31997Y0221(01)

SPECIAL REPORT No 1/97 on the Commission Decisions of 10 April 1996 and 20 November 1996 on the clearance of accounts 1992 and certain expenditure for 1993, accompanied by the Commission's reply (97/C 52/01)

CONTENTS

Paragraph Page

1. INTRODUCTION . 1.1-1.5 2

2. METHODOLOGY OF THE COMMISSION . 2.1-2.9 2

Staff resources . 2.1-2.2 2

Risk analysis . 2.3-2.5 2

Execution of the work . 2.6 3

Working methods . 2.7-2.9 3

3. CONCILIATION BODY . 3.1-3.7 4

4. EVALUATION OF THE AMOUNTS DISALLOWED . 4.1-4.25 4

General remarks . 4.1-4.4 4

Flat-rate corrections . 4.5-4.11 5

The overruling of the Financial Controller's Decision: beef public storage in Ireland . 4.12-4.21 7

Ad hoc corrections . 4.22-4.25 7

5. CASES WHERE NO CORRECTIONS WERE APPLIED . 5.1 8

6. CONCLUSION . 6.1-6.3 8

ANNEX: TABLES 1-4 . 10

The Commission's reply . 12

1. INTRODUCTION

1.1. Following the resolution of the European Parliament of 17 March 1995 that decisions by the Commission on the clearance of the accounts of the European Agricultural Guidance and Guarantee Fund, Guarantee Section be the subject of the discharge procedure, the President of the European Parliament requested the European Court of Auditors by letter of 26 June 1996 to produce a report on the Commission Decision of 10 April 1996 concerning the clearance of the accounts of the 1992 expenditure and part of the 1993 expenditure (1).

1.2. This decision made ECU 825,1 million of corrections (see Table 1) and included two individual corrections concerning 1993 expenditure (ECU 106,9 million). Owing to the fact that part of the 1992 expenditure was still subject to the conciliation procedure as at April 1996 (see paragraph 3.6), the Commission also took a second decision concerning the 1992 clearance. This second decision, which made ECU 70,0 million of corrections (see Table 2) and amended that of 10 April 1996, was taken on 20 November 1996 (2).

1.3. Article 5 (2) (b) of Council Regulation (EEC) No 729/70 (3) requires the Commission to clear the annual accounts submitted by the Member States of payments made by them under the Guarantee Section of the EAGGF. The aim is to ensure that only expenditure legally incurred in conformity with Community rules is financed. To achieve this, the Commission department responsible (DG VI) carries out audits of the expenditure, using generally accepted auditing standards. Any errors or weaknesses detected are notified to the Member States in writing. The Member States have the opportunity to reply. Bilateral meetings are also held between the services of the Commission and the Member States to enable the parties concerned to clarify their respective positions. After this bilateral phase, the Commission notifies its conclusions, which opens the way for the conciliation procedure (see paragraphs 3.1-3.7). The Commission then finalizes its summary report (4) (5), which is presented to the EAGGF Committee of Member States' representatives. The final clearance decision is taken by the Commission as a college.

1.4. Significant changes have been made to the clearance of accounts procedure by Council Regulation (EC) No 1287/95 (6), which amends Council Regulation (EEC) No 729/70 (7). These changes will only come into effect for the EAGGF year beginning 16 October 1995.

1.5. Both Decisions mentioned in paragraph 1.2 are the subject of this report, which is the second one produced by the Court on the clearance of the accounts of a particular financial year (8). To produce it the summary reports of the Commission were examined, together with its audit mission reports and correspondence with Member States. The work of Financial Control (DG XX) was reviewed and discussions held with representatives of the conciliation body. No audit visits were made to the Member States by the Court.

2. METHODOLOGY OF THE COMMISSION

Staff resources

2.1. The Commission service responsible for the 1992 clearance work continued to be constrained by the human resources available. A maximum of 40 staff, allocated by markets, were available (of which 19 were at 'A` grade) to express an opinion on a total of more than ECU 30 000 million of expenditure incurred in 12 Member States. Even though the work programme submitted to the EAGGF Committee (9) stated that the number of staff available was insufficient to ensure coverage of all significant measures, there was no redeployment of staff to this task.

2.2. The European Parliament, in its resolution giving discharge to the Commission in respect of EAGGF clearance of accounts for the 1991 financial year (10), agreed with the Commission's recognition of the need for more staff for clearance of accounts work and called for the redeployment of 15 extra staff or, if this were not possible, to come forward to the budgetary authority with an appropriate proposal. To date no action has been taken to do this.

Risk analysis

2.3. The lack of sufficient staff meant that risk analysis was a vital consideration for allocating these scarce resources. The risk analysis was first carried out for each budget line by market and not at a central level. The priorities were set, and the work planned, according to the staff resources per market.

2.4. Moreover the Court considers that:

(a) export refunds, represented a third of the budget and is an area which the Commission's analysis recognized as being a high risk. For six of the Member States with the largest amounts of export refunds (DK, D, F, IT, NL, UK) the audits undertaken were in the context of physical controls and/or prefinancement and did not cover the supporting documentation in the paying agency;

(b) in Italy, only olive oil and sunflower seeds were audited in the oils and fats sector. Aid for soya, considered by the Commission to be as risky as both olive oil consumption aid and production aid, had more expenditure (ECU 428 million compared with ECU 368 million and ECU 219 million respectively) and yet was not audited. Moreover this budget line was not audited in the context of the clearance of accounts of any year in any Member State during the period 1988 to 1992;

(c) in Germany and the United Kingdom, only dried fodder was audited out of the oils and fats measures, yet expenditure on this was low and risk assessed as low or average. Peas and beans (D, UK) and rape seed (D) were far more deserving of scarce audit resources, as based on the results of the risk analysis.

2.5. The risk analysis was not put into practice in the milk sector because the available resources were engaged in following up the consequences of the Council Decision on the milk quotas in Italy, Spain and Greece. There was only limited coverage of other measures in the milk sector and no audit plan identifying priorities for the audit of 1992 expenditure was prepared.

Execution of the work

2.6. For various reasons, the work programmed was not carried out to the extent necessary:

(a) in a mission to Belgium only 64 of the planned 129 substantive tests were exhaustively carried out. In addition a particular aspect was examined in a further 60 cases;

(b) in Ireland, it was impossible during the first mission to carry out a satisfactory examination of beef export refund files because of the disarray in the supporting documents. A second mission took place in September 1993, inter alia, to test the corresponding files, but, due to the continuing disarray and the lack of time, this was not done. The resulting report concluded that, as the supporting documents were not in good enough order to allow the Commission's auditors to complete the work, Ireland received an unequal, more favourable treatment when clearing the accounts. A further mission in October 1994 concerning the prefinancing regime noted some improvement, but did not carry out the substantive tests. No corrections have been made in this context;

(c) the audit plan for oils and fats to Italy foresaw a mission for dried fodder which did not take place in the context of the 1992 clearance;

(d) in the oils and fats sector a systems audit in Denmark concluded that substantive tests on actual payments needed to be carried out in order to obtain assurance as to the regularity of the expenditure. No such tests have taken place;

(e) a mission report to Greece concerning the 1991 financial year identified the need for further work on olive oil consumption aid because of cases of fraud detected by the national Olive Oil Control Agency. No follow-up was made until 1996 and as a result there was no audit for this measure in Greece in connection with 1992;

(f) a financial correction of DM 5,8 million (ECU 3,1 million) was made on beef export refunds in Germany. The amount of the correction finally decided was based on figures provided in a report from the German authorities. Although the Commission service responsible planned to check the basis for these figures on-the-spot, this was not done.

Working methods

2.7. A significant delay (18 months) sometimes occurred between missions and the report conveying the results to the Member State. This undermined the Commission's position in applying corrections, as was noted by the conciliation body (cases FR/009, IT/010, UK/011 and IR/013: beef public storage).

2.8. Staff in charge of markets, except for that of export refunds, do not follow up systematically all potential corrections, including replies from Member States. A correction for casein in the Netherlands, already agreed by the Dutch authorities, would have been forgotten had the Court not brought this to the Commission's attention in January 1996.

2.9. There is no minimum limit for corrections. The Commission's 1992 Decision contains at least three corrections of less than ECU 1 and one of less than ECU 0,05. Such corrections are not material. However, in the case of the correction for olive oil production aid in Spain, Financial Control, after checking the correction, found that an extra Pta 3,3 million (ECU 20 000) needed to be disallowed. Although DG VI agreed with this analysis, in view of the 'small` amount involved and the fact that the procedure for the renotification of the Member State would be so cumbersome, it did not alter the correction.

3. CONCILIATION BODY

3.1. Commission Decision 94/442/EC (11) created a conciliation procedure which was available to Member States for the first time in the context of the clearance of accounts of the financial year 1992.

3.2. The conciliation body is an independent body, made up of four members and a President, who are experts in EAGGF Guarantee matters. They are appointed by the Commission, after consulting the EAGGF Committee.

3.3. The aim of this body is to reconcile the diverging positions of the Member State concerned and the Commission before the Commission takes its eventual decision on the correction to be made. The body produces a report for each case. However, the Commission is not bound by its conclusions.

3.4. A request for conciliation is admissible only where the financial correction recommended by DG VI either exceeds ECU 0,5 million, or represents more than 25 % of the Member State's total annual expenditure for that budget line.

3.5. In relation to financial year 1992, 34 cases were submitted to the body, (see Table 3). Three cases (two Greek, one Danish) were not considered because the Commission had already removed the correction. A further three were not considered because the cases were not submitted within the time limit set out in Commission Decision 94/442/EC (12) (two Italian, one Portugese). The results of the 28 cases actually considered by the body are detailed in Table 4.

3.6. The existence of the conciliation body has brought extra discipline to the work of the Commission's auditors as it obliges them to analyse their findings more thoroughly. However, the conciliation procedure has caused additional delay to the clearance procedure. The supplementary decision regarding 1992, taken in November 1996, was, in the view of the Commission, necessary, because 13 cases were still before the body in March 1996. Three of these cases (ECU 17,2 million) could still not be finalized by that time and will be cleared as part of the 1993 clearance decision.

3.7. A further aim of the creation of the conciliation body is to reduce the number of cases referred to the European Court of Justice. By the end of 1996 it is too early to assess whether this has been achieved.

4. EVALUATION OF THE AMOUNTS DISALLOWED

General remarks

4.1. The clearance of the 1992 expenditure included two corrections concerning 1993 expenditure (ECU 106,9 million). It would seem more appropriate to deal with these two corrections within the 1993 clearance procedure.

4.2. In its opinion No 6/95 on the Commission's clearance decision for the financial year 1991 (13) the Court stated that priority should be given to audit methods which provide an objective basis for a correction, such as extrapolations of results found by statistical sampling methods.

4.3. The Commission replied that it had introduced advanced computer tools, which would, inter alia, allow statistically valid sampling techniques. However, there is little evidence of this in respect of the work carried out to support the 1992 clearance decisions.

4.4. The Commission's clearance of 1992 expenditure led to three general types of correction: flat-rate corrections, ad hoc corrections (a refusal to finance certain expenditure) and routine accounting corrections.

Flat-rate corrections

4.5. As in respect of 1991, many corrections were applied using flat-rate corrections. Out of a total of ECU 895,1 million (see Tables 1 and 2), ECU 336 million (37,5 %) were flat-rate corrections in the 1992 clearance decisions.

4.6. The Commission has established an internal grid leading to flat-rate corrections of 2 %, 5 % or 10 %, depending on the seriousness of the systems weakness identified. In the view of the Commission remedial action - if effective - by the Member State concerned may only be taken into account in cases where there is doubt as to which of the three categories (2 %, 5 % or 10 %) to apply. Higher rates may be applied in exceptional circumstances. An internal document, agreed by the EAGGF Committee, governs the use of these flat-rate corrections (14).

4.7. Recourse to the flat-rate corrections is necessary in cases where the actual financial impact of systems weaknesses cannot be established. However, the flat-rate corrections were applied inconsistently, in particular the rewarding of subsequent improvements, which were sometimes not even tested. Thus, a correction was reduced or even withdrawn if the Member State concerned could show that remedial action had been taken to correct any identified weaknesses, even if such remedial action had taken place long after the year under examination.

4.8. This practice fails to take account of the conclusions of the European Parliament in its resolution A4-0058/96 giving discharge to the Commission in respect of the EAGGF clearance of accounts for the 1991 financial year (15). This resolution deplored the practice of assessing the rates of financial corrections on the basis of criteria other than objective evidence of the actual loss borne by the Community budget in the year under review.

4.9. In the following cases the Court's audit revealed that the flat-rate corrections were not based on the loss, or assessed risk of loss, to the EAGGF budget in the year under examination:

(a) Olive oil production aid: Greece

The implementation of this measure has been repeatedly criticized by the Commission (e.g. 1990 and 1991 clearance) and the catalogue of fundamental weaknesses were found to persist in 1992. The 10 % flat-rate correction imposed (Dr 5 252 million, ECU 17,2 million see Table 2) is the maximum normally allowed under the Commission internal grid. However, the summary report (16) states that the Commission may reduce this percentage under a future clearance procedure if the recently introduced changes in management and control constitute a significant response to the need to protect Community funds against the risk of irregularity and fraud. Such a reduction should not be made until the required improvements are in place, and the assessed risk to the EAGGF budget is reduced.

(b) Cotton production aid: Greece

The opinion on the 1991 clearance decision (17) describes how the 25 % flat-rate correction for cotton in Greece could be reduced to 10 % if the Greek authorities strengthened the control system and re-established cooperation with the Commission by 31 December 1995. The same 25 % correction was originally proposed in relation to 1992. However, the eventual decision for 1992 (18) reduced the 1992 correction to 10 %. The 1991 correction was similarly revised downwards, thus leading to a positive correction in 1992 to reimburse the 'overcorrection` in 1991. The reason given for this was that the enquiries undertaken in 1995 and 1996 confirmed that, as from the 1996/97 campaign, the control system is working correctly and that there has been a good collaboration between the Greek authorities and the Commission. In particular this is the case with regard to the conducting of ex-post controls of the marketing years concerned. The Court's auditors found evidence in the Commission's reports that improvements were only effectively in place from 1993/94 and that the 1992/93 campaign was carried out almost without any effective controls. The combined effect of these reductions is Dr 29 405 million (ECU 96,2 million).

(c) Beef public storage:

(i) Germany

The flat-rate correction was limited to 2 % (DM 19,59 million, ECU 10,1 million: see Table 2) largely because the German authorities had taken measures after 1992 to improve the deficiencies in their application of the system.

(ii) Italy

The clearance audit work for this measure was carried out in 1990 and 1991 in Italy. The Commission limited the corrections to 10 % of the expenditure for 1990 and 1991 (total Lit 54 927 million, ECU 28,4 million) 'in view of the Italian authorities efforts`. No correction has been made for 1992, even though the improvements made in Italy (which commenced at the earliest in 1993) were considered by the Commission to be unsatisfactory.

(iii) France

The Commission limited the flat-rate correction to 5 % (FF 189,4 million, ECU 29,2 million) because of the efforts of the French authorities to improve their buying-in and monitoring procedures. The plan to implement such improvements was only submitted to the Commission in March 1994.

(iv) United Kingdom

The Commission originally limited the correction to 5 % because of the efforts made subsequently by the UK authorities to improve monitoring procedures. This was eventually reduced to 2 % (£ 4,98 million, ECU 6,1 million) following a reexamination of the British control system. This correction was applied to 1991 expenditure only. No correction was applied to 1992 expenditure, even though the audit missions took place during EAGGF year 1992 and the weaknesses discovered existed in both years. Moreover, the United Kingdom authorities only undertook a radical overhaul of their system after discussions with the Commission at the end of financial year 1992.

(d) Withdrawal of peaches: Greece

The mission reports for this measure describe significant systems' weaknesses and a 10 % flat-rate correction was at first proposed for 1992, 1993 and 1994. However a letter from the Commission to Greece announced that no correction was proposed for 1992 or 1993 because missions to Greece in August 1995 had revealed improvements in the Greek system. The letter states, however, that serious weaknesses still persist. Thus, no correction was applied for 1992, even if the budget was exposed to serious risks in that year.

(e) Grubbing-up of vines: Greece

The Commission originally proposed a correction of 10 % but finally imposed a 2 % (Dr 61,1 million, ECU 0,2 million: see Table 2) flat-rate correction, after considering the conciliation body's report. However:

(i) any improvements (which in any case the Commission deems to be inadequate) took place well after 1992;

(ii) the vine register, which is the fundamental means by which this measure can be controlled and which the Member State was obliged to maintain, is still not in existence.

4.10. There were also cases where flat-rate corrections were not necessary as it was feasible to determine the exact loss to the Fund or the exact degree of non-compliance with a Regulation. For instance, in the case of production refunds for starch and for sugar in Germany, the Commission applied a 5 % flat-rate financial correction in both cases (DM 5,2 million, ECU 2,7 million). The case concerns certain firms submitting applications for refund certificates after the processing of the starch (or sugar), contrary to the provisions of the Regulations. However, it would have been more accurate to determine the amount of production refunds actually affected in this way.

4.11. So that Member States accept flat-rate corrections, there must be coherence and transparency in their application:

(a) in the case of production refunds for starch and for sugar in Germany and notwithstanding the facts referred to in paragraph 4.10, the 5 % flat-rate correction was applied simply because such a rate had been applied in previous years, (before the introduction of the criteria established by VI/216/93 Rev. 1 (19)) without considering whether it was still appropriate, in the light of these criteria;

(b) for the beef special premium, the systems in Ireland and Spain contained fundamental weaknesses, particularly in Ireland, where weaknesses included the inability to determine the precise numbers of animals claimed and unjustified retagging of animals. Yet the flat-rate corrections applied were 5 % for Spain (ECU 0,96 million) and only 2 % for Ireland (ECU 0,91 million). Moreover, in view of the seriousness of the weaknesses both cases merited a higher correction.

The overruling of the Financial Controller's Decision: beef public storage in Ireland

4.12. Between 1990 and 1993 Commission services investigated public storage of beef and veal in Ireland. The Irish authorities did not contest the extent of the deficiencies found, but stressed that they took immediate steps to remedy the situation as early as 1991.

4.13. DG VI wrote to Ireland stating that a 5 % flat-rate correction would be imposed for 1990 and 1991 only. This commitment was made without consulting the Financial Controller, thus undermining the position of the Commission in cases such as this where the Financial Controller takes a different view.

4.14. In this case, the Financial Controller considered that the weaknesses discovered were of a fundamental nature and merited a 10 % correction for both 1990 and 1991. The Irish authorities were notified accordingly. Ireland submitted this case to the conciliation body, as did Italy, France and the United Kingdom who had also received increased corrections for this measure. The conciliation body stressed that it would favour a differentiation of the rate of correction in line with the year from which improvements had been made. However by letter of 13 November 1995 DG VI informed the Irish authorities that it could not see any scope for reducing the corrections.

4.15. The Irish authorities considered the 10 % correction for both 1990 and 1991 to be financially and politically unacceptable. They set about demonstrating how they had improved their system from 1991 onwards.

4.16. These improvements were never tested by the Commission and there is conflicting evidence concerning the implementation dates of these improvements. The Commission documents indicate that nothing of substance was carried out before early autumn 1991, which is part of EAGGF year 1992. The Irish authorities vouched that their Control Enquiry Teams had carried out 21 unannounced inspections during EAGGF year 1991. This was the system improvement for that year.

4.17. The final decision on the clearance of accounts submitted for the approval of the Financial Controller did differentiate between 1990 (10 % correction) and 1991 (5 % correction). This reduction represented £ Irl 24 million (ECU 30,3 million). No differentiation was made for the other three Member States mentioned in paragraph 4.14.

4.18. This differentiation led directly to the Financial Controller withholding his approval, citing the need to verify the improvements which were supposed to have taken place.

4.19. By its unpublished Decision of 10 April 1996 (C(95) 2786, final) the Commission overruled the Financial Controller's decision to withhold approval for the clearance of the 1992 accounts.

4.20. The Commission took the decision to overrule the Financial Controller, citing the Irish authorities' 'willingness to significantly improve its control system` and how they had 'started to put into place the measures necessary to remedy the deficiencies found`.

4.21. The principle of recognizing improvements in systems is justifiable, but any reduction in flat-rate corrections can only take effect from the date of such improvements. Furthermore, there are no corrections at all proposed for 1992, although it is clear from a mission report relating to 1992 expenditure that there were still significant deficiencies in that year.

Ad hoc corrections

4.22. The Court is concerned by the manner in which certain ad hoc corrections were arrived at.

4.23. The correction for late payments of the beef special premium in Italy was reduced by Lit 2 855 million (ECU 1,5 million) on the basis of further documentation from the Member State. However, this documentation did not justify the reduction.

4.24. The Commission had detected an error rate of 10,53 % in two Nomos in Greece for the ewe premium for 1992. In its reply to the opinion on the 1991 clearance decision (20), the Commission stated that the 1992 correction would be made in that year's decision. However, the Commission has only applied a 5 % (Dr 221,5 million, ECU 0,73 million: see Table 1) correction for the two Nomos.

4.25. For the cereals co-responsibility levy in Denmark the Commission, using figures sent by Denmark to Eurostat, calculated that Denmark needed to pay the levy on an extra 710 000 tonnes. However, the Member State contested these figures (calculated in accordance with the same method used and accepted for all Member States) and provided other data to reduce the deficit to 206 000 tonnes. The Commission accepted the revised figures without question. Furthermore the Danish authorities informed the Commission that standing crops are not subject to the levy (based on a discussion of experts in 1988 for which there exists no evidence) and demanded that a further 147 840 tonnes be reduced. The actual correction only relates to 58 160 tonnes which makes a difference of Dkr 49,3 million (ECU 6,7 million).

5. CASES WHERE NO CORRECTIONS WERE APPLIED

5.1. There were cases where, in spite of evidence of weaknesses in Member States' systems which put the EAGGF budget at risk, no correction was made in respect of 1992.

(a) Significant work was carried out by the Commission into the prefinancing of beef and cereals export refunds. The many weaknesses resulting from these enquiries will only be taken into account in the 1993 clearance decision, according to the Commission. No correction has been made on the 1992 expenditure, even though the weaknesses also affected that year.

(b) An enquiry on the application of physical controls at export also revealed many deficiencies. The Commission proposed no corrections in respect of 1992, although it reserves its position regarding subsequent years if recommended corrective measures are not applied. It must be emphasized that:

(i) the Member States had to comply with the relevant Regulations from 1 October 1990 and that, as the audits carried out in all Member States in 1992 and 1993 found the application of these Regulations to be unsatisfactory, a financial correction should have been made;

(ii) in the domain of export refunds the proportion of EAGGF funds at risk and their absolute value vary significantly from Member State to Member State. To make no financial corrections at all amounts to unequal treatment of Member States;

(iii) DG VI wrote to Member States in early 1994 stating that no correction would be made if corrective action were taken by 1 July 1994. Financial Control was not consulted on this commitment, which has a significant impact on the budget.

(c) In Belgium a Commission audit mission concerning export refunds revealed payment delays, which would normally lead to a correction. In this case no correction was made.

(d) Both the mission report and the letter to the national authorities in the case of Germany beef special premium conclude that the very serious weaknesses undoubtedly led to an increased and undue charge on the Fund. However, no corrections were applied. No justification for the non-application of penalties was provided, except that 1992 was the last year in which Germany administered the scheme in this way.

(e) In Limburg (NL) 500 of the 800 ewe premium claims were lost. It was not considered useful therefore to conduct a file examination. No corrections were imposed. It is not acceptable that adequate tests could not be carried out.

(f) The 1992 accounts of 19 of the 44 paying agencies were audited by private audit firms. Their main tasks were to reconcile the annual declarations of expenditure to the underlying accounting records and to review the internal organization of the paying agency. The private firms provided useful system descriptions and identified serious deficiencies in respect of 1992. However, no financial corrections were made.

6. CONCLUSION

6.1. Article 5 (2) (b) of Council Regulation (EEC) No 729/70 (21) states that the accounts of a particular year must be cleared by the end of the following year. Thus, as for the 1991 decision, the 1992 clearance decision was three years late.

6.2. The introduction of the conciliation body may be considered as a positive development. It should be emphasized that the body's existence has imposed a greater discipline on the work of the Commission's auditors. However, it caused some additional delay to the clearance procedure in respect of 1992 (see paragraph 3.6).

6.3. Further improvements could be achieved by:

(a) redeploying the Commission's staff in order to allocate the necessary staff resources to the audit work (see paragraphs 2.1-2.2);

(b) applying a centrally performed risk analysis in order to allocate staff to key areas (see paragraph 2.3);

(c) applying flat-rate corrections only when it is not possible to ascertain the actual loss to the EAGGF budget. Their use should be transparent and equitable (see paragraphs 4.10-4.11);

(d) ensuring that there is no financial incentive to carry out what a Member State was already obliged to do under existing legislation (see paragraphs 4.7-4.9 and 4.12-4.21).

This report was adopted by the Court of Auditors in Luxembourg at its meeting of 9 January 1997.

For the Court of Auditors

Bernhard FRIEDMANN

President

(1) Commission Decision 96/311/EC (OJ No L 117, 14. 5. 1996, p. 19).

(2) Commission Decision 96/701/EC (OJ No L 323, 13. 12. 1996).

(3) OJ No L 94, 28. 4. 1970.

(4) VI/6355/95 of 27 March 1996.

(5) VI/5112/96 of 14 June 1996.

(6) OJ No L 125, 8. 6. 1995.

(7) OJ No L 94, 28. 4. 1970.

(8) OJ No C 10, 15. 1. 1996.

(9) VI/193/93.

(10) OJ No C 141, 13. 5. 1996, p. 107, paragraph 7.

(11) OJ No L 182, 16. 7. 1994.

(12) OJ No C 10, 15. 1. 1996.

(13) Doc. No VI/216/93 Rev. 1, 12. 9. 1994.

(14) OJ No C 141, 13. 5. 1996, p. 107, paragraph 7.

(15) VI/6355/95 of 27 March 1996.

(16) OJ No C 10, 15. 1. 1996.

(17) Commission Decision 96/701/EC (OJ No L 323, 13. 12. 1996).

(18) Doc. No VI/216/93 Rev. 1, 12. 9. 1994.

(19) OJ No C 10, 15. 1. 1996.

(20) OJ No L 94, 28. 4. 1970.

ANNEX

>TABLE>

>TABLE>

>TABLE>

>TABLE>

THE COMMISSION'S REPLY

1. INTRODUCTION

1.1-1.5. The Court's report concerns the decision on the clearance of the 1992 financial year, and some expenditure relating to the 1993 year. The Commission performed the major part of the work underlying this decision between March 1993 and March 1994. Certain enquiries required additional work and missions on the spot in later months. The Commission's working methods have continued to be developed and refined since 1994, and, as the Court notes in paragraph 1.4, significant changes to the procedure were proposed by the Commission in 1994 which, after consultation of Parliament and adoption by Council of the necessary legal bases, came into force in the 1996 financial year.

The Court's comments concern essentially the financial consequences decided by the Commission following its audit of the 1992 year. Corrections are only one aspect of the clearance procedure, if the most visible to the public. In addition to the ex-post audits which in some, but not all, cases give rise to corrections, the Commission undertook many ex-ante missions aimed at recommending improvements to the Member States' procedures before payment took place, and thus to avoid control failures, to reduce the risk of losses to Community funds, and hence to avoid the reasons for correction of the expenditure. For example, a considerable effort was deployed from 1993 onwards to encourage Member States to put the integrated control system into place properly, and to monitor the procedures in place for verifying the claims introduced under the reform of the CAP.

The Court's report concentrates more on the amounts of the corrections, and less on the other aspects of the clearance work, such as preventive audits, or the often extensive work undertaken in order to determine the amount at which the correction should be established, or the issues discussed during the conciliation procedure.

2. METHODOLOGY OF THE COMMISSION

Staff resources

2.2. Since 17 April 1996, the date of the European Parliament's resolution regarding 15 extra posts, DG VI has allocated three additional posts to that task, being its entire increase of staff posts for the 1996 year. The Commission is seeking a solution to the need for additional staff and will raise the matter in its negotiations in respect of the 1998 budget.

There is a general lack of resources for the Commission's controls in the Member States. In addition to staff for the clearance procedure, the Commission lacks staff for the corps of inspectors in the wine sector and the fruit and vegetable sector. These two corps are responsible for monitoring the controls over quality and labelling, regarding both commercial activities in the two sectors and produce presented for Community subsidies. Each corps has at present a staff of two, and each needs at least three more if any meaningful programme of work is to undertaken. These requirements are not included in the abovementioned 15 posts.

Risk analysis

2.3. The Court refers to only one element considered during the allocation of staff and planning of the Commission's work, being the weighting of the expenditure on each budget line by the evaluated risks of the measure and of each Member State's administrative procedures. The final plan was established at central level, and took account of this analysis, but also many other elements such as the need to undertake preventive audits in new areas of expenditure, and of the obligation to have a reasonable coverage of the many areas at risk if the Commission's clearance decision on the year's expenditure as a whole is to be meaningful. The plan as adopted contained a reasonable balance between these considerations. One objective of the reform of the clearance procedure, referred to in paragraph 1.4 of the Court's report, is to allow the Commission to concentrate on fewer enquiries at any one time, as its enquiries will no longer need to be linked to an annual clearance decision. This reform did not come into effect until the 1996 financial year, so all work plans up until the 1995 year were established on the same basis as for 1992.

2.4. (a) As is explained in paragraph 4.2.1 of the summary report, it was fully justified to give priority to the verification of the application of Regulation (EEC) No 386/90 and of the prefinancing regime, as these are the key controls over export refunds. Most of the paying agencies concerned were examined under the programme of financial audits, and in view of the lack of staff resources in DG VI, Financial Control is undertaking a programme of audits in 1994 and 1995 accounts of these agencies.

(b) Aid for soya was managed under the same dispositions as production aid for oil seeds, the latter having been examined by the Commission in the 1990, 1991 and 1992 clearances. The soya regime was abolished on 30 June 1992, and replaced by the 'fast-track` aid to surfaces cultivated in oil seeds. In view of the high risk presented by a new measure, and the fact that this measure was the precursor to the CAP reform, it was seen as a greater priority than a continuation of audits of the superseded soya production aid, and was examined in depth by both DG VI and DG XX auditors.

(c) The Commission endeavours to achieve equality of treatment by examining the implementation in all Member States principally concerned before drawing its final conclusions. For this reason it undertook its enquiry into dried fodder in all six Member States concerned by this measure, including the United Kingdom and Germany. The fact that expenditure in these two Member States was increasing rapidly was a further justification for examining the control measures in each. The Commission does not therefore agree that dried fodder was less deserving for priority in 1992, particularly as rape seed had been examined in the 1988 and 1989 clearances in Germany, and was assessed as a low risk, and as an examination of the ex-post scrutinies of peas and beans undertaken by the UK showed that national controls had been thorough.

2.5. The importance given by the Commission to ensuring the full introduction of the milk quota regime in Italy, Spain and Greece cannot be contested. The interest in this matter of both the Parliament and Council demonstrates the priority which this subject deserved, in the full knowledge that the Commission would be unable to examine other measures in the milk sector until later financial years.

Execution of the work

2.6. The Commission's annual work plans are necessarily indicative. Some enquiries need to be extended when deficiencies are found on the spot, so the annual programme has to be rescheduled accordingly, and new priorities appear during the year. There is no reserve of human resources available to absorb unforeseen activities. The work plan is a planning instrument, and cannot confine the clearance work to a programme to be followed blindly. Unfinished work scheduled for one year's clearance must necessarily be completed in a following year, if still justified under that year's review of priorities.

(a) It is correct that staff resources are not always adequate to complete all desirable work.

(b) The improvements required of the Irish authorities in its supporting documents and files were verified in the subsequent mission in October 1994, and the Commission was satisfied that the records permitted an audit to be carried out. It is correct, however, that the planned substantive tests were not completed.

(c) Due to lack of staff resources, dried fodder was examined in a mission under the 1993 work plan, instead of that for 1992.

(d) A further mission to Denmark to examine the supporting documents concerning oils and fats was not undertaken, in view of the limited staff resources, the relatively low level of expenditure, and the fact that all expenditure declared by Denmark had been audited under the financial audit programme.

(e) Staff resources do not permit the undertaking of all desirable missions. The Commission considers that the irregularities found by the Olive Oil Control Agency are evidence that national controls were effective. Accordingly, priority was given to missions in the sectors of olive oil production aid and of cotton, where national controls had been found to be particularly weak. Member States would be deterred from reporting irregularities if these reports led to an intensification of clearance audits.

(f) The Commission made three missions to examine the export of beef from Germany, and in particular the outcome of enquiries undertaken by the German authorities into the aid claimed. Following the Commission's findings, the German authorities were requested to undertake further enquiries, and presented in July 1995 their detailed report on these enquiries. After complementary information was requested and obtained, the Commission had a sound basis for appraising the conclusions of the report submitted, and for evaluating the financial consequences which amounted in total to DM 30,0 million, including the DM 5,8 million mentioned by the Court. Furthermore, in paragraph 4.2.9.5 of its summary report, the Commission reserved the right to recover further sums if the ongoing enquiries undertaken by the German authorities were not satisfactorily completed and the necessary recoveries made.

Working methods

2.7. The delay in conveying results was an exception rather than the rule, as it concerned a fraud and irregularities enquiry into the public storage of beef which was a lengthy procedure involving six Member States. In view of the need to demonstrate an equitable approach, and to keep the findings confidential, the results were not communicated formally until the enquiry was well advanced. It should be stressed that findings were brought to the national authorities' attention at the time of the on-the-spot visits, and in subsequent meetings in Brussels.

2.8. The point concerns a mere ECU 6 000, and appears contradictory to its comments on immaterial corrections in its following paragraph.

2.9. One should distinguish between the numerous arithmetical corrections made to the figures declared by the Member States, which can often be small, and the refusals to finance part of the expenditure declared because of failures to comply with Community legislation, which are almost always significant. While no minimum has been set for the 'compliance` corrections, the Commission gave some guidance in its decision setting up the conciliation procedure, when it decided that corrections of less than ECU 0,5 million, or 25 % of the budget line concerned for that Member State would not be eligible for conciliation. Furthermore, the guidelines concerning flat-rate corrections set 2 % of the expenditure under the measure concerned as the lowest rate of flat-rate correction.

The Commission is prepared to examine further the issue raised by the Court. It is aware that the many corrections imposed on Member States serve to prevent the development of a constructive relationship between the Commission and the Member States' services as regards the constant improvement of the effectiveness of their control procedures. The limitation of 'compliance` corrections to only those cases where there have been particularly serious failures in controls, warranting a significant financial consequence, may well improve the receptiveness of these services to the Commission's recommendations. But in taking such a line, the Commission risks criticism for not systematically making a correction for every imperfection, as is made by the Court in paragraph 4.7 of this report.

As regards the particular case mentioned by the Court, the error in calculating the correction was found after the final conclusions had been notified to the Member State, and after the conciliation procedure had been opened. The delay which would be caused to the clearance procedure by a new notification was not worth such a small sum.

3. CONCILIATION BODY

3.6. The existence of a conciliation procedure inevitably lengthens the clearance procedure, as up to six months can elapse between the date of notification of a correction and the end of the time limits applying to conciliation. The cause for the delay to the complementary decision, taken in November 1996, was the need to conduct additional enquiries or to obtain further information, in particular in the cases concerning Greece.

The Commission considers that the conciliation procedure has operated satisfactorily in its first year, in particular as time is necessarily required for the Commission and the members of the conciliation body to introduce such an innovation. The implementation of the procedure is kept under constant review, and any adjustments which prove necessary will be made after the three-year trial period for which this procedure was introduced.

4. EVALUATION OF AMOUNTS DISALLOWED

General remarks

4.1. Two items of 1993 expenditure were included with the 1992 clearance, both particularly significant cases of irregularity (imported olive oil in France, ECU 7,7 million, and cereals storage in Italy, ECU 99,2 million), because it was considered that clearly irregular sums of this magnitude should be recovered by the Community budget in the shortest possible time.There is no legal obstacle to this approach.

4.2-4.4. The Commission's computer tools allow it to undertake a wide variety of analyses and methods for the selection of transactions or beneficiaries to audit. Statistical sampling of the whole of the declared expenditure was required in the financial audits of the paying agencies, but not in all the compliance audits, as such sampling is not appropriate for the many audits which aim to examine and test the operation of a control system, particularly when only a small number of regions, or customs offices can be examined in the time available.

As regards the evaluation of financial corrections, the examination of supporting documents rarely allows the loss to Community funds to be evaluated, as the goods or animals which were inadequately checked by the Member State are no longer present at the time of the Commission's audit. The Commission therefore adopts the generally accepted approach of first analysing and testing the system and procedures and then evaluating the risks. Because of the heavy burden of staff resource costs in testing a sufficient number of beneficiaries selected by statistical sampling methods throughout a Member State's territory, and because of the problems in finding physical evidence, the Commission has in many cases necessarily relied on its best assessment of appropriate flat-rate financial corrections, based on the results of its systems tests and risk evaluations. Following the Court's remarks on the 1991 clearance, the European Parliament accepted, in its resolution on the 1991 discharge, that such financial corrections are a legitimate means of recovering funds.

Flat-rate corrections

4.7. The Commission does not agree that flat-rate corrections were applied inconsistently. The rates of correction are decided with a great deal of care, and in most significant cases after close scrutiny by the independent conciliation body. Corrections initially proposed have been reduced when the Member State could demonstrate that the risk to the fund was lower than that initially assessed, and, whenever staff resources allow, the Member States' arguments are verified on the spot. In the example of the correction for production aid for cotton, in paragraph 4.9 (b) of the Court's report, the Commission undertook five further missions, and held four meetings of a joint working group, to verify the measures taken to improve controls and to determine the actual losses incurred in the 1992 year. In one case, that of the public storage of beef in Ireland (paragraph 4.17), it was unable to undertake further enquiries as purchases of beef into intervention had stopped. It would not be reasonable to refuse to give a Member State the benefit of doubt as the reasons were beyond its control. Furthermore, in the one case of withdrawing a correction in 1992, a correction was finally proposed for the 1994 year instead of for the 1992 and 1993 years as initially envisaged (see paragraph 4.9 (d).

4.8. The Commission does not agree that any correction was made on a basis other than an objective evaluation of the actual loss to Community funds. The evidence supporting this evaluation was examined in great detail, and tested as far as staff resources and circumstances permitted.

4.9. The Commission considers that all corrections are based on the assessed risk of loss to the EAGGF budget.

(a) Olive oil production aid: Greece

A correction of 10 % of every year's expenditure from 1990 is a severe consequence of the deficiencies in the Greek control system. The Commission will continue to apply the criteria as described in paragraph 4.6 of the Court's report when deciding on the amount of the corrections, in the light of the evidence available regarding each individual year. Effective remedial action will be taken into account in cases of doubt as to the real risks to the Fund in the years concerned, as provided for under these criteria.

(b) Cotton production aid

In its clearance decision on the 1991 year, the Commission decided a correction of 25 % of the expenditure declared, with a reserve permitting the rate of correction to be reduced to 10 % if the reinforcement of controls was brought into operation before 31 December 1995. This rate of correction was exceptionally severe, and reflected the fact that the reality of a 20 % increase in the quantity of cotton declared for aid in the 1991/92 was put into question. The severe financial consequences were one factor which brought the Greek authorities to improve their control systems. They also undertook enquiries into the reality of the quantities declared for aid in 1991/92, in particular by comparing the quantities of ginned cotton sold with the equivalent quantities of raw cotton declared. Whenever there was a discrepancy between these two quantities, a recovery of the aid on the discrepancy was initiated. The recoveries initiated totalled Dr 1 208 million. The results of this enquiry demonstrated to the Commission's satisfaction that the risk of loss due to the deficient control systems in 1991 and 1992 did not exceed 10 %.

The Commission's enquiry into irregularities in the cotton sector took place between September and December 1992. The Greek authorities took immediate steps to improve their controls over the cotton sector, with effect from November 1992. These measures largely put a stop to the irregularities which had occurred in the previous marketing year. The Commission undertook numerous missions aimed at checking the effectiveness of the control systems, at verifying the conduct and results of ex-post controls of beneficiaries in the 1992/93 marketing year, and at recommending further improvements to the control procedures.

The Commission is satisfied that the Greek authorities did begin to improve their procedures as from November 1992, and that they undertook enquiries into the reality of the cotton on which aid had been claimed in the previous marketing year, and to demonstrate that the risk of loss to the Fund did not exceed 10 %. In these circumstances, a financial correction of 25 % could not be maintained.

(c) Beef public storage

(i) Germany

In the interest of fairness the Commission always recognizes the fact in its summary report when a Member State does improve its system following a mission as attaining such improvements is one of the objectives of the clearance procedure. As allowed under the guidelines, improvements can influence the rate of correction when there is doubt as to the rate to be applied. In this case, there was little doubt as to the rate of correction to be applied. An examination of the mission reports, and of the documents presented during the bilateral meetings and the conciliation procedure shows that the weaknesses found in the system were not sufficiently severe as to justify a rate of correction higher than 2 %, and, unlike the other Member States corrected under this clearance decision, no meat of irregular quality was found during the Commission's enquiries.

(ii)-(iv) Italy, France and the United Kingdom

The corrections followed an anti-fraud enquiry into the quality of meat in intervention at the time of the enquiry in 1990 and 1991 (Italy), 1991 (UK) and 1992 (France). In some cases a later mission was made to examine meat selected earlier which required thawing. The enquiry could not be repeated in the following years, as intervention purchases ceased in early 1993, so there were no findings on which to base a financial correction. Intervention reopened in mid-1996, and purchases are currently being examined in clearance audits.

(d) Withdrawal of peaches: Greece

Corrections were proposed for the 1994 financial year only, as this was the year actually covered by the enquiry.

(e) Grubbing up of vines: Greece

The rate of correction was reduced from 10 % to 2 %, subsequent to the report by the conciliation body, as the Greek authorities had complied with all control requirements other than the constitution of a vine register, and had verified 100 % of all aid claims before and after the grubbing-up of vines, followed by a second check of 1 % of the related files.

4.10. The flat-rate correction of 5 % on the two measures in Germany had been confirmed by the Court of Justice in 1993, and so was maintained in each clearance decision until the German authorities amended their irregular system.

4.11 (b) There are fundamental differences between the two cases, not only because Ireland applied the on-farm scheme whereas Spain adopted the slaughter model but also because Ireland took good note of the Commission's criticisms and rectified them from the 1992 year (payable 1993) while Spain did not. To this must be added the fact that Ireland conducted on-farm inspections at a rate of about 50 % in 1991, far exceeding the regulatory minimum of 10 %, and also based its system on a cattle identity scheme - factors which greatly reduced the risk of payment on non-existent animals. Spain on the other hand singularly failed to guarantee the security of its system for certifying slaughter. Neither country failed to conduct the controls actually specified, the problem lying more in the conception and quality of controls rather than their physical execution. Higher corrections were therefore not considered appropriate. The responsible services continue to monitor the procedures in both Member States.

The overruling of the Financial Controller

4.13. DG VI now systematically invites the Financial Controller's observations before communicating proposed corrections to the Member States, and there is an agreement on the procedure to be followed whenever the two services continue to hold differing views.

4.14. Between the DG VI letter of 13 November 1995 and the Commission's clearance decision of 27 March 1996, the Irish authorities presented voluminous correspondence supporting their contentions.

4.16. The deficiencies in the quality of meat found in intervention stores was ascribed by the Commission mainly to the weakness of the Irish system which relied on a permanent presence of its officials in the intervention centres, but with neither effective supervision, nor rotation of its officials. The improvements claimed by the Irish authorities to have been introduced as from the beginning of the 1991 financial year aimed at remedying this weakness, and reducing the consequent risk to the Fund, by assuring an effective supervision and control over their staff in the field, and at ensuring consistent and correct grading of meat at purchase. The unannounced inspections, mentioned by the Court, were part of the reinforced control claimed by the Intervention Agency's management. The effectiveness of these claimed improvements could not be tested as no subsequent enquiry was undertaken in Ireland before 1993, when intervention purchases ceased.

4.17. No Member State other than Ireland could claim to have improved their control system in the year being corrected, which is why no differentiation of the corrections was decided for other Member States. There was no doubt that Ireland had introduced some improvements. The doubts were raised on the sufficiency of these improvements to justify applying a lower rate of correction. The Commission finally gave the Irish authorities the benefit of the doubt.

4.18-4.19. The effectiveness of the improvements could not be tested, as intervention purchases had ceased in 1993, and at the time of the Commission's decision, there was no expectation of intervention reopening. The Commission considered it inequitable to penalize Ireland for an inability to verify this effectiveness for a reason beyond Ireland's control, and therefore overruled the Financial Controller. Due to the BSE scare, intervention did reopen shortly after the Commission's decision, and an enquiry on the spot has now confirmed a very substantial reinforcement of controls.

4.21. The Commission does not disagree with the principle that corrections should be reduced only from the year in which improvements are made. The reduction in the flat-rate correction for 1991 was decided precisely for the reason that the claimed improvements in Ireland dated from that year. No reduction was made for the other Member States concerned, as their improvements were made in a later year.

The corrections followed an enquiry into the quality of meat in intervention at the time of the enquiry. The enquiry could not be repeated in the following years as intervention purchases had ceased in 1993, and so there were no findings on which to base a financial correction in later years.

Ad hoc corrections

4.22. All corrections are arrived at with great care, and on the basis of the information available at the time of decision.

4.23. According to the Italian authorities, a significant proportion of the late payments proposed for correction in the 1992 clearance arose as a result of commune by commune enquiries undertaken in Sicily from 1988 onwards, the intention being to safeguard Community funds by extensive controls prior to payment. This argument was accepted by the Commission as a valid reason for delaying payment, providing that this explanation could be evidenced. Unfortunately the central authorities in Italy encountered great difficulties in obtaining information from Sicily in 1996, several years after the event. Initially proposed corrections for late payments of livestock subsidies totalled Lit 20 767 million and, after examining the evidence supplied, were reduced to Lit 14 925 million, including a reduction of Lit 2 855 million for special beef premium.

It is true that information supplied by Italy in respect of special beef premium was not as clear as that produced (and which the Court examined) for the other schemes but, in the knowledge that the extensive enquiries had been undertaken in Sicily and had caused payment delays, Italy was given the benefit of the doubt. It is not in the Community's interest to oblige Member States to effect payments when a doubt exists as to their regularity. This doubt certainly existed in Sicily.

4.25. The calculations aiming at verifying the reasonableness of the levies collected by every Member State were based on the statistics communicated by the Member State to Eurostat, taking into account subsequent corrections of these statistics. The clearance of the accounts unit did not have the resources to check the validity of the figures by undertaking an audit of the Member States' statistical services. The Commission confirms that it had agreed, in 1988, that standing crops were not subject to the levy. In a mission of May 1994 it verified the reasonableness of the Danish figures in this respect. Furthermore, it did examine the evidence submitted for the reduction of the deficit to 206 000 tonnes based on updated statistical information. The reduction of the deficit from 710 000 to 206 000 tonnes represents 5,5 % of cereals production.

5. CASES WHERE NO CORRECTIONS WERE APPLIED

5.1 (a) The Commission cannot systematically extend the results of its enquiries back to the earliest financial year remaining uncleared at the time of the enquiries. If it did, clearance decisions would be delayed indefinitely, as yet more enquiries would be completed while the bilateral and conciliation procedures were being undertaken for earlier enquiries. The prefinancing enquiry examined customs procedures in force in the 1993 and 1994 financial years, and the results are being dealt with under the clearance of those years. There is no legal obstacle to this approach.

(b) The enquiry into physical controls aimed at verifying whether the Member States were complying with Regulation (EEC) No 386/90, which required that a minimum of 5 % of exports attracting refunds were physically examined by the customs services. The Commission's auditors found no significant failures to comply with this percentage, and so there was no firm legal basis for a correction. They did, however, find that the nature of the examination varied widely, and so DG VI wrote to the Member States in early 1994 to inform them of the nature of examinations which it considered effective, and insisted that such examinations be undertaken uniformly by all Member States, stating its intention to propose financial corrections if these standards were not respected. A firmer legal basis for DG VI's position was provided by Regulation (EC) No 163/94, and by Regulation (EC) No 2221/95 applicable from 1 January 1996. Financial corrections can only be imposed if the Member State has failed to comply with a legal provision, and these legal provisions cannot be applied retrospectively. The letter of 1994 serves to provide a certain interpretative basis for corrections as from the 1995 year, but not retrospectively, and an enquiry has been underway since early 1996 to verify compliance with its terms. The approach adopted by the Commission led to some improvement in the quality of customs controls in a much shorter time than it takes for formal legislation to be adopted. Financial Control is now systematically consulted before the audit conclusions are communicated to the Member States.

(c) The delays were justified by the Belgian authorities by ongoing Court proceedings for half the ECU 1 million concerned, and by the need to complete supplementary controls on the other half. The Commission cannot sanction a Member State for not paying disputed or doubtful claims.

(d) Apart from the on-farm inspections made by the Commission services, most of the negative comment arises from examination of claims and inspection statistics relating to the 1990 claim year payable in 1991. The mission report deals in much more detail with the 1991 financial year when expenditure was indeed corrected. The German authorities took several important administrative steps to bring about improvements within days of the audit mission of May 1991 when deficiencies initially came to light (i.e. in time for the retention period for 1991 claims payable in 1992). Germany conducted on-farm inspections which were effective, and at a rate superior to the minimum 10 % required. In these circumstances, a correction of the 1992 expenditure was not considered justified.

(e) To support their claim for allowing the ECU 1,0 million concerned, the Dutch were able to present lists substantiating that there had been no loss to the Fund, and further informed the Commission that their Agriculture Ministry's own audit body had conducted, on a national basis, an audit of 80 % of ewe premium claims for completeness and accuracy and that payments were made correctly.

Taking into account the explanations and documentation supplied and being of the opinion that the risk of undue charge on the Fund was minimal, the Commission services proposed no correction, but conveyed the unsatisfactory nature of the situation and that similar occurrences in the future would be unacceptable.

(f) Private firms audited the financial systems and procedures of the paying agencies, but were precluded from verifying the compliance of the underlying expenditure with the Community regulations, as Article 9 of Regulation (EEC) No 729/70 requires that only officials mandated by the Commission may undertake such checks.There were in any case, in 1992, no explicit legal requirements determining the systems which paying agencies should adopt, and therefore no legal basis for corrections. The Commission took measures to provide this legal basis, and for requiring Member States to comply with detailed guidelines on financial control systems. This legal basis came into force in the 1996 financial year, and compliance by the Member States is being closely monitored by the Commission.

6. CONCLUSIONS

6.1-6.3. The Commission remains open to debate on improvements to the guidelines it currently applies when assessing the amounts of the corrections, and on the degree to which the corrections can be used as a pedagogical tool. While the Commission has the duty to recover misspent money, not every recommendation for improvement in the Member States' procedures can justify a financial correction.

The Commission notes with interest the Court's view in paragraph 2.10 of its report that there should be a minimum limit to corrections. This view is compatible with the principles of the reform of the clearance procedure, adopted by the Commission in 1993 following the 'Belle` report, receiving the favourable opinion of the Parliament and the legal bases being adopted by Council in 1995. This reform aimed at directing the clearance work towards the prevention of non-compliance by the Member States, and so to the avoidance of the conflicts which arise from corrections which are decided long after the Member State has already improved its procedures. A preventive audit will naturally result in a number of recommendations for improvements, but this does not necessarily mean that there had been a risk to the Fund of such significance that a correction must be made. There is a good case for applying corrections only in those cases of serious failures to comply with Community rules, which justify a material correction.

6.1. Experience over many years has shown that the time limit for taking the clearance decision was impossible to respect under the procedures in force before 1992, and with the addition of the conciliation procedure, the time required became even longer. This is one of the reasons for reforming the clearance procedure, a reform which was applicable from the 1996 financial year.

6.3. The Commission agrees with the objectives underlying the Court's views expressed in this conclusion. However:

(a) The Commission is seeking a solution to the need for additional staff, and will raise the matter in its negotiations regarding the 1998 budget. There is a lack of resources for the Commission's controls in the Member States, regarding both clearance of accounts and the corps of inspectors;

(b) it considers that risk analysis was applied centrally to allocate staff to key areas of the 1992 clearance procedure, as far as was compatible with the requirement to clear the entire expenditure declared and to conduct preventive audits of new measures with significant future expenditure;

(c) it is rare that the actual loss to the Community budget caused by a control deficiency can be ascertained, so the Commission must rely on its assessment of the risk of this loss, and apply flat-rate corrections, which the Parliament agrees is a valid approach;

(d) the existence of a risk of financial corrections is itself a particularly effective incentive for Member States to comply with the Community Regulations. The guidelines provide that improvements in control systems be allowed as a mitigating factor when there is doubt as to the level of correction to be applied. The corrections finally decided must stand up to legal scrutiny by the Court of Justice, and must represent a fair and equitable assessment of the seriousness of the infringement and of the real loss to the Fund. A great deal of care is taken by the different Commission services, by the conciliation body and by the college itself to arrive at this assessment, without losing sight of the principal objective, which is to bring the Member States to put into place procedures which fully protect the Fund from the risk of fraud and irregularity.