, in principle, such a proposal. In particular, it should guarantee that, once a commitment of a global nature has been contracted, the individual monitoring of the implementation of this commitment at the level of the award of individual contracts, and at the level of the implementation of payments for the latter will be effectively assured by the authorising officer, the financial controller and the accounting officer. It might even be desirable, in time, to have more than two levels of commitment in order to cover all the stages of the process by which the institution allocates the appropriations it has at its disposal in a more and more complex and irreversible manner. For example, a particular project financed within the framework of development aid necessitates a decision of principle by the Commission, then the signature of a financing agreement and then, eventually, the award of the contract.

The shortcomings of the existing and proposed provisions

Preliminary clarification

2.21. The Financial Regulation, even when amended according to the Commission's views, is extremely confused on the subject of commitments. There are, however, two fundamental distinctions that can be identified in these texts:

- firstly, the distinction between the legal act, which leads to expenditure from the budget (what Article 36 (1) calls 'measure`), and the act of budgetary procedure, which may correspond to it (commitment proposal issued by an authorising officer and approved by the financial controller). Let us refer to the act, or the measure, as the 'legal commitment` and the act of budgetary procedure as the 'budgetary commitment`,

- secondly, the distinction between commitments at the global level, a level at which the recipients of Community funds are not identified by name or with a definite amount, and commitments for the benefit of designated recipients, for a definite amount, within the budget set by the commitment at the global level. Let us refer to the first type as the 'global commitment` and the second as the 'specific commitment`.

2.22. The above two distinctions overlap, with the result that, in the end, there are four types of commitment:

- global legal commitments,

- global budgetary commitments,

- specific legal commitments,

- specific budgetary commitments.

If the Financial Regulation (and the Implementing Rules) and the amended texts envisaged by the Commission were to keep strictly to these four categories and state, every time they referred to commitments, which category they meant, some clarity would already have been restored to the matter. In this respect, the wording of Article 36 of the Financial Regulation, both in the present version and in the version proposed by the Commission, is unsatisfactory, as it is full of ambiguities on the nature, whether global or specific, legal or budgetary, of the commitments referred to in this Article.

The basic questions

2.23. Once this problem of terminology has been settled, the basic questions remain. The Court has identified six:

(a) whether an internal decision, entailing no undertaking towards an identifiable third party, has to give rise to a budgetary commitment;

(b) the sequence of legal commitments and the corresponding budgetary commitments;

(c) the possibility that a draft decision can have the force of a legal commitment;

(d) whether it is advisable or not to provide for several budgetary commitments, following one from another, corresponding to the various levels, global and specific, of the corresponding legal commitments;

(e) the generally most advisable level, global or specific, of budgetary commitment and, therefore, of the financial controller's approval;

(f) the entering of commitments in separate statements, or not, according to their level, whether global or specific.

2.24. The texts which are currently in force, or even if amended according to the Commission's views, give no unequivocal or acceptable answers to these questions, a confusion which is obscured but, in reality, exacerbated by the fact that the provisions determining the choice between the various foreseeable solutions are dispersed over various parts of the Financial Regulation and the Implementing Rules. There is a need to clarify carefully the answers to be given to these various questions.

Question (a): does an internal decision have to give rise to a budgetary commitment?

2.25. According to one interpretation of a commitment, only acts giving rise to a debt to a third person or likely to give rise to one in an irreversible manner constitute or entail a 'commitment`.

2.26. This interpretation is too restrictive. It is much more useful and effective to stipulate that every act, every decision, which provides for any budgetary expenditure must be treated as a legal commitment and, therefore, be accompanied by a budgetary commitment. The essential criterion is whether or not the act, even when of a purely internal character, means that a certain amount in terms of appropriations is to be reserved for its implementation and will thus become unavailable for all other uses. Seen from this point of view, an internal decision by the institution must be considered a commitment, in the same way as the signature to an agreement, a contract or the allocation of a subsidy.

Question (b): the sequence of legal commitments and the corresponding budgetary commitments

2.27. Article 36(1) states a principle of a general nature which is immediately contradicted by paragraph 2. The general principle is that budgetary commitments come before legal commitments. Article 36(1) is, in fact, the basic provision concerning commitments and deals with all the categories of commitments defined in paragraph 2.22 above.

2.28. Article 36(2) considers the case of decisions taken by the Commission (decisions of principle, therefore of global legal commitment), and Article 53 of the Implementing Rules, which refer to these same decisions of principle, would stipulate, if the Commission's draft is followed, that the 'prior approval` of the commitment proposal by the financial controller must be given 'when the institution has approved the draft decision`.

2.29. In reality, by means of these imprecise and contradictory provisions, the Commission intends to reassert the system whereby global budgetary commitments most often follow global legal commitments, whilst specific budgetary commitments, however, always precede specific legal commitments. In its explanatory memorandum for the amendment to the Financial Regulation, the Commission in fact shows by several examples that, in the majority of cases, global budgetary commitment does not occur until after the decision of principle by the institution, but nevertheless mentions the case of research, which gives rise to draft decisions by the institution accompanied by a commitment proposal 'which is generally, but not always, approved by the financial controller`.

2.30. In fact, the division of powers between the Commission and the authorising officers must be clarified, particularly in the light of the Commission's normal practice. Basically, it is the Commission, as an institution, that holds the powers to implement the budget (Article 205 of the EC Treaty), i.e. the powers attached to the functions of the authorising officer. However, the Commission may delegate the power to commit expenditure to its officials and always delegates the power to authorise expenditure (Internal Rules for the implementation of the budget, Article 3). If the Commission has not delegated its powers to commit expenditure, it must, before deciding, consult a number of departments. In particular, the financial controller must give his opinion (but not yet his approval) before the proposal is submitted to the Commission for decision (Internal Rules, Article 8). It is nevertheless clear that, in principle, budgetary commitment and the approval of the financial controller do not precede the Commission's decisions of principle, even when the latter would have a financial impact.

2.31. Article 53 of the Implementing Rules clearly summarises this system while applying it to all the institutions. In fact, it confirms that the prior approval of the financial controller is given 'when the institution has approved the draft decision`. The approval follows the decision, but remains 'prior` in the sense that there is no budgetary commitment until the approval is given.

2.32. This system is acceptable under certain conditions which the Financial Regulation must specify. Since, in the case of the institution's decision of principle, the financial controller gives his 'approval` after the decision has been taken but is nevertheless consulted on his opinion before this same decision, the Commission, in its proposal, has introduced a new distinction between the 'agreement` of the financial controller - given on the basis of a draft decision - and the 'approval` - given after the decision has been taken (the proposed Article 53 of the Implementing Rules). The consequences of this agreement have yet to be defined - under which conditions it must be given, and in what way it differs in this respect from the prior approval.

2.33. In addition, regarding the proposed reform which aims to create, in a more explicit manner, two levels of commitment, the articles of the Implementing Rules which concern global commitments (Articles 52 and 53) should, for reasons of greater clarity, be transferred to the Financial Regulation.

Question (c): does a draft decision constitute a commitment?

2.34. In view of what has been said above, the wording of Article 52 of the Implementing Rules is inexplicable and incorrect. This Article states that 'draft decisions of a general nature taken by the institution . . . shall constitute measures which may give rise to expenditure` and therefore, pursuant to Article 36(2) of the Financial Regulation, 'constitute commitments`. A draft decision is not a decision, is therefore not a 'measure` and does not therefore constitute a 'commitment`. This strange provision derives from the fact that, in certain cases, the financial controller is obliged, in advance, to give his approval to the draft decision, contrary to the generally applicable rule explained in paragraphs 2.30 and 2.31. It is clear that this advance action is conditional on, and subordinate to, the subsequent adoption of a decision conforming to the draft which he has approved. This anticipation of approval is obviously not intended to put the decision itself into effect before its adoption by the institution. The cause of this imperfect wording of Article 52 lies in interference and confusion between the decision procedure on the one hand and the budgetary commitment procedure on the other.

Question (d): a single budgetary commitment or a multi-stage commitment?

2.35. The question is whether the system of a single budgetary commitment should be retained, be it global or specific, or whether a double system of budgetary commitment should be introduced, according to a formula analogous to the one which was introduced under the Eighth EDF. The formula opted for by the Commission in its proposal remains that of the single budgetary commitment, but one which would apply sometimes at the global level and sometimes at the specific level.

2.36. However, this choice is not obvious from a reading of the proposal, which is insufficiently clear in this respect. Article 36(1), which is general in scope, would allow the conclusion that there is always a global budgetary commitment - which is not the case. The wording of paragraph 2 of the same Article leaves room for doubt as to whether the budgetary, or legal, nature of the Commission's decisions 'constitutes commitments of expenditure`. As for 'the legally binding commitments` mentioned in paragraph 1 of the Article, it is not apparent from reading it whether they are each accompanied by a corresponding budgetary commitment, which is not the case according to the opinion of the Commission. As for cases where the budgetary commitment would be entered into only at the specific level, the new proposed provisions do not mention this. There are, therefore, serious shortcomings in the wording of the provisions relating to budgetary commitments.

2.37. It may be said in favour of the choice of the single budgetary commitment that it is a simpler solution in respect of management. The establishment of a double level of budgetary commitment would further complicate the management. However, a balanced answer to this problem can only be given when taking the answers to the two following questions into consideration, and these answers lead clearly, as will be seen, to the recommendation of the - clearer - formula of double budgetary commitment.

Question (e): if it is a single commitment, is it preferable for the budgetary commitment to be at the global or specific level?

2.38. It is clear that, for the Commission, the most frequent form will be the global budgetary commitment, since the Commission has even omitted to discuss the alternative. On this assumption, the actual approval by the financial controller will be given only at the global level. The fact that both the proposal relating to the Financial Regulation (Article 36) and the draft amendment to the Implementing Rules (Articles 52 and, above all, 53) fail to mention any budgetary commitment procedure or any intervention by the financial controller at the specific level, is a clue, if not proof. But above all, the explanatory memorandum of the proposal to amend the Financial Regulation leaves no doubt: 'As the financial controller has already approved the primary commitment, there appears to be no need for him to act a second time when the individual legal commitments are made to guarantee the reliability of the accounts, especially as these commitments are then submitted to him as supporting documents for the payments` (paragraph 8.3, seventh subparagraph).

2.39. This solution and the argument which the Commission supplies for it are unacceptable, since, if such a solution were to be applied, there would no longer be any a priori control of contracts. That is the level, however, at which the risks of fraud and irregularities, or even failings in sound financial management, are most significant and not, obviously, at the level of the commitment of a programme, a stage at which the contractors or aid recipients are totally unknown. 'Retrospective correction` by means of checks on payments is not a satisfactory solution, firstly because payments are, generally, much more numerous than contracts, secondly because the prior nature of the checks would then be lost, and, lastly, because the legal commitment, even if irregular at internal level, would nevertheless continue to be binding on the Community. The Court does indeed suggest the abolition of all prior approvals (paragraphs 5.5 and 5.12), but only if linked to the implementation of a serious alternative solution to the latter, which should be studied very carefully.

Question (f): the exhaustivity of the accounting of commitments

2.40. In the event of multi-stage (global and then specific) budgetary commitments, distinct and separate statements would need to be drawn up for global and specific commitments, just as distinct as the statements of commitments and the statements of payments are at present. At each level the accounts would show the full amounts of appropriations used and the balance available.

2.41. The system, applied by the Commission, of a single budgetary commitment, whether global or specific, raises questions about the methods for drawing up the accounts of the use of appropriations and the revenue and expenditure account. The Commission does not propose to change the present system at all in this respect: the accountancy for commitments would total up all the commitments without distinction, whether global or specific. Insofar as the Commission specifies and improves the accounting for budgetary operations, this is not done horizontally by the level of use of the appropriations, but vertically, by a system of entering payments against a specific commitment in a number of stages; from specific commitments to the global commitment; from the latter to the allocations of appropriations to the appropriate budget heading. However, as the booking of the specific commitments against the global budgetary commitment would 'be registered by the authorising officer in the central accounts` (Article 36(2), second indent; see also Article 53, fourth indent, of the Implementing Rules) and as, in certain cases, the booking of the specific commitment would have to be made directly against the allocation to the budget heading in the absence of a global budgetary commitment, it is hard to see how the central accounts could avoid systematically creating the two levels for recording commitments. If the level of global commitment was, in certain cases, purely and simply 'left out`, paradoxical situations could arise where the total of specific commitments would be greater than total global commitments. The accounting system to be used would have to be defined in precise detail and the computer system adapted accordingly.

3. TOWARDS AN ACCRUALS-BASED ACCOUNTING SYSTEM

The recording of assets and the drawing-up of the balance sheet

3.1. Within the context of the development of the Community accounting system towards an accruals basis, the way in which assets and liabilities are recorded is of considerable significance. It is particularly difficult to record fixed assets owing to the complexity of the task of taking stock of them, their vast number, their geographical distribution, the various methods used for their acquisition and the lack of any interinstitutional harmonisation enabling consolidated accounts to be drawn up. The entry in the accounts of assets makes it necessary to introduce into the Financial Regulation procedures guaranteeing that the recording of all the items of property and movements affecting them is reliable. Improvements could be made to the current system (Articles 65 to 68).

3.2. The procedures for collecting information and for entering data in the property inventory system should be uniform and should to this end be governed by an interinstitutional agreement, which is not the case under the current arrangements. The procedures should interlink the systems, making it possible to trace an item of property from its purchase and its entry in the inventory up to its removal from the inventory, and at the same time guaranteeing the appropriate transcription of these transactions in the financial statements and thereby ensuring that the transactions concerned may be audited. The responsibility of the parties involved should be specified in the legislation, both that of the accounting officer and that of the administrative departments responsible for collecting and processing the information.

The valuation of assets

3.3. It is apparent from reading the Community financial statements, in particular the explanatory notes accompanying the consolidated balance sheet (11), that tangible fixed assets and other assets are entered in the balance sheet at their purchase value. The counterpart entry appears on the liabilities side, under the 'own capital` item, which changes from one financial year to the next, in line with the assets to which it corresponds.

3.4. Insofar as the cost of purchasing movable and immovable property is included in the budgetary expenditure of the financial year of purchase and does not give rise to any subsequent depreciation, some degree of confusion may result from this accounting method, aggravated by the fact that revaluations of land and buildings are regularly booked to the consolidated balance sheet. The Commission has attempted to overcome these drawbacks and to move closer to the objectives of a classic accruals-based system of accounting by introducing, in Article 70a of the Financial Regulation and in Articles 131a and 131b of the Implementing Rules, the possibility of entering the depreciation of assets in the accounts by writing down their value or by making provisions, operations which are entered solely in the balance sheet and thus have no impact on the budget. This initiative is useful provided that the ensuing mechanisms are clearly specified, i.e. the methods which are applied for writing down the assets' value or for making provisions, and their calculation and accounting bases. The possibility of revaluing certain assets has already been applied in the past (12) although the Financial Regulation makes no provision therefor. The proposed reform would formalise this practice (Article 131a of the Implementing Rules), but with insufficient precautions, since such revaluations usually run counter to the accounting principle of prudence.

The accounting treatment of advances and payments by instalment

3.5. When examining the Community financial statements and the financial regulations governing their drawing-up, the Court has on several occasions drawn the Commission's attention to the need to make a clear distinction between the various legal natures of the payments made.

3.6. Payment is an action which presupposes for its beneficiary a counterpart represented either by a service rendered (work, services, supplies) in accordance with a contract, or by a document justifying the eligibility of the expenditure, in the case of subsidies. Non-repayable payments do not therefore create any debt.

3.7. In contrast to payment, an advance is a payment made either on a provisional basis (repayable advance, for example Article 46(2)), or prior to the existence of any counterpart, the latter being expected subsequently. An advance thus has the effect of creating for the benefit of the Community a debt of equal amount which will not disappear until the advance has been repaid, the contractual service required has been performed or proof has been provided, in the form of supporting documents, that the expenditure has become eligible. Unlike the payment, the advance therefore has the effect of creating a debt recorded in the balance sheet. It follows that an advance, which is not an item of budgetary expenditure, may not be booked as such. That is why Article 71 of the Financial Regulation stipulates that 'any advance shall be entered in a suspense account` prior to settlement.

3.8. Then there is the notion of payment by instalment. The Financial Regulation refers to it (Article 46(1)) but does not define it. It is, however, accepted that an instalment, usually a limited proportion of the overall amount payable, most of which will be effected subsequently on the basis of the requisite supporting documents, is itself regarded as equivalent to a payment despite the fact that when the instalment is paid, its counterpart does not yet exist. It is thus usual in a works contract for a payment by instalment - for example, 10 % of the contractual amount - to be made upon signature of the contract; this instalment is intended to cover the contractor's immediate cash requirements, in particular the cost of setting up the construction site. It is traditional for the instalment to be thus classified in terms of both the accounts and the budget as a genuine payment; this can be justified by the fact that it involves a small amount, it is paid at the very outset and its counterpart is expected to be imminent. Like all other payments, the instalment is directly booked as budgetary expenditure and thus does not create any debt.

3.9. If these basic definitions are accepted, there are many gaps or weaknesses in the financial legislation as regards advances and payments by instalment. The first task would be to include in the Financial Regulation a clear definition of these two terms, which are in fact regularly interchanged.

3.10. The second, more difficult, task would be to specify the cases in which payments by instalment may be effected and the cases in which advances may be made. Given how widespread the practice of paying huge advances has become in the Communities' budgetary management, it is regrettable that the Financial Regulation tackles this question under the rather secondary aspect of advances to personnel (Article 46(2)). Three observations should be made in this respect.

3.11. Firstly, just because an item of expenditure is subject to subsequent clearance (in particular in the case of the EAGGF Guarantee Section) or may, by sampling, form the subject of external audits which may give rise to adjustments to the amounts approved and paid (as in the case of shared-cost research contracts), this does not mean that there are grounds for denying that these payments, which run some risk of being revised, constitute real budgetary expenditure. If that line of argument were followed, virtually all budgetary expenditure would thus have to be treated as advances to be settled, since an excess payment could give rise to recovery of the undue payment or be booked as a deduction from a subsequent payment.

3.12. The second observation concerns the designation and the processing of the sums paid to the Member States within the framework of the common agricultural policy. The Member States prefinance the expenditure and the Commission reimburses them on the basis of the monthly statements of expenditure which they submit. These sums therefore constitute expenditure, and the term 'advances`, as used by Article 99 of the Financial Regulation, should be deleted, since it was a term which was warranted only when the Community prefinanced the expenditure.

3.13. The third observation concerns operational expenditure, such as that of the Structural Funds. The accounting principle used for the preparation and the presentation of the consolidated accounts of the European Union (13) mentions the term 'payments on account` while the basic regulations governing these funds only use the term 'advances`. The terminology should be harmonised, using the term which best corresponds to reality.

3.14. These cases are a good illustration of the ambiguous nature of the texts concerned and their internal contradictions. The amendments proposed in this area do not make any improvement.

The treatment of interest

3.15. The implementation of a large number of Community measures involves the payment of advances to beneficiaries. It is often the case that the funds advanced are not immediately used but remain on deposit in bank accounts where they earn interest. As mentioned earlier, the Court has on several occasions (14) drawn the Commission's attention to the variety of methods used to deal with interest for the different types of measures by the various departments. The fact is that the Financial Regulation and the Implementing Rules fail to provide satisfactory answers as to how interest should be treated in the various cases.

3.16. The existing or proposed provisions do not offer any general solution, as they deal only piecemeal with specific cases:

- the proposed Article 22(4)(a) deals only with interest generated by advances paid to subcontractors,

- Article 49 refers only to the payment of interest in the case of the recovery of an undue payment of Community aid. Not only does the question of recovery of undue payments arise in many other areas than that of Community aid, but also, even in this area, the case of the recovery of undue payments is far from being the only one that gives rise to the generation of interest,

- Article 111(4) deals only with the interest received on the external aid transit accounts.

3.17. The fundamental problems have therefore been overlooked or merely touched upon. These problems are mainly as follows:

(a) does the ownership of interest depend upon the account holder?

(b) does it depend on the ownership of the funds paid?

(c) how should the interest which has been received be allocated?

In the Financial Regulation the question of account ownership is dealt with only in Article 111, that of the ownership of the funds in Article 49, that of allocation of interest in Articles 22(4)(a) and 111. Furthermore, the solutions adopted by these articles do not always appear to be appropriate.

3.18. The main question for the Court is not that of the account holder but rather that of the ownership of the funds paid. It thus matters little whether the Commission transfers the funds intended for the implementation of a programme to an account in its name or pays an advance to an account held by the beneficiary country - Member State or non-Member State: as long as the funds paid are still owned by the Commission, it should also receive the interest earned. The ownership of any benefits arising follows that of the principal.

3.19. This situation could only change if the funds were paid directly to the aid recipient, supposing that aid is involved. Even in this case, a distinction would probably have to be made between aid which is widely spread over a large number of recipients, each of whom receives only a small amount, and larger amounts of aid. In the first case it is understandable that the fact that these funds assume the nature of an advance for a brief period may be disregarded and that the Commission may waive very small and scattered amounts of interest, as suggested, moreover, by Article 95 of the Implementing Rules.

3.20. As regards the allocation of interest, when it is considered, on the basis of the previous rules, to be owned by a third party, the way it is allocated does not concern the Community. The interest earned by funds still owned by the Commission should, for the sake of the principle of budgetary universality, be classed as general budgetary revenue, without being allocated. The practice followed in certain programmes, such as PHARE, whereby the interest earned by the funds paid to them by the Commission is allocated to the programmes managed by the subcontractors, and the proposed Article 22(4)(a), which confirms this practice, are thus incorrect. The same is true of Article 111, which, moreover, uses the vague term 'project`; this warrants further explanation: does this just mean projects or does the term also cover more extensive measures such as programmes? In general, Article 111 is not precise about the ownership of deposited funds and on the link between this ownership and the allocation of the interest earned on these funds.

4. AWARDING OF CONTRACTS

The role of the Advisory Committee on Procurements and Contracts

4.1. If the successive increases in the threshold amounts at and above which the contract award procedures change are examined, it can be seen that the size of these increases, when compared with inflation, has by and large been acceptable. This is not the case, however, for one of the most important thresholds, if not the most important, that of the amount at and above which the opinion of the Advisory Committee on Procurements and Contracts (ACPC) is required. According to the Commission's draft concerning the Implementing Rules, this threshold increased by 900 % (a tenfold increase) between 1973 and 1996 (year of the draft), whereas inflation for the same period was only just under 400 % (an increase of slightly less than fivefold). Since the 1993 review, the Commission's wish would mean an increase of + 185,7 %, compared with a cumulative inflation figure of just + 8,8 % over the same period.

4.2. The fact that it is much easier for the Commission to amend the Implementing Rules than the Financial Regulation thus enables it to reject the useful provisions that limit its freedom of management. That is why consideration should be given to the question of whether at least the threshold at which the ACPC must be consulted should not be reincorporated into the Financial Regulation.

The case of external aid

4.3. Title IX of the Financial Regulation contains special provisions applicable to external aid which stipulate that the normal provisions of Titles I to VI, Title XI and Part III of the Financial Regulation apply to external aid, except where the derogations laid down in Title IX apply. This means that the provisions applicable to each operation in the field of external aid need to be checked. Unfortunately, the wording of Title IX is ambiguous in some cases, which makes determining the provisions applicable difficult on occasions.

4.4. The provisions of Title IX concerning the awarding of contracts (Section III, Articles 112 to 119) are particularly ambiguous. For example, Article 113 is worded: 'the procedure to be followed for the award of works, supply or service contracts shall be specified in the financing agreement or the contract, subject to the following principles`. It does not state explicitly in which articles these principles may be found, given that Articles 114 to 119 all contain principles. It can thus be concluded that the provisions of Articles 114 to 119 apply to the procedures for the awarding of contracts in cases where a financing agreement or a contract has been signed beforehand, under the conditions stipulated in Article 106(1). This must therefore be interpreted as meaning that in the absence of a prior financing agreement or contract, as defined in Article 106(1), the normal procedures for the awarding of contracts set out in Title IV of the Financial Regulation apply.

4.5. Based on this reading of the Financial Regulation, contracts awarded directly by the Commission, in the absence of a prior financing agreement or contract specifying the procedures to be followed, are subject to the provisions of Articles 56 to 64(a) of the Financial Regulation, including the obligation to submit any contract valued at over ECU 42 000 to the ACPC. The Commission does not follow this procedure, however: instead it follows the provisions of Article 119, which it takes in isolation, i.e. it regards the latter as independent of Article 113, and considers that only service contracts awarded in the interests of the Commission (this concept is not defined in the regulations and may be interpreted in many different ways) are subject to the rules laid down in Title IV. The Commission then applies the provisions concerning the use of restricted invitations to tender and the awarding of contracts by private treaty (Article 118) to all contracts awarded in the field of external aid, whether a prior financing agreement or contract exists or not.

4.6. In short, in the field of external aid, there is a large volume of contracts for which the procedures followed by the Commission could be deemed irregular. This unsatisfactory situation must be remedied, but the proposed amendments to Articles 112 and 113 are not conducive to this.

4.7. More generally, Title IX must be revised thoroughly to take account of the new nature of the Community's external aid programmes. The current text, adopted in 1990, is greatly influenced by the Mediterranean Protocol procedures. These, however, are now being replaced by the different procedures being applied under the MEDA programme. Likewise, the development of the procedures for the PHARE and TACIS programmes, or for the MED programme concerning the networks of projects between organisations in Europe and the Mediterranean countries, calls for a complete overhaul of Title IX, so that a framework can be set up that is geared to the new situation.

Misuse of subcontracting

4.8. The explanatory memorandum for the proposal asserts that, as far as subcontracting is concerned, 'rules must be devised to ensure that the accounts for subcontracted operations are perfectly transparent`.

4.9. In fact, the provisions that the Commission wishes to see incorporated for this purpose into the Financial Regulation (Article 22(4)(a)) and into the Implementing Rules (Article 5(b)), assuming that the desired transparency would thereby be slightly improved, have the drawback that they deal with matters that extend far beyond cases of subcontracting and that should have been settled within a much more general framework (treatment of interest generated by funds entrusted to third persons, clear identification of administrative costs, clauses to be stipulated in the contracts, accounting policies and recovery of undue payments).

4.10. Above all, the proposal does not examine the field of application of subcontracting and the restrictions that might be put on this, and even goes as far as to assert that 'subcontracting is a fact of life which cannot be called into question per se`. This is the most debatable point of the existing Regulation and of that proposed.

4.11. The argument invoked to justify subcontracting - i.e. 'for some Commission departments which are short of staff, it is the only way to carry out certain policies` - is in itself already questionable in cases where the problem could be solved by redeploying staff.

4.12. The crux of the matter does not lie there, however. The essential point, which is not mentioned in the existing Regulation or the version amended according to the Commission's views, is the obligation which the Treaty (Article 205) places on the Commission alone to 'implement the budget on its own responsibility`.

4.13. Admittedly, this obligation does not prohibit the Commission from entrusting certain subordinate implementation, study or technical-assistance tasks to consultants by means of service contracts. Such contracts have always existed, and the Court of Auditors has not systematically objected to this practice. The practice becomes questionable, however, when, by virtue of such contracts, the Commission divests itself of its public authority powers, which, by their very nature, cannot be entrusted to third persons, and hence not to subcontractors.

4.14. This practice has reached dangerous and reprehensible proportions in certain Community policy areas in recent years, particularly in the case of the PHARE and TACIS programmes for aid to Central and Eastern Europe, the newly independent States and Mongolia. The same is also true of the MED programmes to assist Mediterranean countries. Under the framework of these measures, of whose political importance everyone is aware, there are cases of subcontracting which are unauthorised, sometimes because they are decided on the basis of irregular procedures or sometimes, indeed primarily, because they relate to essential aspects of management, such as the organisation of invitations to tender, including the drafting of the related documents, notably the terms of reference of contracts, and, more generally, to the Commission's discretionary powers, which, according to the Court of Justice, are types of powers that may not be delegated (Meroni judgment of 13 June 1958). Furthermore, there are also cases of subcontracts being subcontracted, where the initial contractor has entrusted part of his contractual tasks to other experts. Conflicts of interests have been observed in many cases, even to the point where some consultancies replied to invitations to tenders that they themselves had drawn up, or where some consultancies' representatives were on the administrative board of the body responsible for overseeing the performance of the contracts awarded to the very same consultancies.

4.15. Amongst the consequences of these cases of misuse of procedures, the following should be noted: the indirect manner in which the departments in Brussels, and even the Commission delegations on the spot, become aware of the problems and what has been implemented, since the staff rely increasingly on countless reports sent to them by the technical assistants; the tendency for the consultants contributing to the implementation of certain programmes to form themselves, as a group, into a 'socio-professional category` sui generis, within which there is an increasing number of cases of consultants competing and at the same time colluding with each other; lastly, the existence of cases where some firms are charged with implementing certain measures and with taking part in controlling them as well.

4.16. It is therefore imperative to incorporate a provision into the Financial Regulation that delivers a clear and strong message to the effect that the Institutions, and especially the Commission, are forbidden to subdelegate powers that belong preeminently to the public authority. It would be expedient to reiterate also that recourse, albeit legitimate, to authorised agents may in no way diminish the Commission's institutional powers as enshrined in the Treaties.

4.17. Admittedly, the concept of 'responsibilities` or 'powers of the public authority` would still have to be precisely defined. The Court is aware that only time will enable the boundaries of this concept to be specified. However, even if it were still not possible to define it in detail, this would not constitute an excuse for not immediately incorporating the principle into the Financial Regulation. The fact of incorporating it straight away would give rise to an in-depth legal examination, which would gradually determine the implications.

5. THE RENEWAL OF THE CONTROL

The respective roles of the various officials responsible for control

Introduction

5.1. The procedures for managing and controlling revenue and expenditure are applied in three stages: authorisation, implementation and a posteriori control. The Court has repeatedly drawn attention, notably in its Opinion No 1/89 (15), to the unsatisfactory nature of these procedures and the imbalance that exists between the authorising officers, the financial controller and the accounting officer. The situation still needs to be greatly improved and warrants a major review of the definition of their functions. This review should be geared towards moving the Community away from the spending culture that has so often been criticised in the Court's reports, which have, for example, highlighted the large proportion of appropriations committed at the end of the budgetary year, and redirecting it towards a management culture that thinks in terms of objectives and which is concerned more with the quality of expenditure than with its volume.

Departures from the present system

5.2. The reasons prompting criticism of the current situation are numerous and have hardly altered since 1989. The authorising officer must be primarily responsible for the quality, regularity and effectiveness of the measures undertaken by the Community. The authorising officer's role and responsibilities are not accorded the importance they merit, however, and the former may thus tend to rely on the involvement of the financial controller, who, by giving his prior approval, attests to the legality and regularity of the proposed measures. The Financial Regulation specifies in abundant detail, in fact, all of the points that the financial controller must ascertain have been strictly observed before granting his approval, whereas it is the authorising officer who clearly must first have complied with these points himself before issuing a commitment proposal or a payment order. However, the Financial Regulation, paradoxically, makes almost no reference to the authorising officer's obligations in this regard.

5.3. Where an authorising officer is opposed to the withholding of approval by the financial controller, he is entitled to have such a refusal overruled by the institution's supreme authority. In its proposal, the Commission suggests shortening the time-limits for the overruling procedure by supplementing the provisions of Article 39 of the Financial Regulation. This addition consists in requiring the supreme authority to take its decision before 15 February of the year n + 1, the time-limit for closing the accounts for the year n.

5.4. This might well create several problems, however, the most serious of which is the retroactive nature of the decision to overrule that has been incorporated into subparagraph 3 of the proposed Article 39. For example, the authorising officer may be tempted to continue to enter into commitments after the decision to withhold approval has been taken in the expectation that this refusal will be overruled. He might hope that commitments that are obviously affected by irregularity or illegality will then be validated.

5.5. The systematic practice of obtaining prior approval is no longer adhered to. Because management is becoming increasingly complex and the number of transactions is rising every year, the Commission has resorted to making do, as regards payment orders, with approval being granted on the basis of samples, a practice which does not comply with the Financial Regulation as it stands. This new practice is unsatisfactory, as it wavers halfway between the two possible systems. It does admittedly allow the control work to be speeded up, but renders prior control fictitious. As for completely abolishing the requirement of prior approval, this would not be any more acceptable unless it were to be replaced by other procedures that provided at least equivalent safeguards and that could also involve financial officials other than the financial controller.

5.6. As stated in the abovementioned 1989 Opinion, the impact of financial control is uneven: it is most effective when appropriations directly managed by the Commission (e. g. administrative expenditure) are involved, but weak when it comes to operating expenditure that is decentralised down to the level of the Member States. The financial controller does not grant approval to payments to final recipients within systems involving management by the Member States, but merely to the overall payment to the Member State's administration. For this reason, the development of agreements being signed with financial control and internal audit bodies in the Member States constitutes an innovation whose implementation will need to be monitored.

5.7. Lastly, the Commission's proposal to extend the financial controller's remit would not necessarily guarantee an improvement in the situation. Even if it seems pertinent to have the financial controller involved in analysing the accounting, inventory and financial-management systems used by the authorising officers (proposal to amend Article 24 of the Financial Regulation) and to submit the analysis of the financial management to him (proposal to amend Article 79 of the Financial Regulation), his true powers in this area are not clearly defined. Unless his role is defined clearly, the 'consultation` and 'submission` referred to in Articles 24 and 79 respectively do not involve any new power or new competence as regards the information involved. Moreover, in its study of the Community's financial systems published in 1981 (16), the Court has already pointed out the latent conflict existing between the power to grant approval, and thus authorise expenditure, and the responsibility for subsequently evaluating the same expenditure in accordance with the criteria for sound financial management.

5.8. The Commission proposal would confer the task of carrying out the Institution's internal audit on the financial controller (proposal to amend Article 24 of the Financial Regulation). This is an ambitious proposal, but its consequences should be defined more clearly. If it were to be adopted it would result in the de facto ending of the prior approval procedure. Furthermore, the text does not specify which approach the financial controller should favour; the wording of the current Financial Regulation favours an approach directed towards evaluating the internal control. Internal audit is a wider concept, however, in particular because its powers include, inter alia, the independent evaluation of whether the internal control systems are appropriate and effective.

5.9. Since the accounting officer operates at the end of the chain of control, he, like the authorising officer, is likely to rely unduly on the financial controller's prior approval and not to exercise all of his control prerogatives, thus exacerbating the imbalance referred to earlier. The Commission has, admittedly, endeavoured to involve the accounting officer in control activities to a greater degree by proposing that Article 25 of the Financial Regulation should be supplemented with the requirement that the accounting officer be consulted beforehand when the accounting systems of financial management used by the authorising officers are set up and altered. The criticisms that can be levelled at this proposal are the same as those that apply to the proposal analysed above concerning the strengthening of the financial controller's role. Unless a clear definition is given of the accounting officer's role, which appears to be one of straightforward 'consultation`, devoid of either new competence or new power, then it will not be possible to hold him responsible for any weaknesses that he has detected.

A new definition of the role of the financial officials

5.10. Faced with this situation, the Court has a number of recommendations to make, some of which have already been put forward in the 1989 Opinion. These focus on a radical reform of the financial controller's role and on making the accounting officer and authorising officer more accountable.

The authorising officer

5.11. The first reform would involve placing greater emphasis on the powers of the authorising officer, who, as already stated above, would be the person primarily responsible for the quality, regularity and efficiency of management. The texts should state clearly the precise extent of his powers: first and foremost, to ascertain the impact of the measures envisaged and the quality of the financial management, but also to carefully assess expenditure and adhere to the budgetary limits, to check the legality and regularity of the measures proposed, particularly to take great care to avoid any risk of partiality on the part of his staff in respect of the tasks they have to deal with and, where appropriate, to penalise, without fail, any infringements in this area.

The financial controller

5.12. On the express condition that the authorising officer's role is strengthened in the manner described above, the financial controller's duties could develop into those of a real internal auditor. Full discharge of this function would presuppose the financial controller's independence, which could be guaranteed only if he were to report to the highest level of the institution, i.e. to its President. This independence could have nothing but a beneficial effect on the impact of his reports and the influence that he could exert on systems and procedures. His traditional financial controller role, involving a priori control of measures proposed by authorising officers, would be greatly altered. As a precaution, the financial controller could apply the prior approval procedure. However, he would have to determine the cases in which it would be expedient to exercise this right and, this being so, would himself select which operations needed to be subject to prior approval. He could thus target the operations he chose to select on those areas most open to the various risks referred to further on. To offset this, the a posteriori audit of the measures carried out and the continuous audit of the systems and procedures would need to be strengthened and guided by the systematic use of techniques for analysing risks, including those of irregularity, fraud, error and mismanagement. Each audit would result in a formal report being issued and communicated to all interested parties. The change in the role of the financial controller should, furthermore, be used as an opportunity to give consideration to a new way of organising the control functions and could lead to the work carried out by the various parties involved, in particular the Inspectorate General, being rationalised.

5.13. The financial controller would also have to verify the quality of the financial data and financial statements and, in so doing, would cover the entire financial field, starting with the commitment procedure and ending with the presentation of the financial statements. His task of notifying the institution of serious anomalies of which he has become aware would continue in its present form (Article 40(1) of the Implementing Rules). It should not, however, move in the direction proposed by the Commission, which seeks to grant the financial controller prerogatives to initiate disciplinary proceedings, which should not be within his remit. Moreover, the Staff Regulations are the only possible framework for such a reform. The Court of Auditors, in its capacity as external auditor, would be notified of the financial controller's work programmes and audit reports and would have access to his working files. As for the role of the financial controller cum internal auditor in validating Community operations carried out at Member State level, consideration of this matter at both political and technical level needs to move in the direction of developing the cooperation procedures in the area of audit (17). This will enable a chain of coordinated and consistent audits to be set up throughout the procedure for implementing Community action.

The accounting officer

5.14. Lastly, the accounting officer's powers should be enhanced. He should first and foremost play a central role in compiling and generating accounting, financial and management information and in its dissemination to the authorising officers. In order to enable the latter to manage their operational activities as well as possible, the abovementioned information should also cover the resources made available to them under Part A of the budget.

5.15. Article 25 of the Financial Regulation assigns to the accounting officer the task of drawing up the financial statements. So that he may do this satisfactorily, he is not subject to any particular obligation in this regard, other than that of guaranteeing that the financial statements are presented objectively (Article 16 of the Implementing Rules). Yet objectivity is only one of the financial statements' qualities. After defining the objectives of the financial statements, it would be more appropriate to list all of the qualities that must be evident, e. g. those set out by Intosai (they must be understandable, relevant, reliable, material, comparable, timely, etc.) (18). Even if the Commission has proposed extending the accounting officer's role in various areas, his role with regard to generating and presenting accounting and financial data has yet to be strengthened.

5.16. The accounting officer should be empowered to check that the formalities stipulated in the Financial Regulation have been adhered to before any payment is made and any operation is entered in the accounts. This would undoubtedly help to reduce the proportion of payments for which there are no supporting documents. The 1989 Opinion contained a similarly oriented proposal that provided for reducing the period of time within which an Institution had to take a decision on the final discharge to be given to the accounting officer to six months from the date of submission of the revenue and expenditure account, rather than allowing the period of two years currently permitted in Article 77 of the Financial Regulation. This step would prove particularly useful within this renewed framework in that it would enable the discharge authority to have all the necessary data at its disposal before taking the decision to grant the discharge relating to the implementation of the budget.

The Court of Auditors' rights of access

5.17. Some specific regulations, notably in the fields of the common agricultural policy and traditional own resources, refer to the Court's rights of access to information in restrictive terms, suggesting that these rights are more limited than those of the Commission. At present, the Financial Regulation does not state clearly that the Court of Auditors' powers of audit are exactly equal in extent to those of the Commission; in the new revised version it should.

Compensation of financial damage suffered by the Community

5.18. Financial damage suffered by the Community arises from a number of sources. Sometimes a shortcoming, negligence or a mistake on the part of one or more Community staff gives rise to a financial loss or a loss of income, but it is not possible to consider any third party to be answerable to the Community for the amount in question. Sometimes the error committed by a department consists in making an undue payment to a recipient who must as a result be regarded as a debtor to the Community in respect of the amount which he was not entitled to receive.

5.19. In this respect, the Financial Regulation has a serious shortcoming in that it fails to state explicitly the principle whereby any sum paid in error must be reimbursed by the person who received it (i.e. the principle of 'recovery of undue payments`) and therefore be the subject of a systematic recovery procedure. A principle of this kind would constitute an expedient addition to Article 28(1), the basic Article on which the entire budgetary system of revenue is constructed.

The liability to disciplinary action and to payment of compensation of each Institution's financial officials

5.20. The financial and the disciplinary aspects of the question of liability of financial officials need to be dealt with separately. The financial officials' liability to payment of compensation consists in a duty, in the event of an error on their part, to repay the financial damage suffered by the Community out of their own pocket, whether or not there is a third-party debtor (see paragraph 5.18). Provision for this form of liability already exists in Title V of the Financial Regulation. But it would be expedient to spell out in this Title exactly how serious the fault that has caused the loss must be in order to warrant the person who committed the error being made personally liable for it; this Title should also specify the period of time that should apply before time-barring comes into effect and the administration is thus prohibited from holding the person liable. Furthermore, Title V refers for the most part to the Staff Regulations applicable to officials and other servants of the European Communities. It is clear that the latter Regulations are not equipped to deal with the question of the liability to payment of compensation of authorising officers, financial controllers and accounting officers and that the existing rules, based on Article 209 of the EC Treaty which makes provision for such liability, are therefore highly unsatisfactory.

5.21. What is more, these rules, implementation of which has been left to the institution concerned, have to date hardly ever been applied. For this reason, the institution to which the official in question belongs ought to refer the matter to an independent body: a special body for budgetary and financial discipline could be set up which, provided that there was the possibility of bringing an appeal before the Court of Justice, could give a ruling on the financial officials' liability to payment of compensation.

5.22. There remains the question of these same financial officials' liability to disciplinary action, which is also dealt with by Title V of the Financial Regulation. With regard to this matter, the following steps should be taken (19): a detailed definition of the substance and scope of the powers of the authorising officers, financial controllers and accounting officers should be established, provision should be made for obligations requiring the institutions to check how these individual powers have been assumed, and a strengthened procedure for penalising these financial officials should be set up. Obviously, the proposals made here should apply mutatis mutandis to administrators of imprest accounts.

5.23. In its proposal, the Commission makes provision for penalising officials who have acted as authorising officers when not duly authorised to do so. The proposed Article 22(4) refers to Title V for this purpose. The latter, however, deals only with the liability of staff who are empowered to act as authorising officers, financial controllers or accounting officers and the text as it stands cannot therefore be invoked to penalise staff who are not empowered to act in this capacity. It is therefore necessary either to incorporate the concept of de facto authorising officers into Title V, or, preferably, to make provision in Article 22 whereby, as far as their liability to payment of compensation and to disciplinary action is concerned, de facto authorising officers are to be treated as actual authorising officers for the purpose of Title V.

Overcoming the weaknesses in adapting to computerisation

Introduction

5.24. The introduction of computer systems affects not only the way in which managerial tasks are performed by the institutions, but also the way in which the external audit of Community finances is carried out and, consequently, the manner in which the Court plans and carries out its audit work. Moreover, the Financial Regulation and the Implementing Rules are not equal to the demands of this development. Even outside any computer context (see paragraph 5.17), the definition of the Court of Auditors' rights of access to information is more restricted than that of the Commission. Furthermore, the effect of computerising management tasks has been to reduce further the concrete scope for the Court to exercise its rights to their fullest extent.

Access to supporting documents

5.25. Of the questions that concern the Court more particularly, access to supporting documents is especially important. Various provisions should be revised in this area. In particular, identification of and access to supporting documents that have actually been used as a basis for approving accounting transactions, which are dealt with in Article 23 of the Financial Regulation, must be facilitated by appropriate provisions and, at the very least, must not be made more difficult by technical developments, one consequence of which is that some of these documents are recorded on computer, instead of on paper.

5.26. For the Court, given the fact of computerisation, its main concern is no longer the transmission of supporting documents on a systematic basis, as provided for in Article 84, but rather obtaining guaranteed comprehensive access, whether on the spot or by means of electronic transmission, to all of the documents stored in computer memories. This right, which would put into operation the general right of access to information provided for in Article 188c(3), subparagraph 2, of the EC Treaty, is currently not stated explicitly in Articles 83 to 87 of the Financial Regulation.

5.27. Lastly, Article 21 of the Implementing Rules, which lays down the conditions for the approval of accounting practices, does not specify the rules for its concrete application in the context of computerisation.

Notes

(1*) Translator's Note: Although the French text is identical for both paragraphs, the English version of Article 7(2)(b) differs slightly from that of Article 7(1)(a): 'the appropriations provided for the headings concerned in the budget for the following financial year do not cover requirements.`

(2**) Translator's Note: Since the French text quoted here is that of the original Article 99, as it appears in the 1977 Financial Regulation, the corresponding English version is used too.

(3) OJ C 10, 15.1.1996.

(4) The Fourth, Seventh and Eighth Council Directives on:

- the annual accounts of certain types of companies (Fourth Directive - OJ L 222, 14.8.1978),

- consolidated accounts (Seventh Directive - OJ L 193, 18.7.1983),

- the approval of persons responsible for carrying out the statutory audits of accounting documents (Eigth Directive - OJ L 126, 12.5.1984).

(5) International Organisation of Supreme Audit Institutions.

(6) Special Report in support of the Statement of Assurance concerning activities financed from the general budget for the financial year 1995 (OJ C 395, 31.12.1996), paragraphs 3.55 to 3.56.

(7) Opinion No 1/89 of 9 February 1989, paragraph 8.1. (OJ C 72, 20.3.1989, p. 7) and Opinion No 4/94 of 8 December 1994, paragraph 3.2.3, (OJ C 383, 31.12.1994, p. 18).

(8) OJ L 374, 31.12.1988, p. 1.

(9) OJ L 193, 31.7.1993, p. 20.

(10) Council Regulation (EEC) No 2052/88 of 24 June 1988 (OJ L 185, 15.7.1988, p. 9), as amended by Council Regulation (EEC) No 2081/93 of 20 July 1993 (OJ L 193, 31.7.1993, p. 5).

(11) Revenue and expenditure account and balance sheet relating to operations under the 1996 budget of the European Union (Volume IV), Introductory comment to Explanatory Note A, page 51 (document SEC(97) 403 EN).

(12) Revenue and expenditure account and balance sheet relating to operations under the 1996 budget of the European Union (Volume IV), Explanatory Note A1, page 51 (document SEC(97) 403 EN).

(13) Revenue and expenditure account and balance sheet relating to operations under the 1996 budget of the European Union (Volume IV), accounting principle 2.5, page 8 (document SEC(97) 403 EN).

(14) Most recently in the Annual Report concerning the financial year 1995 (Chapter 11, paragraphs 11.58 to 11.67) published in OJ C 340, 12.11.1996.

(15) Paragraph 4.2 of the European Court of Auditors' Opinion No 1/89 (OJ C 72, 20.03.1989).

(16) OJ C 342, 31.12.1981, p. 1.

(17) Subject to respecting the recommendation made in the Court's Annual Report concerning the financial year 1995 on the necessity to verify the quality of the work performed by third persons, paragraph 0.27 (OJ C 340, 12.11.1996).

(18) The characteristics cited in brackets have been taken from the communication entitled 'Qualitative characteristics of government financial reports`, presented by Intosai's Accounting Standards Committee at the XVth Incosai (International Congress of Supreme Audit Institutions) in September 1995.

(19) This paragraph is based on paragraph 3.7 of the 'Report to the reflection group on the functioning of the Treaty on European Union` presented by the Court in May 1995 as part of the preparation for the Intergovernmental Conference.