52014DC0060

REPLIES OF THE COMMISSION TO THE SPECIAL REPORT OF THE EUROPEAN COURT OF AUDITORS

"THE RELIABILITY OF THE RESULTS OF THE MEMBER STATES' CHECKS OF THE AGRICULTURAL EXPENDITURE"

Executive Summary

Common reply to paragraphs I –IX:

The ECA has, for a number of years, criticised the Commission's reliance on the Member States' statistics on the results of their controls as a source of information for calculating a residual error rate.  In the first place, the Commission was aware that it could not place full reliance on the control statistics from the Member State as the only indicator of the level of error in agriculture spending.  This is why, for many years and until the 2012 Annual Activity Report (AAR) it had applied a 25% "safety margin" to the residual error rate calculated on the basis of the statistics.  This was regularly contested by the Court. Secondly, in its 2012 AAR, DG AGRI took on board the concerns about the control statistics and changed its approach after consulting with the central services of the Commission responsible for methodological issues.  The Court, in its annual report for 2012, did not criticise this approach and rather, expressed itself positively thereon and recommended that a similar approach be applied for the EAFRD.

Moreover, the new horizontal regulation on the control, financing and monitoring of the CAP in force as from January 2014, provides for a significant increase in the responsibilities of the Certification Bodies in the Member States with regard to delivering an opinion on legality and regularity of the expenditure including validating the member States' control results which are communicated to the Commission. 

In consequence, the Court's report deals with a methodology which has changed and is still evolving. Rather, therefore, than concentrate on historic elements, the Commission would prefer to concentrate on the present and future and on the establishment, including with the constructive input of the Court, of a sound methodology.

III. For its 2012 Annual Activity Report DG AGRI has developed a method for calculating the residual error rate that also takes into account its own audits and those of the Court. This method has been used for direct payments (around 70 % of the CAP expenditure) in respect to financial year 2012 and will be extended from 2013 financial year to other CAP expenditure and in particular, to rural development.

IV. The establishment of control systems is governed by precise EU rules, complemented by Commission guidelines as appropriate.  For instance, paying agencies shall perform administrative checks on 100 % of the claims before payment, and at least 5 % of the claims shall be controlled on-the-spot. The conformity of the control systems with the rules is checked by DG AGRI auditors and the risks for the EU budget that may derive from the identified weaknesses in the control systems are covered by financial corrections.

The paying agencies are also responsible for compiling the results of their controls into statistical control data to be sent annually to the Commission.  The certification bodies must systematically test the accuracy of these compiled statistics, notably via reconciliation with individual transactions, and give an opinion on them.

Finally, for claim year 2012, DG AGRI has tested, for area related payments, the transmission of the individual control data via the IT system that is used for payments (so called CATS).  The test has been successful and this automatic transmission of control data will be extended to most of the CAP measures and expenditure from claim year 2014.  This will prevent inaccuracies in the control statistics resulting from inadequate compilation of the individual data.

V. The Commission's own audits indicate that in some Member States control systems are not in line with the rules and do not detect all errors that could be detected.

The Commission was therefore aware that it could not place full reliance on the control statistics from the Member State as the only indicator of the level of error in agriculture spending.  This is why, for many years and until the 2012 AAR it had applied a 25% "safety margin" to the residual error rate calculated on the basis of the statistics.  This was regularly contested by the Court without however their assessment on the extent to which it was insufficient.

Nevertheless, DG AGRI took on board the concerns about the control statistics and changed its approach (see reply to paragraph III) after consulting the central services of the Commission responsible for methodological issues.  The Court, in its annual report for 2012, did not criticise this approach and rather expressed itself positively thereon and proposed that it be extended to other expenditure areas within the CAP and in particular, to rural development.

VI. The new method described in the reply to paragraph III provides for the Commission to adjust the error rate resulting from statistics from the Member States where Commission and/or the Court have identified deficiencies in the systems for detecting irregular expenditure.

VII. The Commission was also aware that there were limitations to the scope of the work which the certification bodies were requested to carry out. This is why, as from claim year 2014 in application of the CAP reform, certification bodies will have to carry out much deeper work on the reliability of the controls by the paying agencies.

VIII. In spite of certain shortcomings the information provided is considered useful for monitoring purposes and as an indication of the error detected by the Member States in the course of their administrative and on-the-spot checks. As already stated above, a new method is used by the Commission to arrive at a more informed estimate of the level of error than could be derived solely from the Member States' control statistics.

IX. The Commission does not claim that the adjustment it makes to the Member States' control statistics is statistically valid but considers it to be the best estimate in the absence of any other more reliable data.  This approach was the result of an intensive internal discussion not only within DG AGRI but also with the central services of the Commission.

The Commission notes that, in its Annual Report for 2012, the Court recommends that :

"The Commission, in DG AGRI's annual activity report, applies a similar approach for EAFRD as for decoupled area aid, where the Commission takes account of the results of its own conformity audits in assessing the error rate for each paying agency."

X.

The Commission shares the opinion of the Court regarding the necessary improvement of the administrative and on the spot cheks.  The adequacy of those checks is verified/assessed during the conformity audits and recommendations on the improvements are provided to the Member States.  The Commission also regularly informs Member States of its most common audit findings with a view to encouraging best practices.

The Commission considers that it implements this recommendation on a permanent basis as set out below and it has strengthened certain tools in the framework of the CAP reform.

Where the Commission, and indeed the Court, detect weakness in the management and control systems in the Member States it uses all tools at its disposal in order to induce the Member States to remedy the situation.  These include the conformity clearance procedure which results in financial corrections that not only protect the EU budget but also act as a powerful incentive to the MS, the requirement for Member States to implement action plans addressing the deficiencies.  Interruption of payments has also been used for the EAFRD and the new Horizontal Regulation on the control, financing and monitoring of the CAP, which comes into force in January 2014, includes a reinforced mechanism for suspension of payments for Member States who fail to implement a remedial action plan.

Area based aid schemes in general and the LPIS (Land Parcel Identification System) in particular are subject to a continuous rolling out of audits in order to detect deficiencies. An ineffective LPIS is taken very seriously by the Commission and Member States are required to implement urgent changes which are closely monitored by the Commission via follow-up audits. At the same time, any risk for the EU budget is covered by financial corrections. 

First indent: The guidelines on controls to be performed will be updated in the framework of the CAP reform, in parallel to the establishment of delegated and implementing acts. They will be discussed and explained to both paying agencies, which will have to apply them and to certification bodies, which will have to check their application by the paying agencies.

The Commission will also review the guidelines provided to Member States for the compilation of the statistics and in particular endeavour to harmonise the best practices between the different sectors.  Guidelines will, in particular, be reviewed in the context of the automation of the statistics as well as in the context of the implementing rules for the CAP reform.

The Commission will continue to request Member States to improve their internal procedures for verification of the control data sent.

Second indent: The new Horizontal Regulation on the control, financing and monitoring of the CAP, which comes into force in January 2014, will require the certification bodies to deliver an opinion on the legality and regularity of transactions.  This will entail re-performance of a representative sample of transactions already controlled by the paying agency and the validation (or not) of the resulting control statistics.  It is expected that this additional work will result in an improvement in the quality of the Paying Agencies control data and coupled with the Certification Body opinion will produce improvements in the degree of assurance which can be drawn.

Third indent: The Commission emphasises that the various deadlines for sending the statistics have been established for a number of operational reasons and not purely for AAR reporting purposes.  The Commission notes that a balance has to be reached on the optimal time for receiving the data bearing in mind the heavy administrative burden its compilation places upon the Member States in order to avoid multiple updated transmissions.

In 2012 the Commission conducted a pilot exercise for the automatic transmission of the control data for area related payments.  That data was thus transmitted in July 2013 and the initial assessments of its completeness and quality are positive.  Adjustments will be made in order that all relevant control data will be automated in this manner.  This will limit the risk of wrongly compiling errors and ensure that a single methodology is used in all MS.

In the new Horizontal Regulation for the control, financing and monitoring of the CAP, a provision was introduced which enables the Commission to suspend payments to Member States in respect of those control data which have not been sent within the specified deadlines.

Finally, with the new work of the certification bodies on legality and regularity the Commission will get a more robust opinion on the reliability of the controls themselves and a reinforced assurance on the error rates.

Fourth indent: The new Horizontal regulation provides that the certification bodies will have to systematically test, on representative samples for EAGF and for EAFRD, the reliability of the administrative and on-the-spot-controls performed by the paying agencies. The guidelines prepared by the Commission aim at ensuring that this work is conducted on a robust statistical methodology, that its results will reinforce assurance on the legality and regularity of the transactions and provide a valid estimate of the residual error rate per member state and per Fund. This would allow the Commission to progressively reduce the need for the individual adjustments it currently makes to the statistics provided by the Member States.

In the meantime, the Commission will apply the new approach referred to in the reply to paragraph III and will implement the recommendation in the Court's annual report for 2012: "The Commission, in DG AGRI's annual activity report, applies a similar approach for EAFRD as for decoupled area aid, where the Commission takes account of the results of its own conformity audits in assessing the error rate for each paying agency."

Regarding its own audit activities, the Commission will continue to focus on systemic audits of the paying agencies, on a risk basis and in the framework of the conformity clearance procedure, to ensure that risks resulting from identified weaknesses in the control systems are corrected.  Audit activities will also cover the new tasks of the certification bodies on legality and regularity, to provide technical guidance and contribute to achieving the objective of a reinforced assurance.

Observations

Common reply to paragraphs 28-36:

DG AGRI has already in the AAR 2012 stated that "…Member States control statistics need to be considered with a degree of caution".  DG AGRI has already undertaken to extend the integrated approach taken for decoupled direct aids in the 2012 AAR to other areas of CAP expenditure for the 2013 AAR – i.e. the Member States' statistics will be adjusted by taking account of the results of DG AGRI's own audit findings, the Court's and Certification Bodies' findings.

30. The Commission is following-up the Court’s observation on the effectiveness of the supervisory and control systems in the PAs concerned in the context of its ongoing conformity procedures.

For 2011 the Commission notes that most of the errors referred to were relatively small in financial terms and mainly concerned small differences in the re-measurement of land parcels carried out by the Court (42 of the errors found by the Court concern very small differences in parcel re-measurement and 24 errors were below 2%).

For 2012, similarly, most of the quantifiable errors were relatively small in financial terms and also concerned very small differences in the re-measurement of parcels by the Court (36 of the errors found by the Court were less than 5% of which 23 were below 2%).

32. When calculating the residual error rate, the Commission takes account of the potential impact of the weaknesses affecting the data reported by the MS.

First indent: The Commission is bound to follow the definition of a farmer in Article 2 of Council Regulation (EC) No. 73/2009.

Second indent: While the Commission does not always share the appreciation of the Court's auditors as to the exact eligibility of areas, where it finds that a Member State fails to correctly record eligible area, the issue is followed up through its clearance of accounts procedure.

Third indent: In the specific case referred to by the Court, the paying agency concerned considered their replacement of parcels to be a correction of "obvious errors" – a concept which is provided for in the EU legislation/guidelines.  As stated in its previous replies, the Commission does not share the view of the Court concerning the application of the obvious error concept in case of misplacement of parcels.

Fourth indent: Paying agencies are not always responsible for how the payment entitlements should be calculated.  In a number of cases the instructions on how to calculate entitlements are established by the competent authority in the Member State.

Fifth indent: While the Commission does not always share the appreciation of the Court's auditors as to the exact eligibility of areas, where it finds that a Member State fails to correctly record eligible area, the issue is followed up through its clearance of accounts procedure.

Box 1 – Example of incorrect data in the LPIS.

According to the Italian authorities, the Lombardia issue has been rectified as from 2011. The Commission is pursuing this through the conformity clearance procedure of accounts.

The Commission is aware of the issues related to the eligibility of pasture areas in Spain. The Spanish authorities were requested to act on this matter and presented the LPIS Improvement Plan including, notably, measures to implement the application of an eligibility coefficient to pasture parcels and to ensure systematic update of LPIS with the results of on the spot checks.  The Commission will continue to monitor the implementation of this plan by the national authorities.

The Commission has also found some cases in Hungary where LPIS was not updated with the results of the on the spot checks and has requested the Hungarian authorities in writing to take corrective action in this respect.

33. The weaknesses referred to by the Court have already been reported in previous years. The corresponding recommendations are being implemented since 2010 in collaboration with the Member States and followed up where necessary in the context of conformity procedures.

34. For 2011 the Commission notes that most of the errors referred to were relatively small in financial terms (half of the errors are less than 5%).

35. When calculating the residual error rate, the Commission takes account of the potential impact of the weaknesses affecting the data reported by the MS.

Box 3 – Example of insufficient quality of a Member State's on-the-spot checks

 This issue has been followed in an enquiry the outcome of which is as follows:

- The Paying Agency decided – after having reassessed the case - to recover the support already paid for the specific project;

- The verification procedure/check lists were updated in order to exclude that an inspector could not notice that an agricultural construction has mostly the characteristics of a house;

- A re-performance of the checks (ex post) has been carried out over all projects checked and approved by the same inspector which approved the non-eligible one;

- A re-performance of the checks (ex-post) has been carried out over all the projects having similar characteristic of the non-eligible project;

- Those ex-post controls showed some minor irregularities which are now following the standard contradictory steps and will probably involve recoveries of part of the support already paid;

- Nevertheless, as the last re-performance of the checks only concerned the projects involving new constructions and not also restructuring of existing buildings the Commission asked the Italian authorities to re-verify those projects. The results for these final ex-post checks will be sent by 30 November 2013.

Box 4 – Example of undetected breach of public procurement rules

As indicated in the reply to the Annual Report 2012, Commission services will follow the Court's findings up vis-à-vis the national authorities with a view to protecting the financial interest of the EU and recovering any undue payments.

Concerning Romania, the Commission has in its audits also found substantial weaknesses, and a clearance of account procedure is on-going.

36. The Commission has recognised this deficiency itself and therefore in its 2012 AAR has changed the way in which it draws assurance from Member States' control statistics.  Therefore, it takes into account other audit opinions, including ECA audit evidence, so as to reflect the extent of errors which are not detected by Member States and thus not reported in their statistics.

38. The Commission is in the process of automating the transmission of the control statistics (with a successful pilot exercise in July 2012) and is, in this context, preparing more detailed guidelines and procedures both internally and for the Member States. See also reply to paragraphs 39 – 41.

39. DG AGRI has not defined a precise cut-off date for compilation of the statistical reports as it prefers to receive updated statistics even the following year. However, Member States are in the Rural Development Committee constantly reminded of the importance of providing the statistics. Nevertheless an "intermediary" cut-off date could possibly be introduced.

DG AGRI verifies the completeness of statistics through the Certification Bodies' annual report which it scrutinises and which may result in follow-up conformity clearance procedures.

It is correct that the data is not complete and that the error rate is calculated only on the basis of the expenditure for which control statistics are available.  However, the Commission calculates that only around 0.6% of the amount of direct aid payments in respect of claim year 2010 were paid after the deadline of 30 June 2011 and therefore were possibly unreported via the control statistics.  As a consequence, the impact of this 0.6% on the precision of those control statistics as a basis of the error rate calculation is very limited.

Box 5 – Examples of incomplete or inaccurate data reported by some of the paying agencies audited

See reply to paragraph 92.

40. As mentioned by the Court, the current guidelines do not highlight statistics re-submission. Nevertheless, during the period between the sending of the statistics by the Member State and their use by the Commission in the AAR, the latter carries out consistency and plausibility checks and takes contact with Member States were necessary in order to verify possible anomalies. Updates are also requested in the Rural Development Committee, as well as during audit missions and bilateral meetings, if deemed appropriate.

Moreover, requiring systematic updates of the statistics would create an additional administrative burden not only on the Member States but also on the Commission due to having to repeat its verification work and incorporate the amended data into the complex calculation base for the residual error rate.  There is ultimately a very limited window of opportunity to receive, verify and process any revised data in order to be able to incorporate it into the annual activity report of the Director General.  The Commission emphasises that the control statistics and the error rate that is derived from them are not the annual accounts of the Member State and their accuracy does not have to be established to the same extent.  Rather, they are used to give an indication of where the error lies and, even before the Commission introduced its new method for adjusting the error rate, it applied a significant margin of appreciation to them in order to reflect the likelihood of their being understated. 

Finally, a pilot exercise was carried out with regard to the automation of statistics for area aids for claim year 2012.  This work commenced already in 2011 with the necessary advance notification of the technical requirements to the Member States and the resulting data for claim year 2012 was transmitted and automatically processed into the Commission's database by 15 July 2013.  The initial Commission assessments of the quality have been positive.  When all necessary review has been completed, work will commence on the extension of the automation to the other main expenditure areas of the CAP and is expected to be in place for the claim year 2014.  In this context, the Commission is preparing more detailed guidelines and procedures both internally and for the Member States.

41. The Commission is attentive to not excessively increasing the administrative burden on the Member States. Nevertheless the automation of the transmission of the control statistics should improve the traceability of the checks made.

Common reply to paragraphs 42 - 46:

The Commission is aware that progress has still to be made in some paying agencies in the compilation and verification of the control data.  Indeed in most of its own audits the Commission tests the reconciliation of statistical reports with individual control data and frequently observes inconsistencies.

This is the reason why the Commission has initiated an automatic procedure that allows paying agencies to communicate their individual control data with the system used for the individual payments.  A pilot exercise covering area related payments in all Member States for claim year 2012 (communication 15 July 2013) has been successful and the method will be extended to other CAP measures.

Such automation reduces the administrative burden and suppresses all compilation errors.

44.

Fourth indent: The inclusion of additional information in relation to previous years could only lead to an overstatement of the level of error.  Even if information is reported in the incorrect table, the residual error rate for Rural Development as a whole would not be affected.  Thus it can be concluded that neither of the two errors cited resulted in the level of error being understated.

Box 6 – Examples of inaccurate presentation of the number or value of the on-the-spot checks carried out

In the Polish example the checks were carried out, the paying agency did identify and fully record the error but inadvertently misclassified part of it as an administrative rather than as an on-the-spot check.  

Box 7 – Examples of errors found and having a significant impact on the error rates reported to the Commission

Second paragraph: In the example with regard to Bulgaria, while the paying agency reported erroneously, it had in fact overstated the error rate.

Common reply 47 – 50 The number of checks carried out by the certification bodies is in line with the sample size requested by the Commission in its guidelines to those bodies.  The Commission has, however, for a number of years, also considered that improvement was needed with regard to the quality of the Member States checks, how they reported on them and how better assurance could be given by the certification bodies.  At the end of 2009, it provided for a voluntary reinforcement of assurance exercise by certification bodies which would require them to re-perform a representative sample of the paying agency on-the-spot checks. This possibility was only taken up by a small number of paying agencies with rather mixed results.  However, in the context of the CAP reform, political agreement has been reached on a provision in the new Horizontal Regulation for the financing, management and monitoring of the CAP, which requires the certification bodies to deliver an opinion on the legality and regularities of the transactions underlying the expenditure.  This will entail a cradle to grave review and re-performance of a representative sample of transactions (a range of 60-170 transactions per Fund depending on a number of factors including population size).

The Commission therefore considers that the findings were addressed in the legislative initiatives it had already taken and which will be adopted at the beginning of 2014.

Box 8 – Examples of certification bodies’ weaknesses in not covering by their checks all the EU funding requirements

ECA's findings will be taken into account in the certification team risk analysis. 

51. Whenever necessary, recommendations were made to the concerned Certification Bodies to enhance the quality of their work for the future.  Findings and recommendations were also shared and made available to all Certification Bodies through an expert group meeting held in September 2012.  Some certification bodies have a good understanding of the checks to be carried out and their purposes, whilst other do not sometimes go deep enough in their testing and stay at a rather formalistic level of assessment.

The Commission considers that most of the findings were minor and did not call into question the assurance that can be derived by the Commission in the AAR in this regard.

Common reply to paragraphs 52-54:

As a rule, the Commission continues to review the work of the certification bodies via the analysis of the documents related to the financial clearance of accounts as well as audit missions to certification bodies. In the light of the lessons drawn from the voluntary application of the "reinforcement of assurance procedure", the guidelines for the certification bodies are being reviewed to provide for the  new responsibility of certification bodies to cover legality and regularity of expenditure.

Common reply to paragraphs 56 -57:

The Commission shares the Court's view that the sample currently checked by the certification bodies is too small.  To this effect, the draft guidelines for the new certification bodies' tasks underpinning the opinion on legality and regularity set out a statistically valid model for performing the work and have been expanded as to the nature of the work to be carried out by the certification bodies to ensure a complete cradle to grave test of each transaction selected.  Furthermore, it is intended that there will be closer supervision of the work of the certification bodies in the future.

58. The Commission considers that its information system is suited for its intended purposes. In particular, the information from the Member States serves the needs of operational and audit services. The regulatory deadlines are those which will be maintained and observed. 

59. The Commission collects information on the results of the administrative and on-the spot checks for claim year n which is paid to the farmers from December n until June n+1 and then reported to the Commission in July n+1.  The AAR for financial year n+1 is drawn up in March of n+2.

The Commission notes that rather than use the statistics for monitoring Member States implementation, it uses them to check inter alia if the number of controls respects the regulatory requirements.  A further purpose for the statistics is their use as a management tool for the Member States.  They are a useful source of information on errors detected which in a given year should be corrected. 

60. For direct payments, the payment period for claim year n is 15/10/n to 30/06/n + 1 (the payment period therefore falls within financial year n+1 which is the subject of the AAR for n+1 drawn up in March n+2.).  The control statistics cannot therefore be sent until the claim year has been processed.  While there are some payments made after this date, normally they are of a relatively low number and of a low financial impact.  The draft legislation for the CAP reform proposes to align the payment period and deadline for the EAGF IACS to that of the EAFRD-IACS.  One reason is to have a better link between claim year and financial execution.  The 15 July deadline also enables the Member States to make any adjustments necessary before the following claim year.

61. The deadlines for sending statistics on controls carried out are laid down by the relevant EU legislation. It is not always possible to obtain statistics updates in time to be included in the AAR.

62. The Commission notes that the only statistics which can be considered to be "not fully relevant" are those for EAFRD non-IACS for which the period in respect of which the statistics are reported does not fully correspond to the financial year of the AAR.

63. The data referred to by the Court was sent electronically directly to the CATS data base in a pilot exercise in July 2013.  All of the data for area direct aids was received in this manner and thus the Commission's scope for carrying out checks has been enhanced.  However it is noted that the Member States formal financial declarations (including the "x-tables" referred to by the Court) in respect of the transactions underpinning those statistics is sent in February n+1 – just 4 months after the end of the financial year in question.  The timing of the annual declarations juxtaposed with the timing of the Director-General's Annual Activity Report which has to be signed by end of March n+1, renders it impossible to carry out an in depth reconciliation of the x-table data and that of the statistics sent the previous July.

Common reply to paragraphs 66 - 68.

The Commission considers that it is not possible to systematically verify all the data presented in statistics during the conformity audits carried out in the Member States.  DG AGRI estimates that there are over 900 000 controls carried out by the Member States each year which underpin the control statistics and it would therefore be impossible to carry out the extent of verification demanded by the Court.

The audits are to cover the essential risks. While some work is done in this regard, it is not possible to pursue all issues found in the statistics.  Checking of control statistics is not the core object of the conformity audits which focus, rather, on the functioning of the management and control systems.  Furthermore, a large number of the audits for the year concerned are carried out after the AAR has been completed.

Depending on the delay given by the EU legislation and the audit procedure, it is not always possible to obtain statistics updates in time to be included in the AAR. The wide range of elements which already have to be checked and audited by the Commission's auditors in DG AGRI, renders it impossible, in the context of the available resources, to follow the Court's suggestion, without negatively affecting the core business. Nevertheless reinforced attention to the statistical reports will underlie the work to be carried out by the certification bodies in the context of the opinion on legality and regularity. 69. As already explained above, the purpose of the conformity clearance procedure is not primarily to check the Member States' control statistics. It is, rather, to check their management and control systems – and also, to following up the Court's own DAS findings, and to correct risk for the EU budget.

70. The Commission has always been aware that the Member States' estimate of irregular payment was not fully reliable.  This is why other evidence is taken into consideration in order to support the annual statement of assurance of the Director General of DG AGRI.

71. While the Commission agrees that the samples referred to by the Court were not sufficiently representative, the impact of the practices employed resulted in the residual error rate being overstated in most cases.

72. In the draft guideline for certification bodies for post-2013 a statistically valid model for building up the audit assurance has been put forward which paying agencies will also be advised to apply. The assurance is proposed to stem not only from the system verification results, but also from a more sound substantive approach (and where possible, a dual purpose test).

Common reply to paragraphs 73 -78:

The Court has criticised the Commission for using the data reported by Member States on the controls which they have carried out.  As a consequence, the Commission has adapted its approach and then applied its own appreciation to this data in order to give an assessment of elements (e.g. wrong interpretation or application of the EU and/or national legislation, deficient controls) which may mean that the statistics do not present the full picture.  This approach has been approved by the central services of the Commission and is considered to be an appropriate means of increasing the assurance which can be derived from the work carried out by the Member States under shared management.

According to the revised methodology, available information from ECA audits, certification body findings and Commission's own audits is integrated in an overall assessment of the situation at paying agency level and taken into account when aggregating at Member State and EU-27 level.

As indicated in the reply to the Court's Annual Report 2012, the new integrated approach, which was used for 2012 for decoupled direct aids only, will be further developed and extended as much as possible to the other CAP measures for the year 2013, and in particular to rural development.  By doing so, the Commission will be in line with the Court's annual report for 2012 where it is recommended that: "The Commission, in DG AGRI's annual activity report, applies a similar approach for EAFRD as for decoupled area aid, where the Commission takes account of the results of its own conformity audits in assessing the error rate for each paying agency."

78. As already indicated in the reply to the Annual Report 2012, the new integrated approach which was used for 2012 for decoupled direct aids will be further developed and extended as much as possible to the other CAP measures for the year 2013.

By doing so, the Commission considers that it will apply the recommendation of the Court in its Annual Report for 2012 that: "The Commission, in DG AGRI's annual activity report, applies a similar approach for EAFRD as for decoupled area aid, where the Commission takes account of the results of its own conformity audits in assessing the error rate for each paying agency."

Common reply to paragraphs 79-81:

The Commission agrees that the level of error estimated by the two institutions cannot be compared. 

It is noted that the residual error rate estimated by DG AGRI in 2012 for EAGF, falls within the upper and lower error limits of the Court's estimate of error.

82. The methodologies used by the two institutions to calculate the error rate are different. As the Commission needs information at the level of the individual paying agencies it has thus developed its own methodology for that purpose.

83. As the Court notes in paragraph 76, the adjustments made by the Commission for decoupled direct aids in its 2012 AAR resulted in a significant increase in the estimated residual error rate which the Commission considers as adequate to take account of the weaknesses in the Member States’ control statistics.

The Commission notes that in its 2012 Court's annual report the Court recommended that this approach should be extended to all agriculture expenditure.

84. In spite of reliability issues, the Commission considers that there is a value in the results of the more than 900 000 controls carried out by the Member States and that this work cannot simply be disregarded.

See also reply to paragraph 82.

87. Since the Court may, as illustrated, apply a different assessment of the quantitative impact of errors to that determined by the Commission and notified to Member States, it will therefore always estimate a higher error rate.  This does not mean that the Commission, which is the institution which is responsible for the implementation of the legislation, under the control of the European Court of Justice, should align its appreciation of the seriousness of an error to that of the Court. 

In determining the amount of a correction, the Commission takes into account the principle of proportionality, the type and extent of irregularity and financial implications of the shortcomings found. Purely formalistic errors do not always endanger sound financial management.

88. As indicated in its reply to the Annual Reports 2011 and 2012, the Commission is of the opinion that the respect of cross-compliance obligations does not constitute an eligibility criterion for CAP payments and, therefore, the controls of these requirements do not pertain to the legality and regularity of the underlying transactions.  Cross-compliance is a mechanism by which farmers are penalised when they do not respect a series of rules which stem in general from other policies than the CAP and apply to all EU citizens independently of the CAP.  Thus, the Commission considers that penalties imposed for violations of cross-compliance requirements should not be taken into account for the calculation of the error rates for the CAP.

CONCLUSIONS AND RECOMMENDATIONS

90. The Commission wishes to underline that its treatment of the Member States' statistics was significantly changed for the 2012 AAR.  DG AGRI had, up to 2011 used a 25% safety margin in order to compensate for incompleteness of Member States' statistics but this approach had been criticised by the Court as being unreliable.  Therefore, the Commission, in its 2012 AAR, undertook further evaluation of the statistics using information available to it from the certification bodies, the Commission's own audits and those of the Court.  Where there was a divergence between the appreciation of those audit bodies and the level of error indicated from the Member States control statistics, an adjustment was applied to reflect what the Commission considered to be a more realistic level of error.  Following the Court's recommendation in its 2012 Annual report, this approach is being applied to other expenditure areas for 2013 AAR, and in particular, to rural development.

91. When calculating the residual error rate, the Commission takes account of the potential impact of the weaknesses affecting the data reported by the Member States.

Recommendation 1

The Commission shares the opinion of the Court regarding the necessary improvement of the administrative and on the spot checks. The adequacy of those checks is verified / assessed during the conformity audits and recommendations on the improvements are provided to the Member States.  The Commission also regularly informs Member States of its most common audit findings with a view to encouraging best practices. 

The Commission implements this recommendation on a permanent basis as set out below and it has strengthened certain tools in the framework of the CAP reform.

First indent: Where the Commission, and indeed the Court, detect weakness in the management and control systems in the Member States it uses all tools at its disposal in order to induce the Member States to remedy the situation.  These include the conformity clearance procedure which results in financial corrections that not only protect the EU budget but also act as a powerful incentive to the Member States, the requirement for Member States to implement action plans addressing the deficiencies, interruption of payments used for the EAFRD.  The new Horizontal Regulation on the control, financing  and monitoring of the CAP, which comes into force in January 2014, includes a reinforced mechanism for suspension of payments for Member States who fail to implement a remedial action plan.

Second and third indents: Area based aid schemes in general and the LPIS in particular are subject to a continuous rolling out of audits in order to detect deficiencies.  An ineffective LPIS is taken very seriously by the Commission and Member States are required to implement urgent changes which are closely monitored by the Commission via follow-up audits. 

The Commission launched in 2012 an exercise with the Member States to identify the root causes of error for Rural Development. All Member States have been requested to set up action plans with a  view to reducing the error rates. Improvement of the quality of  Member States' controls is an integral part of the plan.

Common reply to paragraphs 92 and 93.

The new Horizontal Regulation on the control, financing  and monitoring of the CAP, which comes into force in January 2014, will require the certification bodies to deliver an opinion on the legality and regularity of transactions.  This will entail re-performance of a representative sample of transactions already controlled by the paying agency and the validation (or not) of the resulting control statistics.  It is expected that this additional work will result in an improvement in the quality of the paying agencies' statistics.

Nevertheless, the Member States will be requested to carry out further verification of the data which is transmitted.  An important step was taken in summer 2012 in respect of the control statistics for claim year 2012 for direct payments; the statistics were transmitted and input directly by the Member States into the Commission's CATS database and the initial assessments of the exercise were very positive.  Following a full review, and any necessary adjustments the automatic transmission will be rolled out for other important CAP expenditure area.

Recommendation 2

The Commission will review the guidelines provided to Member States for the compilation of the statistics and in particular endeavour to harmonise the best practices between the different sectors.  Guidelines will, in particular, be reviewed in the context of the automation of the statistics as well as in the context of the implementing rules for the CAP reform.

Second paragraph:

This part of the recommendation is addressed to the Member States. The Commission will continue to request Member States to improve their internal procedures for verification and compilation of the control data sent.

94. The Certification Bodies carry out the verifications requested of them by the EU rules and this entails the checking of a rather limited number of transactions.  The new Horizontal Regulation on the control, financing and monitoring of the CAP which will be adopted at the beginning of 2014 requires certification bodies to carry out significantly more checks to a greater depth in order to deliver an opinion on the legality and regularity of the expenditure.

Recommendation 3

First paragraph:

The COM considers the recommendation is  addressed by the new Horizontal Regulation on the control, financing  and monitoring of the CAP, in force from January 2014 and  will require the certification bodies to deliver an opinion on the legality and regularity of transactions.  This will entail re-performance of a representative sample of transactions already controlled by the paying agency and the validation (or not) of the resulting control statistics.  Detailed guidelines are well advanced setting out how this work should be done.

First indent: The guidelines propose the re-performance by the certification body of a representative sample of paying agencies checks.

Second indent: Re-performance has already been proposed by the Commission for the claim year 2014.

Third indent: It is expected that this additional work will result in an improvement in the quality of the paying agency statistics and coupled with the certification body opinion will produce improvements in the degree of assurance which can been drawn.

95. See replies to paragraphs 58 to 62.

A balance has to be reached on the optimal time for receiving the data bearing in mind the heavy administrative burden its compilation places upon the Member States in order to avoid multiple updated transmissions.

96. Desk reviews are limited in scope and are not intended to verify each individual data item.  Such a work is not possible with proportionate effort (the statistics are made notably on the basis of some 900 000 on-the-spot checks carried out).

97. See reply to paragraphs 66 – 69.

Recommendation 4

The Commission emphasises that the various deadlines for sending the statistics have been established for a number of operational reasons and not purely for AAR reporting purposes.  The Commission notes that a balance has to be reached on the optimal time for receiving the data bearing in mind the heavy administrative burden its compilation places upon the Member States in order to avoid multiple updated transmissions.

The timeliness of the reception of the statistics has been reviewed by the Commission.  In 2012 it launched a pilot exercise for the automatic transmission of the control statistics for direct payments.  That data was thus transmitted in July 2013 and the initial assessments of its completeness and quality are positive.  The necessary adjustments will be made in order that all relevant statistics can be automated in this manner.  This will limit the risk of reporting errors and ensure that a single methodology is used in all MS.

Furthermore, in the political agreement on the Horizontal Regulation for the control, financing and monitoring of the CAP, a new provision was introduced which enables the Commission to suspend payments to Member States in respect of those statistics which have not been sent within the specified deadlines.

The European legislator has made the clear choice to further improve the existing assurance building blocks in the context of the CAP reform and accordingly the Commission has no intention to carry out a full verification of the data underlying the control statistics which would also undermine the principle of shared management.

While some work is carried out on verifying control statistics this is not the primary purpose of the conformity clearance procedure.  It is rather to check their management and control systems – and to following up the Court's own DAS findings.  The wide range of elements which have to already be checked and audited by the Commission's auditors in DG AGRI within the resources available render it impossible to follow the Court's suggestion without negatively effecting its core business including impacting negatively on the level of financial corrections recovered in the protection of the EU's financial interests.  Nevertheless, elements of the underlying data are checked during the Commission's audits in order to address doubts over certain elements and are also used to prepare the sample of transactions to be checked on- the- spot. Furthermore, reinforced attention to the statistical reports will underlie the work to be carried out by the certification bodies in the context of the opinion on  legality and regularity.

Common reply to paragraphs 98 - 99:

For the AAR 2012 for decoupled direct aids (representing 65% of the CAP expenditure) the Commission adopted a new approach described in its reply to paragraph 69. The Commission's synthesis report states that DG AGRI will extend the approach to other CAP expenditure areas for 2013 AAR. This extension is also recommended by the Court in its 2012 Annual Report.

The Commission does not claim that the adjustment it makes to the Member States' control statistics is statistically valid but considers it to be the best estimate in the absence of any other more reliable data.

The Commission’s estimate of the residual error rate is not only based on the Member States’ control statistics but also take into account other relevant information based on professional judgment, notably the Commission’s and the Court’s own audit results and the findings of the certification bodies.

Recommendation 5

Second paragraph:

From claim year 2014 the certification bodies will have to systematically test, on representative samples for EAGF and for EAFRD, the reliability of the administrative and on-the-spot-controls performed by the paying agencies.  The guidelines prepared by the Commission aim at ensuring that this work is conducted on a robust statistical methodology and that its results will reinforce assurance on the legality and regularity of the transactions and will provide a valid estimate of the residual error rate per member state and per Fund.  DG AGRI will increase its supervision of the Certification Bodies in order to get assurance on the reliability of their new work on legality and regularity.  This would allow the Commission to progressively reduce the need for the individual adjustments it currently makes to the statistics provided by the Member States.

In the meantime and for the sake of a necessary stability, the Commission will apply the revised integrated approach used for decoupled direct payments in DG AGRI 2012 AAR and extend it to other CAP measures, in particular, to rural development, as recommended by the Court in its 2012 Annual report.