52007SC0335

[pic] | COMMISSION OF THE EUROPEAN COMMUNITIES |

Brussels, 21.3.2007

SEC(2007) 335

COMMISSION STAFF WORKING DOCUMENT

Annex to the REPORT FROM THE COMMISSION Member States' replies to the Court of Auditors' 2005 Annual Report {COM(2007) 118 final} Analysis of Member States' replies to the Court of Auditors' 2005 Annual Report

TABLE OF CONTENTS

Introduction 3

1. Summary of Member States’ replies 4

2. The 2005 Statement of Assurance 64

2.1 DAS Findings 64

2.2 Exchange of information between the Court and the Member States 68

Introduction

The European Court of Auditors (‘the Court’) published its Annual Report on the implementation of the budget concerning the financial year 2005 (‘2005 Annual Report’) on 31 October 2006[1]. In the report, the Court presented its Statement of Assurance (‘the DAS’) and the supporting information, including observations concerning management in Member States.

As obliged by the Financial Regulation[2], the Commission informed Member States immediately of such observations as well as the findings identified by the Court during its audits and attributed to Member States. Member States were invited to submit their replies by 24 November 2006. This working document presents in more detail the analyses of the replies and supporting information.

Section 1 contains two tables. Table 1.a provides an overview of general comments made by the Member States. Table 1.b lists the observations in the Annual Report mentioning one or more Member States. Where Member States have made comments on the Court's observations, these comments are summarised in the second column of the table. The replies have been shortened so that only main issues raised by the Member States are included in the summary. It is not meant to present a full overview of Member States’ positions on specific policies or cases of irregularity.

Section 1 is based solely on the replies submitted by Member States concerning observations in the Court’s 2005 Annual Report. It presents the views of Member States only.

Section 2 contains an analysis of the Court’s DAS findings as well as the exchange of information between the Court and Member States.

1 . SUMMARY OF MEMBER STATES’ REPLIES

The Commission sent a letter to each Member State on 23 October 2006, attaching:

- A list of points in the 2005 Annual Report specifically concerning the Member State in question.

- A list of DAS findings for the financial year 2005 attributed by the Court to the Member State in question.

Member States were asked to comment on the observations made by the Court in the 2005 Annual Report. They were also asked to state for each finding: 1) if they agreed fully, partly or not at all with the Court’s appreciation of the facts, 2) the reason for the finding occurring (in the case of full or partial agreement with the Court) or the reason for not agreeing with the Court, 3) information on follow-up to the finding made by the Court. Member States were also given the opportunity to make general remarks.

All Member States replied. Most replied within or shortly after the deadline set by the Commission (24 November 2006). Some Member States did not comment on all observations or DAS findings made by the Court.

The general remarks are summarised in table 1.a below. The replies to observations by the Court in the 2005 Annual Report are summarised in table 1.b (own resources), 1.c (agricultural policy) and 1.d (structural actions). An analysis of the reactions to the DAS findings is presented in section 2.

TABLE 1.A. GENERAL REMARKS |

Member State | Reply |

Austria | The Court's statement that, "The reports of certifying bodies… cannot be relied upon to provide assurance as to the legality and regularity of payments… as claims for EU aid are usually not checked on the spot by the certifying body" (Chapter 5 – Common Agricultural Policy) seems debatable. We can at least confirm that in 2005 the certifying body for Agrarmarkt Austria (AMA) accompanied on-the-spot inspections on 14 farmers by the technical audit service of the payment office in order to assess the quality of the inspections (inspector's approach, procedure for obtaining data, drafting of on-the-spot inspection report). An activity report was then drawn up, a summary of which was also incorporated in the 2005 certification report. Primary surveys of aid applicants by the certifying body, which the Court apparently has in mind, would, however, create an extra layer of inspections, which would seem to make little sense from the point of view of avoiding overregulation of inspection procedures. Moreover, we would point to the trouble the Court went to in 2005 to advocate a single audit system. Under such a system the Court could build on an existing inspection chain and additional primary checks would be superfluous. |

Czech Republic | The Czech Republic has sent to the ECA its observations and explanations concerning the TOR inspection report, expressing disagreement with many points of the evaluation of the findings. The Czech Republic expected the ECA to deliver an opinion on these cases of dispute and inform the Czech Republic accordingly. However, this expectation has not been met. An audit should primarily examine the system, focusing on the examination and evaluation function for example on the following aspects: compliance with legal provisions, measures and established procedures in the activities performed by the inspected entity; early identification of risks and adoption of corresponding measures to avoid or mitigate the risks; effective management controls with reliable and timely operating, financial and information functions; testing the effectiveness of the internal control system in the context of responding to changing economic, legal, operating and other conditions; assessment of whether all resources available in the customs administration departments, i.e. both material and human resources, are used adequately in order to achieve the best results; verification of whether measures proposed on the basis of a notification following a management control or own measures were adopted, etc. 3. The Czech Statistical Office (ČSÚ) has not received the precise wording of the findings in points CZ11-CZ14 and is therefore unable to provide more specific answers. |

Denmark | The Danish authorities are happy to see that every year since the statement of assurance was introduced the Court of Auditors has, with some reservations, considered the EU’s accounts to give a true and fair view of the Communities’ revenue and expenditure and financial position. On the whole, therefore, this part of the statement of assurance has been positive every year. It is clearly unsatisfactory, however, that for 12 years running the Court of Auditors has given a statement of assurance subject to major reservations regarding the greater part of the EU’s expenditure budget. But this is not due to some permanently deadlocked disagreement, because despite its repeated reservations the Court of Auditors does acknowledge the progress that has been made in recent years in the financial administration and internal control of the EU budget. In the Danish view the main steps that could be taken at this stage in order to make further progress in the financial administration and internal control of the EU expenditure budget are the following: A more uniform system of internal control and auditing of Community appropriations at all levels of administration, as proposed by the Court of Auditors in its opinion on the single audit model, which has been followed up in the Commission’s roadmap and action plan for an integrated internal control framework. The implementation of the recently revised Financial Regulation for the EU budget, which introduces a simpler, more flexible and more transparent system of budget administration, and incorporates part of the Commission’s roadmap and action plan for an integrated internal control framework. Better presentation of the main control results in the Commission’s annual activity reports and declarations and in the Court of Auditors’ annual reports on the implementation of the budget. Error rates should be published for payments under the main headings of the expenditure budget, so that the estimated financial consequences of the errors noted can be seen. Greater cooperation between the Court of Auditors and the national audit bodies (in Denmark the National Audit Office (Rigsrevisionen)); findings by the national bodies that have implications for the EU should be integrated systematically into the Court of Auditors’ own auditing and statement of assurance process. A broadening of the commitment and responsibility of the Council, and hence the Member States, with regard to the financial management and control of the budget. A standing specialist working party should be set up, under Council or Commission chairmanship and with representation from the Court of Auditors, which would be able to provide more continuous and systematic follow-up to the Court’s annual report and the Member States’ replies as summarised by the Commission. This would also benefit the Council’s own annual discharge procedure. Annual report for 2005 The Danish authorities find that: The Court should expand the statement of assurance (the DAS) so as to make it clearer in which policy areas and in which countries the problems are greatest. This would make it easier to tackle the difficulties. The Court should considerably expand the description of the DAS method in the annual report, which is very short (paragraphs 1.59 and 1.60); this could be done in a separate annex. The description of the method could for example give a more detailed explanation of the interaction between the four sources of evidence, based partially on estimates, which the Court uses when it undertakes its assessments of specific aspects of the implementation of the budget and decides whether to enter an auditor’s reservation. The disagreement between the Commission and the Court of Auditors regarding the treatment of multiannual activities and the financial correction of errors (recoveries) should be clarified further (see in particular paragraphs 1.63-1.65, 2.18, 5.59 and 6.41 of the annual report). The Commission should provide a summary of the nature and extent of recoveries completed or still outstanding. The question of “associated risk” and thus the assessment and management of risk and the achievement of “reasonable assurance” regarding the different sections of the budget should be clarified further (see in particular paragraphs 2.2, 5.44, 5.46, 7.2, 7.8 and 7.28 of the annual report). Discharge procedure The Danish authorities would like to see a standing working party set up to consider questions of financial management and control, which among other things would be of assistance to the Council before, during and after the discharge procedure. |

Germany | Germany has not filled in table 2.2 but makes reference to the replies already provided by the German authorities when responding to the ECA's statements of preliminary findings. These replies are available to the Commission and Germany therefore does not find it reasonable that Member States are requested to provide the same information repeatedly. |

Hungary | It is regrettable that for the 2005 financial year the European Court of Auditors was once more not able to issue the declaration of assurance for the implementation of the European Union budget without reservations. It is important to arrive at a situation where the Court is able to issue a positive declaration of assurance in respect of the implementation of the EU budget, as this would certify that the public money has been used properly. To this end, Hungary has been supporting and would like to be involved in the creation of a Union-level integrated internal control framework that could help make the utilisation of Union resources more transparent, efficient and in conformity with the rules. The Court of Auditors' report states that, although the European Commission's internal control system continued to improve in the 2005 financial year, the existing supervisory and control systems needed to be further strengthened at the Commission, in the Member States and in beneficiary countries. The Court of Auditors proposes further improvements in the supervision and control systems in the areas of full compliance with guidelines and annual activity reports, analysis of the impact of internal control standards and processing of special indicators relating to the legality of transactions and their conformity with the rules. In line with this, more attention needs to be paid to seeing that uniform guidelines and procedures are created at Union level and - in the Hungarian view - both the Commission and the Court of Auditors should play a key role in this. As regards structural measures (regional policy, employment and social policy, regional development and fisheries), in the assessment of the Member States' supervision and control systems the Court of Auditors reported inspection errors in every programme checked. The errors included such things as incorrect audit trails, management audits being carried out irregularly or incorrectly, and shortcomings in audits that involve taking samples. The Court of Auditors concluded that the Commission supervision of the control mechanisms operated by the Member States was not effective, some programmes (between 1994 and 1999) were wound up when they should not have been, while others were wound up unduly late, and, in addition, the procedures for making financial corrections and reporting recovery measures did not conform to the rules. In order to avoid a repetition of these problems, we think it important to ensure that better preparations are made for winding up the 2000-2006 programmes; arrangements should be made at Union level for a more uniform distribution of audit activities and for more efficient application of the financial correction procedures. As regards the errors and shortcomings found in respect of regional policy, a summary of commonly occurring shortcomings would be welcomed, principally those occurring in several Member States, discovered in the course of the audits carried out by the Commission and the Court of Auditors; this should be made available to the Member States (on a regular basis), so that the Hungarian authorities can learn from the mistakes of the past, thus making for management and control systems for the 2007-2013 programming period that operate correctly, effectively and efficiently. For the purpose of further supporting the setting up of management and control systems that operate correctly, effectively and efficiently, we would welcome more detailed specification at Union level of the set of criteria relating to assessment in accordance with Article 71 of EC Regulation 1086/2006 (maybe as part of the guidelines issued by the Commission), with the Court of Auditors involved in devising the criteria. In its report on the 2005 financial year, the Court of Auditors found that, since the net financial corrections are restricted (as a consequence of the irregularities discovered by Commission auditing), there is nothing to encourage the Member States to take measures to prevent irregularities or to improve their management and control systems. We agree that greater efforts need to be made by both the Commission and the Member State to prevent irregularities and, where they do arise, to investigate them efficiently and effectively and apply the appropriate financial corrections. At the same time, we would observe that, when it comes to preventing irregularities and handling those that do arise efficiently and effectively, a suitably operating supervision and financial management and control system is a far better tool than the negative incentive of higher financial corrections being imposed by the Commission. In the same vein, we would welcome the production of an "irregularity catalogue" (maybe with the involvement of the Court of Auditors) containing uniform correction measures to be applied by the Member States and the Commission for the main types and instances of irregularity. The findings relating to Hungary in the Court's 2005 report are for the most part well-founded, although the extent of the errors arising and identified in the area of agricultural policy is not borne out in all respects by MVH's internal audit experience. For example, the finding at the end of the third paragraph on page 19 of the report stating that a quarter of area aid involves considerable overpayment is open to question. In the case of Hungary, the issue of establishing the reduction coefficient has yet to be resolved between the Commission and the MVH (see Annex 4), so the accounting errors are often due to this difference of opinion and to the lack of a uniform method and approach. Hungary therefore agrees with the Court of Auditors' proposal that the Commission's DGs should do more to assess the operation of the internal control systems and the use, in accordance with uniform practice and a uniform approach, of the applicable calculation procedures. It is necessary both to devise uniform implementing provisions and, when it comes to auditing, to ensure standards and procedural practice that are in tune with the specific characteristics of the EU. |

The Netherlands | The Netherlands is pleased to learn that the Integrated Administration and Control System (IACS), where properly applied, is an effective control system for limiting the risk of error or irregular expenditure. However, the European Court of Auditors also notes that expenditure as a whole still shows significant errors. The European Court of Auditors indicates that, under the IACS, animal premiums present a greater risk than area aid. The main reasons for this are frequent animal movements and the complex conditions for the animal premium schemes. The Netherlands acknowledges the importance of good monitoring and control systems. They are a precondition for proper control over the legality of expenditure financed by the European Agricultural Guidance and Guarantee Fund, Guarantee Section. However, a legitimate distribution of aid also requires clear and simple Community rules and regulations. The conditions for granting aid and the control measures required to audit it are often, by their very nature, complex. This can get in the way of correctly and fully applying Community regulations. The Netherlands would therefore stress the great importance of simplifying Community rules and regulations. As a general remark the Netherlands would point out that it is difficult to compare data from different Member States because the IACS questionnaire is open to different interpretations. |

Spain | The Report refers to matters arising from inspection visits whose final conclusions have not been received by Spain. It therefore seems that the conclusions contained in the 2005 Annual Report in some cases depend on an appraisal of the statements and comments submitted by the Member State following the provisional report concerning the inspection visit. Reference could therefore usefully be made to the situation of the specific procedures (whether final or provisional conclusions of the inspection visits are concerned, and whether comments have been made by the Member State). |

The United Kingdom | The United Kingdom states that while it agrees with the need for greater propriety on EC expenditure it agrees with the Commission’s comments on the need to change the single transaction audit methodology employed by the ECA when applying it to the multi-annual European programmes. Until a more appropriate methodology is found there will continue to be errors found and no recognition of any corrective action taken, as the single transaction audit is time limited. Another point is that the differing interpretations of regulations by the Directorates-General, the ECA and OLAF make conformity to the regulations an almost impossible task. Neither do they help in the simplification of regulations and therefore the likelihood of the ECA giving an overall positive assurance of the accounts or enhance the attraction of entering into a contract of confidence. Slightly more detailed comments are: The United Kingdom will continue to work with the Commission and the ECA to improve the systems with relevance to proportionate action. The ECA 2005 Audit Report highlights deficiencies in the reporting of Irregularities and suggests that the European Commission is not following its own Regulations in respect of the imposition of financial corrections and also suggests that an increased level of checking is required of the day to day transactions. This causes concern in the United Kingdom as it may result in a disproportionate response by the European Commission; The United Kingdom is concerned with the continued variation in interpretation of the Regulations by each successive audit mission which has cast aside previously agreed control methods. This is causing an unnecessary burden on the delivery bodies who are continually having to adapt their systems to the ever changing requirements of the auditors; Changes to the interpretation of the Regulations are now being made far too late in the operational delivery of the programmes. The United Kingdom does not see any reason to carry out additional and even retrospective work to provide ever-increasing levels of assurance as a result of changes in the interpretation of the Regulations. Satisfying these additional requirements will require additional resources as the current delivery methods are already stretched managing the current programmes and designing and implementing the 2007-2013 programmes There appears to a lack of coordination in the timing of various audit missions to the United Kingdom. This has been promised in the past, however it is not evident in the United Kingdom and in Scotland there have been three audit missions running simultaneously. The new Regulations embrace the principle of proportionality. The United Kingdom sees as a key principle that will determine roles and responsibilities in respect of the delivery of the new programmes. Once the roles and responsibilities have been agreed with the Commission this should be the benchmark against which the United Kingdom is judged. The United Kingdom requires to ensure that the ECA are in agreement with this position. |

TABLE 1.B. SPECIFIC OBSERVATIONS IN THE 2005 ANNUAL REPORT |

Paragraph | Observation in the 2005 Annual Report | Member State reply |

CHAPTER 1 – THE STATEMENT OF ASSURANCE AND SUPPORTING INFORMATION |

1.11 | Experience in public administrations undertaking similar reforms5 shows that projects of a scale and complexity similar to the modernisation of the Communities' accounts inevitably face a variety of problems which complicate and delay their complete implementation. Against this background, important progress has been achieved: - on 1 January 2005, the Commission moved the general accounting of the EU from cash-based to accruals-based accounts. New accounting rules6 and methods, a new harmonised chart of accounts and new consolidation tools were introduced in all Community institutions and agencies; - stakeholders were provided with individual and consolidated accounts which comprise the new obligatory elements and contain more detailed information about the resources controlled by the different entities, the costs of their operations and the cash flow operations. _______ 5 For instance at the United Nations Headquarters and the Organisation for Economic Cooperation and Development or in Australia, Denmark, France, Spain, Sweden, the United Kingdom and the United States. 6 The new set of 15 accounting rules adopted by the Commission's Accounting Officer are based on the International Public Sector Accounting Standards (IPSAS) and for accounting transactions that are not yet covered by the IPSAS, on the relevant International Accounting Standards (IAS) / International Financial Reporting Standards (IFRS). | Denmark informs that it has made a change to accruals-based accounting with effect from 2005. The national budget is to be accruals-based from 2007 onward. Due to thorough preparation and gradual introduction in trial departments from 2003 onward, the changeover did not give rise to any major difficulties. Denmark finds that the challenge going forward is to ensure that the new accounting information and the accruals-based budget produce gains in terms of management. This is being done in part by means of incentives for greater economy which have been built into the new system of allocation of State funds and the rules for depreciation of assets. In order to realise the full benefit, however, the accruals-based accounting figures will also have to be allocated at a meaningful and uniform level of activity, so that it becomes clear to the decision-makers who have to determine priorities across the different branches of government expenditure what effect the resources appropriated have had. Denmark takes the view that as the Commission progresses in the use of accruals-based accounting and budgeting principles it could usefully try to spell out management implications of this kind. |

Sweden has sent information on the introduction of a new accountancy model in the 1990s. The reform comprised a change in the rules in force at the time (i.e. the Accountancy Regulation), the launch of a comprehensive training campaign, the production of a manual, the establishment of an accountancy plan for national authorities, and further development of the systems support for accounting. |

CHAPTER 3 – BUDGETARY MANAGEMENT |

3.14 | Most of the cancellations under the n+2 rule relate to the European Social Fund. Moreover, in practice new commitments from the performance reserve9 were - as in 2004 - in some cases allocated to the programmes from which commitments were cancelled, thereby partly negating the effect of the procedure10. _______ 9 4 % of the appropriations allocated to each Member State were placed in reserve for distribution to the best performing programmes in three tranches from 2004 to 2006. 10 For example, in the case of 79 million euro decommitted from a programme following the year n+2 rule in the Netherlands, 24 million was allocated to the same programme from the Performance Reserve. | The Netherlands state that the cancellations under the N+2 rule and the allocation of the performance reserve fall under different annual instalments and can therefore not be compared. The performance reserve was allocated as laid down by the Regulation and was agreed to by the Monitoring Committee. |

CHAPTER 4 – REVENUE |

4.4 | The Court's audit of the accounts cannot cover undeclared imports and imports that have escaped customs surveillance. However, its audit work included an evaluation of supervisory and control systems, both at the Commission and in Member States, to assess whether they gave reasonable assurance of completeness. It consisted of a review of the organisation of customs supervision and of the national systems for accounting for traditional own resources in eight Member States3 (see paragraph 4.22); an examination of the effectiveness of the mutual assistance arrangements (paragraphs 4.10-4.12); an examination of the Commission's accounts for traditional own resources; and an analysis of the flow of duties in order to gain reasonable assurance that the amounts recorded were complete and correct. ______ 3 Belgium, the Czech Republic, Germany, Ireland, Luxembourg, Malta, the Netherlands, and the United Kingdom. In addition, the traditional own resources accounting systems were reviewed in Italy, Poland and Sweden. | The Czech Republic has in its reply indicated that it does not agree with the ECA's observation concerning a) the complexity of the system, b) the separation of the accounting system from the customs clearance system, and c) the absence of any controls. |

The Netherlands has in its reply presented its reactions to a number of detailed remarks made by the ECA in the statement of preliminary finding sent to the Netherlands concerning these issues. The Netherlands has followed up on the ECA findings. |

Sweden refers to its reply to point 4.22. |

The United Kingdom refers to the reply it has provided directly to the ECA on 7 April 2006. |

4.7 | As in previous years the Court's audit and the Commission's inspections (see paragraph 4.9) found systematic problems with the B accounts in a number of Member States4. Some errors are related to the conditions for entry in the accounts. In Member States whose B accounts represent 34 % of the balance, customs debts that are partly secured are nevertheless entered in full in the B accounts, leading to delay in making the secured part available to the Commission. Other errors arise because Member States have not adapted their accounting systems so as to record the appeal or recovery accurately, or because there is insufficient internal control over the compilation of reports. 22,7 million euro of potential duties remain under discussion between the Commission and Germany as a result of such a problem. ____ 4 Belgium, the Czech Republic, Germany, Ireland, Italy, Greece, Spain, Finland and the United Kingdom. | Belgium makes reference to a judgment made by the European Court of Auditors in 2004 (C-274/04) which has clarified the issues raised by the ECA. The Belgian authorities are currently trying to bring their procedures in line with this judgement. |

The Czech Republic notes that the ECA found three possible irregularities: a) establishing duty in the B-account in the case of deliveries of post items, b) a missing proof of guarantee in a case, and c) claiming only a part of the duty when challenged. Concerning a), a new contract of mandate is being negotiated between the Customs Administration and the Czech Post which should remedy the detected irregularity. Concerning b), the decision on the discharge of the customs debt was repealed. Concerning c), only after administrative proceedings was it possible to conclude that the debtor should be granted remission of a part of the duty originally established. |

Finland says that the clearance and correction systems need to be replaced in their entirety with new applications. This change will, however, have to be fitted into the schedules for the general overhaul of the integrated systems. Nevertheless, with the current – partly manual – system, most of the traditional own resources have been made available to the Commission within the time limits stipulated. |

Germany agrees with the first part of the ECA's observation but notes that the finding made in the second part ("Other errors arise …. As a result of such a problem.") was not made during these checks. |

Greece notes that the remarks made regarding the keeping of the B account in the Court of Auditors’ Annual Report for 2005 are not specific, but on the basis of the observations put forward by the Commission after it carried out its own checks, the Customs Administration itself has given customs offices new guidance where necessary on the way in which the B account is to be kept. |

Ireland states that the Commission's auditors noted a significant improvement, i.e. much closer alignment of the A and B account, during their visit to Ireland in October 2006. The issue of partly secured debts being entered in the B account did not arise in Ireland. |

Italy has taken measures to improve the computerised accounting system, including better alignment with the ordinary account (A-account). |

The United Kingdom does not fully agree with the ECA. Present legislation permits debts that are not secured or contested to be entered in the B Accounts and does not differentiate between whole or partial debts. The UK authorities sought clarification by intervention on several points in ECJ case C-275/04. The Court of Justice gave its decision on 5 October 2006 but it does not address the points raised by the UK. |

4.22 | The systems for customs supervision and for accounting for traditional own resources were generally found to be functioning correctly. However, an examination of Community transit found that many Member States18 did not have effective systems for starting enquiries on time when there was doubt about the arrival of consignments, and there were also considerable delays in later stages of the enquiry and recovery procedures. As a result duties were frequently collected late in such cases. In several Member States19 customs control of goods in temporary storage20 was not sufficient to ensure that the time-limits and other rules in the Community Customs Code were observed. In one Member State21 the intervals between customs audits of economic operators could routinely exceed three years, thus putting traditional own resources at risk because of time-barring. ___ 18 Belgium, Germany, Spain, France, Italy, Latvia, Hungary, Poland, Slovenia and Sweden. 19 The Czech Republic, Germany, Malta and the United Kingdom. 20 Goods that have arrived and been included on a summary customs declaration, but which are awaiting the formalities necessary for them to be assigned a definite customs-approved treatment or use (such as release for free circulation). 21 The United Kingdom. Under the Community Customs Code additional customs duty cannot normally be charged more than three years after the original customs debt was incurred. | Belgium is developing a "Research" module which will be integrated in NCTS in 2007. This should automatically resolve the problems mentioned in point 4.22. |

France has made efforts to respect the delays but admits that problems may have occurred. However, by the end of 2005 procedures were computerized, permitting services to assure better follow-up. |

Germany notes that it was not possible to implement the shorter times for inquiries in connection with the NCTS until all countries involved in the Community/transit system had also been connected to the computerised inquiry procedure. Time limits for notifying principals have been observed since the start of 2006. The aim is to restructure the auditing of transactions and the technical supervision of main customs offices. Germany shares the ECA's view that computerisation of the transit procedure should not mean that economic operators no longer expect to be checked. |

Hungary informs that – in light of the ECA findings – an action plan was drawn up to identify and rectify any measures found wanting. As part of this, every single non-discharged Community transit operation covered by the period under inspection was reviewed by the Hungarian customs authorities and the omissions were rectified. The subsequent improvements to the NCTS now make it possible to carry out the enquiry procedure electronically. Use of this system now ensures that Community transit operations are properly discharged and that the necessary action is taken within the time-limits. |

Italy informs that appropriate measures are currently being taken to improve the computerised management system to ensure inquiry procedures are initiated in a timely way. |

Latvia informs that a national assessment of the application of transit customs procedures over the period 1 May 2004 to 1 January 2006 established 55 occasions where deadlines for entry into accounts of tax debts were not met, including instances where enquiry procedures were not begun on time. |

Malta informs that customs officials are making an extra effort to keep to the temporary storage time-limits by a) monitoring the relevant module in the Electronic Customs Systems which enables the official to determine which goods have been in temporary storage for longer than they should and taking the relevant action, b) requesting that stock held in temporary storage for more than 45 days are sent to a new, separate store. |

Poland states that the findings made by the ECA do no reflect the guidelines laid down by the Ministry of Finance in a letter to the heads of all customs chambers. At present, the likelihood of practices arising which do not comply with the rules for conducting enquiries and collecting duties in Community transit is minimal in view of the fact that the NCTS IT tool is used for those purposes. On 4 November 2005 the Polish customs administration extended the functional scope of this system to phase 3.2.2. Poland has followed up the findings of the Court by instituting proceedings for the formal completion of the transit procedure. |

Slovenia expresses disagreement with the findings regarding the failure of the system for initiating enquiry procedures on time. The ECA’s audit was performed in June 2005. However, samples on which these findings were based were from 2004 (initiated transit procedures), that is from the period which immediately followed the accession of Slovenia to the EU and the establishment of the Central Transit Office. This initial period did in fact witness some delays which occurred upon the initiation of enquiry procedures (mostly of technical nature due to unreliability of the MCC application). We perceived these delays and to that end set up a larger number of internal controls (aimed at initiating enquiry procedures for all required transit procedures on time). Effective performance of these internal controls already at the time of the audit was established also by the ECA – in item 37 it states the following: “The delays noted for the audited files were linked to temporary problems in the set up phase of the Central Transit Office. Enquiries are now dealt with promptly.” With regard to “considerably delayed” enquiries and recoveries, our reply to preliminary findings by the ECA already contained a comment on the representativity of the audit sample. Out of 1757 samples, cases of initiated enquiries were gathered in sample of 22 according to the criteria of exceeding the term (delays were evident from the enquiry record). Furthermore, the majority of “delays established” by the ECA were not actually delays (e.g. in cases in which an IE106 is received, it is not necessary to notify the principal). With regard to collecting duties, it cannot be claimed that frequent delays occurred. This only happened in one case of audit – the customs debt had not been established on time, therefore the debt in the amount of 100 EUR was recovered with a one-month delay, and penalty interests would amount to 2 EUR. |

Sweden informs that a reorganisation in 2004 meant that enquiries concerning transit operations not completed in time were centralised in Malmö. The amount of work in Malmö during the audit period concerned was so great that temporary assistance was needed from clearance offices in other parts of Sweden. As a result processing time was somewhat longer. This also affected the collection procedure which is closely connected to how well the time limits for enquiry procedure have been complied with. The procedures have been improved since the cases concerned have been dealt with. |

The United Kingdom agrees with the Commission that NCTS should provide improvements concerning the timely discharge of transit movements. The United Kingdom informs that it carries out post clearance audits as part of a risk management framework designed to assure that the correct amount of revenue is accounted for at the correct time. Within this framework, certain regimes are high-risk and warrant a higher degree of intervention and assurance. However, the United Kingdom does not accept that this will always warrant a full systems audit. In an attempt to use resources efficiently and target resources to assuring the highest risks, different methods of assurance and intervention according to the level of risks are used. |

4.23 | In its Annual Report concerning the financial year 2004, the Court noted that there were significant differences between Member States as regards the existence and implementation of certain elements of supervisory and control systems at Statistical Offices with respect to the compilation of national accounts. In 2005 the examination was extended to six more Member States22 and a number of weaknesses were identified in the following areas: (a) the performance of a formal and structured risk analysis in respect of the process of national accounts compilation23; (b) the existence of agreements or equivalent arrangements between national accounts departments and units providing basic statistical data, which set out the conditions for the delivery and the quality of data24; (c) the systematic production of "quality reports" accompanying statistical surveys25; (d) the performance of internal audits on the process of statistical data collection and compilation26. ____ 22 Austria, Belgium, Finland, the Czech Republic, the Slovak Republic and Sweden. 23 This element was not fully implemented in any of the Member States visited. 24 This element was fully implemented in Belgium, the Czech Republic, Austria, the Slovak Republic and Sweden. 25 Occurred in Austria, the Slovak Republic, Finland and Sweden. 26 This element was fully implemented in the Czech Republic and Sweden. | Point 4.23 (a): Austria informs that Statistik Austria, as part of the ongoing planning and performance of work for the national accounts, also conducts a regular and comprehensive survey and analysis of potential risks in essential data sources which might have implications for the national accounts. Statistik Austria is also required to involve the national accounts department in the quality control of basic national account statistics ("feedback meetings on quality"). Point 4.23 (c): Statistik Austria has developed a well advanced process with quality reports on statistical products and "feedback" meetings with external experts on the basis of these reports. A quality report on the annual national accounts has since been produced and published on the home page of Statistik Austria. Point 4.23 (d): The basic statistics underlying the calculations in the national accounts are audited internally in the specialised directorates responsible for compiling those statistics. These audits are described in detail in the published quality reports. On this basis individual project officers carry out a series of further plausibility checks at subsequent states of the process of compiling the national accounts. |

Point 4.23 (a): Belgium informs that it has made a serious effort to make progress on this issue (description of procedures underlying the compilation of the national accounts). Certain elements are more advanced than others but nevertheless constitute a first global approach to formal risk analysis. Point 4.23 (d): Belgium bases its national accounts on a number of sources. Substantial input is provided by the Belgian Central Bank where a number of control procedures guarantee the quality and continuity of the statistics. The internal audit of the Bank also covers the entities producing the statistics. Data concerning production (value added) is based on accounting data which is approved by an external auditor before being transmitted to the Central Bank which again performs further quality checks. External data sources must also undergo numerous quality checks before being used as input to the national accounts. |

Point 4.23 (a): Finland does not find it entirely clear what the ECA means by risk analysis in connection with GNI-based own resources, in particular whether this only relates to the crisis contingency plan or also to "normal" operations. |

Point 4.23 (a): The Czech Republic notes the ECA finding that no action plan to follow up risks identified by the National Account Section had been presented. However, in practice some measures have been adopted and partially implemented. Points 4.23 (b) and (c): The Czech Republic informs that these two observations have been complied with. |

Point 4.23 (a): Slovakia points out that the performance of formal and structured risk analysis in respect of the process of national accounts compilation is an exacting process that needs to developed over the long term. Therefore, to reduce the risks associated with the compilation of the national accounts, the Statistical Office of the Slovak Republic has introduced a quality management system defining the main products in the field of national accounts, responsibility for the compilation of the accounts, quality measurement indicators and so on, so that the process of producing the statistical outputs for the purposes of the national accounts is mapped and documented as comprehensively as possible. The system introduced by the Office is already contributing to eliminating the risks associated with the compilation of the national accounts. Point 4.23 (c): The process of preparing and carrying out the statistical surveys and processing the data also involves the drawing up of a technical project, which contains control mechanisms running from checking the input data right up to validation of the data that have been processed. Once the data have been processed, the results are validated on the basis of a set of parameters (non-response, completeness, reliability, grossing-up etc.). Delivery of results is recorded in the form of a receipt record. Point 4.23 (d): As of 24 October 2006, the Statistical Office holds an ISO 9001:2000 quality management system certificate. Prior to the application for the certificate to be issued, two internal audits were carried out at the Office, focussing on the core processes and ancillary processes involved in national statistics. A certification audit was carried out by Bureau Veritas Certification before the certificate was issued. |

Point 4.23 (a): Sweden acknowledges that no fully worked out risk analysis for the National Accounts exists as yet but it is regarded as a priority. Steps have been or are being taken. For instance, back-ups of data bases and the calculation system are available, and a process-analysis has also been started. Point 4.23 (b): A service-level agreement has been drawn up between the National Accounts division and the primary statistics department. Point 4.23 (c): Statistics Sweden's figures are quality-declared and for the most part accompanied by a description of the current situation regarding supplies to the National Accounts. Point 4.23 (d): Internal audits on statistics products are performed at Statistics Sweden by specially designated groups. Similar studies also take place internally within the National Accounts division. |

Table 4.2 | Finland underlines that its long-term aim is to work with the Commission to lift any reservations. The outstanding reservations either pose problems of a theoretical nature or are open to different interpretations from the legal point of view in which case it can take a long time to resolve the issue. The ECA does not always clearly distinguish between the Commission's reservations and the Member States' reservations. |

The Netherlands notes that at the end of 2005 only one reservation was still outstanding. |

Table 4.3 | Austria notes that the outstanding reservation was lifted by letter dated 8.3.2006. |

Finland has forwarded additional information to Eurostat. |

The Netherlands informs that at the end of 2005 two items were still outstanding, and it has submitted revised figures for this. |

CHAPTER 5 – THE COMMON AGRICULTURAL POLICY |

5.4 | In order to obtain assurance as to the legality and regularity of the transactions underlying the Community's accounts, the Court audited the main supervisory and control systems (see paragraph 5.3) and tested a random sample of payments drawn from the expenditure of 25 paying agencies in EU-15 (which were collectively responsible for 66 % of CAP expenditure) (see Table 5.1) and 20 payments under the Single Area Payments Scheme (SAPS)5 in the New Member States. _____ 5 The SAPS replaces all direct aid with a single payment. It involves payment of a uniform amount per hectare of agricultural land. Eight of the ten new Member States (excluding Malta and Slovenia) apply it. | Slovenia notes that in the report for 2005 the European Court of Auditors took account of the statistcs for Slovenia which were forwarded in March 2005. Slovenia subsequently adjusted these statistics, but unfortunately no account was taken of this in the report. When the corrected statistics are taken as a basis, the percentage of overstated declarations of area in 2004 comes to 1.79% (previously 53.5%). |

5.6 | The Court analysed the inspection statistics of IACS for all Member States and audited its implementation in a number of old Member States as well as in five6 of the eight New Member States that have opted for SAPS and in the two New Member States7 that have opted for the traditional direct aid schemes (see Annexes 5.2 and 5.3). ___ 6 The Czech Republic, Lithuania, Hungary, Poland and Slovakia. 7 Malta and Slovenia. | Malta informs that errors found have now been adjusted and stricter checks have been implemented to minimise the risk of similar incidents. |

Poland informs that the IACS statistics submitted in March 2005 were prepared using incorrect data, the European Commission being informed of this in September 2006. The statistics were prepared at a time when not all on-the-spot inspection reports had been entered in the IACS database. The basic data received by the Court of Auditors were different to the data used to prepare the IACS statistics. Verified statistics based on the latest data for applications and inspections carried out in 2004 were forwarded to the European Commission in June and subsequently in September 2006. |

Slovenia has forwarded detailed information on its reaction to all the ECA's finding during its audits. |

5.10 | The Court's findings in Greece indicate that there has been no significant improvement since last year: a) the quality of inspections is low and findings are poorly or not at all documented, reporting of results is unreliable and is not always based on genuine inspections; b) in certain local authorities in Greece, the techniques used when measuring parcels lead to a higher technical tolerance than the maximum allowed (5 %). The financial impact of this practice cannot be quantified, c) farmers' unions control the input of all data into the computer system. None of the data in the system are secure and they can be and are modified by the farmers' unions at any time before payment. The computer system does not record when and why changes to the original data are made. Many of these changes are irregular but cannot be precisely quantified, d) overdeclarations of the areas claimed by farmers continued to be detected by the Court. | Point 5.10 a): Greece cannot accept the Court of Auditors’ observation on the low quality of on-site inspections, because it refers to isolated cases, and does not tally with the overall picture of the country. In any event, the fact that an on-site inspection report may have been completed incorrectly is not representative of the quality of the work of the inspection staff as a whole. Point 5.10 b): Greece points out that the methods of calculating area that are used in inspections are accepted methods in line with those contemplated by Ispra. In addition, methods are sometimes used which are not widely known but which can prove very useful, for example where the boundaries of a parcel are inaccessible and a conventional GPS measurement would be difficult. One such is the Simpson method, in which the area of a parcel can be calculated and a reliable result derived using only the number and coordinates of its vertices. Point 5.10 c): Greece informs that the Union of Agricultural Cooperatives (EAS) provides a computer service only: it enters the primary data. Its operators at all times work under the supervision and in the presence of the responsible civil servants of the Directorate of Rural Development. They carry out a straightforward registration of data, and have no power to process them further. The data are forwarded electronically to the Information Technology Directorate of the Ministry of Rural Development and Food, where the integrated system database is also kept. The software used for the management and implementation of the integrated system in the central database allows a check to be made for any changes to claims from 2004 onward: any change to any claim by any producer can be detected. The Directorates of Rural Development are themselves under the supervision of the OPEKEPE, and in order to make full use of the possibility in the central data base of keeping a record of corrections and changes made to claims, the regional directorates of the OKEPEKE have been sent a computer file that indicates particular producers and asks them to check the legality of any changes appearing in the electronic files in connection with claims and audit documents. Point 5.10 d): Greece states that the cases in which checks carried out by inspectors from the Court of Auditors on the farmers’ files identified discrepancies between the first-level check by the Directorate of Rural Development and the check by the Court of Auditors are isolated cases, and certainly do not provide evidence for the remark that there were “overdeclarations of the areas claimed by farmers”. |

5.11 | The area claimed and for which aid is paid to the farmer may be higher or lower than the area measured by the inspection as long as it lies within a certain technical tolerance. In Slovenia, however, if the area is overclaimed, the 4 % tolerance is incorrectly applied, leading to overpayments. | Slovenia informs that it applied the 4% tolerance in good faith based on its understanding of the rules. However, in order to determine empirically the risk to the Fund of using a 4% tolerance, the most risk-prone applications have been re-calculated. Based on these conclusions, Slovenia concludes that overpayment in these cases corresponds to 2.85% of the total direct area payment. |

5.13 | The Court's audit has shown that amongst the Member States audited which apply the Single Area Payment Scheme (SAPS) and have chosen to use production blocks as reference parcels, only in the Czech Republic does the LPIS contain eligibility data for agricultural parcels. In the others the LPIS only contains eligibility data for the reference parcels. These Member States did however not require the application to be supported with particulars and accompanied with documents that enable each agricultural parcel to be located and measured12 so that the information available to the authorities was not sufficient to ensure that every agricultural parcel can be reliably identified. This failure affected the ability and effectiveness of the checks to detect and prevent errors. ____ 12 As required for claim year 2004 by Article 4 of Commission Regulation (EC) No 2419/2001 (OJ L 327, 12.12.2001, p. 11) and as required by Article 6 of Regulation (EC) No 796/2004 for claim years after 2004. | No reply. |

5.14 | As a consequence, overdeclarations and/or double declarations of agricultural parcels go undetected as long as the reference parcel is not overclaimed. Parcels not claimed or claimed partly only by one farmer can compensate for double or overdeclarations made by another farmer so that the real extent of overdeclarations and/or double declarations inside a reference parcel cannot be correctly established. Therefore, aid reductions and penalties are either not applied or applied at too low a level13. For example, in Hungary and Slovakia this led to arbitrary reductions and sanctions being imposed in cases of reference parcel overclaims, where it was not possible to attribute the responsibility for the overclaim to any particular farmer. In none of the Member States visited was there an obligation to inspect all claimed parcels within the production block. ___ 13 In the absence of any information the financial impact of this failure cannot be established. | Hungary has described its administrative procedure designed to sift out unjustified claims where the physical block is overdeclared. Inter alia, where overdeclaration is less than 3%, the area of the declared parcels is reduced proportionally up to the net area of the respective physical block. Where overdeclaration is more than 3%, an on-the-spot inspection is carried out. In both cases, penalties are applied in accordance with the rules. Farmers have a right of appeal. |

5.15 | During on-the-spot inspections, overdeclarations and under declarations of parcels can be off-set against each other as long as the cadastral area is not exceeded. In Poland, the area measured in excess of the cadastral area of a parcel was used for off-setting area deficits found on other parcels. The financial impact of this error cannot be quantified. | Poland makes reference to the rules governing measurements of parcels. In the light of these rules, payments to areas greater than the area applied for may not be made in any circumstances. During the on-the-spot inspections the parcels declared by the applicants are measured and under no circumstances does offsetting take place for the total area of cadastral parcels. |

5.16 | Further problems were found in: a) Hungary: inspection results were not or not correctly taken into account for the calculation of the SAPS payment14; b) Poland: national top-up payments15 are made under more restrictive conditions than SAPS payments. When national criteria were not met, SAPS payments were reduced in the same way as national payments even though SAPS criteria were met. This is not in line with EU legislation. Moreover, the administrative cross-checks allow a tolerance for differences found between the claimed area and the eligible area registered in the LPIS, which is only permitted for parcels controlled on the spot; ______ 14 In a sample of 30 inspection reports analysed, representing claim payments totalling 23 800 euro, overpayments and underpayments totalling 3 122 euro due to inadequate analysis of inspection reports were found. The differences add up to 13 % of payments. 15 Regulation (EC) No 1782/2003, Article 143(c). | Hungary informs that it has provided the ECA with information on the corrective action being taken in respect of the findings. The rectification of the incorrect assessments and discrepancies is ongoing. |

Poland states that it – on the basis of existing rules and guidelines – assumed that it was not correct to apply a different system of penalties to complementary area payments (CAPs) and that different inspection methods should not be applied to areas declared as single area payments (SAPs). Therefore, if during the inspection an error comes to light making it necessary to apply a penalty involving the exclusion of the application from the aid scheme, such exclusion shall cover all the direct payments for which the applicant applied. In the light of the suggestion put forward by the auditors who carried out the inspection, ARMA has taken steps to amend the IT system and manual of procedures as regards the application and relationship of the system for calculating penalties for SAPs and CAPs. The technical tolerances applied in the system in administrative checks on eligible areas enable the recipients drawn by lot for on-the-spot inspections to be treated in the same way as those which were not. If the recipients have not been drawn by lot for on-the-spot inspections, the areas declared by them are checked only by administrative checks using potentially eligible areas. In similar cases, i.e. cases referred for inspection and cases not selected for inspection, where the technical tolerance is not exceeded, payment of the same amount corresponding to the declared area will be awarded. The same approach has therefore been applied for the whole inspection system, thus ensuring its cohesion. The solution applied does not represent any threat to the fund. |

5.17 | In 5 Member States16, evaluation of risk factors applied was not carried out or not documented for claim year 2004 and not applied to claim year 2005 (relevant for financial year 2006). ______ 16 The Czech Republic, Hungary, Slovakia, Slovenia and Malta. | The Czech Republic states that it does not agree with the finding that a risk assessment had either not been carried out or not documented for claim year 2004. The reasons for this were communicated to the ECA in February 2006. The Czech Republic also does not agree that it failed to apply the 2004 risk assessment to claim year 2005. This is documented and described in methodological guidelines. Following later reactions from the ECA, the Czech Republic believes the ECA now agrees with the Czech authorities and therefore requests point 5.17 to be revised. |

Malta informs that risk factors identified for the financial year 2006 were documented and are therefore being assessed. |

Slovenia informs that no detailed analysis of the results of on-the-spot checks was conducted in 2004. For 2005 Slovenia took into account of the criterion of infringements from previous years and included the most risk-prone farmers and regions in the risk analysis. The ARSKTRP control service drew up an analysis of the results of on-the-spot checks for 2005 and took it into account when drawing up the risk analysis for 2006. Apart from that, it is pointed out that the method of selecting a sample of applications for on-the-spot checks of direct payments to farmers for 2006 has been significantly improved. |

5.18 | Under SAPS if the total aid payable in a Member State exceeded the national financial envelope the aid per hectare had to be reduced by an equivalent proportion. The Member States had to inform the Commission of the rate of reduction by 30 November 2004. The Czech authorities were not able to substantiate the figure for the total area claimed that was used to calculate the reduction of the aid by 1,63 %. Hungary and Lithuania did not communicate the final reduction coefficient by the deadline. | The Czech Republic notes that information on the actual area, which is registered in the IS and for which aid was actually paid under SAPS 2004 (as on the day of audit), was submitted to the auditors. As the total area for SAPS in the Czech Republic exceeded the financial envelope for the given year, the Ministry of Agriculture fixed the reduction coefficient at 0.9837. This coefficient was fixed using the data in the database of applications for single area payments (SAPS) which is part of the information system of the State Agricultural Intervention Fund (SZIF). |

Hungary informs that the deadline was not met as the Hungarian authorities were not in possession of the full information (completely checked data) necessary to determine the redistribution rate. Inspections were delayed due to difficulties in the first aid year. Hungary sent a report to the Commission in May 2005 when inspections were fully completed. |

Lithuania confirmed that it was not possible to establish the exact reduction coefficient by 30 November 2004. Following guidance from the Commission, a reduction coefficient was established on the basis of information available at the time. Adjustments of payments to applicants were made when final checks had been completed. This was not a second payment but simply a correction in the level of the direct payment. |

5.19 | According to the regulations, SAPS aid is to be paid once, between 16 October 2004 and 30 April 2005. Hungary made two payments. There is no legal basis for this practice. | Hungary notes that the Commission has not replied in writing to a letter from Hungary, asking the Commission for its position. At the Management Committee, the Commission representative however did say that the relevant legislation did not seem to preclude the Hungarian interpretation. |

5.21 | The Commisssion's statistics for animal premiums show the number of animals claimed by farmers which inspectors found not to exist or not to be eligible for subsidy. For the largest scheme, the suckler cow premium (see Table 5.3 and Graph 5.4), Member States inspected 14,9 % of the animals claimed, finding 1,8 % of these to be missing or ineligible. Overall, this percentage for cattle shows small variations from Member State to Member State. However, for Italy, Malta and Slovenia it is very high (for the suckler cow premium in Italy, 11,4 % and in Slovenia, 48,2 %, and for the special beef premium in Malta, 11,8 %, in Italy, 21,8 % and in Slovenia, 56,2 %). | Malta informs that this discrepancy resulted from the fact that, for the financial year ending 2005, the number of bovine animals slaughtered before Malta's entry into the European Union was also taken into account. Subsequent checks have shown that the figures resulting from this discrepancy have now been corrected. |

Slovenia informs that the percentages mentioned for Slovenia by the ECA are taken from statistical reports which were sent to the Commission on 31 August 2005. On 17 July 2006 ARSKTRP sent the Commission a corrected statistical report, in which the percentages of unduly paid animal premiums were considerably lower. The errors in the report arose from a misunderstanding of the data required. In the report of 31 August 2005 ARSKTRP gave the percentage of animals for which reductions and exclusions were applied. In the report of 17 July 2006 it gave the percentage of animals for which irregularities were found. In the corrected report the proportion of animals ineligible for the suckler cow premium was 11.77% (previously 48.2%) and for the special beef premium 7.68% (previously 56.20%). |

5.22 | For the sheep and goat premiums the number of overclaimed animals decreased from 8,2 % in 2003 to 6,3 %. Italy and Slovenia report significantly higher levels of error for sheep and goat premiums (10 % and 24,1 % respectively) than the other Member States (1,2 %). The Court found problems in Greece, Spain, France, Netherlands and the United Kingdom. The flock registers are poorly maintained. The registers cannot be relied upon to confirm that the retention period requirements have been met or to reconcile the claim with the number of sheep found on inspection. | France has followed up on the findings made by the ECA and the Commission. Instructions to services have been made more precise. Sanctions are introduced if registers do not exist or are not well maintained. |

Greece points out, as regards the Court of Auditors’ observation that farm registers are not being kept properly, that the Greek authorities have taken all the measures called for in Regulation (EC) 21/2004, such as the tagging and registration of stock, the imposition of penalties, etc., so as to avoid any danger of undue payment of the premium, and also that a joint ministerial decision has been issued, No 263493/27-07-04, providing for supplementary measures for the implementation of a system of identification and registration of sheep and goat farms and their flocks in accordance with Council Regulation No 21/2004. |

The Netherlands notes that the ECA bases this conclusion on one random check, in which this failing was discovered on a single farm. To generalise from this one finding, strikes the Dutch Government as extraordinary. |

Slovenia informs that the percentages mentioned for Slovenia by the ECA are taken from statistical reports which were sent to the Commission on 31 August 2005. On 17 July 2006 ARSKTRP sent the Commission a corrected statistical report, in which the percentages of unduly paid animal premiums were considerably lower. The errors in the report arose from a misunderstanding of the data required. In the report of 31 August 2005 ARSKTRP gave the percentage of animals for which reductions and exclusions were applied. In the report of 17 July 2006 it gave the percentage of animals for which irregularities were found. In the corrected report the proportion of animals ineligible for the sheep and goat premium was 2.74 % (previously 24.1%). |

Spain notes that the ECA's findings concerned shortcomings in the keeping of the holding's farm register: the producer had not entered the movements of the animals in chronological order. However, this did not affect the findings of the inspection. |

The United Kingdom informs that it has encouraged, and will continue to encourage, claimants and keepers to maintain up to date and adequate records. The standard has improved over the years. However, where inspections uncover records which are non-compliant, the keeper/producer is informed and penalty or sanction action initiated as appropriate. |

5.23 | In Malta, aid reductions and penalties applicable to bovine and ovine premiums were systematically calculated incorrectly. | Malta informs that this error occurred because, when calculating the penalties, the Food and Veterinary Regulatory Division official used as the denominator the number of bovine and ovine animals declared by the official rather than the actual number that were found at the time of the checks. This resulted in a difference of MTL 10.20 (EUR 23.76). |

5.25 | The Court has examined nine olive oil production aid payments in Spain, Greece and Italy17 - these Member States have declared some 2 000 million euro. All of the transactions contained either an overpayment and/or one or more formal errors and the two cases in Italy were found to be irregular. The results indicate that there are serious weaknesses in control over the production aid scheme and also more generally with the reliability of the olive cultivation Geographical Information System (GIS). This system of aerial photographs is used to verify the existence of olive tree parcels. ____ 17 One transaction was also audited in Portugal. No significant errors were detected. | Spain cannot accept the ECA's doubts as to the reliability of the olive cultivation GIS as a monitoring instrument for aid management on the grounds of inadequate application of the technical tolerance. The Commission considered acceptable the adjustments made by Spain to the technical tolerance prior to the approval of the olive cultivation GIS in the various Autonomous Communities. The matter will therefore have to be settled between the two institutions. Also, the appropriate calculations were made in two Autonomous Communities applying the technical tolerance in accordance with the Court's guidelines. The results had no practical relevance as regards conflicting declarations and amounts paid so that similarly the risk for the Fund is minimal. |

5.26 | The GIS should have been fully operational from the 2003/2004 marketing year. The Commission was initially responsible for certifying the completion of the GIS but changed the Regulation so that Member States had to declare when the system was completed. Despite the fact that they have confirmed completion, the Court has found that in all Member States the failure to update the GIS data (the alphanumerical database, the real number of olive trees, new plantings and the production potential) clearly demonstrates that the GIS can still not be considered fully operational. This has significant implications because the olive cultivation GIS will be merged with IACS and be used as a basis for the calculation of payment entitlements and for the management and control of the new Single Payment Scheme (SPS)18 (olive growers will be paid on an area basis). ____ 18 Council Regulation (EC) No 864/2004 of 29 April 2004 (OJ L 161, 30.4.2004, p. 48). | Greece notes that after the completion of the geographical information system (GIS) for olive cultivation, procedures and submission dates were laid down for producers’ applications for modification of olive cultivation declarations or for new registration. These applications, which were submitted for the periods 2003/04 and 2004/05 via the producers' organisations, will be processed by contractors; the work has not yet been assigned, however, because of legal difficulties. In order to update the olive-cultivation GIS the Register Management Directorate has also developed an electronic application whereby the alphanumerical reference base is linked to the corresponding cartographic reference base. Interconnection has already been established with the Rural Development Directorates, which can enter changes in the olive cultivation declarations for the 2005/2006 growing period directly into the database. There are 781 117 producers registered in the olive-cultivation GIS database; 468 202 producers submitted applications for aid for the period 2003/04, of which 26 161 also submitted applications for modification of crop declarations. For the renewal of orthophoto maps, a tender procedure has been launched by the Topography Directorate of the Ministry of Rural Development and Food. Under Article 24 of Regulation (EC) No 2366/98, the Member State fixes the minimum size of olive-growing parcels below which no boundaries will be determined. That minimum size may not be greater than 10 ares. Greece points out that under the initial technical specifications of the project to set up the olive-cultivation GIS, which were approved by the JRC at Ispra, it was not compulsory to display small parcels (5 ares) in the cartographic base. However, following agreement with Ispra and the project's technical adviser, Mr Miranda, the technical specifications were modified with effect from May 2002, and the minimum area under olive cultivation for compulsory display, determination of the coordinates of boundary points in the GIS cartographic base and coding is 5 ares. Parcels of over 1 are and up to 5 ares are identified on the cartographic base by a solid circle/isosceles triangle 1.5 mm in diameter, over which the code number of the parcel is shown. As a result of this amendment to the technical specifications, 50% of parcels smaller than 5 ares are displayed in the cartographic base. The result of this change in the technical specifications was that 50% of parcels smaller than 5 ares are displayed in the cartographic base. The identification of parcels smaller than 5 ares is ongoing and will be completed when the registers are updated. |

5.27 | The average yield for the production zone is fixed by the Commission every year. It is difficult for the Commission to verify that the Member States use these averages appropriately to check the eligibility of the production claimed and do not make irregular payments (see for example Italy paragraph 5.25). | Greece informs that a check is made on the quantities of olive oil produced by all olive growers as compared with the yields of the homogeneous production zones. The Ministry of Rural Development and Food issued Ministerial Decision No 302158/11-10-04 determining the yields of the homogeneous production zones, and the criteria for the assessment of the quantities of olive oil accepted for assistance in respect of the period 2003/2004, having regard to the following: a) for yields, the recommendations of the committees on the delimitation of the homogeneous production zones in the Rural Development Directorates; b) the information on quantities of olive oil drawn up for each prefecture in the last 14 years; and c) Articles 15 and 16 of Regulation (EC) No 2366/98 concerning the quantity of olive oil qualifying for aid in accordance with the yields of the homogeneous production zones. The Ministerial Decision states that the coefficient which is multiplied by the yields of the homogeneous production zones in order to determine the quantities of olive oil accepted is to be equal to 2.4 (Article 16 of Regulation (EC) No 2366/98). If the quantity of olive oil claimed by a producer exceeds this limit (number of olive trees multiplied by the yield of the homogeneous production zone multiplied by 2.4), the producer is liable to further checks (under what is known as the incompatible producer procedure). |

5.31 | The financially most important measure in Poland is Less-Favoured Areas, for which 225,8 million euro was paid in 2005. One of the eligibility conditions for this measure is that farmers must apply usual good farming practices. The Polish rural development programme sets out verifiable standards for this, which are checked on the spot as part of the 5 % controls. These verifications found a high level of non-compliance with the standards; 3 281 farms (9,6 % of the farms checked) had one or more infringements. | Poland informed that in 2004, following on-the-spot inspections, 3057 farmers, representing 8.9% of all the farms checked, were found to have failed to comply with usual good farming practices (one or more infringements). All the farms subjected to on-the-spot inspections for compliance with usual good farming practices in 2004 at which infringements were noted were automatically referred for further on-the-spot inspections in 2005 and 2006. This enabled further checks to be made to determine whether those farmers who had received reprimands for failing to comply with usual good farming practices following the first on-the-spot inspection, i.e. in 2004, now met the requirements associated with usual good farming practices. Penalties were imposed on farms that were still not complying with the rules of usual good farming practices (114 farms in 2005 and 259 farms in 2006). |

5.32 | In accordance with Polish laws, in the case of a first infringement, the farmer only receives a warning and is not sanctioned. This was the case for all farmers, thus no recoveries or sanctions were applied. This is not in accordance with EU law: infringements of an eligibility condition of the Council Regulation22 should lead to reductions in the amounts payable. The total value of payments affected was 0,8 million euro. ____ 22 Council Regulation (EC) No 1257/1999 (OJ L 160, 26.6.1999, p. 80). |

5.34 | Given the time needed to finalise clearance procedures, the weaknesses found by the Commission during audits in Member States in 200424 have not yet been the subject of a final decision on whether or not a financial correction should be imposed. In 2005 problems were found relating to prior warning in Belgium. ____ 24 Annual Report concerning the financial year 2004, paragraph 4.12. | Belgium informs that it has reacted to the findings made by the ECA by notifying all involved parties of the appropriate procedure to use when declaring transports. |

5.35 | The Court's audit of physical and/or substitution checks in 11 Member States from 2004 to 2006 concluded that: a) the fundamental requirement that exporters should not have tacit prior warning of such checks at the point of loading had not been systematically complied with (Belgium, Denmark, Germany, France, Italy, Netherlands, United Kingdom); b) there were significant weaknesses in the verification of ingredients used in the manufacture of "Non-Annex I" goods such as biscuits, confectionery, whisky, etc. on which refunds are paid (Belgium, Germany, France25, Italy25, United Kingdom25); c) excessive numbers of physical checks had been carried out on consignments for which the refund claim was less than 200 euro26. According to the rules, these transactions should be excluded unless they are to prevent fraud and abuse; d) substitution checks are not carried out at the point of exit from the EU on significant numbers of consignments which have been customs-sealed despite not having been physically checked at the point of loading27. In particular, Denmark28 and the Netherlands allowed authorised exporters to affix customs seals themselves; e) as the physical presence of customs officials at the point of taking goods under customs control at the customs office of exit from the EU has been replaced by computerised inventory systems in the United Kingdom and certain French ports, the necessary checks that customs seals have not been broken or removed are no longer carried out. ______ 25 A posteriori verification of recipes under Commission Regulation (EC) No 1520/2000 (OJ L 177, 15.7.2000, p. 1). 26 Analysis of data reported to the Commission under Commission Regulation (EC) No 2390/1999 (OJ L 295, 16.11.1999, p. 1) shows that all EUR-15 Member States (with the exception of Greece and Spain, which did not provide the data, and Luxembourg, which had an insignificant number of transactions) had to varying extents carried out checks on such consignments. 27 Affixing customs seals to consignments which have not been subject to a physical check does not contravene current legislation. 28 Denmark abandoned this procedure from 1 February 2006. | Point 5.35 (a): Belgium finds that the controls were based on a very strict interpretation of the legislation (article 5 in Regulation (EC) 2090/2002). Nevertheless, guidelines for staff have been adjusted. Belgium points out that the issue is currently being discussed in the relevant management committee. Point 5.35 (b): Belgium notes that the ECA found certain weaknesses in the systems. It has taken action on to follow-up on these findings (e.g. better risk analysis and better management of data). |

Point 5.35 (a): France states that it ensures that checks are made without prior warning. Operators are only informed of checks when the goods are already in the control of the customs authorities. Point 5.35 (b): France says it cannot comment on this point as it does not know on what basis the ECA has made this observation. Point 5.35 (c): France includes other parameters than the size of the refund claim when deciding on which checks to make (e.g. is the operator new, do declarations seem abnormal). Making checks on small transactions also prevent operators from splitting transactions into smaller parts in order to avoid checks. Several services found this to be the practice of certain operators. Point 5.35 (d): France points out that on the basis of current Community law (Regulation (EC) 2090/2002) substitution checks are only foreseen when goods have not been sealed by a customs office. Point 5.35 (e): France notes that in certain French ports (e.g. Marseille and le Havre) a computer system containing detailed information on transports is a very useful instrument for the customs authorities. Customs officers are still present in the ports and perform checks. The customs authorities have set a target of checking 2 per cent of transports with a customs seal even though such a requirement is not set out in the Regulation (EC) 2090/2002. |

Point 5.35 (a): Germany finds that the ECA's observation does not give a correct picture and that the reservation expressed is not justified. Germany has established and explained in detail that the customs offices concerned conducted their checks on goods at the place notified by the exporter in accordance with Community law and Germany's implementing rules. Point 5.35 (b): Germany notes that the ECA checks did not reveal weaknesses and that the ECA described the German system of checks as good. Germany reacted to certain weaknesses concerning ex post inspections found in the past. During its on-the-spot checks, the ECA found that the structure worked well. Point 5.35 (c): Germany informs that it has detailed administrative rules on conducting physical checks. Under these rules, physical inspections of goods declared under Article 2(2) of Regulation (EC) No 2090/2002 are not to be included in the statistics on compliance with the minimum control rates. Customs offices regularly monitor compliance with the rules. Point 5.35 (d): Germany notes that Community law as it stands makes no provision for substitution checks of customs-sealed consignments presented intact at the customs office of exit. |

Point 5.35 (a): Italy stresses that the findings made by the ECA do not reflect the situation nationally. Point 5.35 (b): Italy notes that the failure to update the recipe register had no negative effects during the clearance stage as officials, before clearance, must ensure that the recipe given in the electronic system tallies with the information given in paper form. Point 5.35 (c): Italy points out that it only had one case of this type due to a customs office acting with a view to preventing fraud and abuse. The finding does therefore not correspond to the real situation on the ground in Italy. |

The Netherlands holds the opinion that controls carried out at the present time comply with Regulation (EEC) No 386/90. The Netherlands has acknowledged the risks pointed out by the ECA concerning customs seals. The risk is covered by including consignments sealed by authorised consignors as part of the population qualifying for substitution checks. |

Point 5.35 (a): The United Kingdom notes that the ECA’s auditors had initially considered that reports produced from the United Kingdom’s CHIEF system gave exporters prior notification of physical checks. Our understanding is that following a subsequent visit to the UK in September 2006 the auditors were satisfied that this is not the case. Point 5.35 (b): The United Kingdom understands that this relates to the procedures applied what manufacturers fail to retain adequate records to support the manufacture of goods on which refund is subsequently paid. The United Kingdom is currently considering the concerns raised by the Court. Point 5.35 (c): Apart from occasional checks to prevent fraud and abuses, the United Kingdom now excludes low value refund consignments from its programme of physical checks. Point 5.35 (d): The use of computerised inventory systems in the United Kingdom has not replaced the physical presence of customs officers but has allowed them to be deployed more efficiently. The United Kingdom is, however, aware of the Court’s concerns and is currently considering the necessary steps required to address them. |

5.45 | With respect to the paying agencies' internal control systems the Court's analysis of the reports of 31 paying agencies33 has shown that most of the major deficiencies found by the certifying bodies related to payment checks and procedures, accounting (including debtors) and delegation of functions34; four paying agencies were found with four or more major deficiencies relating to different accreditation criteria35. The Commission places reliance on the work of the certifying bodies by reviewing their reports and certificates but does not evaluate the work of the certifying bodies on the spot on a systematic basis36. _____ 33 The reports of the 31 certifying bodies from which transactions were tested (see Table 5.1). 34 The Commission has addressed the latter aspect by an enquiry started in 2002. Even though it found widespread weaknesses in the management of delegated bodies by AGEA (Italy) which have persisted, the Commission does not intend to impose a financial correction. Furthermore, no separate Article 8 letter pursuant to Commission Regulation (EC) No 1663/95 (OJ L 158, 8.7.1995, p. 6) has been sent to date in order to follow up the results of the mission to France carried out in 2003. 35 AGEA (payment checks and procedures, debtors, computer security, internal audit, delegated bodies), OPEKEPE (payment checks and procedures, debtors, internal audit, delegated bodies), ARDA (payment checks and procedures, accounting, computer security, delegated bodies), ONIFLHOR (written procedures, debtors, computer security, delegated bodies). The paying agencies concerned declared some 7 418 million euro which are to be further analysed by the Commission. 36 In 2005 only the certifying body for OPEKEPE was evaluated on the spot. In addition, as part of the Commission's review of accreditation ten missions were carried out to new Member States. | France agrees with the ECA that it is regrettable that the Commission has not sent a so-called Article 8 letter as a follow-up the results of the mission to France carried out in 2003. |

5.53 | CAP expenditure managed and controlled by IACS amounts to 25 500 million euro in EU-15 and to 1 400 million euro in the New Member States. IACS, where properly applied, is an effective control system for limiting the risk of error or irregular expenditure. However, this is not the case in Greece. In the New Member States audited by the Court systems are not yet reliable: in particular the information recorded in LPIS together with the information requested in the aid applications was found insufficient to systematically ensure the correct identification of agricultural parcels. This failure affects the efficiency of administrative controls and of on-the-spot inspections and measurement tolerances are wrongly applied. | Greece states that the criticisms and remarks in the Court of Auditors’ report relate to isolated cases, and Greece does not agree with the Court’s observation regarding the proper application of IACS in Greece. |

5.73 | The Court examined the implementation of the Slaughter Premium Scheme in the context of the follow-up of the Court's Special Report No 6/2004. The slaughter premium scheme is one of the premium schemes relying on the bovine identification system47. The premium is paid for the slaughter or export from the EU of adult cattle and calves. _____ 47 The Court examined its implementation by the Commission and in selected Member States (Germany (North Rhine-Westphalia), Spain (Aragon), France, Ireland, Italy, the Netherlands and the United Kingdom (England)). | Germany has followed up on the observations made by the ECA. |

Ireland makes reference to point 5.74 in the 2005 Annual Report where the ECA stated that "only Ireland and the United Kingdom (for animals slaughtered in Northern Ireland) had arranged to exchange information automatically between the relevant databases. |

The United Kingdom welcomes the ECA's comments which recognise that the United Kingdom and Ireland have set up procedures to exchange information automatically between their relevant databases. |

5.74 | The design of the scheme facilitates adequate control over payments in respect of animals which are slaughtered in the Member State of payment. However checks relating to the minority of animals (1,2 % on average in the regions examined) which are slaughtered in one Member State and for which a premium is paid in another are not as rigorous. The transfer of information between Member States relating to such cases is incomplete and does not ensure that all eligibility criteria for the premium are verified before payment. Germany and the Netherlands paid the premium in some cases where confirmation of slaughter from other Member States had not been received before payment. For the Member States visited, only Ireland and the United Kingdom (for animals slaughtered in Northern Ireland) had arranged to exchange information automatically between the relevant databases. | The Netherlands informs that the Member State in question limited itself to replying that the verification requests had been passed on to its competent authorities. When repeated requests failed to elicit a conclusive reply, the [Dutch] paying agency discontinued its efforts to obtain information concerning 2002. The Dutch paying agencies honour applications for premiums for bovine animals slaughtered in other Member States only if satisfactory proof is provided that all conditions have been met, which includes presenting the proof of slaughter issued by other Member States. |

The United Kingdom refers to its comment to point 5.73. |

Table 5.2 | The Netherlands notes that the table shows Netherlands with a relatively high rate of applications with errors. It is pointed out that: a) The Netherlands has good risk analysis arrangements which increase the likelihood finding anomalies; b) The high error rate has prompted the Netherlands to intensify its field inspections, c) Checks on whether land was actually used as arable land may have been reported differently by other Member States. |

Slovenia points out that data concerning Slovenia has been significantly revised. For instance, the share of applications with errors has been revised down from 53.5 per cent to 1.8 per cent. |

CHAPTER 6 – STRUCTURAL OPERATIONS |

6.15 | The Court found certain types of error appearing systematically in the projects for certain programmes. Examples are: - ERDF Objective 1 United Kingdom (South Yorkshire): declaration of EU grant exceeding the co-financing rate without proper justification; - ESF EQUAL Community Initiative Spain: inadequate supporting evidence for the expenditure declared because the training organisations in receipt of funding failed to clearly identify the ESF activity in their accounting systems; - FIFG France: the verification and checking activities carried out by the Direction Départementale des Affaires Maritimes prior to the payment of an invoice were not based on adequate supporting documentation. | France disagrees with the ECA's analysis. It was regrettable that not all documentation seemed to be available at the DDAM but the full dossier was available at the Fisheries department. France has forwarded additional information to the ECA. This cannot be considered a systemic error. |

The United Kingdom informs that many of the areas/recommendations of the ECA's findings have been addressed, withdrawn or further clarification sought. It states that although it supports strict financial control, it is disappointing that South Yorkshire should receive such adverse comments on the management of ERDF. It should be recognised that most errors are usually small amounts, paid as ineligible expenditure (rather than fraud against the EC budget), most of which are corrected later. |

6.26 | The Court found a material level of errors in the project expenditure declarations across all its sample of programmes for the 1994 to 1999 period2. Of the 65 projects examined, 33 were affected by material error. The most frequently occuring errors were similar to those found in the sample of current programmes (paragraph 6.14 above). ____ 2 The programmes audited were: ERDF Objective 1 programmes in Spain, Ireland (Industrial Development) and Portugal (Regional Development); ESF Objective 1 Germany (Sachsen), ESF Objective 3 Sweden and EAGGF-Guidance Objective 5a Denmark (Improving Efficiency of Agricultural Structures). | Germany has taken on board or already implemented the ECA suggestions. However, it points out that, for the most part, the auditing standards used for the 1994-1999 period were first laid down in connection with Regulation (EC) No 438/2001. |

Ireland has made a description of its setup of Enterprise Ireland (EI) which was one of the bodies responsible for the implementation of the R&D measure under the Industry OP 1994-1999. It is inter alia mentioned that EI clearly stated eligible expenditure in its grant agreements with project promoters. EI project officers visited project promoters to inspect all invoices. Every expenditure claim was supported by an independent Audit Certificate and other required documentation. Procedures for payment and monitoring of projects existed. |

Portugal does not support all the findings of the ECA. Portugal disagrees in about half the cases and either fully or partially agrees in the remaining cases. |

Sweden does not share the Court's finding concerning a material level of errors as regards the Swedish programme inspected by the Court and was prompted by the Court's comments to perform a comprehensive follow-up audit, according to which the Court's criticisms are largely unfounded. The substance of many of the supporting documents which according to the Court of Auditors were lacking was identified in the follow-up audit. Sweden considers that most of the Court's observations may have been prompted partly by misconceptions as to how the Swedish administrative system operates and partly by the fact that not all of the supporting documents were available at the time of the Court's visit. Given the fact that several years had elapsed since the project was concluded, the supporting documents concerned were not always to be found with the beneficiaries but were held in the archives of the competent authority. In most cases, Sweden has therefore been able to produce the supporting documents and only in a few cases identified shortcomings of the type referred to by the Court. |

6.27 | For three programmes in the audit sample (ERDF Objective 1 Ireland (Industrial Development), ESF Objective 1 Germany (Sachsen) and ESF Objective 3 Sweden) some of the beneficiaries selected for audit had not retained the supporting evidence for their declared expenditure3. ____ 3 The Commission has proposed to relax the provisions concerning the time period for retention of supporting documents for the next programming period (2007 to 2013). In its Opinion No 2/2005, the Court has pointed out the serious limitations this would place on its audit prerogatives and on the scope for checks on programmes at the end of the period. | Germany informs that in four cases no supporting evidence was provided in the time available. The authorising authority has since received the files relating to three of these cases. The ECA was informed of this situation and other details regarding the cases. The fourth case occurred because of insolvency that arose after the funding conditions had been met or the funding agreement had expired. In this case, the Structural Funds Regulation (stating that all documentation must be kept under this regulation) and German bankruptcy law (stating that the insolvency practitioner has power of attorney over all documents) are in conflict. Therefore problems cannot be ruled out when the ECA or the Commission conducts ex post checks in relation to the clearance procedure. However, Germany believes the retention period required had already expired when the checks were conducted [it seems Germany disagrees with the ECA's interpretation of Article 23 of Regulation (EEC) No 2082/93]. |

Ireland points out that Community law says documents should be available for three years but it is not clear when this period starts. Regulation (EEC) No 2082/1993 defines the period following the last payment in respect of any operation which the Irish authorities have understood to be "any project" and not "any programme". The final payments in question were made in 1999 or earlier, and Ireland therefore does not believe that the final beneficiaries concerned were legally required under Community law to retain supporting documents until now. |

Sweden states that it may be said that the said documents were lacking in some few cases. Many of the supporting documents that the Court expected to obtain from the beneficiaries were held by the competent authorities instead. In Sweden's opinion the Regulation then in force (Article 23(3) of Regulation (EC) No 2082/93) cannot be interpreted to mean that the supporting documents concerned necessarily had to be stored with the individual beneficiaries for that period. |

6.28 | Similarly to the 2000 to 2006 programmes, certain errors also occurred systematically in the projects audited for certain individual 1994 to 1999 programmes. Examples are: - ERDF Objective 1 Ireland (Industrial Development):unclear allocation of overheads between co-financed and non co-financed projects; - ESF Objective 3 Sweden: declaration of costs for services provided by companies related to the project promoters, but with no underlying justification of the amounts charged; - EAGGF-Guidance Denmark: failure to inform the farmers that the aid was co-financed by the EU and therefore subject to specific eligibility rules. | Ireland admits that at the time of the audit some document had been discarded and in some cases it was therefore not possible to provide explanation of the overhead allocation methodology and to justify overhead charges. However, Ireland is satisfied that in all cases precise and prudent methods of overhead calculation in accordance with best practice were used as the calculation methodology had been certified by an independent firm of auditors. |

Sweden does not share the Court's criticism that errors occurred systematically. In its follow-up audit, Sweden found that most of the shortcomings referred to by the Court did not exist and that neither had there been any systematic errors. Sweden does not accept the example quoted by the Court. The files of the decision-making authority show that the supporting documents concerned were requested, received and authenticated. Justification for the amounts charged therefore existed. However, Sweden notes that in purely general terms it can be said that there are risk factors in cases where services are provided by companies related to the project promoters and that these have to be managed by the competent authority. In this respect, Sweden therefore intends to scrutinise the example concerned more intensely. |

6.30 | The Court conducted a limited examination of nine Cohesion Fund projects which led to no material observations. However, the Court also noted that the annual declaration of the Director-General of Regional Policy for 2005 includes a reservation on the supervisory and control systems for the Cohesion Fund in Spain (the principal beneficiary of the Cohesion Fund). | Spain informs inter alia that as regards the reservation expressed by DG Regional Policy, the relevant bodies in Spain (including the Government Audit Department) are in ongoing contact with that DG and are taking measures to resolve the matter. |

Table 6.1 | Greece notes that on the basis of its audit of expenditure cofinanced by the ERDF as part of the operational programme for competitiveness, the Court of Auditors finds in its report (Table 6.1) that there are systematic shortcomings in the progress made with sample checks by the Financial Audit Committee (EDEL). Greece does not agree with the Court's finding and has provided the Commission with details on why it rejects the Court's finding. |

Table 1.c. Agriculture policy |

Paragraph | Member State | Observation | Reply |

CHAPTER 9 – PRE-ACCESSION STRATEGY |

9.8 | For Sapard, the Court's audit comprised tests of control and tests of transactions in Bulgaria and Romania on five projects in each country, covering the main measures implemented at the time of the audit. These audits involved verifying the project documentation in the respective paying agencies and the legality and regularity of the projects on-the-spot. The audit also involved an appraisal of the supervisory and control systems including the Commission's review of the certifying body's reports. | Romania informs that – during its future audit missions - the Romanian Certifying Body shall take into account the risks identified by ECA and shall assess the way in which the implementing authorities ensure the fulfilling of recommendations. |

9.10 | The Sapard transactions audited by the Court were affected by significant errors. The Court found that changes in procedures for approving Sapard projects in Bulgaria had not been given the necessary prior approval by the Commission. Sapard expenditure of 3,4 million euros in respect of 53 projects was therefore ineligible. In both Romania and Bulgaria, the Court found projects where beneficiaries did not comply with their contractual obligations. In particular, one project audited in Romania was found to have been ineligible for EU financing because the essential conditions governing the use of the grant aid facilities were not met. Furthermore, certain requirements had not been adequately checked by the paying agencies leading to cases of acceptance and payment of projects without adequate evidence to ensure compliance with eligibility criteria. | Bulgaria cannot agree that the SAPARD transactions are affected by significant errors, neither in number nor in individual value. Some bona fide technical errors occurred at early stages in implementation but were rapidly identified by the combined vigilance of the National Authorising Officer, National Fund and SAPARD Agency. The application of the “limited time procedure” does not affect the eligibility of the projects. All approved projects have been subject to the standard accredited eligibility checks at the implementation level and further controlled dully before payment. Accordingly there are no significant risks resulting from the application of the “limited time procedure” and the respective expenditure is eligible and in compliance with the Multi-Annual Financing Agreement. |

Romania informs that the Paying Agency for Rural Development and Fishery has analysed the situation of the project mentioned by the ECA. The Paying Agency considers that the explanations received from the beneficiary are reasonable and in accordance with the accreditation procedures. Considering that some corrective and preventive measures were already taken by the Romanian authorities for diminishing the risks identified within this report, the Directorate of Public Internal Audit within the Ministry of Public Finance will reassess all areas perceived with high risk and, if the case may be, will introduce into the 2007 audit Annual Plan the necessary audit engagements |

9.15 | With the granting of EDIS in the new Member States, specific implementing agencies were accredited to implement the Phare programme without the Commission's ex-ante control of tendering and contracting. However, the Court found cases9, where the accredited implementing agencies were not the actual contracting authorities and did not always exercise the type of control foreseen in the EDIS accreditation. Nevertheless, the process of implementing EDIS was a major step towards improving management capacity and control systems in the new Member States. ____ 9 Latvia, Poland. | Poland points out that the competent bodies involved in the ongoing introduction of the EDIS provided detailed explanations on the recommendations and instructions sent by the European Commission in respect of the audit carried out from December 2005 to January 2006 to check that the EDIS had been correctly introduced in Poland. As a result the European Commission confirmed the explanations provided by the Polish authorities and stressed that all bodies involved with the EDIS should function properly and that the National Authorising Officer and the European Commission should be informed of any changes of substance in the system for implementing Phare programmes and transitional measures. |

9.16 | The EDIS accreditation process for Phare and ISPA in Bulgaria and Romania was further delayed. Romania submitted its EDIS application at the end of 2005, while Bulgaria had not submitted an application by that date. Both countries had planned to have EDIS in place by January 200610. Furthermore, in the case of ISPA, the Court noted a combination of incompatible management and control functions within the Ministry of Finance in Bulgaria, while the Managing Authority for Infrastructure as National ISPA Coordinator in Romania was inadequately resourced. ___ 10 See Annual Report concerning the financial year 2004, paragraph 8.18. | Romania informs that the EDIS accreditation for ISPA and PHARE programmes proved to be a very complex and time demanding process. In order to have a sound financial management system it was necessary to strengthen the existing ISPA and PHARE bodies through developing procedures, hiring additional staff, staff training, strengthening the control functions, etc. and this process took more time then initially forecasted. Romania notes that the European Commission issued on the 28th of June [2006] the decision for waiving the ex-ante control in the case of management of ISPA Programme. In the case of PHARE, the Romanian authorities are in the final stage of implementing the verification audit recommendations. Romania says that the Managing Authority for Infrastructure (MA INFRA) has strengthened its capacity and currently operates with an increased number of personnel. MA INFRA has 66 positions allocated, out of which, by October 2006, 42 job positions were filled. The recruitment contest for the remaining vacancies will take place by the end of the year 2006. |

9.17 | The Court's audit in Bulgaria and Romania found that the Sapard systems included the key concepts but their functioning showed weaknesses: (a) the documentation underlying public tenders in Romania did not ensure that bids were received within the deadlines and were duly examined; (b) no appropriate supporting documents were provided for the reimbursement of certain amounts11 to the beneficiaries, which meant that the paying agency was not able to ensure whether they had been really incurred and/or were eligible (Romania). In Bulgaria, some expenditure is reimbursed (by Sapard) on a flat rate basis. The precise composition and justification of these amounts could not be established; (c) systems to check the reasonableness of prices (such as a price data base) affecting eligible amounts were not yet in place in Romania and they were not fully operational and documented in Bulgaria; (d) some invoices for high value items were settled by the beneficiary in cash (Bulgaria), which is more difficult to verify and therefore presents a higher risk. ____ 11 Labelled as "actualisation costs" and "costs for unforeseen circumstances". | Point 9.17 (b): Bulgaria notes that regarding the observation on reimbursement on a flat rate basis, payment authorisation is given for the lower between contracted and invoiced values. Flat rates are applied only at the implementation level before project approval. Therefore at the stage of payment, at most only invoiced amounts will be co-financed by the EU; Point 9.17 (c): Bulgaria notes that the initially accredited procedures of the SAPARD Agency through the Acts of Accreditations of the Competent Authority 2000 and 2003 were the basis for the Decisions of the European Commission conferring management of aid to the Bulgarian Authorities. In these procedures the SAPARD Agency has envisaged all necessary controls to ensure compliance with MAFA requirements: completeness and legitimacy controls, private procurement (3 offers) control, eligibility and financial criteria compliance controls, on-the spot controls, business plan analysis and project viability control. In addition, following the recommendation given within the first Conformity clearance procedure, in 2004 the controls performed by the SAPARD Agency were further elaborated as to reflect the Commission Auditors’ comments. The SAPARD Agency introduced modifications in their procedures, namely additional controls aiming at strengthening their compliance with the principle of sound financial management – modification of the three offers checks for private procurement over 10 000 euro setting up that detailed justification would be required in all cases where the chosen offer is not the least expensive one, introduction of a new procedure for comparing the prices of the assets applied for with the information from the SAPARD Agency’s referent prices database. These modifications were presented to the Commission in May 2004 and were acknowledged as relevant corrective measures of the Bulgarian Authorities within the conformity clearance procedure. Point 9.17 (d): Bulgaria informs that Bulgarian legislation does not impose limits on values that can be settled in cash. Additionally SAPARD regulations do not impose any further limits on types of payments. The SAPARD Agency executes additional accounting and documentary checks on the beneficiaries, which supplement the standard accredited administrative checks on invoices and cash receipts in order to verify that amounts paid in cash are executed in conformity with the Bulgarian legislation. |

Romania informs that similar findings were also presented by the Romanian Certifying Body in its 2005 Audit Report. For all findings, the Certifying Body issued recommendations to be implemented by the Sapard Agency. Point 9.17 (a): The Sapard Agency took a series of actions, inter alia developing an improved manual of procedures for the public procurement of works, goods and services based on the European Commission manual. It also took action in cases where conflicts of interest had been found. Point 9.17 (b): Romania has included a detailed description of its practices concerning supporting documents. On a specific case found by the ECA, Romania believes that the necessary legal documents for justifying the unforeseen costs exist. Point 9.17 (c): Romania informs that a price data base had to be developed – a process which proved very time consuming. In August 2006, the created data base was tested. |

9.18 | In the Court's Annual Report concerning the financial year 200412 it was highlighted that the quality of the assessment of the Sapard applications, and the related payment claims, carried out at the end of the contracting period presented a higher risk due to time pressure and possible staff shortages. The importance of this risk was confirmed by the situation found in Hungary in 2005, where the procedure for project approval was modified by reducing the checks on project applications. The changes were applied from April 2004 without the Commission having granted its prior approval. The relevance of this risk is therefore reiterated. ___ 12 Paragraph 4.70. | Hungary notes that the error committed consisted in only informing the Commission with considerable delay of the changes in procedure. Organisational and Operational Rules in force as of May 2005 clarify the relevant functions and who is responsible for what, so the risk of any repetition of such a case has been averted. The Commission accepted this explanation provided by the Hungarian authorities. |

9.21 | The Court has published a Special Report relevant to the pre-accession strategy in the last twelve months concerning the Phare Investment projects in Bulgaria and Romania (Special Report No. 4/2006). It can be found in the Court's website (www.eca.europa.eu). | Romania refers to the Commission replies in the special report. It notes that, regarding the industrial parks, new jobs were created and a higher degree of usage is anticipated for the end of 2007. Experience related to the management of the industrial parks will improve in time. |

2 . THE 2005 STATEMENT OF ASSURANCE

The European Court of Auditors is obliged by the Treaty to provide the European Parliament and Council with a statement of assurance as to the reliability of the accounts and the legality and regularity of the underlying transactions (also known as ‘the DAS’).

As regards the legality and regularity of the underlying transactions, the Court for the 2005 DAS based its audit opinion on “a consolidation of the specific assessments concerning own resources and each of the six operational chapters of the 2000-2006 financial perspective[3]. These specific assessments […] are based on four sources of evidence:

(a) An examination of the way in which the supervisory and control systems set up both in the Community institutions and in the Member States and third countries work;

(b) Testing of samples of transactions for each major area by carrying out checks down to final beneficiary level;

(c) An analysis of the annual activity reports and declarations by the Directors-General and of the procedures applied in drawing them up;

(d) Where possible, an examination of the work of other auditors who are independent of Community management procedures.” [4]

In the following, a more detailed analysis of the Court’s findings underlying the 2005 DAS is presented.

Section 2.1 presents a quantitative analysis of findings underlying the 2005 DAS and Member States' reaction to these findings whereas section 2.2 examines the timetable for exchange of information between the Court and the auditees (Commission and Member States).

2.1 DAS FINDINGS

The Commission is responsible for the implementation of the budget, but in practice not only the Commission manages EU funds (centralised management) but also Member States (shared management), third countries (decentralised management) and other international organisations (joint management). The vast majority of funds are implemented by Member States[5].

Controls exist at each level of management in order to ensure that payments to the final beneficiaries are legal and regular. In order to verify that controls are in place and working well, the Court goes on the spot to audit procedures and transactions in the Commission, authorities in Member States as well as final beneficiaries in Member States and third countries.

The findings made by the Court are split into four groups, each relating to the sources of evidence used by the Court when establishing its DAS opinion:

(1) Weaknesses in the supervisory systems and controls - these findings cover issues such as lack of 5 % controls for structural actions, use of inappropriate sampling methods, lack of risk analysis, lack of documentation of checks made, unjustified delays in closure of programmes and inappropriate set-up of roles and responsibilities among authorities performing controls.

(2) Formal findings - cases where the Court concludes that for instance tender procedures have not been respected, publicity of EU participation is inadequate or payments were late. The irregularity does not have any financial impact – or it is not possible to quantify such an impact - on the EU budget.

(3) Substantive findings - cases where the irregularity is considered to have a financial impact on the EU budget. Typical examples are the farmer who declares a too big area for a field or the manager of a structural fund project claiming ineligible expenditure.

(4) Other findings – these may relate to the accounts or they are other observations concerning for instance the declarations made by Directors-General in the annual activity reports.

Category (1) findings relate to pillar (a) in the Court's method, i.e. the examination of supervisory and control systems. Category (2) and category (3) findings relate to pillar (b), i.e. substantive testing of representative samples, whereas category (4) findings cover all other issues.

The Court's findings vary in importance. It is not possible to quantify the relative importance of formal findings or findings concerning the supervisory systems and controls. However, for substantive findings, the importance is measured by the potential financial impact on the EU budget.

The real financial impact on the EU budget may be less than the potential, either because the Court may – after having received further information and documentation from the auditee – consider that proceedings had been legal and regular after all, or because the Commission or the Member State may recover the amount in error or introduce other corrections (exchange of ineligible projects/expenditure for eligible projects/expenditure, providing missing documentation etc.).

Depending on the circumstances, the real financial impact may not be possible to calculate until several years after a finding was made. As the Court is obliged to deliver an annual opinion and therefore cannot await final confirmation of the real financial impact, it must base its opinion on the estimated potential financial impact at the time of the audit.

As an input to the 2005 DAS, the Court calculated an error rate for each substantive finding showing the amount estimated to be in error in relation to the total value of the transaction audited.

In many cases, such error rates are easy to calculate. If the Court measures a field and finds the area is 3 per cent smaller than declared by the farmer, then the error rate is considered to be 3 per cent. Likewise, if the Court considers expenditure claimed on a structural fund project to be ineligible, then the error rate is calculated as the ineligible amount as a percentage of the total amount paid out from the EU budget.

In other cases, it is more difficult to assess the error rate, e.g. when documents are missing at the time of the audit. In such cases, the Court often sets the error rate to 100 per cent. If the auditee later submits the documents to the Court and the Court finds that these documents justify fully or partially the payments made, the Court may withdraw the finding or reduce the error rate.

The Commission invited Member States to state their position on the findings underlying the 2005 DAS in order to compile the present report[6]. As regards substantive findings, Member States replied on around 60 per cent of the findings underlying the 2005 DAS. Germany declined to reproduce its replies to the Court's statements of preliminary findings[7], Italy provided replies to only a small number of findings and Spain provided replies to the majority but not all findings.

The lack of replies to some findings seems to be an effect of some Member States not being able to collect the information within the short deadline set by the Commission, indicating the need for appropriate follow-up procedures. The absence of a reply to findings for the purpose of this summary report should not be interpreted as acceptance of the Court's findings, as the Member States had in some cases contested them during the contradictory procedure.

The response rate for this report, measured as the share of findings replied to, was much higher for agricultural policy (almost 90 per cent) than for structural actions (around 50 per cent). Member States did not seem to systematically reply to small findings rather than big findings measured by value of transaction or value of amount considered to be in error.

In the case of substantive findings attributed to Member States within shared management, error rates concerning 2005 DAS findings were – as a rule – significantly lower for findings relating to agricultural policy than for findings relating to structural actions, cf. table 2.b .

Table 2.b. Distribution of error rates for substantive DAS findings relating to agricultural policy and structural actions in 2005. Total = 100 for each category.

Error rate |

Less than 2 per cent | 2 - 10 per cent | 11 - 99 per cent | 100 per cent | Total |

--------------------------------------- Per cent --------------------------------------- |

Agriculture policy Of which: Finding relating to size of area | 24 28 | 45 63 | 32 9 | 18 0 | 100 100 |

Structural actions Of which: Finding relating to missing documentation | 13 7 | 7 1 | 36 32 | 44 60 | 100 100 |

Note: All substantive findings relating to Member States are included.

Substantive findings with low error rates (less than 2 per cent) were twice as frequent within agricultural policy as within structural actions. The opposite was the case for substantive findings where the full amount of the transaction audited was considered to be in error (i.e. errors rated at 100 per cent) which were more than twice as frequent within structural actions as within agricultural policy.

Within structural funds, around 3 in 4 of the substantive errors with errors rated at 100 per cent related to cases where the Court did not find the necessary documentation available when visiting the final beneficiary on the spot.

Member States accepted findings made by the Court more frequently within agricultural policy than within structural actions, cf. table 2.c. The difference between the two sectors was however not as pronounced as in the case of the 2004 DAS findings.

Table 2.c. 2005 and 2004 (figures in brackets) substantive findings accepted or not accepted by Member States. Total = 100 for each category.

Member State accepts, even if only partly | Member State does NOT accept | Total |

--------------------------- Per cent ---------------------------- |

Agriculture policy Of which: Finding relating to size of area | 67 (85) 83 (98) | 33 (15) 17 (2) | 100 100 |

Structural actions Of which: Finding relating to missing documentation | 54 (47) 38 | 46 (53) 62 | 100 100 |

Note: "Findings relating to missing documentation" were not identified in the analysis of the 2004 DAS findings.

Within agricultural policy, 6 in 10 of the substantive findings related to the final beneficiary having declared a bigger area than measured by the Court during its visit. Member States tended to accept these findings to a large extent.

Within structural actions, 5 in 10 of the substantive findings related to the Court not finding the appropriate documentation (audit trail) during its visit to the final beneficiary. Member States tended to not accept these findings, either because documents had since been produced, or because the auditee believed that the Court's audit had taken place after the deadline for keeping documents as defined by the legislation.

2. 2 EXCHANGE OF INFORMATION BETWEEN THE COURT AND THE MEMBER STATES

When the Court has completed an audit mission to a Member State, it analyses information collected and sends a so-called statement of preliminary finding (‘SPF’) to the supreme audit institution in the Member State which coordinates a reply with the relevant national authorities. The Court usually invites the recipient to reply within a deadline of 6-8 weeks. When the recipient has replied, the Court in most cases sends another letter with its analysis of the reply, indicating if the findings are maintained.

The Commission services closely follow and assist in this contradictory procedure with the Member State and seek further information to elucidate difficult cases. For public procurement and state aid issues, the opinion of the competent specialist departments or the Legal Service sometimes has to be obtained.

The Court bases itself on the findings when assessing the legality and regularity of the transactions underlying the accounts. The final assessment is presented in the Annual Report which – concerning the financial year 2005 – had to be published by 31 October 2006 at the latest[8]. Due to time needed for translation of the report and the contradictory procedure with the Commission, the Court already adopted its first draft of the report in May 2006.

The Court must ensure that audit missions are carried out and letters sent in due time before the preparation of the Annual report. This is complicated, not least because time is often needed for translating supporting documents as well as letters and analyses.

An analysis of the SPFs containing substantive findings underlying the 2005 DAS shows that the Court carried out 7 in 10 of its missions to Member States in 2005 with the remaining missions taking place up to April 2006, cf. figure 2.a. The Court sent its last SPF to Member States in July 2006, i.e. one month after the Court had adopted its draft Annual report. On average, the time between the Court's last mission underlying an SPF and the actual transmission of the SPF was almost 4 months.

The different shades of grey in figure 2.a indicate the extent to which it was possible in practice for the Court to take into account replies from Member States in the final Annual Report. The darker the shading, the less possible it was.

Figure 2.a. Distribution of missions, SPFs and Member States' replies concerning SPFs

[pic] |

Note: The figure is based on data for all statements of preliminary findings (SPFs) sent by the Court to Member States in relation to all findings underlying the 2005 DAS.

Around 3 in 10 SPFs were sent by the Court to Member States in May 2006 or later, i.e. at the same time or later than the adoption of the draft report. The late transmission of SPFs meant that only one quarter of the Member States' replies were available before the adoption of the draft report. At the end of June - when the final phase of the contradictory procedure began – about half the replies were still outstanding.

Member States replied to the vast majority of SPFs sent by the Court. They respected the deadline of replying within 2 months in about half the cases and submitted most of the remaining replies within an additional month, cf. table 2.d . On average, Member States needed 66 days to prepare a reply to the Court.

Member States’ performance was best for SPFs concerning agricultural policy, followed by own resources and lastly structural actions. A reply was prepared within 2 months in almost 6 in 10 SPFs concerning agricultural policy, whereas this was only the case for 3-4 in 10 SPFs concerning structural actions.

Table 2.d. Distribution of replies from Member States/Commission to the Court concerning SPFs underlying the 2005 DAS and the 2004 DAS (figures in brackets).

60 days or less | Between 60 and 90 days | 90 days or more | Total | Average |

------------------------------- Per cent -------------------------------- | Days |

Member States Agricultural policy Structural actions Own resources | 49 (36) 59 (50) 35 (25) 43 (…) | 42 (50) 35 (50) 53 (50) 43 (…) | 9 (14) 6 (0) 12 (25) 14 (…) | 100 100 100 100 | 66 (71) 62 (64) 69 (76) 70 (…) |

Note: The table is based on calendar days between the date in the Court’s SPF letter and the date in the Member States’/Commission’s letter of reply. Last year’s analysis of SPFs did not include data concerning own resources.

[1] The report was published in the Official Journal C 263 of 31.10.2006. It is available on the Court’s website: www.eca.eu.

[2] Article 143(6) in The Financial Regulation applicable to the general budget of the European Communities, Council Regulation (EC, Euratom) No 1605/2002 of 25.6.2002.

[3] Agriculture, structural measures, internal policies, external actions, administration and pre-accession strategy.

[4] Point 1.59 in the 2005 Annual Report.

[5] In the financial year 2005, almost 80 per cent of EU funds were managed by Member States.

[6] In accordance with article 143(6) in the Financial Regulation applicable to the general budget of the European Communities, the Commission sent a letter to each Member State (including the two acceding countries Bulgaria and Romania) on 23 October, inviting them to: 1) comment on observations made by the Court in the 2005 Annual Report concerning the Member State in question, and 2) inform whether the Member State agreed fully, partially or not at all with the specific DAS findings made by the Court in that country. Member States were also invited to inform whether they had taken any action to follow up on the finding made by the Court.

[7] Germany did not reply on the individual findings, on the ground that the Commission had already received the replies that were sent by the German National Supreme Audit Institution to the Court at an earlier stage.

[8] Following the latest revision of the Financial Regulation, the Annual Report shall be published at the latest on 15 November, i.e. the Annual Report concerning the financial year 2006 will be published by 15 November 2007 at the latest.