Court of Justice of the European Communities (including Court of First Instance Decisions)

Provisional text

JUDGMENT OF THE COURT (Fourth Chamber)

3 September 2020 (*)

(Appeal – European Agricultural Guarantee Fund (EAGF) and European Agricultural Fund for Rural Development (EAFRD) – Expenditure excluded from EU financing – Expenditure incurred by the Czech Republic – Regulation (EC) No 555/2008 – Articles 19 and 77 – Wine market – Regulation (EC) No 1122/2009 – Article 33 – Support for rural development – Area-related aid – Decoupled direct aid – Cross-compliance checks – Traditional on-the-spot checks and on-the-spot checks by remote sensing – Burden of proof – One-off and flat-rate corrections – Doubts concerning the effectiveness of the checks – Risk analysis – Deficiencies)

In Case C‑742/18 P,

APPEAL under Article 56 of the Statute of the Court of Justice of the European Union, brought on 27 November 2018,

Czech Republic, represented by M. Smolek, O. Serdula, J. Pavliš and J. Vláčil, acting as Agents,

appellant,

the other parties to the proceedings being:

European Commission, represented by Z. Malůšková, K. Walkerová and J. Aquilina, acting as Agents,

defendant at first instance,

Kingdom of Sweden,

intervener at first instance,

THE COURT (Fourth Chamber),

composed of M. Vilaras, President of the Chamber, S. Rodin (Rapporteur), D. Šváby, K. Jürimäe and N. Piçarra, Judges,

Advocate General: J. Kokott,

Registrar: C. Strömholm, Administrator,

having regard to the written procedure and further to the hearing on 29 January 2020,

after hearing the Opinion of the Advocate General at the sitting on 12 March 2020,

gives the following

Judgment

1        By its appeal, the Czech Republic seeks to have set aside point 2 of the operative part of the judgment of the General Court of the European Union of 13 September 2018, Czech Republic v Commission (T‑627/16, not published, EU:T:2018:538; ‘the judgment under appeal’), by which the General Court dismissed its action seeking annulment of Commission Implementing Decision (EU) 2016/1059 of 20 June 2016 excluding from European Union financing certain expenditure incurred by the Member States under the European Agricultural Guarantee Fund (EAGF) and under the European Agricultural Fund for Rural Development (EAFRD) (OJ 2016 L 173, p. 59; ‘the contested decision’), in so far as that decision concerns the Czech Republic.

 Legal context

 Regulation No 1975/2006

2        Article 26(1) and (4) of Commission Regulation (EC) No 1975/2006 of 7 December 2006 laying down detailed rules for the implementation of Council Regulation (EC) No 1698/2005, as regards the implementation of control procedures as well as cross-compliance in respect of rural development support measures (OJ 2006 L 368, p. 74), as amended by Commission Regulation (EC) No 484/2009 of 9 June 2009 (OJ 2009 L 145, p. 25) (‘Regulation 1975/2006’), provided as follows:

‘1.      Administrative checks shall be carried out on all applications for support or payment claims, and shall cover all elements that it is possible and appropriate to control by administrative means. The procedures shall require the recording of control work undertaken, the results of the verification and the measures taken in the event of discrepancies.

…

4.      Administrative checks relating to investment operations shall include at least one visit to the operation supported or the investment site to verify the realisation of the investment.

However, Member States may decide not to carry out such visits for smaller investments, or where they consider that the risk that the conditions for receiving aid are not fulfilled, or that the reality of the investment has not been respected, is low. That decision and its justification shall be recorded.’

3        Article 27 of that regulation provided:

‘1.      Member States shall organise on-the-spot checks on approved operations on an appropriate sampling basis. These shall be, as far as is possible, carried out before the final payment is made for a project.

2.      The expenditure controlled shall represent at least 4% of the eligible public expenditure that has been declared to the Commission each calendar year and at least 5% of the eligible public expenditure declared to the Commission over the whole programming period.

3.      The sample of approved operations to be checked in accordance with paragraph 1 shall take into account in particular:

(a)      the need to check an appropriate mix of types and sizes of operations;

(b)      any risk factors which have been identified following national or Community checks;

(c)      the need to maintain a balance between the axes and measures.

4.      The results of the on-the-spot checks shall be evaluated to establish whether any problems encountered are of a systemic character, entailing a risk for other similar operations, beneficiaries or other bodies. The evaluation shall also identify the causes of such situations, any further examination which may be required and the necessary corrective and preventive action.’

4        Article 28 of that regulation was worded as follows:

‘1.      Through the on-the-spot checks, the Member States shall endeavour to verify the following:

(a)      that the payment claims submitted by the beneficiary can be supported by accounting or other documents held by the bodies or firms carrying out the operations supported;

(b)      for an adequate number of expenditure items, that the nature and timing of the relevant expenditure comply with Community provisions and correspond to the approved specifications of the operation and the works actually executed or services delivered;

(c)      that the use or intended use of the operation is consistent with the use described in the application for Community support;

(d)      that the publicly funded operations have been implemented in accordance with Community rules and policies, especially the rules on public tendering and relevant mandatory standards established by national legislation or established in the rural development programme.

2.      The on-the-spot checks shall cover all the commitments and obligations of a beneficiary which can be checked at the time of the visit.

3.      Except in exceptional circumstances, duly recorded and explained by the national authorities, the on-the-spot checks shall include a visit to the operation or, if the operation is intangible, to the operation promoter.

4.      Only checks meeting the full requirements of this Article may be counted towards the achievement of the control rate set out in Article 27(2).’

 Regulation No 479/2008

5        Article 15 of Council Regulation (EC) No 479/2008 of 29 April 2008 on the common organisation of the market in wine, amending Regulations (EC) No 1493/1999, (EC) No 1782/2003, (EC) No 1290/2005, (EC) No 3/2008 and repealing Regulations (EEC) No 2392/86 and (EC) No 1493/1999 (OJ 2008 L 148, p. 1), provides as follows:

‘1.      Support may be granted for tangible or intangible investments in processing facilities, winery infrastructure and marketing of wine which improve the overall performance of the enterprise and concern one or more of the following:

(a)      the production or marketing of products referred to in Annex IV;

(b)      the development of new products, processes and technologies related to the products referred to in Annex IV.

2.      Support under paragraph 1 at its maximum rate shall be limited to micro, small and medium-sized enterprises within the meaning of Commission Recommendation 2003/361/EC of 6 May 2003 concerning the definition of micro, small and medium-sized enterprises [(OJ 2003 L 124, p. 36)]. For the territories of the Azores, Madeira, the Canary Islands, the smaller Aegean islands within the meaning of Council Regulation (EC) No 1405/2006 of 18 September 2006 laying down specific measures for agriculture in favour of the smaller Aegean islands [(OJ 2006 L 265, p. 1)] and the French overseas departments, no size limits shall apply for the maximum rate. For enterprises that are not covered by Article 2(1) of Title I of the Annex to Recommendation 2003/361/EC with less than 750 employees or with a turn-over of less than EUR 200 million, the maximum aid intensity shall be halved.

Support shall not be granted to enterprises in difficulty within the meaning of the Community guidelines on State aid for rescuing and restructuring firms in difficulty.

3.      The eligible expenditure shall exclude the elements referred to in Article 71(3)(a) to (c) of [Council] Regulation (EC) No 1698/2005 [of 20 September 2005 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD) (OJ 2005 L 277, p. 1)].

4.      The following maximum aid rates in relation to the eligible investment costs shall apply to the Community contribution:

(a)      50% in regions classified as convergence regions in accordance with [Council] Regulation (EC) No 1083/2006 [of 11 July 2006 laying down general provisions on the European Regional Development Fund, the European Social Fund and the Cohesion Fund and repealing Regulation (EC) No 1260/1999 (OJ 2006 L 210, p. 25)];

(b)      40% in regions other than convergence regions;

(c)      75% in the outermost regions in accordance with Council Regulation (EC) No 247/2006 of 30 January 2006 laying down specific measures for agriculture in the outermost regions of the Union [(OJ 2006 L 42, p. 1)];

(d)      65% in the smaller Aegean islands within the meaning of Regulation (EC) No 1405/2006.

5.      Article 72 of Regulation (EC) No 1698/2005 shall apply mutatis mutandis to support referred to in paragraph 1.’

 Regulation No 555/2008

6        Commission Regulation (EC) No 555/2008 of 27 June 2008 laying down detailed rules for implementing Regulation No 479/2008 as regards support programmes, trade with third countries, production potential and on controls in the wine sector (OJ 2008 L 170, p. 1), as amended by Commission Regulation (EC) No 702/2009 of 3 August 2009 (OJ 2009 L 202, p. 5) (‘Regulation No 555/2008’), provides as follows, in Article 19(1) thereof:

‘Support shall be paid once it is ascertained that either a single operation or all the operations covered by the support application, according to the choice made by the Member State for the management of the measure, have been implemented and controlled on the spot.

…’

7        Article 77(3) and (5) of that regulation provides:

‘3.      Except for the cases where systematic on the spot checks are foreseen by Regulation (EC) No 479/2008 or by this Regulation, the competent authorities shall perform on-the-spot checks by sampling an appropriate percentage of beneficiaries/producers on the basis of a risk analysis in accordance with Article 79 of this Regulation.

…

5.      Concerning measures foreseen by Article 15 of Regulation (EC) No 479/2008, Articles 26, 27 and 28 of Regulation (EC) No 1975/2006 shall apply mutatis mutandis.’

 Regulation No 1122/2009

8        Commission Regulation (EC) No 1122/2009 of 30 November 2009 laying down detailed rules for the implementation of Council Regulation (EC) No 73/2009 as regards cross-compliance, modulation and the integrated administration and control system, under the direct support schemes for farmers provided for [by] that Regulation, as well as for the implementation of Council Regulation (EC) No 1234/2007 as regards cross-compliance under the support scheme provided for the wine sector (OJ 2009 L 316, p. 65), as amended by Commission Implementing Regulation (EU) No 1368/2011 of 21 December 2011 (OJ 2011 L 341, p. 33) (‘Regulation No 1122/2009’), states as follows, in recitals 37 and 40 thereof:

‘(37)      The minimum number of farmers to be checked on-the-spot under the various aid schemes should be determined. In the case where Member States opt for the application of the various livestock aid schemes, an integrated holding-based approach should be foreseen in relation to farmers applying for aids under those schemes.

…

(40)      The sample of the minimum rate of on-the-spot checks should be drawn partly on the basis of a risk analysis and partly at random. The competent authority should establish the risk factors as it is in a better position to choose the relevant risk factors. To assure relevant and efficient risk analysis, the effectiveness of the risk analysis should be assessed and updated on an annual basis taking into account the relevance of each risk factor, comparing the results of randomly and risk-based selected samples and the specific situation in the Member States.’

9        Article 26(1) of that regulation provides as follows:

‘Administrative controls and on-the-spot checks provided for in this Regulation shall be made in such a way as to ensure effective verification of compliance with the terms under which aids are granted and of the requirements and standards relevant for cross-compliance.’

10      Article 30(1) of that regulation provides as follows:

‘The total number of on-the-spot checks carried out each year shall cover at least 5% of all farmers applying respectively for the single payment scheme, the single area payment scheme or area-related payments under specific support. The Member States shall assure that on-the-spot checks cover at least 3% of the farmers applying for aid under each of other area-related aid schemes provided for under Titles III, IV and V of [Council] Regulation (EC) No 73/2009 [of 19 January 2009 establishing common rules for direct support schemes for farmers under the common agricultural policy and establishing certain support schemes for farmers, amending Regulations (EC) No 1290/2005, (EC) No 247/2006, (EC) No 378/2007 and repealing Regulation (EC) No 1782/2003 (OJ 2009 L 30, p. 16)].’

11      Article 31(1) and (2) of that regulation provides as follows:

‘1.      Control samples for on-the-spot checks under this Regulation shall be selected by the competent authority on the basis of a risk analysis and representativeness of the aid applications submitted.

To provide the element of representativeness, the Member States shall select randomly between 20% and 25% of the minimum number of farmers to be subject to on-the-spot checks as provided for in Article 30(1) and (2).

However, if the number of farmers to be subject to on-the-spot checks exceeds the minimum number of farmers to be subject to on-the-spot checks as provided for in Article 30(1) and (2), the percentage of randomly selected farmers in the additional sample should not exceed 25%.

2.      The effectiveness of the risk analysis shall be assessed and updated on an annual basis:

(a)      by establishing the relevance of each risk factor;

(b)      by comparing the results of the risk based and randomly selected sample referred to in the second subparagraph of paragraph 1;

(c)      by taking into account the specific situation in the Member State.’

12      Under Article 33 of Regulation No 1122/2009:

‘On-the-spot checks shall cover all the agricultural parcels for which aid is requested under aid schemes listed in Annex I to Regulation (EC) No 73/2009. Nevertheless, the actual determination of the areas as part of an on-the-spot check may be limited to a sample of at least 50% of the agricultural parcels for which an application has been submitted under the aid schemes established in Titles III, IV and V of Regulation (EC) No 73/2009 provided that the sample guarantees a reliable and representative level of control both in respect of area checked and aid claimed. When this sample check reveals anomalies the sample of agricultural parcels actually inspected shall be increased.

Member States may make use of remote sensing in accordance with Article 35 and Global Navigation Satellite Systems techniques where possible.’

13      Article 57 of that regulation, entitled ‘Basis of calculation in respect of areas declared’, makes the following provision in paragraph 3 thereof:

‘Without prejudice to reductions and exclusions in accordance with Articles 58 and 60, in the case of applications for aid under area-related aid schemes if the area declared in a single application exceeds the area determined for that crop group, the aid shall be calculated on the basis of the area determined for that crop group.

However, without prejudice to Article 30 of Regulation (EC) No 73/2009, if the difference between the total area determined and the total area declared for payment under aid schemes established in Titles III, IV and V of Regulation (EC) No 73/2009 is less than or equal to 0.1 hectare, the area determined shall be set equal to the area declared. For this calculation, only over-declarations of areas at crop group level shall be taken into account.

The second subparagraph shall not apply where that difference represents more than 20% of the total area declared for payments.’

14      Article 58 of that regulation, entitled ‘Reductions and exclusions in cases of over-declaration’, provides as follows:

‘If, in respect of a crop group, the area declared for the purposes of any area-related aid schemes exceeds the area determined in accordance with Article 57, the aid shall be calculated on the basis of the area determined reduced by twice the difference found if that difference is more than either 3% or two hectares, but no more than 20% of the area determined.

If the difference is more than 20% of the area determined, no area-linked aid shall be granted for the crop group concerned.

If the difference is more than 50%, the farmer shall be excluded once again from receiving aid up to an amount equal to the amount which corresponds to the difference between the area declared and the area determined in accordance with Article 57 of this Regulation. That amount shall be off-set in accordance with Article 5b of Commission Regulation (EC) No 885/2006 [of 21 June 2006 laying down detailed rules for the application of Council Regulation (EC) No 1290/2005 as regards the accreditation of paying agencies and other bodies and the clearance of the accounts of the EAGF and of the EAFRD (OJ 2006 L 171, p. 90)]. If the amount cannot be fully off-set in accordance with that article in the course of the three calendar years following the calendar year of the finding, the outstanding balance shall be cancelled.’

 Background to the dispute and the contested decision

15      The background to the dispute was set out by the General Court in paragraphs 1 to 4 of the judgment under appeal and may, for the purposes of the present proceedings, be summarised as follows.

16      In three audit missions which took place from 10 to 14 September 2012, from 12 to 16 November 2012 and from 8 to 12 September 2014 (‘the audit’), the Commission investigated the compatibility of payments made by the Czech authorities under the European Agricultural Guarantee Fund (EAGF) with the rules of EU law relating to cross-compliance, investments in the wine sector and decoupled direct aid respectively.

17      On 20 June 2016, it adopted the contested decision, excluding from European Union financing certain expenditure incurred by the Member States under the EAGF and the European Agricultural Fund for Rural Development (EAFRD), specifically, so far as the Czech Republic is concerned, expenditure amounting to:

–        EUR 29 485 612.55 in respect of the cross-compliance requirements for the financial years 2011 to 2014;

–        EUR 636 516.20 in respect of the aid scheme for investment in the wine sector for the financial years 2011 to 2014;

–        EUR 462 517.83 in respect of the decoupled direct aid scheme for the financial years 2013 to 2015.

 The procedure before the General Court and the judgment under appeal

18      By application lodged at the Registry of the General Court on 31 August 2016, the Czech Republic brought an action seeking annulment of the contested decision, relying on four pleas in law: the first three alleging infringement of Article 52(1) of Regulation (EU) No 1306/2013 of the European Parliament and of the Council of 17 December 2013 on the financing, management and monitoring of the common agricultural policy and repealing Council Regulations (EEC) No 352/78, (EC) No 165/94, (EC) No 2799/98, (EC) No 814/2000, (EC) No 1290/2005 and (EC) No 485/2008 (OJ 2013 L 347, p. 549), and the fourth alleging infringement of paragraph 2 of that article.

19      By the judgment under appeal, the General Court dismissed all the pleas put forward, with the exception of the third branch of the fourth plea. As a result, it annulled the contested decision in so far as it excludes payments made by the Czech Republic under the EAGF of EUR 6 356 909.30 in respect of cross-compliance checks for the financial year 2011.

 Forms of order sought by the parties before the Court of Justice

20      The Czech Republic claims that the Court should:

–        set aside point 2 of the operative part of the judgment under appeal and the corresponding part of the judgment;

–        annul the contested decision in so far as it excludes expenditure amounting to EUR 462 517.83 in connection with the single area payment;

–        annul the contested decision in so far as it excludes expenditure amounting to EUR 636 516.20 in connection with investments in the wine sector; and

–        order the Commission to pay the costs.

21      The Commission contends that the Court should:

–        dismiss the appeal; and

–        order the appellant to pay the costs.

 The request to have the oral part of the procedure reopened

22      Following the delivery of the Advocate General’s Opinion, the Czech Republic, by a document lodged at the Court Registry on 30 March 2020, applied for the oral part of the procedure to be reopened, pursuant to Article 83 of the Rules of Procedure of the Court of Justice.

23      Under that provision, the Court may at any time, after hearing the Advocate General, order the reopening of the oral part of the procedure, in particular if it considers that it lacks sufficient information or where a party has, after the close of that part of the procedure, submitted a new fact which is of such a nature as to be a decisive factor for the decision of the Court, or where the case must be decided on the basis of an argument which has not been debated.

24      In support of its application, the Czech Republic submits that the Advocate General’s Opinion is based, so far as the third ground of appeal is concerned, on manifestly incorrect facts which are inconsistent with the claims made and the evidence presented in the proceedings before the Court.

25      Nevertheless, it should be noted that, under the second paragraph of Article 252 TFEU, it is the duty of the Advocate General, acting with complete impartiality and independence, to make, in open court, reasoned submissions on cases which, in accordance with the Statute of the Court of Justice of the European Union, require his or her involvement. The Court is not bound either by the Advocate General’s Opinion or by the reasoning on which it is based (judgment of 5 March 2020, Foundation for the Protection of the Traditional Cheese of Cyprus named Halloumi v EUIPO, C‑766/18 P, EU:C:2020:170, paragraph 35 and the case-law cited).

26      It should also be recalled that the Statute of the Court of Justice of the European Union and the Rules of Procedure make no provision for the submission of observations in response to the Advocate General’s Opinion. Disagreement with the Advocate General’s Opinion cannot therefore in itself constitute grounds justifying the reopening of the oral procedure (judgment of 5 March 2020, Foundation for the Protection of the Traditional Cheese of Cyprus named Halloumi v EUIPO, C‑766/18 P, EU:C:2020:170, paragraph 36 and the case-law cited).

27      In the present case, it is apparent from the application for the oral part of the procedure to be reopened that, by that application, the Czech Republic is in fact seeking to respond to the interpretation made by the Advocate General of the facts and law underlying the third ground of appeal. However, as is apparent from the case-law cited in the previous paragraph, that is not a ground which justifies reopening the oral part of the procedure. Moreover, those facts and that law have been debated at length between the parties to the appeal, in particular during the written part of the procedure.

28      The Court considers, after hearing the Advocate General, that it has all the necessary information to rule on the appeal.

29      Having regard to the foregoing considerations, there is no need to reopen the oral part of the procedure.

 The appeal

30      The Czech Republic puts forward four grounds of appeal. The first, second and fourth grounds of appeal allege errors of law in relation to the burden of proof, Article 33 of Regulation No 1122/2009 and Articles 19 and 77 of Regulation No 555/2008 respectively. The third ground of appeal alleges an error of law and distortion of the facts and of the subject matter of the dispute in relation to the General Court’s finding concerning the point at which the Czech authorities implemented the national aid programme.

 The first ground of appeal

 Arguments of the parties

31      By its first ground of appeal, the Czech Republic alleges that the General Court erred in law in its application of the settled case-law in relation to the burden of proof in the EAGF field. In the first place, it alleges that the General Court erred in law in paragraph 22 of the judgment under appeal inasmuch as it held that on-the-spot checks by remote sensing and traditional on-the-spot checks must give rise to rates of irregularities which are ‘in principle similar’.

32      The Czech Republic asserts, in essence, that the comparison of the error rates shown in the two checking methods is irrelevant, given the inherent objective differences between those methods in the manner in which the control sample is selected.

33      More specifically, the Czech Republic argues that the risk analysis for the purposes of remote sensing cannot entail a rate of irregularities comparable to that of the risk analysis for the purposes of traditional on-the-spot checks, since the targeting of the risk analysis for the purposes of remote sensing is objectively less precise.

34      That is the result, first, of the fact that the risk analysis used for remote-sensing checks takes place at an earlier point on account of the requirements provided for in page 36 of the common technical specifications of 2013 imposed by the Commission and is based on less up-to-date data than the risk analysis carried out for traditional on-the-spot checks.

35      Secondly, the risk analysis for the purposes of on-the-spot checks by remote sensing gives rise to the selection of control samples in the form of zones of 30km by 20km in which all the agricultural parcels are checked irrespective of the risk profile of the farmers in that zone, whereas the risk analysis for the purpose of traditional on-the-spot checks concentrates specifically on farmers ‘posing a risk’. It thus follows that the targeting of the risk analysis for the purposes of remote sensing is, in the Czech Republic’s view, reduced in relation to the risk analysis carried out for the purposes of traditional on-the-spot checks.

36      Thirdly and lastly, inasmuch as the zones containing less than 25% agricultural area cannot be checked by remote sensing means, in accordance with the requirements laid down by the Commission, some high-risk farmers cannot be subject to such a check, and therefore they can only be the subject of traditional on-the-spot checks.

37      Therefore, in the light of the objective differences between those two methods, a lower rate of irregularities in the risk analysis for the purposes of remote sensing cannot give rise in itself to serious and reasonable doubts as to the proper functioning of a Member State’s control system.

38      It follows that the objective impossibility of comparing the targeting of the methods of risk analysis also prevents determination of the maximum discrepancy authorised between the rates of irregularities noted between the two methods.

39      The Czech Republic also points out that, although the second paragraph of Article 33 and Article 35 of Regulation No 1122/2009 authorise the Member States to check the agricultural area by means of two methods of control without giving preference to one of those two methods or imposing supplementary conditions on their use, the fact remains that an effective control system is characterised by the use of those two methods of control as a complement to one another. Requiring those two, objectively different, methods to produce comparable results would lead a Member State in practice either to have to reduce the targeting of the risk analysis for the purposes of traditional on-the-spot checks or to have to abandon remote-sensing checks, in which case the provisions of that regulation permitting the use of that control method would be rendered redundant.

40      Moreover, the Czech Republic submits that the General Court did not address the Commission’s argument according to which serious and reasonable doubts are raised by the fact that the sample selected randomly for the purposes of the check by remote sensing shows a higher rate of irregularities than a sample selected, with a view to a check by remote sensing, on the basis of a risk analysis.

41      In the second place, the Czech Republic claims that the General Court committed the same error in paragraphs 23 and 24 of the judgment under appeal in holding that the Czech Republic had not put forward any specific evidence which could demonstrate that the different rates of irregularities found were attributable to the nature of the methods concerned or to the risk profile of the operators checked and not to shortcomings as regards the choices made by the national authorities.

42      Having regard to the foregoing, even if the Commission’s findings gave rise to serious and reasonable doubts, it is unnecessary, in order to dispel those doubts, to explain the reasons for the discrepancy between the rates of irregularities found. In order to dispel those doubts, it is sufficient for the Member State to demonstrate the effectiveness of the functioning of the control system and the appropriate use of both methods of control. In this connection, the Czech Republic argues that it submitted specific evidence in the audit and in the proceedings before the General Court that proves clearly that the risk analysis complied with the requirements of Regulation No 1122/2009. The Commission has never refuted that claim.

43      The Commission contends in response, in the first place, that the General Court’s finding in paragraph 22 of the judgment under appeal that it is reasonable to expect traditional on-the-spot checks and remote-sensing checks to give rise to rates of irregularities which are ‘in principle similar’ is materially correct.

44      It submits, in essence, that the arguments set out by the Czech Republic to justify the discrepancies between the rates of irregularities of the different methods of on-the-spot checks are inconsistent with Article 26, Article 31(2) and Article 33 of Regulation No 1122/2009.

45      First, Article 26 of that regulation requires that, irrespective of the method used, on-the-spot checks must ensure effective verification of compliance with the terms under which aids are granted.

46      Secondly, it follows from an interpretation of Article 26 of Regulation No 1122/2009 in conjunction with Article 31(2) and Article 33 thereof that the use of those methods must not lead to a difference in treatment between farmers. According to the Commission, such a difference in treatment would be unavoidable if the probability of detecting errors were higher among those who had been checked by remote sensing than among those who had been subject to a traditional on-the-spot check. The comparison of the error rates found between the methods of on-the-spot checks is an indicator of effectiveness that is clearly appropriate in order to compensate for that risk of different treatment, and it enables a shortcoming in the control system to be identified.

47      Nevertheless, the Commission, being aware of the specific features of each method of on-the-spot checks, contends that if a greater proportion of irregularities is detected by one method rather than another over several years, it is for the Member State to explain the reason for those discrepancies, failing which those discrepancies are deemed to indicate that one method is less effective or that an inappropriate control sample was selected, a fortiori where the Member State has carried out approximately 50% of the on-the-spot checks by remote sensing. According to that institution, obtaining comparable rates of irregularities is a means of demonstrating the effectiveness of those checks and cannot be regarded as an end in itself. The Commission is of the opinion in the present case that the discrepancy found indicates a shortcoming in the control system that is such as to give rise to serious and reasonable doubts.

48      In the second place, the Commission submits that the explanations provided by the Czech Republic, based on the characteristics inherent to the control methods used, are too general to dispel the Commission’s serious and reasonable doubts. It argues that the Czech Republic thus has not provided specific factual evidence demonstrating that the control method chosen was effective. More specifically, it has explained neither precisely why the selection of the sample or the remote-sensing method led to a significant discrepancy between the rates of irregularities for several years, nor the manner in which it altered the risk analysis and the sampling to remedy that fact.

49      Lastly, the Commission recalls that the Czech Republic, being aware of the problems caused by the differences between the rates of irregularities, acquired new analysis software in 2013, which had a positive impact in respect of the applications concerning 2014.

50      In the reply, the Czech Republic submits that, first, Article 26 of Regulation No 1122/2009 requires that the control system as a whole be effective, not that there be an identical or similar error rate found in connection with the various types of control.

51      Secondly, it submits that, during the audit, it gave the Commission explanations on the high-risk nature of the applicants in a sample selected on a random basis, and that it also submitted documents containing a more in-depth analysis.

52      Thirdly, even if the Court were to find the comparison of the rates of irregularities between the two control procedures to be such as to give rise to a serious and reasonable doubt as regards the effectiveness of the control system, the Member State cannot be required to give explanations as to the reasons underlying the discrepancy between those rates. According to the Czech Republic, that Member State can only be required to provide evidence concerning, inter alia, the manner in which farmers are selected for checking and the manner in which the risk analysis for both control methods is carried out, including an assessment and annual updating of the risk analysis in accordance with Article 31(2)(b) of Regulation No 1122/2009. It is that very evidence which the Czech Republic submitted to the Commission in the audit. It thus supplied a detailed description of the risk analyses in respect of both control methods, including the risk factors used and their relevance, and an assessment and annual updating of the effectiveness of those analyses, which take account of the lessons learned.

53      Fourthly and lastly, it denies that its acquisition of software in 2013 may support the argument that its risk analysis was inadequate. The progressive improvement of risk analysis by means of acquiring software cannot be regarded as recognition of the failings of the status quo. In this connection, it observes that the risk analysis for the purposes of remote-sensing checks has its objective limits, as a result of which it can never equal in effectiveness the risk analysis for the purposes of traditional on-the-spot checks, without that fact being such as to affect the effectiveness of the control system as a whole.

54      In the rejoinder, the Commission repeats that the comparison of the error rates detected in the samples is an appropriate indicator of the effectiveness of the methods of control selected. The method based on representative samples is drawn from Article 31 of Regulation No 1122/2009, according to which the effectiveness of the risk analysis is evaluated by comparing the results of the sample selected according to risk with those of the randomly selected sample. In the view of that institution, that is the only way of ensuring that remote-sensing checks are of equivalent effectiveness to traditional on-the-spot checks.

55      That provision does not specify whether, for the purposes of comparing those two results, it is necessary to make a distinction according to whether the results were obtained by means of a traditional on-the-spot check or by remote sensing so that it must be found that those control procedures are supposed to be of guaranteed equivalent effectiveness.

56      In addition, the Commission repeats that, in paragraph 24 of the judgment under appeal, the General Court correctly found that the evidence submitted by the Czech Republic did not suffice to explain the discrepancy in the rates of irregularities and to dispel the doubts that the Commission had as to the effectiveness of the remote-sensing checks. The Commission also states that the improvement of results due to an effort to eliminate the lacunae that it had pointed out demonstrates that that was within the capability of the Czech Republic.

 Findings of the Court

57      By its first plea, the Czech Republic in essence submits, first, that the General Court erred in law in paragraph 22 of the judgment under appeal in holding that on-the-spot checks by remote sensing and traditional on-the-spot checks must give rise to rates of irregularities which are ‘in principle similar’ and that it is reasonable, in this connection, to expect that the rates of irregularities resulting from such checks at national level do not show any discrepancies, such as those in the present case, consisting in rates of irregularities three or four times greater found in connection with traditional on-the-spot checks as against those found in connection with remote-sensing checks. According to the Czech Republic, contrary to the approach taken by the General Court, the mere fact that a difference is noted between those rates of irregularities cannot serve as a basis for serious and reasonable doubts as to the effectiveness of its control system.

58      Secondly, the Czech Republic alleges that the General Court erred in law in paragraphs 23 and 24 of the judgment under appeal in examining whether that Member State had dispelled the Commission’s doubts to the effect that where such discrepancies are found, it is for the Member State to provide detailed evidence justifying those discrepancies. That court did so by setting out for that purpose a series of factors to which that Member State could refer and holding that the Czech Republic had not put forward any specific evidence which could demonstrate that the discrepancies observed are attributable to those factors rather than to shortcomings as regards the choices made by the national authorities in respect of the implementation of remote-sensing checks.

59      In this respect, it is settled case-law of the Court that, in order to prove the existence of an infringement of the rules on the common organisation of agricultural markets, the Commission is not required to demonstrate exhaustively that the checks carried out by the national authorities are inadequate or that the figures supplied by them are incorrect, but to provide evidence of the serious and reasonable doubt it entertains concerning the checks carried out by the national authorities or the incorrect nature of the figures those authorities have supplied. It is for the Member State to adduce the most detailed and comprehensive evidence that its checks are actually carried out and its figures are genuine and, if appropriate, that the Commission’s assertions are incorrect (judgment of 6 November 2014, Netherlands v Commission, C‑610/13 P, not published, EU:C:2014:2349, paragraph 58).

60      Under Article 33 of Regulation No 1122/2009, Member States may make use of remote sensing in accordance with Article 35 of that regulation and Global Navigation Satellite Systems techniques where possible.

61      Nevertheless, on-the-spot checks, which may be traditional or made by remote-sensing means, must, according to Article 26 of that regulation, be made in such a way as to ensure effective verification of compliance with the terms under which aids are granted and of the requirements and standards relevant for cross-compliance.

62      It follows that, provided that their choice ensures effective verification, the Member States are authorised to choose either traditional on-the-spot checks or on-the-spot checks using remote sensing, or, as the case may be, a combination of both methods of control.

63      Furthermore, Article 30(1) of Regulation No 1122/2009 sets out the minimum rate of all the farmers who must be subject to on-the-spot checks each year.

64      In this respect, Article 31(1) of that regulation stipulates that the selection of control samples for the purposes of on-the-spot checks must be carried out on the basis of a risk analysis and representativeness of the aid applications submitted. To provide the element of representativeness, the Member States are to select randomly between 20% and 25% of the minimum number of farmers to be subject to on-the-spot checks as provided for in Article 30(1) and (2) of that regulation.

65      To assure relevant and efficient risk analysis, Article 31(2) of that regulation provides that the effectiveness of the risk analysis is to be assessed and updated on an annual basis, first, by taking into account the relevance of each risk factor, secondly, by comparing the results of the risk-based selected sample and a randomly selected sample and, thirdly, according to the specific situation in the Member State concerned.

66      It is therefore necessary to examine whether the fact that the on-the-spot checks by remote sensing and the traditional on-the-spot checks result in rates of irregularities which are dissimilar is, in itself, such as to raise a serious and reasonable doubt concerning whether the risk analyses on which those checks are respectively based are relevant.

67      In the first place, it follows from paragraphs 50 to 55 above that none of the articles of Regulation No 1122/2009 which are mentioned in those paragraphs require the Member States to ensure that the traditional on-the-spot checks and the on-the-spot checks by remote sensing give rise to rates of irregularities which are in principle similar.

68      In the second place, the risk analysis on which a remote-sensing check is based must, as the Czech Republic submits without that being refuted by the Commission, be carried out before the Member State has chosen to use that method of checking, which is not the case in respect of traditional on-the-spot checks.

69      Thus, on-the-spot checks by remote sensing are based, on account of the characteristics inherent to that type of check, on a risk analysis using less up-to-date data than the data used for the risk analysis for traditional on-the-spot checks.

70      It follows that, due to that characteristic, inherent to checking by remote sensing, that method of checks may, in principle, lead to the occurrence of rates of irregularities that diverge from those found following traditional on-the-spot checks.

71      In the third place, according to the Czech Republic’s claims, which have not been refuted by the Commission, remote sensing may be carried out only on zones of 30km by 20km, the zone checked must contain at least 25% agricultural land and the check must cover all the farmers in that zone.

72      Therefore, in order for on-the-spot checks by remote sensing to result in the same rates of irregularities as those resulting from traditional on-the-spot checks, they should be used not only for zones of 30km by 20km containing at least 25% agricultural land, but also target solely farmers selected on the basis of a risk analysis with a view to carrying out traditional on-the-spot checks.

73      It follows from all the foregoing that the mere discrepancy between the rates of irregularities found following traditional on-the-spot checks and those found following checks by remote sensing was not, alone, capable of giving rise, on the part of the Commission, to serious and reasonable doubts as to the effectiveness of on-the-spot checks by remote sensing.

74      Since the General Court therefore erred in law, the first ground of appeal must be upheld.

 The second ground of appeal

 Arguments of the parties

75      By its second ground of appeal, the Czech Republic in essence alleges that the General Court misinterpreted Articles 33, 57 and 58 of Regulation No 1122/2009 in paragraphs 38 and 40 of the judgment under appeal in holding that, where over-declarations are found representing between 0.1 and 2 hectares or 3% of the area determined, it is for the competent national authority to increase the control sample until no over-declaration of that magnitude is found.

76      More specifically, that Member State argues, in the first place, that Articles 57 and 58 of Regulation No 1122/2009 are irrelevant in the context of increasing the control sample, inasmuch as the thresholds of 0.1 hectare, 2 hectares or 3% of the area determined provided for in those articles relate to the consequences to be inferred from the definitive results obtained in the check.

77      In the second place, neither the wording of Article 33 of Regulation No 1122/2009 nor that of the Commission’s instructions on methodology make provision for a threshold of anomalies detected above which automatically increasing the control sample can be justified. In addition, that article and those instructions do not state the extent to which the control sample must be increased.

78      The Czech Republic therefore argues for a purposive interpretation of that article to the effect that increasing the control sample is necessary only with a view to ensuring effective checking and that that sample is representative. Thus, increasing such a sample is necessary only if the nature of the results and the knowledge of the other uninspected parcels of the farmer concerned suggest that there might be errors in respect of other parcels. It submits that such an interpretation is supported by the use of the term ‘anomalies’ in the plural in the first paragraph of Article 33 of that regulation, which should be understood as requiring that the irregularities are of a certain seriousness and not isolated and insignificant findings.

79      In the third place, the issue of increasing a sample in order to ensure effective checking and that that sample is representative must be examined, in practice, in the context of an assessment of all the circumstances of a particular situation. Assessment on a case-by-case basis, it argues, rather than one-size-fits-all solutions, constitutes, according to the Court in its judgment of 17 April 2018, B and Vomero (C‑316/16 and C‑424/16, EU:C:2018:256, paragraph 70), one of the fundamental approaches of EU law.

80      In the fourth place, the approach taken by the Czech authorities did not lead to any risk for EU funds, since it is only in the event that the competent authorities consider with a sufficient degree of certainty that errors in the agricultural area declared by the farmer concerned are no longer to be anticipated, in accordance with Articles 57 and 58 of Regulation No 1122/2009, that the controlled sample is not increased.

81      Lastly, in the fifth place, the procedure required by the General Court would require that the control sample be increased without any added value.

82      The Commission contends in response that, even if it agrees with the General Court’s finding in paragraph 42 of the judgment under appeal, it does not accept certain aspects of the reasoning for that judgment, in particular in paragraphs 37, 38 and 41 thereof, and it therefore requests the Court of Justice to maintain the judgment under appeal and to reject the second ground of appeal, while altering the reasoning for that judgment.

83      The Commission admits, in the first place, that, as regards paragraphs 37 and 38 of the judgment under appeal, Articles 57 and 58 of Regulation No 1122/2009 clarify the consequences following from a finding of errors in the declaration of the agricultural area eligible for aid but are not directly linked to increasing the control sample.

84      In the second place, it submits that Article 33 of Regulation No 1122/2009 does not contain any threshold below which there is no requirement to increase the sample size. In paragraphs 39 and 40 of the judgment of 28 January 2016, Slovenia v Commission (T‑667/14, EU:T:2016:34), the General Court rejected the interpretation adopted by the Slovenian authorities that, when the percentage error was less than 3%, the sample was not automatically increased, but it was for the inspector to decide whether or not it was necessary to extend the check to the whole holding.

85      The Commission submits that that approach must be endorsed, since nothing in the wording of Article 33 of Regulation No 1122/2009 supports the Czech Republic’s arguments. More specifically, that provision does not fix any threshold above which increasing the sample is obligatory and under which increasing that sample is at the discretion of the checking authority.

86      Moreover, it submits that, in accordance with Article 33 of Regulation No 1122/2009, read in the light of recital 44 thereof, the option for the Member States to restrict carrying out on-the-spot checks to a sample representing at least 50% of the agricultural parcels constitutes an exception to the rule that all the agricultural parcels for which an aid application has been submitted must be subject to on-the-spot checks, and it must therefore be strictly interpreted. Consequently, the Czech Republic’s choice to consider it superfluous to increase the control sample where the difference between the area of the parcels declared and that determined is less than 3% runs counter to the conditions for the application of the option to restrict the checks set out in Articles 26 and 33 of Regulation No 1122/2009.

 Findings of the Court

87      By its second ground of appeal, the Czech Republic argues, in essence, that the General Court erred in law in paragraph 38 of the judgment under appeal inasmuch as it did not interpret Article 33 of Regulation No 1122/2009 to the effect that the control sample was to be increased only in the event of more serious irregularities, which was a matter for the assessment of the national authorities. In addition, it argues that the General Court incorrectly interpreted Articles 33, 57 and 58 of that regulation in paragraphs 38 and 40 of the judgment under appeal in holding that increasing the control sample is obligatory not only where an over-declaration is found suggesting a significant possibility that the threshold of 3% or 2 hectares of the area declared will be exceeded, but also in the event of an over-declaration suggesting a significant possibility that the threshold of 0.1 hectare will be exceeded.

88      In paragraph 38 of the judgment under appeal, the General Court held that where a check concerning a sample of parcels covered by a single aid application reveals over-declarations representing between 0.1 and 2 hectares or 3% of the area determined, it is for the competent national authority to increase the control sample, in accordance with Article 33 of Regulation No 1122/2009, until no over-declaration of that magnitude is found.

89      In paragraph 40 of that judgment, the General Court noted that the Czech Republic stated that the 261 farmers in respect of which checks were not increased fell within the group of farmers who had submitted applications in which there were over-declarations not exceeding the 3% threshold, which did not, in the General Court’s view, rule out the possibility that a reduction in the aid to the level corresponding to the area determined, without a penalty, would be necessary.

90      As is apparent from paragraphs 30 to 36 of the judgment under appeal, the General Court based that finding on the text of Article 33 of Regulation No 1122/2009, Article 57(3) of that regulation and Article 58 thereof.

91      In the first place, as the Advocate General noted in point 61 of her Opinion, it follows from the very wording of Article 33 of Regulation No 1122/2009 that, where they have found that the sample check reveals anomalies, the national authorities do not, as a rule, have any discretion. Bound by a strict obligation, they must thus increase the control sample.

92      It must be added in this respect that, in accordance with recital 39 of that regulation, it is only in the situation – which is not claimed to exist in the present case – where it is impossible to increase the control sample in such a way as to guarantee a reliable and representative level of assurance that it is permissible not to increase that sample.

93      The Czech Republic’s argument that the presence of the term ‘anomalies’ in the plural in the first paragraph of Article 33 of Regulation No 1122/2009 means that only a finding of serious irregularities obliges the national authorities to increase the control sample must be rejected, since it is apparent from Article 2(10) of that regulation that the concept of ‘irregularities’, which is analogous to that of ‘anomalies’, does not refer only to serious irregularities but is, on the contrary, defined in that provision as ‘any non-respect of the relevant rules for the granting of the aid in question’.

94      It follows from the foregoing that Article 33 of Regulation No 1122/2009 must be interpreted to the effect that the national authorities are required to increase the control sample where the check carried out revealed anomalies, irrespective of the degree of such anomalies.

95      In the second place, as is apparent from their respective wording, Articles 57 and 58 of Regulation No 1122/2009 relate, first, to the basis of calculation in respect of areas declared and, secondly, to the reductions and exclusions in cases of over-declaration. By their scope, they clarify the consequences stemming from a finding of errors in the declaration of the agricultural area eligible for the grant of aid.

96      On the other hand, those articles do not in any way provide that, in certain cases, the Member States are bound by an obligation to increase the control sample. That obligation is set out in Article 33 of the regulation.

97      It follows from the foregoing that by referring, in paragraphs 38 and 40 of the judgment under appeal, to Articles 57 and 58 of that regulation to justify the obligation on the national authority to extend the control sample where a check covering a sample of parcels reveals over-declarations representing between 0.1 and 2 hectares, or 3% of the area determined, the General Court erred in law.

98      Nonetheless, since in order to reject the arguments directed against the ground of correction relating to the minimum rates of on-the-spot checks, in paragraph 42 of the judgment under appeal, the General Court correctly relied upon the interpretation of Article 33 of Regulation No 1122/2009 made in paragraph 30 of the judgment under appeal, the Czech Republic’s second ground of appeal must be rejected as unfounded.

 The third ground of appeal

 Arguments of the parties

99      By the third ground of appeal, the Czech Republic alleges that the conclusion reached by the General Court, in paragraph 49 of the judgment under appeal, that the Czech Republic’s argument concerning the aid paid for investments realised after the implementation of the national aid program but before the approval of the individual aid application, was ineffective, is vitiated by an error of law and a distortion of the facts and subject matter of the dispute.

100    It observes that, on several occasions during the administrative procedure, the Commission criticised it for granting aid applications for investments that had already been realised, presenting for that purpose an example of an investment realised on 29 September 2009. It also claims that the Commission took issue with it for paying such aid before the examination and approval or the introduction of the individual aid applications.

101    The Czech Republic observes that, during that procedure, the Czech authorities admitted making errors in connection with eight investments realised before the approval of the national aid programme, and provided information on all the other cases in which the investment aid had been paid for operations implemented after the approval of that programme but before the introduction or approval of the individual aid application. The unresolved disagreement with the Commission concerned the lawfulness of the aids in relation to those latter investments.

102    In its pleadings submitted in the action for annulment, the Commission argued that the payment of aid for investments already realised constituted an infringement of EU law. That complaint thus formed part of the proceedings before the General Court, the correction imposed by the contested decision covering, inter alia, the retroactive financing of those investments.

103    The Czech Republic accepts in this connection that the General Court correctly defined the subject matter of the proceedings in paragraph 43 of the judgment under appeal. It notes, however, that in paragraph 47 of that judgment, the General Court incorrectly interpreted point 7.1.1.1. of the Commission’s Summary Report when it found that the Commission had taken issue with the Czech Republic for financing an investment project although that investment had been completed in its entirety before the national aid programme had even been put in place. The Czech Republic observes that the project at issue, namely that realised on 29 September 2009, was realised after the approval of the national aid programme.

104    The Commission contends that the third ground of appeal should be rejected.

 Findings of the Court

105    By its third ground of appeal, the Czech Republic in essence challenges the General Court’s rejection of part of its arguments raised against the correction concerning retroactive payments as being ineffective.

106    It should be noted that, according to the Court’s settled case-law, it follows from the second subparagraph of Article 256(1) TFEU and from the first paragraph of Article 58 of the Statute of the Court of Justice of the European Union that the Court of Justice has no jurisdiction to establish the facts or, in principle, to examine the evidence which the General Court accepted in support of those facts. Provided that such evidence has been properly obtained and the general principles of law and the rules of procedure in relation to the burden of proof and the taking of evidence have been observed, it is for the General Court alone to assess the value which should be attached to the evidence produced before it. That appraisal therefore does not, save where that evidence has been distorted, constitute a point of law which is subject, as such, to review by the Court of Justice (judgment of 27 February 2020, Lithuania v Commission, C‑79/19 P, EU:C:2020:129, paragraph 70).

107    There is such distortion where, without recourse to new evidence, the assessment of the existing evidence is clearly incorrect. However, such distortion must be obvious from the documents on the Court’s file, without there being any need to carry out a new assessment of the facts and the evidence. Moreover, where an appellant alleges distortion of the evidence by the General Court, he or she must indicate precisely the evidence alleged to have been distorted by that court and show the errors of appraisal which, in his or her view, led to that distortion (judgment of 27 February 2020, Lithuania v Commission, C‑79/19 P, EU:C:2020:129, paragraph 71).

108    In paragraph 48 of the judgment under appeal, the General Court found that the Czech Republic did not refer to specific circumstances or evidence such as to demonstrate the veracity of its claim that solely in eight cases had investments received an aid although they had been realised before the national aid programme had been implemented.

109    In paragraph 49 of that judgment, the General Court found that the Czech Republic’s arguments in relation to aid paid for investments realised after the implementation of the national aid programme, but before the approval of the individual aid application, were ineffective.

110    It observed in that regard in paragraph 50 of that judgment that, according to point 7.1.1.1. of the Summary Report, the correction at issue was based exclusively, first, on the fact of investments having received an aid even though they had been realised before the national aid programme had been implemented and, secondly, on the fact that the Czech authorities had not supplied evidence demonstrating that those investments concerned only eight cases.

111    Inasmuch as the Czech Republic claims that the rejection of that argument as ineffective is vitiated by an error of law, on the ground that the General Court distorted the facts apparent from the contents of point 7.1.1.1. of the Summary Report by holding that the Commission had imposed a financial correction ‘exclusively’ because of investments realised before the implementation of the national aid programme, the Czech Republic still has to prove that distortion, as required by the case-law cited in paragraph 107 of the present judgment.

112    Without referring to a specific part of that summary report or other documents produced before the General Court, that Member State merely claimed that it was apparent – contrary to what was found by the General Court – from the correspondence exchanged with the Commission before the adoption of the contested decision and from the Summary Report, from the reference in point 7.1.1.1. of that summary report to an example of an investment allegedly realised after the implementation of the national aid programme, and from a statement in the defence submitted before the General Court, that the financial correction had also been imposed on account of investments in the wine sector realised after the implementation of the national aid programme but before the submission of the aid application.

113    It follows that since the General Court had observed in paragraph 50 of the judgment under appeal that no finding had been made in the Summary Report regarding those investments, it was able, correctly, to find that the Czech Republic’s arguments relating to the aids paid for such investments were ineffective.

114    In those circumstances, the third ground of appeal must be rejected as unfounded.

 The fourth ground of appeal

 Arguments of the parties

115    By its fourth ground of appeal, the Czech Republic submits that the General Court erred in law in finding, in paragraph 56 of the judgment under appeal, that Article 77(5) of Regulation No 555/2008 did not allow investments in the wine sector to be checked by means of sampling.

116    In the first place, according to the Czech Republic, Article 19 of Regulation No 555/2008 does not constitute a lex specialis with regard to Article 77(5) of that regulation, since that provision does not represent a general rule but a special provision laying down the specific procedures for checks in the field of investment operations. That provision refers expressly to Article 15 of Regulation No 479/2008, which does not concern all of the wine sector but only investments.

117    In the second place, the Czech Republic claims that the General Court’s interpretation in paragraph 61 of the judgment under appeal renders meaningless the reference to Article 27 of Regulation No 1975/2006 in Article 77(5) of Regulation No 555/2008, since it renders it impossible to proceed in accordance with that reference.

118    In the third place, it argues that that reference to Article 27 of Regulation No 1975/2006 was of a decisive nature since, in 2013, the Commission was obliged to make a legislative amendment consisting, in particular, in deleting that reference from the text of Regulation No 555/2008.

119    Lastly, in the fourth place, the Czech Republic submits that the Commission cannot impose on a Member State a correction that results from the unclear wording of a rule of which it itself is the author. It points out that according to paragraph 60 of the judgment under appeal, Regulation No 555/2008 ‘lays down two requirements that differ in scope’.

120    The Commission contends in response that the reference to ‘all the operations’ in Article 19 of Regulation No 555/2008 means that the Member States must carry out checks on all investments and not only on a minimal sample. In addition, in paragraph 56 of the judgment under appeal, the General Court refers to Article 77 of that regulation as a whole, and not only to Article 77(5). Moreover, Article 19 of that regulation constitutes a lex specialis with regard to Article 77 of that regulation, since, as the General Court noted in paragraph 56 of the judgment under appeal, the obligation set out in in Article 19 of Regulation No 555/2008 constitutes precisely one of the cases where, under Article 77(3) of that regulation, ‘systematic on the spot checks are foreseen …by this Regulation’.

121    Furthermore, it is clear from Article 77 of Regulation No 555/2008 that Articles 26 to 28 of Regulation No 1975/2006, which lay down the detailed rules for control procedures in respect of rural development measures financed by the EAFRD, apply mutatis mutandis to investments in the wine sector. Article 26 of Regulation No 1975/2006 provides for administrative checks, namely checks carried out by the Member States before on-the-spot checks are carried out, on all applications for support or payment claims and which may be regarded as constituting a ‘first check’ carried out by the Member States.

122    The Commission also takes the view that Articles 27 and 28 of Regulation No 1975/2006 contain rules in greater detail on the sample of expenses controlled, which must represent ‘at least 4% of the eligible public expenditure’, and on the effective content of the on-the-spot checks. In addition, the scope of the provisions of Regulation No 1975/2006, which covers a wide range of rural development measures, is much broader than that of Articles 19 and 77 of Regulation No 555/2008, which extends only to investments in the wine sector. It points out that, while some of the rural development measures covered by Regulation No 1975/2006 fall within the scope of ‘investments’, many of those measures are of a different nature.

 Findings of the Court

123    By its fourth ground of appeal, the Czech Republic in essence alleges that the General Court erred in law as regards the interpretation of Article 19 and Article 77(5) of Regulation No 555/2008 by ruling out the possibility that on-the-spot checks could be performed by sampling.

124    In that regard, in paragraphs 54 to 56 of the judgment under appeal, the General Court held in essence, following a reading of Article 19 of Regulation No 555/2008 in conjunction with Article 77 thereof, that Article 19 of that regulation constitutes a lex specialis in relation to Article 77 thereof and that Article 19 lays down a systematic requirement for an on-the-spot check before any payment of investment aid.

125    It is therefore necessary to interpret those provisions in order to examine the merits of the present ground of appeal.

126    In this connection, it is settled case-law of the Court of Justice that it is necessary, in interpreting a provision of EU law, to take into account not only the wording of the provision concerned, but also its context and the general scheme of the rules of which it forms part and the objectives pursued thereby (judgment of 5 July 2018, X, C‑213/17, EU:C:2018:538, paragraph 26).

127    In the present case, it must be observed that Article 77 of Regulation No 555/2008, entitled ‘General principles’, is included in Chapter I, entitled ‘Principles of control’, of Title V, entitled ‘Controls in the wine sector’, of that regulation. Under Article 77(3), the Member States are to perform on-the-spot checks on beneficiaries of aid by sampling, ‘except for the cases where systematic on the spot checks are foreseen by … [that] Regulation’. Furthermore, Article 77(5) of that regulation requires, in respect of investment measures in the wine sector, as defined in Article 15 of Regulation No 479/2008, now Article 103u of Regulation No 1234/2007, the application mutatis mutandis of Articles 27 and 28 of Regulation No 1975/2006, which lay down the specified procedures for on-the-spot checks of those investments by sampling.

128    Article 19 of Regulation No 555/2008, entitled ‘Financial management’, included in Section 6, entitled ‘Investments’, of Chapter II, entitled ‘Eligible measures’, which falls under Title II, entitled ‘Support programmes’, of that regulation, provides in the first subparagraph of paragraph 1 thereof that support is to be paid once it is ascertained that either a single operation or all the operations covered by the support application, according to the choice made by the Member State for the management of the measure, have been implemented and controlled on the spot.

129    As the Advocate General observed in point 77 of her Opinion, it must be noted that the wording of the first subparagraph of Article 19(1) of Regulation No 555/2008 does not provide for any obligation on the Member States to carry out systematic controls. That provision merely indicates that Member States must carry out on-the-spot checks of the measures implemented.

130    That literal interpretation is also consistent with the context and general scheme of that regulation.

131    In that regard, first, it follows from the title and the position of Article 19 of Regulation No 555/2008 in that regulation that the purpose of that article is to govern the procedure for the financial management of aid for the purposes of investments in the wine sector, and not the procedure for on-the-spot checks in respect of those investments. That article provides that the payment of the aid is dependant a prior on-the-spot check of the single operation or all the operations covered by the support application, but without specifying the procedure for that on-the-spot check. The wording and the purpose of that provision therefore do not preclude those operations from being represented by a sample that has been controlled on the spot.

132    Secondly, as the Advocate General observed in point 78 of her Opinion, it can be inferred from the general scheme of Regulation No 555/2008 that if that regulation requires Member States to make systematic checks, it does so expressly. It follows that, since no express obligation to that effect is included in the first subparagraph of Article 19(1) of that regulation, no obligation of that nature may be imposed on the Member States.

133    It follows that Article 19 of Regulation No 555/2008 is not capable of forming a basis for a derogation from Article 77(3) of that regulation, which provides for the performance of on-the-spot checks by sampling except, inter alia, where systematic on-the-spot checks are foreseen by that regulation.

134    It follows from all the foregoing considerations that the General Court erred in law in holding, in paragraph 56 of the judgment under appeal, that Article 19 of Regulation No 555/2008 requires the Member States to perform systematic checks.

135    In those circumstances, the fourth ground of appeal must be upheld, without it being necessary to examine the other arguments put forward in connection with that ground.

 The action before the General Court

136    In accordance with the second sentence of the first paragraph of Article 61 of the Statute of the Court of Justice of the European Union, the Court of Justice may, after quashing the decision of the General Court, itself give final judgment in the matter, where the state of the proceedings so permits, or refer the case back to the General Court for judgment.

137    In the present case, the Court of Justice does not have all the information necessary to rule definitively on the merits of the contested decision in so far as, by that decision, the Commission imposed, first, a one-off correction of EUR 462 517.83 in respect of the financial years 2013 to 2015 on the grounds, inter alia, of several deficiencies of various natures in the risk analysis, and, secondly, a flat-rate correction of EUR 636 516.20 in respect of the financial years 2011 to 2014 on the grounds both of payments made in connection with investments finalised before the support application was approved and insufficient on-the-spot checks so far as concerns investments financed in the wine sector.

138    As regards the one-off correction of EUR 462 517.83, it is apparent from paragraph 19 of the judgment under appeal that the Commission, as well as basing the contested decision on evidence of which the General Court’s assessment is criticised as incorrect in law in the first ground of appeal, also based it on the finding of divergences between the rates of irregularities shown by the on-the-spot checks by remote sensing based on a risk analysis and those shown by that same type of check, but based on samples selected at random.

139    Although the comparison of those rates of irregularities might be such as to give rise to serious and reasonable doubts as to the effectiveness of the on-the-spot checks by remote sensing, in particular as to whether the risk analysis on which those on-the-spot checks were based was relevant, the General Court did not examine the evidence which could have been provided by the Czech Republic in order to demonstrate, in that context, the effectiveness of its control system.

140    Furthermore, as regards the flat-rate correction of EUR 636 516.20, the Court does not have available to it the necessary information to determine the part of that correction which is affected by the upholding of the fourth ground of appeal, since that correction is based on several grounds.

141    It follows that the case must be referred back to the General Court in order for that court to rule on the matters set out in paragraphs 139 and 140 of the present judgment.

 Costs

142    Since the case is being referred back to the General Court, it is appropriate to reserve the costs.

On those grounds, the Court (Fourth Chamber) hereby:

1.      Sets aside the judgment of the General Court of the European Union of 13 September 2018, Czech Republic v Commission (T‑627/16, not published, EU:T:2018:538), in so far as, by that judgment, the General Court rejected the plea in law concerning the one-off correction of EUR 462 517.83 in respect of the financial years 2013 to 2015, to the extent to which it concerns the ground relating to deficiencies in the risk analysis, and the plea in law concerning a flat-rate correction of EUR 636 516.20, in respect of the financial years 2011 to 2014, to the extent to which it concerns the ground relating to insufficient on-the-spot checks so far as investments financed in the wine sector are concerned;

2.      Dismisses the appeal as to the remainder;

3.      Refers the case back to the General Court of the European Union;

4.      Reserves the costs.

[Signatures]

*      Language of the case: Czech.

© European Union
The source of this judgment is the Europa web site. The information on this site is subject to a information found here: Important legal notice. This electronic version is not authentic and is subject to amendment.