Language of document : ECLI:EU:T:2021:568

Provisional text

JUDGMENT OF THE GENERAL COURT (Second Chamber, Extended Composition)

15 September 2021 (*)

(Environment – Regulation (EC) No 443/2009 – Implementing Regulation (EU) No 725/2011 – Implementing Decision (EU) 2015/158 – Implementing Decision (EU) 2019/583 – Carbon dioxide emissions – Testing methodology – Passenger cars)

In Case T‑359/19,

Daimler AG, established in Stuttgart (Germany), represented by N. Wimmer, C. Arhold and G. Ollinger, lawyers,

applicant,

v

European Commission, represented by K. Talabér-Ritz and A. Becker, acting as Agents,

defendant,

APPLICATION under Article 263 TFEU for annulment of Commission Implementing Decision (EU) 2019/583 of 3 April 2019 confirming or amending the provisional calculation of the average specific emission of CO2 and specific emissions targets for manufacturers of passenger cars for the calendar year 2017 and for certain manufacturers belonging to the Volkswagen pool for the calendar years 2014, 2015 and 2016 pursuant to Regulation (EC) No 443/2009 of the European Parliament and of the Council (OJ 2019 L 100, p. 66), in so far as it excludes, as regards the applicant, the average specific emissions of CO2 and CO2 savings attributed to eco-innovations,

THE GENERAL COURT (Second Chamber, Extended Composition),

composed of S. Papasavvas, President, V. Tomljenović, F. Schalin (Rapporteur), P. Škvařilová-Pelzl and I. Nõmm, Judges,

Registrar: J. Palacio González, Principal Administrator,

having regard to the written part of the procedure and further to the hearing on 1 February 2021,

gives the following

Judgment

 Legal context

1        In the context of the European Union’s objective to reduce carbon dioxide (CO2) emissions from light-duty vehicles while ensuring the proper functioning of the internal market, the European Parliament and the Council of the European Union adopted Regulation (EC) No 443/2009 of 23 April 2009 setting emission performance standards for new passenger cars as part of the Community’s integrated approach to reduce CO2 emissions from light-duty vehicles (OJ 2009 L 140, p. 1).

2        In order to achieve that objective, Article 4 of Regulation No 443/2009 provides that, for the calendar year commencing on 1 January 2012 and each subsequent calendar year, each manufacturer of passenger cars is to ensure that its average specific emissions of CO2 do not exceed its specific emissions target determined in accordance with Annex I to that regulation or, where a manufacturer is granted a derogation under Article 11 of that regulation, in accordance with that derogation.

3        The determination of a manufacturer’s specific emissions target is made in accordance with Article 4 of Regulation No 443/2009, read in conjunction with Annex I to that regulation. Moreover, for the purposes of determining a manufacturer’s average specific emissions of CO2, Member States identify the data referred to in Article 8 of Regulation No 443/2009, read in conjunction with Annex II to that regulation, in particular the CO2 emissions for all new passenger cars registered in their territory during the previous year, as determined in the context of the type-approval of vehicles and designated in the certificate of conformity provided for in Article 18 of Directive 2007/46/EC of the European Parliament and of the Council of 5 September 2007 establishing a framework for the approval of motor vehicles and their trailers, and of systems, components and separate technical units intended for such vehicles (Framework Directive) (OJ 2007 L 263, p. 1).

4        The European Commission records that data in a public register. In accordance with Article 8(4) of Regulation No 443/2009, the Commission also establishes, by 30 June of each year, a provisional calculation of the average specific emissions of CO2, the specific emissions target and the difference between those two figures during the preceding calendar year, for each manufacturer, and communicates that data to those manufacturers.

5        After a period of three months from that notification, during which manufacturers may notify the Commission of any errors, the Commission, by 31 October each year, confirms or amends the data provisionally calculated, in accordance with Article 8(5) of Regulation No 443/2009. The Commission publishes the final data in the form of the list referred to in Article 10 of that regulation which indicates, for each manufacturer, the target set for the preceding calendar year, the average specific emissions of CO2 in the preceding year and the difference between those two figures.

6        If a manufacturer’s average specific emissions of CO2 exceed its target for the same calendar year, the Commission imposes an excess emissions premium laid down in Article 9 of Regulation No 443/2009. When levying that premium, the Commission relies on the data defined and formally established in accordance with Article 8(5) of that regulation.

7        The aim of Regulation No 443/2009 is not only to protect the environment and reduce CO2 emissions from new passenger cars but also to ensure the proper functioning of the internal market and in particular to encourage investment in new technologies. Therefore, with a view to promoting the long-term competitiveness of the European automotive industry, the regulation ‘actively promotes eco-innovation and takes into account future technological developments’ (see recital 13 of Regulation No 443/2009).

8        Consequently, Article 12 of Regulation No 443/2009, regarding eco-innovations, provides that CO2 savings achieved through the use of innovative technologies are to be taken into account, on certain conditions. To that end, they are deducted from the specific CO2 emissions of the vehicles in which those technologies are used in the calculation of a manufacturer’s average specific emissions of CO2.

9        On 25 July 2011, the Commission adopted Implementing Regulation (EU) No 725/2011 establishing a procedure for the approval and certification of innovative technologies for reducing CO2 emissions from passenger cars pursuant to Regulation No 443/2009 (OJ 2011 L 194, p. 19).

10      In order to benefit from CO2 savings based on an innovative technology in the determination of its average specific emissions of CO2, a manufacturer may apply to the Commission for approval of an innovative technology as an eco-innovation. To do so, it must submit an application for the approval of an innovative technology as an eco-innovation containing the information listed in Article 4 of Implementing Regulation No 725/2011. The application must, according to Article 4(2)(e) of Implementing Regulation No 725/2011, include a methodology to be used for demonstrating the reduction of CO2 emissions attributable to the innovative technology, or, where such methodology has already been approved by the Commission, a reference to the approved methodology. That testing methodology must, according to Article 6(1) of Implementing Regulation No 725/2011, provide results that are verifiable, repeatable and comparable and be capable of demonstrating in a realistic manner the CO2 emissions benefits of the innovative technology with strong statistical significance and, where relevant, take account of the interaction with other eco-innovations. According to Article 6(2) of Implementing Regulation No 725/2011, the Commission must publish guidance on the preparation of testing methodologies for different potential innovative technologies meeting the criteria in Article 6(1).

11      The Commission therefore published a document entitled ‘Technical Guidelines for the preparation of applications for the approval of innovative technologies pursuant to Regulation No 443/2009’ (‘the Technical Guidelines’). Chapter 4 of those guidelines, entitled ‘Testing methodologies’, provides that the application for the approval of an eco-innovation must include a testing methodology that provides accurate and verifiable results. It is clear from that chapter that an applicant may choose between two different approaches: the comprehensive approach and the simplified approach. Under the comprehensive approach, an applicant must develop a testing methodology and, where necessary, base that methodology on data and hardware. That documentation should, in principle, be provided as part of the application for approval and should be assessed by an independent and certified body for the purposes of the verification report. Under the simplified approach, an applicant can use predefined functions and averaged data given in the Technical Guidelines.

12      The Commission then assesses the application in accordance with Article 10 of Implementing Regulation No 725/2011 and, if necessary, adopts a decision approving the innovative technology as an eco-innovation. That decision specifies the information required for the certification of CO2 savings in accordance with Article 11 of Implementing Regulation No 725/2011, subject to the application of the exceptions to the right of public access to documents provided for in Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents (OJ 2001 L 145, p. 43).

13      A vehicle manufacturer which, in order to achieve its specific emissions target, wishes to benefit from a reduction in its average specific emissions of CO2 through CO2 savings resulting from an eco-innovation under Article 12 of Regulation No 443/2009 may then, with reference to the Commission’s decision on the approval of a specific eco-innovation, request the national approval authority referred to in Directive 2007/46 to certify the CO2 savings achieved by the use of that eco-innovation in its vehicles, in accordance with Article 11(1) of Implementing Regulation No 725/2011. The CO2 savings which are certified for the vehicle types are referred to both in the corresponding type-approval documentation issued by the national approval authority and in the certificate of conformity of the vehicles concerned, issued by the manufacturer.

14      With regard to the certification of CO2 savings by the national approval authorities and the taking into account of certified CO2 savings to determine a manufacturer’s average specific emissions of CO2, Article 12 of Implementing Regulation No 725/2011 provides for a review of certifications by the Commission on an ad hoc basis. The detailed provisions for that ad hoc review and any consequences are defined in paragraphs 1 to 3 of that article.

 Background to the dispute

15      On 2 December 2013 and 6 May 2014, Robert Bosch GmbH filed two applications for the approval of two types of high efficient alternators as eco-innovations. The first application, which is at issue in the present case, concerned high efficient alternators with high efficiency diodes (‘HED alternators’). In that application, Robert Bosch sought approval for various models of HED alternators numbered PL 3Q-130 HED, PL 3Q-150 HED, EL 6-140 HED, EL 7-150 HED, EL 7-150 Plus HED, EL 7-175 Plus HED, EL 8-180 HED, EL 8Q-180 HED and EL 8Q-190 HED as eco-innovations. In making its application, Robert Bosch followed the simplified approach provided for in the Technical Guidelines. In particular, it supplied a protocol of testing measurements for each model of alternator. In those measurement protocols, it gave details of the testing measurements used for the alternators. The protocols stated that some of the alternators had been ‘prepared’ (EL 7-175 Plus HED and EL 8Q-190 HED), that the amount of grease had been reduced by 25% (PL 3Q-130 HED and PL 3Q-150 HED), that they had been optimised (EL 7-150 HED) or that covering disks had been added (EL 8Q-190 HED). The purpose of preparation, which may consist of reducing the grease of the alternator bearings, replacing the sealing ring of the ball bearings with a cover plate or by adding covering disks, the function of which is to eliminate excess grease as quickly as possible, is to reduce the run-in phase. The generic term for the various preparation methods, including run-in, is preconditioning.

16      Pursuant to Article 4(2)(g) and Article 7 of Implementing Regulation No 725/2011, the application filed by Robert Bosch included a verification report dated 14 November 2013 (‘the verification report’) from an independent and certified body, TÜV SÜD Industrie Service GmbH (‘TÜV SÜD’). In accordance with Article 7(2)(c) of Implementing Regulation No 725/2011, TÜV SÜD verified the testing methodology used by Robert Bosch. It went on to confirm in the verification report that that testing methodology was appropriate for certifying the CO2 savings from the eco-innovation. The testing methodology verified was that described in the efficiency measurement requirements of the Verband der Automobilindustrie (Automobile Industry Federation, Germany; ‘VDA’) No 0 124 90A 0GB, 2 November 2010 edition. Paragraph 6.1.1 of those measurement requirements is worded as follows:

‘6.1.1 Run-in of ball bearings/ “prepared ball bearings”

New ball bearings must be run in before the alternator efficiency measurement so that accurate measurements can be made.

A one-hour run-in at a speed of n = 10 000 rpm with full load and a generator voltage of UG = 13.5 V is required. This run-in can also be replaced by a long-time full-load characteristic test (for example, “RB-warm” with the alternator operating for at least four hours at different speeds).

The run-in described above (one-hour full load with n = 10 000 rpm or the long-time full-load characteristic test) is only a minimum requirement. After this run-in, the variation of the friction of the ball bearings is still so high that precise efficiency measurements are not possible.

In order to reduce the variation due to friction in the ball bearings and obtain accurate efficiency measurements, the design stage must use “prepared ball bearings”. The use of “prepared ball bearings” is also recommended for factory measurements (quality control indicator).

“Prepared ball bearings” means ball bearings with an A-side and a B-side without a sliding seal and with the amount of grease reduced by around 20% compared with new bearings. Prepared ball bearings simulate bearings that have been well run in.’

17      In addition, according to the verification report, based on the evidence supplied by Robert Bosch, TÜV SÜD verified whether the CO2 savings achieved by the innovative technology was more than the threshold of 1 g CO2/km referred to in Article 9(1) of Implementing Regulation No 725/2011.

18      On 30 January 2015, the Commission approved both types of alternators by adopting Implementing Decision (EU) 2015/158 on the approval of two Robert Bosch GmbH high efficient alternators as the innovative technologies for reducing CO2 emissions from passenger cars pursuant to Regulation No 443/2009 (OJ 2015 L 26, p. 31).

19      The applicant, Daimler AG, is a German car manufacturer which fits certain passenger cars with Robert Bosch high efficient alternators.

20      Since 2015 and 2016, the applicant has been fitting alternator models HED Bosch EL 7-150 Plus HED and Bosch EL 7-175 Plus HED (‘the alternators at issue’) to some of its vehicles.

21      In accordance with Article 11 of Implementing Regulation No 725/2011 on the certification of CO2 savings achieved through eco-innovations, the applicant applied for and obtained certification from the Kraftfahrt-Bundesamt (Federal Office for Motor Vehicle Traffic, Germany; ‘the KBA’) for the CO2 savings achieved through the use of the alternators at issue in some of its vehicles.

22      In its Implementing Decision (EU) 2018/144 of 19 January 2018 confirming or amending the provisional calculation of the average specific emission of CO2 and specific emissions targets for manufacturers of passenger cars for the calendar year 2016 pursuant to Regulation No 443/2009 (OJ 2018 L 25, p. 64), the Commission took into account the CO2 savings achieved by the alternators at issue.

23      In 2017, the Commission, on the basis of Article 12 of Implementing Regulation No 725/2011 on the review of certifications, carried out an ad hoc review of the certifications of the applicant’s CO2 savings achieved through the use of the alternators at issue.

24      The Commission found that the CO2 savings certified by the KBA were much higher than the CO2 savings that could be demonstrated by applying the testing methodology which, according to the Commission, was prescribed by Article 1(3) of Implementing Decision 2015/158, read together with the annex to Commission Implementing Decision 2013/341/EU of 27 June 2013 on the approval of the Valeo Efficient Generation Alternator as an innovative technology for reducing CO2 emissions from passenger cars pursuant to Regulation No 443/2009 (OJ 2013 L 179, p. 98).

25      By letter of 7 March 2018, the Commission informed the applicant of the discrepancies found and set a deadline of 60 days for it to provide proof that the certified CO2 savings were correct.

26      Between 16 March and 24 July 2018, various exchanges relating to the Commission’s findings took place between the Commission, the applicant and the manufacturer of the alternators at issue, which, following a change of company name and then a sale by Robert Bosch of its starter and alternator division, was now SEG Automotive GmbH.

27      By letter of 22 October 2018, entitled ‘Notification of the withdrawal of eco-innovation CO2 savings attributed to Daimler AG vehicles fitted with Bosch HED EL 7-150 and 175 Plus efficient alternators’, the Commission, in essence, informed the applicant that, following the exchanges with the applicant and with the manufacturer of the alternators, it had concluded that the differences found in the level of CO2 savings were due to the different testing methodologies used. Consequently, the Commission informed the applicant that the CO2 savings certified by reference to Implementing Decision 2015/158 could not be taken into account for the calculation of its average specific emissions of CO2 in 2017. Finally, the Commission requested the applicant to verify the list of the vehicles concerned and to notify it of any errors or omissions within one month of receipt of that letter.

28      By letter of 22 November 2018, the applicant confirmed the list of vehicles concerned and disputed the Commission’s findings contained in the letter of 22 October 2018.

29      On 21 December 2018, the applicant brought an action for annulment of the letter of 22 October 2018, which was registered under Case T‑751/18.

30      By letter of 7 February 2019, the Commission replied to the observations made by the applicant in its letter of 22 November 2018.

31      On 3 April 2019, the Commission adopted Implementing Decision (EU) 2019/583 confirming or amending the provisional calculation of the average specific emission of CO2 and specific emissions targets for manufacturers of passenger cars for the calendar year 2017 and for certain manufacturers belonging to the Volkswagen pool for the calendar years 2014, 2015 and 2016 pursuant to Regulation No 443/2009 (OJ 2019 L 100, p. 66; ‘the contested decision’). Recital 13 of that decision states that the certified CO2 savings attributed to the alternators at issue should not be taken into account for the calculation of the applicant’s average specific emissions.

32      By order of 22 January 2020, Daimler v Commission (T‑751/18, EU:T:2020:5), the Court dismissed the action for annulment of the letter of 22 October 2018 as inadmissible on the ground, in particular, that that letter was not an act open to challenge.

 Procedure and forms of order sought

33      By application lodged at the Court Registry on 14 June 2019, the applicant brought the present action.

34      The defence, the reply and the rejoinder were lodged at the Court Registry on 2 September, 16 October and 28 November 2019 respectively.

35      On a proposal from the Second Chamber, the Court decided, pursuant to Article 28 of its Rules of Procedure, to assign the present case to a Chamber sitting in extended composition.

36      On a proposal from the Judge-Rapporteur, the Court (Second Chamber, Extended Composition) decided to open the oral procedure and, by way of measures of organisation of procedure pursuant to Article 89 of the Rules of Procedure, invited the parties to reply in writing to several questions before the hearing.

37      The parties complied with those requests within the periods prescribed.

38      The applicant claims that the Court should:

–        annul the contested decision, in so far as it specifies, in relation to the applicant, in Article 1(1), read together with Annex I, Tables 1 and 2, columns D and I, the average specific emissions of CO2 and the CO2 savings achieved through eco-innovations;

–        order the Commission to pay the costs.

39      The Commission contends that the Court should:

–        dismiss the action;

–        order the applicant to pay the costs.

 Law

40      In support of its action, the applicant relies on five pleas in law. The first plea alleges infringement of the second subparagraph of Article 12(1) of Implementing Regulation No 725/2011, read together with Article 1(3) of Implementing Decision 2015/158, due to the application of an incorrect Willans’ factor; the second plea alleges infringement of Article 12(1) of Implementing Regulation No 725/2011, read together with Article 1(3) of Implementing Decision 2015/158 and Article 6(1) of Implementing Regulation No 725/2011, due to the lack of a specific preconditioning in the ad hoc review; the third plea alleges infringement of Article 12(2) of Implementing Regulation No 725/2011 due to the certified CO2 savings not being taken into account for the calendar year 2017; the fourth plea alleges infringement of the right to be heard and the fifth plea alleges a breach of the obligation to state reasons.

41      At the hearing, the applicant withdrew the first plea in law, formal note of which was taken in the minutes of the hearing. As a result, it is no longer necessary to examine that plea.

 The second plea, alleging infringement of Article 12(1) of Implementing Regulation No 725/2011, read together with Article 1(3) of Implementing Decision 2015/158 and Article 6(1) of Implementing Regulation No 725/2011, due to the lack of a specific preconditioning in the ad hoc review

42      In the first place, the applicant claims that Implementing Decision 2015/158 provides for a specific preconditioning.

43      More specifically, the applicant infers from the wording of Implementing Decision 2015/158 that it provides for preconditioning. The Commission must, therefore, be regarded as aware that Robert Bosch had carried out the tests on which it based its application for the approval of its eco-innovation using alternators that had undergone a specific preconditioning. It was clear from the documentation supplied by Robert Bosch to the Commission that the alternators at issue had undergone preconditioning.

44      In the second place, the applicant claims that the general legal requirements applicable to the testing methodology necessitate a specific preconditioning.

45      The objective of Regulation No 443/2009 is to encourage the development and use of effective eco-innovations in real-world conditions. That requires the alternators to be preconditioned to correspond to a run-in of around 250 hours. The Commission should therefore have applied a testing methodology using preconditioning. To test an alternator without preconditioning, as the Commission did, would produce results representing only around 1% of the standard overall lifespan of the alternator. Furthermore, the applicant considers test results during that period to be generally inaccurate and variable.

46      That approach is also consistent with the provisions of the testing procedure under Regulation (EC) No 715/2007 of the European Parliament and of the Council of 20 June 2007 on type approval of motor vehicles with respect to emissions from light passenger and commercial vehicles (Euro 5 and Euro 6) and on access to vehicle repair and maintenance information (OJ 2007 L 171, p. 1), namely the New European Drive Cycle or worldwide harmonised light vehicles test procedure, which serves as a common benchmark for measuring CO2 savings achieved through the use of eco-innovations.

47      In the third place, the applicant submits that specific preconditioning of high efficient alternators authorised as eco-innovations in accordance with the manufacturer’s instructions forms a compulsory integral part of the testing methodology to be applied to those eco-innovations, to ensure that the results of that testing methodology meet the criteria of being verifiable, repeatable and comparable.

48      The Commission challenges the applicant’s arguments and maintains, in essence, that neither Implementing Decision 2013/341 nor Implementing Decision 2015/158 expressly prescribes preconditioning as a part of the prescribed testing methodology.

49      In the first place, the Commission maintains that Implementing Decision 2015/158 is a legal act of general application that must be interpreted in accordance with objective criteria. In its view, even if documents relating to the origins of that decision could be used as a reference for its interpretation, they would only contain information generally available, for example, the summary descriptions of applications for approval of an eco-innovation published on its website.

50      Aside from the fact that any knowledge or intention on the part of the Commission of a specific preconditioning carried out by the author of the application is of no relevance to the interpretation of Implementing Decision 2015/158, the Commission asserts that it was not aware that the testing methodology used by Robert Bosch involved the alternators at issue being preconditioned. A mere reference in the verification report to the measurement requirements of 2 November 2010 is insufficient. In any event, there are no details in paragraph 6.1.1 of those requirements as to the specific nature, conditions or duration of any preconditioning required thereunder.

51      It follows that even if the measurement requirements or other details of tests carried out had been included in the application for the approval of the alternators at issue as eco-innovations, and even if the Commission had been able to deduce from that that the author of the application had carried out a specific preconditioning in order to measure the CO2 savings achievable through the innovative technology used in those high efficient alternators, that does not mean that the Commission (implicitly) approved a preconditioning in Implementing Decision 2015/158 or that that particular type of preconditioning formed a part of the testing methodology prescribed in the annex to that decision.

52      The Commission recalls that the basis for the application for approval of the eco-innovation in question in the present case was only the simplified approach under the Technical Guidelines. That simplified approach is based exclusively on Chapter 5 of those guidelines. Any departure from the guidance set out therein must, according to the Commission, be justified.

53      In the second place, the Commission argues that the general legal requirements relied on by the applicant to define a testing methodology for eco-innovations do not constitute a valid reason for concluding that either Implementing Decision 2013/341 or Implementing Decision 2015/158 is unlawful, or that those implementing decisions should be read as meaning that the testing methodology that they prescribe must include the specific preconditioning sought by the applicant.

54      In the third place, the Commission maintains, in essence, that if the duration and the nature of the preconditioning of an eco-innovation not governed by the legal act in question had to conform with the instructions of the relevant manufacturer, the comparable nature of the testing methodology thus required could not be guaranteed, which would be contrary to Article 12(1) of Regulation No 443/2009 or Article 6 of Implementing Regulation No 725/2011. It considers that such an interpretation would be likely to lead to each manufacturer prescribing a specific preconditioning that allowed the CO2 savings achieved through the use of the eco-innovation in question to be measured at its peak performance.

55      The Commission adds that, from the point of view of legal certainty and equal treatment, a specific preconditioning not expressly prescribed as forming part of the applicable testing methodology cannot be allowed. It takes the view that testing methodology conditions which affect the test results are only valid if they are expressly mentioned in the legal act in question. The Commission considers that unregulated conditions must not be applied and that any regulatory gap cannot be filled by means of manufacturer’s instructions.

56      It must be recalled that Article 12 of Implementing Regulation No 725/2011 provides that the Commission is to ensure that the certifications and the CO2 savings attributed to individual vehicles are verified on an ad hoc basis. Where it finds that there is a difference between the certified CO2 savings and the savings it has verified using the relevant testing methodology or methodologies, the Commission is to notify the manufacturer of its findings. The manufacturer may within 60 days of receipt of the notification provide the Commission with evidence demonstrating the accuracy of the certified CO2 savings. Where that evidence is not provided within the indicated time period, or it finds that the evidence provided is not satisfactory, the Commission may decide not to take the certified CO2 savings into account for the calculation of the average specific emissions of that manufacturer for the following calendar year.

57      It must also be recalled that Article 6(1) of Implementing Regulation No 725/2011 provides that ‘the testing methodology referred to in [Article 4(2)(e)] shall provide results that are verifiable, repeatable and comparable. It shall be capable of demonstrating in a realistic manner the CO2 emissions benefits of the innovative technology with strong statistical significance and, where relevant, take account of the interaction with other eco-innovations’.

58      In the present case, the differences noted by the Commission were due to the fact that the CO2 savings achievable through the use of the alternators in question were determined according to a specific preconditioning for the purposes of certification by the KBA, but without any preconditioning for the ad hoc review carried out by the Commission. It must be observed, in addition, that Implementing Decisions 2013/341 and 2015/158 are silent on the question of whether the alternators at issue had undergone preconditioning.

59      It must be noted that the applicant, in accordance with the third subparagraph of Article 12(1) of Implementing Regulation No 725/2011, provided the Commission with evidence demonstrating the accuracy of the certified CO2 savings with regard to the testing methodology which it advocates and in which the alternators at issue had undergone a specific preconditioning.

60      The information on the file in the present case also leads to the conclusion that Robert Bosch had used a testing methodology with specific preconditioning for the approval of the alternators at issue as eco-innovations. The representatives of SEG Automotive stated that the legal predecessor to that company, Robert Bosch, had obtained the results that appear in the documents submitted on the basis of tests in which the bearings in the alternators had been fitted with covering disks and had also been degreased. Similarly, it is clear from the exchanges between the representatives of the applicant, the Commission and SEG Automotive that the latter’s representatives stated that a specific preconditioning had been carried out. The file also shows that the checks carried out by the Commission in order to approve the alternators at issue as eco-innovations were made on the basis of the verification report. That report states, at page 5, that Robert Bosch obtained the test results on the basis of the VDA efficiency measurement requirements No 0 124 90A 0GB, 2 November 2010 edition. Those requirements expressly state that test results are not reliable without preconditioning.

61      It must also be noted that the testing measurement protocols in the file relating to Implementing Decision 2015/158 state that one of the alternator models at issue, namely model EL 7-175 Plus HED, had undergone specific preconditioning. The reference ‘with prepared ball bearings’ appears in measurement protocol 10s01855T 5 of 13 October 2010 and in measurement protocol 10s01855 of 13 October 2010. Those testing measurement protocols also show that five other alternator models which were the subject of Implementing Decision 2015/158 but which are not at issue in the present case had undergone preconditioning (see paragraph 15 above).

62      It must also be borne in mind that the applicant explained at the hearing, without being contradicted by the Commission, that preconditioning was standard industry practice. The Commission itself explained that, following the order of 22 January 2020, Daimler v Commission (T‑751/18, EU:T:2020:5), it had been contacted by around 10 alternator manufacturers who preconditioned their alternators in various ways and wished to know where they stood in that regard.

63      However, as the parties have confirmed, the preconditioning method may differ substantially according to the technical specifications of each alternator (number of hours’ run-in, degreasing, covering disks, and so forth), regardless of whether they are made by the same manufacturer or by different manufacturers.

64      In the present case, the Commission argues simply that the testing methodology advocated by the applicant was not permitted. The Commission merely stated, in its letter of 22 October 2018, that the applicant had not applied the ‘official’ testing methodology laid down by Implementing Decisions 2013/341 and 2015/158 and, in the contested decision, that preconditioning was neither prescribed nor permitted by those decisions.

65      In that regard, it must be recalled that a testing methodology that uses preconditioning is, as stated in paragraph 62 above, standard practice in the alternator industry and is the norm in the surrounding legislation such as Regulation No 715/2007, which provides for testing methodologies that reproduce real driving conditions.

66      In addition, it must be noted that it is apparent from paragraph 6.1.1 of the VDA measurement requirements No 0 124 90A 0GB, 2 November 2010 edition, that, in order to reduce variation due to friction in the ball bearings and obtain accurate efficiency measurements, the design stage must use ‘prepared ball bearings’. The use of ‘prepared ball bearings’ is also recommended for factory measurements (see paragraph 16 above).

67      Furthermore, the testing methodology advocated by the applicant was reviewed by TÜV SÜD which, in its capacity as an independent and certified body, verified and confirmed, pursuant to Article 7(2)(c) of Implementing Regulation No 725/2011, that Robert Bosch’s testing methodology was appropriate for certifying the CO2 savings from the alternators at issue and that it met the minimum requirements specified in Article 6(1) of that regulation.

68      It follows that the methodology advocated by the applicant must be regarded as relevant and appropriate for the purposes of the assessment in question. There is nothing in the file to lead to the conclusion that such a testing methodology would produce results that were not verifiable, repeatable and comparable for the purposes of Article 6(1) of Implementing Regulation No 725/2011. On the contrary, when the Commission notified the applicant of its questions about the certified CO2 savings, the applicant produced evidence demonstrating the accuracy of those savings, in accordance with Article 12(1) of Implementing Regulation No 725/2011. It is in fact clear from the file that, following the Commission’s questions, SEG Automotive carried out new measurements on the alternators at issue, which showed that the efficiency (the relationship between the electrical power produced and the mechanical power absorbed) of those alternators was at the expected level and the value obtained corresponded to the efficiency values confirmed by TÜV SÜD in the verification report. SEG Automotive concluded that the differences between the tests carried out by the Commission and the tests it carried out itself were due to the fact that the Commission had not used any preconditioning.

69      It is common ground between the parties that the testing methodology used by the Commission during the ad hoc review did not include any preconditioning. Furthermore, that method was not prescribed by the legislation and was not the industry standard. That testing methodology was therefore fundamentally different from the testing methodology used by Robert Bosch and advocated by the applicant.

70      In that regard, it must be borne in mind that Article 12 of Implementing Regulation No 725/2011 does not define the testing methodology to be used when carrying out an ad hoc review. However, it can be seen from recital 13 of Implementing Regulation No 725/2011 that, as part of the ad hoc review, the Commission must verify whether the certified savings are consistent with the level of savings resulting from the decision approving a technology as an eco-innovation. The testing methodology used by the Commission does not, in reality, amount to a verification of the applicant’s certified savings as against the savings resulting from Implementing Decision 2015/158. On the contrary, by using a different testing methodology, the Commission effectively made it impossible to compare the certified savings with the savings set out in Implementing Decision 2015/158.

71      The Commission justifies its testing methodology on grounds of equal treatment and legal certainty. However, it can be seen from the foregoing that the Commission’s testing methodology consists of treating different situations in the same way, which does not safeguard the principle of equal treatment which, as a general principle of EU law, requires that comparable situations are not be treated differently and that different situations are not be treated in the same way unless such treatment is objectively justified (see judgment of 20 June 2019, ExxonMobil Production Deutschland, C‑682/17, EU:C:2019:518, paragraph 90 and the case-law cited). Therefore, if the Commission does not take into account the specific technical features of each alternator or the way in which it has been preconditioned, its testing methodology, which consists of carrying out tests on alternators that have not been preconditioned, is liable to favour some car manufacturers and to disadvantage others.

72      Similarly, the Commission’s approach cannot be justified for reasons of legal certainty. The principle of legal certainty requires that rules of law be clear, precise and predictable as regards their effects, in particular where they may have unfavourable consequences for individuals and undertakings (judgment of 22 April 2015, Poland v Commission, T‑290/12, EU:T:2015:221, paragraph 50). Given that the ad hoc review could have serious consequences for car manufacturers, that the testing methodology used by the Commission in the present case is not defined clearly and precisely either in Implementing Decision 2015/158 or in any other legislation and that that methodology does not constitute standard industry practice, it cannot be regarded as an appropriate means of safeguarding the principle of legal certainty.

73      Those findings cannot be called into question by the fact that Robert Bosch had made its application for approval of the alternators at issue as eco-innovations using the simplified approach under the Technical Guidelines.

74      Even if an application is made using the simplified approach under the Technical Guidelines, the Commission must check the application before approving it and, if appropriate, raise objections concerning the appropriateness of the testing methodology, as part of the complete examination of the application which it must carry out under Article 10(2) of Implementing Regulation No 725/2011. As mentioned in paragraphs 15 and 16 and 60 to 63 above, the Commission had access to information, in particular that contained in the verification report prepared by TÜV SÜD, which it could not disregard without rendering ineffective the report specified in Article 4(2)(g) and Article 7 of Implementing Regulation No 725/2011. In addition, the Commission had access to the information in the testing measurement protocols from which it was apparent that the alternators at issue had undergone preconditioning in order to be approved as eco-innovations, especially given that the use of preconditioning was standard industry practice. Therefore, if there had been questions about or objections to the preconditioning, the Commission should have raised enquiries with Robert Bosch in that regard before adopting Implementing Decision 2015/158. In any event, the fact that Robert Bosch used the simplified approach under the Technical Guidelines is not a valid reason why objections about the testing methodology were only raised at the time of the ad hoc review.

75      The same goes for the argument that Robert Bosch did not, in any event, provide specific details of the preconditioning during the procedure for approval of the HED alternators. As is clear from paragraph 74 above, it is during that procedure that the Commission is entitled to raise objections or request further details in relation to the testing methodology, which, in the present case, included a specific preconditioning as mentioned in the verification report. In the absence of any objections or requests for further details during that procedure, the Commission cannot validly use a different testing methodology in the ad hoc review it carries out on the applicant.

76      It is also necessary to dismiss the Commission’s argument that information that is not publicly disclosed cannot be regarded as forming part of Implementing Decision 2015/158 and Implementing Decision 2013/341 as they are decisions of general application. In that regard, it is sufficient to note that, under Article 10(1) of Implementing Regulation No 725/2011, the Commission must publish ‘the summary description of the innovative technology and the testing methodology referred to in [Article 4(2)(c)]’. The fact that the description is ‘summary’ necessarily means that its publication cannot be comprehensive in nature.

77      In the light of the foregoing, it must be held that the Commission failed to carry out the ad hoc review in accordance with Article 12 of Implementing Regulation No 725/2011 and that it erred in law by excluding a testing methodology that involved preconditioning.

78      This plea must therefore be upheld as well founded, which is sufficient to annul the contested decision in accordance with the form of order sought by the applicant.

79      The Court nevertheless considers it appropriate to examine the third plea.

 The third plea, alleging infringement of Article 12(2) of Implementing Regulation No 725/2011 due to the certified CO2 savings not being taken into account for the calendar year 2017

80      The applicant considers the wording of Article 12(2) of Implementing Regulation No 725/2011 to be clear and unambiguous: it is only permissible not to take the CO2 savings into account for the following calendar year, that is to say, in the present case, for the year 2019. The applicant takes the view that that provision expresses the principle of the protection of legitimate expectations and must enable the manufacturer to carry out reliable ‘fleet planning’ by taking into account, in its CO2 savings, the CO2 savings achievable through eco-innovations. Not taking the certified CO2 savings for past years into account (ex tunc) would breach that principle.

81      The Commission disputes the applicant’s arguments. It maintains, in essence, that its method of proceeding complies with Regulation No 443/2009 and with Article 12 of Implementing Regulation No 725/2011.

82      The Commission takes the view that, under that legislation, it has to take account of manufacturers’ CO2 performance for the preceding calendar year. In practice, it should only consider the data supplied by Member States a certain time after the end of the calendar year. That is also the case for data on eco-innovations.

83      It follows that, when the Commission finds that the certified emissions from an eco-innovation are not verifiable, that should be taken into account for the year preceding the one when that discovery was made. Therefore, in the Commission’s view, the reference to the ‘following calendar year’ should be understood to mean that it is not entitled to exclude the CO2 savings that it has already confirmed in a formal decision for a given year.

84      The Commission maintains, in that regard, that reasons pertaining to the protection of legitimate expectations prevented it from making a retrospective amendment, for example, to the applicant’s average specific emissions of CO2 for the calendar year 2016 which were determined in Implementing Decision 2018/144, by taking into account the CO2 savings at issue in the present case when it found, during 2018, that the accuracy of the certifications in question could not be demonstrated either as part of its review or by the applicant.

85      Furthermore, the use of the phrase ‘are no longer taken into account’ in Article 12(3) of Implementing Regulation No 725/2011 confirms that the certified CO2 savings achieved through the use of eco-innovations which were taken into account in the past to define a manufacturer’s emission performance can, according to Article 12 of Regulation No 443/2009, ‘no longer’ be taken into account from the moment that it is established that they cannot be verified. The Commission maintains that to apply the applicant’s reasoning would mean, in practice, that it would have to adopt a decision in which it took CO2 savings into account even when it knew that they were not verifiable.

86      The Commission considers that such an interpretation of Article 12 of Implementing Regulation No 725/2011 conflicts with Regulation No 443/2009, which calls for the most accurate databases possible to determine the emission performance of all manufacturers. The Commission refers, in that regard, to recital 25 of that regulation and to recital 4 of Commission Regulation (EU) No 1014/2010 of 10 November 2010 on monitoring and reporting of data on the registration of new passenger cars pursuant to Regulation No 443/2009 (OJ 2010 L 293, p. 15). Thus, according to the Commission, if it were to apply Regulation No 443/2009 in that way, it would infringe Article 12 of that regulation which presupposes that the CO2 savings achieved through the use of eco-innovations that are taken into account for a manufacturer’s emission performance ‘must make a verified contribution to CO2 reduction’.

87      It must be recalled that the wording of Article 12(2) and (3) of Implementing Regulation No 725/2011 is as follows:

‘2.      Where the evidence referred to in paragraph 1 is not provided within the indicated time period, or it finds that the evidence provided is not satisfactory, the Commission may decide not to take the certified CO2 savings into account for the calculation of the average specific emissions of that manufacturer for the following calendar year.

3.      A manufacturer for which the certified CO2 savings are no longer taken into account may apply for a new certification of the vehicles concerned in accordance with the procedure laid down in Article 11.’

88      It must be observed, in that regard, that, in accordance with Article 8(4) and (5) of Regulation No 443/2009, every year the Commission provisionally calculates for each manufacturer the average specific emissions of CO2, the specific emissions target and the difference between those two figures in the preceding calendar year (see paragraphs 4 and 5 above). The Commission must confirm or amend the provisional calculations by 31 October each year, pursuant to Article 8(5) of Regulation No 443/2009. That calculation is made on the basis of information provided by Member States for the preceding calendar year.

89      It is against that background that the Commission interpreted Article 12(2) of Implementing Regulation No 725/2011 in considering that the CO2 savings certified for the 2018 calculation of average specific emissions of CO2 should not be taken into account for the preceding calendar year, namely 2017. However, even if there is a certain logic to that interpretation, it must be held that it is contrary to the clear and unambiguous wording of Article 12(2) of Implementing Regulation No 725/2011. It is not plausible for the phrase ‘following calendar year’ to be interpreted as actually referring to the preceding calendar year. Such an interpretation also raises questions in view of the principle of legal certainty (see paragraph 72 above).

90      In that regard, it must be held that the Commission’s interpretation, apart from lacking clarity and precision, is unfavourable to the applicant given that the contested decision has serious retroactive consequences for the applicant whereas it should have affected only ‘the following calendar year’.

91      Neither is the expression ‘are no longer taken into account’ in Article 12(3) of Implementing Regulation No 725/2011 sufficient to cause the clear and unambiguous wording of the phrase ‘following calendar year’ to be given a contrary meaning.

92      In addition, in so far as the Commission submits that Article 12 of Implementing Regulation No 725/2011 should be interpreted in a way that is consistent with Regulation No 443/2009, it must be recalled that although, according to settled case-law, an implementing regulation must also be given, if possible, an interpretation consistent with the basic regulation, that case-law does not apply in the case of a provision of an implementing regulation whose meaning is clear and unambiguous and therefore requires no interpretation (see judgment of 28 February 2017, Canadian Solar Emea and Others v Council, T‑162/14, not published, EU:T:2017:124, paragraph 150 and the case-law cited). Otherwise, the principle that secondary EU law must be interpreted in conformity would serve as the basis for an interpretation of that law contra legem, which is not acceptable (see, to that effect, order of 17 July 2015, EEB v Commission, T‑685/14, not published, EU:T:2015:560, paragraph 31 and the case-law cited).

93      In the light of the foregoing considerations, the interpretation that the ‘following calendar year’ refers to the calendar year following the year of the ad hoc review is the one to be used.

94      The present plea must therefore be upheld as well founded and, there being no need to examine the fourth and fifth pleas in this action, the contested decision must be annulled in so far as it specifies, in Article 1(1), read in conjunction with Annex I, Tables 1 and 2, columns D and I, the average specific emissions of CO2 and the CO2 savings from eco-innovations for the applicant.

 Costs

95      Under Article 134(1) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs, if they have been applied for in the successful party’s pleadings.

96      Since the Commission has been unsuccessful, it must be ordered to pay the costs, in accordance with the form of order sought by the applicant.

On those grounds,

THE GENERAL COURT (Second Chamber, Extended Composition)

hereby:

1.      Annuls Article 1(1), read together with Annex I, Tables 1 and 2, columns D and I, of Commission Implementing Decision (EU) 2019/583 of 3 April 2019 confirming or amending the provisional calculation of the average specific emission of CO2 and the specific emissions targets for manufacturers of passenger cars for the calendar year 2017 and for certain manufacturers belonging to the Volkswagen pool for the calendar years 2014, 2015 and 2016 pursuant to Regulation (EC) No 443/2009 of the European Parliament and of the Council, in so far as it specifies, in relation to Daimler AG, the average specific emissions of CO2 through CO2 savings achieved through eco-innovations;

2.      Orders the European Commission to bear its own costs and to pay those incurred by Daimler.

Papasavvas

Tomljenović

Schalin

Škvařilová-Pelzl

 

      Nõmm

Delivered in open court in Luxembourg on 15 September 2021.

[Signatures]


*      Language of the case: German.