1.

In 2016, in response to a well-known scandal concerning the actual level of emissions of certain diesel-powered vehicles, the European Commission introduced a real driving emissions (‘RDE’) testing procedure to complement the previous laboratory procedure, the new European driving cycle (‘NEDC’). The latter had been the only testing procedure used up until that point for ‘type-approval’ of new light passenger and commercial vehicles. Moreover, the Commission also set the values in relation to the mass of oxides of nitrogen (‘NOx’) emissions that were not to be exceeded in those RDE tests.

2.

In its judgment of 13 December 2018, upholding in part the actions lodged by the Ville de Paris (‘City of Paris’), the Ville de Bruxelles (‘City of Brussels’) and the Ayuntamiento de Madrid (‘Municipality of Madrid’) (collectively, ‘the respondents’), the General Court partially annulled the Commission’s regulation setting certain values for the RDE tests in relation to the NOx emissions. ( 2 ) In essence, the General Court found that, by setting those values at too high a level, the Commission de facto amended the Euro 6 standard set out by the EU legislature, for which it lacked competence. ( 3 )

3.

In their appeals, the Federal Republic of Germany, Hungary and the Commission (collectively, ‘the appellants’) ask the Court of Justice to set aside the judgment of the General Court. These appeals raise a number of issues, two of which are perhaps more important than the others. The first is of a procedural nature. It concerns the concept of ‘direct concern’ in relation to a Member State’s regional entity that wishes to challenge an EU act. The second is of a substantive nature. What is the scope of the Commission’s room for manoeuvre when it is empowered to amend non-essential elements of a basic instrument through the adoption of implementing legislation?

4.

Before a new vehicle model can be placed on the market in the European Union, the manufacturer must submit it to the type-approval process. This process is intended to certify that a vehicle prototype meets all EU safety, environmental and production requirements (as set out in the main legislation and the relevant regulatory acts). ( 4 )

5.

As far as the present proceedings are concerned, the main piece of legislation in force at the material time was 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). ( 5 )

6.

Under the terms of Article 4(3) of Directive 2007/46:

‘Member States shall register or permit the sale or entry into service only of such vehicles, components and separate technical units as satisfy the requirements of this Directive.

They shall not prohibit, restrict or impede the registration, sale, entry into service or circulation on the road of vehicles, components or separate technical units, on grounds related to aspects of their construction and functioning covered by this Directive, if they satisfy the requirements of the latter.’

7.

Annex IV of Directive 2007/46, entitled ‘Requirements for the purpose of EC type-approval of vehicles’, included, inter alia, a list of applicable regulatory acts.

8.

With regard to pollutant emissions in the case of light passenger and commercial vehicles (Euro 5 and Euro 6), the relevant regulatory act was 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. ( 6 ) Pursuant to its Article 1(1), that regulation ‘establishes common technical requirements for the type approval of motor vehicles … and replacement parts, such as replacement pollution control devices, with regard to their emission’s.

9.

Article 5(3) of Regulation No 715/2007, provides that the specific procedures, tests and requirements for type approval which are designed to amend non-essential elements of that regulation, by supplementing it, must be adopted in accordance with the regulatory procedure with scrutiny. ( 7 )

10.

Article 10 of Regulation No 715/2007 provides, inter alia, that, for passenger vehicles and vehicles for the carriage of persons, the limits of the Euro 5 standard are to apply from 1 September 2009 for type approval, and that from 1 January 2011 new vehicles that do not comply with this standard can no longer be registered, put up for sale or enter into service. It also lays down that the Euro 6 standard limits are to apply from 1 September 2014 for type approval and that from 1 September 2015 new vehicles that do not comply with this standard can no longer be registered, put up for sale or enter into service. Tables 1 and 2 of Annex I to Regulation No 715/2007 set the limits for NOx emissions for a diesel passenger vehicle or vehicle for the carriage of persons categories at 180 mg/km for Euro 5 and 80 mg/km for Euro 6.

11.

Article 14(3) of Regulation No 715/2007 requires the Commission to keep under review the procedures, tests and requirements referred to in Article 5(3) of the same regulation, as well as the test cycles used to measure emissions. If the review finds that these are no longer adequate or no longer reflect real world emissions, they must – according to that provision – be adapted so as to adequately reflect the emissions generated by actual driving on the road. The necessary measures, which are designed to amend non-essential elements of that regulation, by supplementing it, are to be adopted in accordance with the regulatory procedure with scrutiny.

12.

Commission Regulation (EC) No 692/2008 of 18 July 2008 has been adopted in order to implement Articles 4 and 5 of Regulation No 715/2007. ( 8 ) Recital 2 of Regulation No 692/2008 states that new light-duty vehicles must comply with new emission limits, that those technical requirements take effect in two stages, Euro 5 starting from 1 September 2009 and Euro 6 from 1 September 2014. That regulation thus sets the requirements necessary for the type approval of Euro 5 and Euro 6 specification vehicles.

13.

Article 3(1) of Regulation No 692/2008 provides that EC type-approval is granted with regard, inter alia, to pollutant emissions if the manufacturer demonstrates that the vehicles concerned comply with the test procedures specified in various annexes thereto. Article 3(2) of the same regulation clarifies, in essence, that, depending on their characteristics, vehicles are to be subject to different types of tests listed in Figure I.2.4 of Annex I, which are themselves described in various annexes. Article 3(5) of that regulation states that ‘the manufacturer shall take technical measures so as to ensure that the tailpipe and evaporative emissions are effectively limited, in accordance with this Regulation, throughout the normal life of the vehicle and under normal conditions of use’.

14.

In September 2015, the so-called ‘dieselgate’ scandal broke, when the United States Environmental Protection Agency formally accused Volkswagen AG of breaching US emissions standards. That car manufacturer subsequently admitted that a ‘defeat device’ had indeed been installed in a number of diesel-fuelled vehicles worldwide. ( 9 ) That device was able to detect when a vehicle was being tested in a laboratory and could activate its emissions control system for the purpose of complying with NOx emissions standards. ( 10 ) However, once outside of laboratory conditions, that device would switch off the emissions control system, meaning that the vehicle would produce emissions above the US legal NOx limit. It subsequently came to light that other vehicles, from that same manufacturer, and from a number of other manufacturers, also did not meet the standards laid down for the NOx limit set out in the EU legislation. ( 11 )

15.

At EU level, in response to those circumstances, Commission Regulation (EU) 2016/427 of 10 March 2016 amending Regulation No 692/2008 as regards emissions from light passenger and commercial vehicles (Euro 6) ( 12 ) was adopted amongst other legislation. That regulation introduced the RDE test procedure. That new test procedure seeks to better reflect the actual level of pollutant emissions under real driving conditions than the previously used test, the NEDC. The latter assessed the exhaust emissions of cars during their type-approval process in standardised laboratory conditions only.

16.

Shortly thereafter, the Commission adopted Regulation 2016/646 – the contested regulation – which constitutes one of the regulatory acts under the type-approval procedure laid down by Directive 2007/46. The contested regulation supplements the requirements for the RDE tests, setting not-to-exceed (‘NTE’) values in relation to NOx emissions. Those values stem from the application of pollutant conformity factors (‘CFs’) to the pollutant emission limits laid down for the Euro 6 standard in Regulation No 715/2007. ( 13 )

17.

The contested regulation amends Regulation No 692/2008 in a number of respects. In particular, Article 1(2) of the contested regulation replaces the third paragraph of Article 3(10) of Regulation No 692/2008 with the following text: ‘until three years after the dates specified in Article 10(4) and four years after the dates specified in Article 10(5) of Regulation [No 715/2007] the following provisions shall apply’.

18.

In addition, Article 1(6) of the contested regulation and Annex II thereto amend Annex IIIA to Regulation No 692/2008 by, inter alia, inserting points 2.1.1 to 2.1.3 in the latter annex. Point 2.1.1 provides a final conformity factor for the mass of NOx of ‘1 + margin’, with the margin being equal to 0,5. The ‘margin’ is defined as ‘a parameter taking into account the additional measurement uncertainties introduced by the [portable emissions measurement system (PEMS)] equipment, which are subject to an annual review and shall be revised as a result of the improved quality of the PEMS procedure or technical progress’. According to point 2.1.2, by way of exception to the provisions of point 2.1.1, during a period of 5 years and 4 months following the dates specified in Article 10(4) and (5) of Regulation No 715/2007 and upon request of the manufacturer, a temporary CF pollutant of 2.1 for the mass of NOx may apply.

19.

By their actions under Article 263 TFEU, the respondents requested the General Court to annul the contested regulation. The City of Paris also asked the General Court to order the Commission to pay symbolic damages of one euro as compensation for the harm caused to it as a result of the adoption of that regulation.

20.

On 13 December 2018, the General Court delivered the judgment under appeal, upholding in part the actions.

21.

The General Court found the contested regulation to be a regulatory act which was of direct concern to the respondents and did not entail implementing measures within the meaning of the fourth paragraph of Article 263 TFEU.

22.

As regards the merits of the actions, the General Court decided to assess first the grounds alleging a lack of competence, on the part of the Commission, to adopt the contested regulation. In that regard, the General Court held the limit on NOx emissions for the Euro 6 vehicles, as set out in Annex I to Regulation No 715/2007, to be an essential element of that regulation. Therefore, that limit could not be amended by the Commission through an act adopted following the comitology regulatory procedure with scrutiny. The General Court took the view that, in setting out in the contested regulation the NTE values of NOx emissions to be respected during tests under RDE conditions, and by determining CF pollutant compliance factors, the Commission had in practice amended the limit on those emissions for the Euro 6 standard.

23.

The General Court concluded that by adopting the contested regulation, the Commission exceeded the powers conferred on it by Article 5(3) of Regulation No 715/2007 and, as a consequence, infringed Article 4(1) of the same regulation. For reasons of judicial economy, the General Court did not examine the other pleas and arguments put forward by the respondents.

24.

On that basis, the General Court annulled point 2 of Annex II to the contested regulation in view of the fact that the arguments raised by the respondents concerned that part only, which was considered severable from the rest of the instrument. So as not to undermine the legitimate economic interests of both the automotive industry and the consumers who relied on the existing legislation, and to avoid a legal vacuum, the General Court applied Article 264 TFEU. It thus decided to maintain the effects of the provision annulled in relation to the past and also for a future period, not exceeding 12 months, in which to amend the relevant legislation.

25.

Finally, the General Court dismissed the action for damages lodged by the City of Paris on the ground that no damage had been proven.

26.

In its appeal before the Court, lodged on 22 February 2019, the Federal Republic of Germany asks the Court to:

27.

In its appeal before the Court, lodged on 22 February 2019, Hungary asks the Court to:

28.

In its appeal before the Court, lodged on 23 February 2019, the Commission asks the Court to:

29.

For their part, the respondents ask the Court of Justice to dismiss the appeals and order the appellants to pay the costs. In the alternative, they ask the Court of Justice to refer the case back to the General Court so that the latter can examine the grounds of annulment not examined in the first set of proceedings.

30.

In the proceedings in Case C‑177/19 P, Romania, the Slovak Republic, and the European Automobile Manufacturers’ Association (‘the ACEA’) were given leave to intervene in support of the form of order sought by the Federal Republic of Germany. In the proceedings in Case C‑178/19 P, ACEA was given leave to intervene in support of the form of order sought by Hungary. In the proceedings in Case C‑179/19 P, ACEA was granted leave to intervene in support of the form of order sought by the Commission.

31.

The parties and interveners replied to written questions put to them by the Court.

32.

By decision of the President of the Fifth Chamber of 28 January 2021, the three cases were joined for the purposes of the Opinion and the Judgment.

33.

This Opinion will deal with each of the grounds of appeal raised by the three appellants in the following manner.

34.

First, the grounds and arguments concerning the legal standing of the respondents will be examined (A): whether they are directly concerned by the contested regulation (the second part of Hungary’s first ground and Germany’s first and second grounds) (1), and whether the contested regulation entails implementing measures vis-à-vis the respondents (the first part of Hungary’s first ground) (2).

35.

Next, I shall turn to the grounds that concern the substantive assessment made by the General Court (B), by examining jointly the Commission’s single ground and Germany’s third ground, both relating to the Commission’s alleged lack of competence. Finally, I shall conclude by assessing the grounds which criticise the consequences drawn by the General Court from its substantive findings (C): the scope of the annulment (Germany’s fourth ground) (1), and the temporal effects of the annulment (Germany’s fifth ground and Hungary’s second ground) (2).

36.

The second part of Hungary’s first ground of appeal concerns an alleged error of law by the General Court in finding that the respondents were directly concerned by the contested regulation. Hungary maintains that the contested regulation directly concerned vehicle manufacturers and the national authorities charged with the task of checking compliance with the provisions of the contested regulation only. In particular, in Hungary’s view, the General Court has misinterpreted Article 4(3) of Directive 2007/46 in so far as that provision neither concerns nor affects the exercise of power of national authorities in the field of traffic or environmental regulations. That provision has a more limited scope. It aims at ensuring that new vehicles comply with the technical specifications provided therein.

37.

Similar arguments are developed in Germany’s first ground of appeal. Any restriction that the respondents may face with regard to the exercise of their regulatory powers in the field of traffic control follows from other instruments and provisions of EU law, but not from the contested regulation. In particular, the compatibility with EU law of the establishment of areas with limited traffic by local authorities is to be assessed under Directive 2008/50/EC on ambient air quality and cleaner air for Europe. ( 14 ) The scope and effect of the contested regulation are limited to setting standards for motor vehicles type-approval.

38.

Germany’s second ground of appeal also concerns the General Court’s assessment of whether the respondents were directly concerned by the contested regulation. However, by that ground of appeal, Germany criticises the General Court for an allegedly inadequate statement of reasons on that point. In essence, the German Government – supported on this point by the Slovak Government and the ACEA – criticises the General Court for having based its analysis of the possible effects of the contested regulation on the respondents on the provisions of Directive 2007/46 and not on those of that regulation.

39.

According to settled case-law, ‘the condition that a natural or legal person must be directly concerned by the decision against which the action is brought, laid down in the fourth paragraph of Article 263 TFEU, requires two cumulative criteria to be met, namely, first, the contested measure must directly affect the legal situation of the individual and, second, it must leave no discretion to its addressees who are entrusted with the task of implementing it, such implementation being purely automatic and resulting from the EU rules alone without the application of other intermediate rules’. ( 15 )

40.

It is against that background that I shall assess the arguments put forward by the appellants. I take the view that the statement of reasons in the judgment under appeal concerning the direct concern of the respondents for the purposes of Article 263 TFEU is sufficiently clear in order to be fully understandable (1), but indeed is mistaken in its interpretation of the scope of Article 4(3) of Directive 2007/46 (2). That being said, there are other reasons for which the respondents are directly concerned within the meaning of Article 263 TFEU (3).

41.

According to settled case-law, ‘the statement of the reasons on which a judgment is based must clearly and unequivocally disclose the General Court’s thinking, so that the persons concerned can be apprised of the justification for the decision taken and the Court of Justice can exercise its power of review’. ( 16 )

42.

In my view, the reasoning followed by the General Court to arrive at the conclusion that the contested regulation is of direct concern to the respondents is clearly explained. Germany’s contention regarding an alleged inadequate statement of reasons in the judgment under appeal ought therefore to be dismissed.

43.

The General Court devoted a substantial part of the judgment under appeal (paragraphs 41 to 84 thereof) to the assessment of whether the contested regulation had a direct impact on the position of the respondents. It is true that the reasoning followed by the General Court in those passages is, by and large, focused on the meaning of, and the consequences attaching to, Article 4(3) of Directive 2007/46. However, it is also true that the effects allegedly flowing from the contested regulation are a consequence of its interaction with that provision of Directive 2007/46. In essence, the General Court found that, as the respondents have argued, the contested regulation de facto raises the limit of NOx emissions for vehicles to be registered as Euro 6, whereas Article 4(3) of Directive 2007/46 precludes Member States from restricting ‘circulation’ of Euro 6 vehicles on environmental grounds.

44.

The statement of reasons is thus satisfactory: the General Court duly explained why, in its view, the contested regulation altered the legal position of the respondents, and why it did so without requiring the adoption of other acts to that effect.

45.

That said, the real question is whether that reasoning is sound. In other words, the crucial issue is one of the merits of that argument: did the General Court correctly interpret and apply Article 4(3) of Directive 2007/46?

46.

In that regard, I cannot embrace the reasoning followed by the General Court.

47.

Put briefly, as regards the direct concern of the respondents, the General Court based its conclusion on one single element: Article 4(3) of Directive 2007/46. According to the General Court, that provision prevents the authorities of the Member States from banning (or limiting) the circulation of Euro 6 vehicles on environmental grounds.

48.

I cannot agree with that interpretation of Article 4(3) of Directive 2007/46. In my view, that provision only regulates the technical, product standards to be set for type-approval at the moment of the initial placement of the vehicle on the market. It is not meant to, nor is it capable of, preventing the Member States or their local entities, such as the respondents, from introducing measures regulating the subsequent use of vehicles and their traffic in their respective territories, in particular on environmental grounds.

49.

Admittedly, the General Court’s reading of Article 4(3) of Directive 2007/46 appears – at first sight – to be based on the wording of that provision. Indeed, the second subparagraph thereof provides that Member States are not to prohibit, inter alia, the ‘circulation on the road of vehicles’ if they satisfy the requirements of the directive. ( 17 ) Therefore, if a vehicle complies with Euro 6 standards, its circulation cannot be prohibited by the Member States, or by a geographical part of a given Member State, such as a region or a city. If part of a Member State were to ban vehicles that meet the Euro 6 standard from entering certain areas of that State, such as its largest cities or zones within, that could, technically speaking, amount to a restriction on or an impediment ‘to circulation on the road’ within that Member State.

50.

However, when that provision is read in its entirety, in the context of the directive, in the light of the aim pursued by the EU legislature, and with account taken of its origin, its meaning appears to be different, and its scope more limited than that stated by the General Court.

51.

First, that provision’s limited scope is revealed if the reference to the circulation of vehicles in the second subparagraph of Article 4(3) of Directive 2007/46 is not read in isolation, but together with the rest of the provision. According to that entire subparagraph, Member States ‘shall not prohibit, restrict or impede the registration, sale, entry into service or circulation on the road of vehicles, components or separate technical units, on grounds related to aspects of their construction and functioning covered by this Directive, if they satisfy the requirements of the latter’. ( 18 ) The grounds which may bar Member States’ action in that regard are, therefore, only those connected with the construction and functioning of the vehicles (and not primarily environmental or health concerns), and for the purposes of their initial placement on the market – their registration, sale, entry into service or circulation (and not their subsequent use and the regulation thereof).

52.

Second, there is the structural relationship between the two subparagraphs of Article 4(3) of Directive 2007/46. Despite the fact that only the second subparagraph contains a reference to ‘circulation’, there is a clear parallel between the two. Indeed, their wording appears complementary: the first subparagraph lays down a positive obligation (Members States ‘shall’), whereas the second subparagraph lays down a negative obligation (Member States ‘shall not’). ( 19 ) Both obligations are, however, concerned with the same matter, namely ensuring that Member States give effect to the EU-wide procedure for type-approval of vehicles. Accordingly, those two subparagraphs are analogous to a photograph and its negative: picturing the same subject but colour-reversed. It would thus be odd if, within such settings, the scope of the second subparagraph were to suddenly become far broader than the first – which would be the very consequence of the General Court’s interpretation of the provision in question.

53.

Third, the position of the General Court does not appear to be supported by a contextual and purposive analysis of that provision. Indeed, the objective and the material scope of the directive are quite specific. As follows from its recitals 1, 2 and 22, Directive 2007/46 aims at promoting the internal market by establishing a Community approval procedure relating to the type-approval of motor vehicles and their trailers based on the principle of full harmonisation. To that end, the directive establishes – as provided for in Article 1 – a harmonised framework which contains the administrative provisions and general technical requirements for approval of all new vehicles within its scope and of the systems, components and separate technical units intended for those vehicles, with a view to facilitating their registration, sale and entry into service within the (now) European Union.

54.

Although the technical requirements set out by the EU legislature are meant to, inter alia, ensure a high level of environmental protection, ( 20 ) it is rather clear that that instrument is not meant to regulate matters outside those laid down in Article 1 thereof.

55.

The General Court conceded that much, stating that ‘the main object of that directive is the placement on the market of new motor vehicles, that is to say the free movement of certain goods between the Member States, [the directive] does not seek in general terms to provide a framework for legislation governing circulation on the road covering such vehicles and laid down by the public authorities of the Member States, and … it does not fall within the scope of the transport or environmental policy of the European Union’. ( 21 )

56.

Nevertheless, the General Court then adds, in the following paragraph of the judgment, that that ‘does not mean that any provision on the circulation on the road of new motor vehicles would have no place in Directive 2007/46. It is common for a directive, or another directive adopted under that first directive, to contain provisions which do not come under the main objective stated but which are intended to ensure that the provisions adopted have practical effect with a view to achieving that objective’. After illustrating that point with an (perhaps not entirely pertinent) example of directives on public contracts, the General Court states that it ought to give preference to the ‘interpretation which ensures that the provision retains its effectiveness’ and finds that ‘the reference to circulation on the road would have no practical effect if, as the Commission submits, it had the same scope or meaning as “the registration, sale and entry into service” of the vehicles’. ( 22 )

57.

I agree that the directive could, in principle, include provisions aimed at ensuring its effectiveness and, in order to do that, regulate matters not falling within its main object. I also agree that the interpretation of the EU provision in question to be retained should not, as far as possible, render parts of that provision ineffective or redundant.

58.

Nevertheless, the manner in which the General Court applied those principles in the case at hand is difficult to grasp. If the term ‘or circulation’ contained in Article 4(3) of Directive 2007/46 is given the meaning ascribed to it by the General Court, the scope and effect of that directive would ipso facto be significantly broadened, well beyond the subject matter and objective stated in the title, the recitals, and the main provisions thereof. Suddenly, one single word in a single provision transforms Directive 2007/46 into a very different instrument, governing, to some extent, the subsequent use of vehicles and concerned, although indirectly, with the regulation of traffic.

59.

Moreover, were Article 4(3) to be read in such a way, its impact on environmental policy – especially on the capacity of the Member States to combat pollution – would also be quite significant. As the appellants point out, there is a specific EU instrument which concerns that matter: Directive 2008/50 on ambient air quality and cleaner air for Europe. ( 23 ) Article 13 of that directive provides that Member States are to ensure that, throughout their zones and agglomerations, the limit values of various pollutants are not exceeded, inter alia in relation to nitrogen dioxide. Article 23 states that, ‘where, in given zones or agglomerations, the levels of pollutants in ambient air exceed any limit value or target value … Member States shall ensure that air quality plans are established for those zones and agglomerations in order to achieve the related limit value or target value’. The same article states that those plans may themselves contain short-term action plans. In that regard, Article 24(2) of Directive 2008/50 provides that the short-term action plans may, inter alia, ‘include measures in relation to motor-vehicle traffic.’

60.

However, if Article 4(3) of Directive 2007/46 were to be interpreted as introducing limits to the Member States’ ability to regulate the circulation of vehicles, the interaction between the two instruments would be – as an understatement – quite unclear. It would be surprising to learn that the EU legislature, which had prepared the two instruments at the same time, had not introduced, if indeed both of those instruments were meant to relate to the regulation of traffic and the use of vehicles for environmental and public health reasons, some provisions setting out the relationship between the two instruments.

61.

In the light of the above, it seems to me that, despite its statement to the contrary, ( 24 ) the General Court failed to carry out a contextual and teleological analysis of the provision. It did not read that provision in the light of the other provisions, the recitals, or the title of the directive, none of which concerns the regulation of traffic. Nor did the General Court duly take into account the fact that a reference to the circulation of vehicles is not contained in: (i) Article 10(5) of Regulation No 715/2007, which belongs to the same framework; (ii) Article 6(4) of Regulation 2018/858, which repealed and replaced Directive 2007/46; and (iii) any other similar instrument, adopted in the same context and for a similar objective. ( 25 )

62.

Fourth, the General Court did not pay any attention to historical arguments. While in no way compulsory, that method of interpretation could have provided valuable insights, particularly in the light of the Commission’s argument that the second subparagraph of the provision in question (which includes the term ‘or circulation’) was in fact a late addition in the legislative process and was not intended to extend its scope. ( 26 ) The Commission explained that the second subparagraph of Article 4(3) of Directive 2007/46 was meant to constitute an anti-circumvention clause only: once legally type-approved, vehicles cannot be excluded from circulation by means of other measures that constitute disguised restrictions. ( 27 )

63.

What can be gleaned from the travaux préparatoires appears to support the Commission’s arguments regarding the genesis and aim of the provision. The Commission’s Proposal for a Directive ( 28 ) – which indeed did not include a second subparagraph of Article 4(3) ( 29 ) – confirms that the proposed instrument was meant mainly to consolidate and update the acquis in this field, whilst extending it to other types of vehicles. ( 30 ) The new instrument did not intend, therefore, to go beyond the subject matter covered by the previous instruments. That element appears important in so far as Directive 70/156/EEC, of which Directive 2007/46 is a recast, concerned the technical characteristics of a type of vehicle only and did not contain – as the Court has had the opportunity to clarify – any ‘other provision relating to the road traffic rules which are to be complied with by the drivers of motor vehicles’. ( 31 )

64.

Subsequent documents from the same institution ( 32 ) or from the other EU institutions ( 33 ) also confirm that the provision was amended at a rather late stage of the procedure and that the amendment pursued a limited objective of clarification. As far as I can tell, there is no trace, in the preparatory documents, of any discussion regarding a possible extension of the material scope of the directive. ( 34 )

65.

If, as argued, the second subparagraph is meant to complement and clarify the first subparagraph, I also do not see why the alternative interpretation of that provision, put forward by the Commission, Germany and Hungary, would render the provision ineffective, as the General Court stated. ( 35 )

66.

Fifth, nor do I find the other arguments upon which the General Court based its interpretation of the provision in question, in the judgment under appeal, more convincing. The General Court stated that ‘it is inherent in a situation resulting from complete harmonisation, such as that resulting from that directive, that [the Member States’ authorities] cannot, save in particular circumstances, oppose the ordinarily intended use of a product which satisfies the requirements laid down in the harmonising arrangements, since they would otherwise undermine the practical effect of those arrangements’. ( 36 )

67.

The statement is, in the abstract, certainly correct. However, once again, I am not able to embrace the consequences that are supposed to follow, in the General Court’s view, from that statement for the case at hand. Does it mean that the Member States cannot, in principle, adopt rules that limit the use of a Euro 6 vehicle (as a vehicle meant to carry passengers and/or goods) on environmental grounds? That cannot possibly be correct, as the General Court itself recognises when it stated that Member States can still organise car-free days. ( 37 ) However, how then can one and the same harmonising measure allow and prohibit the same type of conduct? There seems to be an inherent contradiction in the reasoning of the General Court.

68.

There appears to be, at the root of that contradiction, a confusion as to the regulatory logic and effects of the EU rules on positive integration (harmonisation) and those on negative integration (free movement rules). Indeed, in the Court’s case-law, a similar issue did arise in the past and was examined from the angle of Article 34 TFEU. The Court has – understandably – rejected an over-restrictive approach in that regard. In Mickelsson and Roos (with reference to jet skis) the Court found that only national laws ‘preventing users of [those vehicles] from using them for the specific and inherent purposes for which they were intended or of greatly restricting their use’ constituted a measure having equivalent effect to quantitative restrictions on imports, and thus needed to be duly justified. ( 38 ) Similarly, in Commission v Italy (motorcycle trailers), the Court found an Italian law that precluded motorcycles from towing trailers to constitute a measure having equivalent effect in so far as it impeded access of those products to the Italian market. ( 39 )

69.

In both cases the national measures were assessed under the rules on free movement, and found to constitute measures having equivalent effect only if they made access to the national market of the product in question impossible or more difficult, because the product could not be used, or its use was greatly reduced. Admittedly, no instrument of harmonisation came into play in those cases. However, in the present cases, the situation is no different in that regard: the instrument(s) of harmonisation invoked by the respondents harmonise only the rules on the type-approval of vehicles for the purpose of their initial placement on the market. They do not govern the subsequent use of vehicles (let alone the local circulation of vehicles) and its regulation for environmental or public health reasons.

70.

Against that background, it is rather clear to me that a provision such as Article 4(3) of Directive 2007/46 does not preclude Member States from deciding that, for some reason in the public interest (relating to, inter alia, public security, public health, road safety or environmental protection), all vehicles or only some types of vehicles (that is, less polluting or smaller vehicles, or those belonging to residents or the public authorities, and so on) can circulate in certain areas (a city centre, national parks, off road, military areas, construction sites, and so on) in general, or at some given moment (on Sunday, during a car-free day, when a market or some cultural or sports event is taking place, and so on). In particular, nothing in that provision appears to limit the respondents’ legal competence to establish low emissions zones (‘LEZs’), a phenomenon which the EU institutions have consistently regarded favourably. ( 40 ) Likewise, Member States can limit – for the same reasons – certain ways in which vehicles are used (speeding, unauthorised parking, driving with children but without using proper equipment, and so on). ( 41 )

71.

Therefore, a local regulation that introduces limits to the circulation of vehicles in some specific areas, even where potentially using parameters on emissions that are stricter than those used for the Euro 6 standard, is not in breach of the EU type-approval legislation. If anything, its compatibility with EU law could be assessed under Article 34 TFEU, on the basis of the test set out in Mickelsson and Roos or in Commission v Italy. However, and yet again, although a certain national measure concerning subsequent use might ultimately be caught as a measure having equivalent effect under Article 34 TFEU, ( 42 ) it certainly does not mean that EU law has positively harmonised that very same issue or area.

72.

In summary, I take the view that the General Court erred in concluding that the respondents were directly affected by the contested regulation through its interaction with Article 4(3) of Directive 2007/46.

73.

However, the judgment under appeal need not be set aside since, in my view, there are other reasons for which the respondents are directly concerned by the contested regulation for the purposes of Article 263 TFEU.

74.

For an EU act to be of direct concern to an applicant, the legal position of that applicant must be affected. That means, in essence, that the EU act in question must alter the applicants’ possession of rights and obligations, of private or public nature. ( 43 )

75.

As far as local entities are concerned, case-law indicates that those entities are directly concerned by an EU act when they are entrusted with powers that are exercised autonomously within the limits of the national constitutional system of the Member State concerned, and the EU act prevents those entities from exercising those powers as they see fit. ( 44 )

76.

Against this background, does the contested regulation prevent the respondents from exercising their (constitutional) powers in the fields of environmental protection and in the regulation of traffic as they see fit?

77.

I admit that, in the present case, the answer to that question is not straightforward.

78.

There is no doubt that the contested regulation, in so far as it permits the marketing of vehicles that allegedly do not comply with certain Euro 6 standards (which could be referred to as ‘false Euro 6’ vehicles), can render the respondent’s legal obligation to combat pollution more difficult. As the respondents have argued and demonstrated, local traffic regulations would typically distinguish between vehicles on the basis of their ‘Euro’ category. ( 45 )

79.

However, as explained in the preceding section of this Opinion, nothing in the contested regulation (or in the legislation which that regulation implements) can be read as legally precluding the respondents from differentiating between less polluting and more polluting vehicles, regardless of the Euro category to which they belong, when regulating local traffic. Certainly, to the extent that the national measures would then need to rely on parameters other than the Euro category, such measures could be more difficult to implement, monitor and enforce. Nevertheless, that is immaterial under the strict understanding of Article 263 TFEU. That amounts to an impact on the respondents’ factual situation, not on their legal position, which is insufficient to make them directly concerned within the meaning of Article 263 TFEU. ( 46 )

80.

That said, I do not think that the analysis on direct concern can really stop there. The legal test devised in the case-law with regard to the direct impact upon local or regional entities, such as the respondents, is whether the challenged EU act prevents those entities from exercising their autonomous powers as they see fit. ( 47 )

81.

It seems to me that there are different ways of approaching that wording. On the one hand, a particularly stringent interpretation could be envisaged, according to which ‘direct concern’ exists only in so far as the challenged EU act makes it impossible for them to exercise their powers. In that case, local entities could only challenge EU acts that require them to either entirely abstain from using their powers, or oblige them to use those powers in a very specific manner, thus leaving absolutely no options whatsoever. On the other hand, a generous interpretation, according to which a mere impact of the entities’ freedom to exercise their autonomous powers would be sufficient to trigger standing, is also possible. In that scenario, local entities could challenge an EU act which has some impact on the room for manoeuvre they enjoy under national law.

82.

I think that the most sensible reading of Article 263 TFEU is in the middle ground between those two extremes.

83.

Admittedly, a rather generous interpretation seems to follow from the very wording of the test devised by the EU Courts: preventing those entities from exercising their powers as they see fit. The test does not refer to absolute impossibility or deprivation of any discretion. On its text, any limitation as to the exercise of their legal powers appears to be caught.

84.

However, such a ‘loose’ interpretation of the test would indeed be difficult to reconcile with the Court’s general case-law under Article 263 TFEU. That case-law requires, inter alia, that a direct causal link between the contested EU act and the alteration in the legal situation of the applicant can be established. ( 48 ) In addition, the EU Courts have also made clear that the mere fact that a local entity has some competence – as a body competent for economic, social or environmental matters in its territory – with regard to the matter regulated by an EU measure of general application, cannot, of itself, be sufficient for that entity to be regarded as ‘concerned’ within the meaning of the fourth paragraph of Article 263 TFEU. ( 49 )

85.

Thus, I do not think that the impossibility to exercise their powers ‘as they see fit’ can in practice be reduced to it not being possible to exercise those powers at all, without entirely discarding existing case-law. Moreover, in such a case, the case-law on standing under Article 263 TFEU would sit (yet again) very uneasily with many other areas of EU law, which continue to move in exactly the opposite direction. ( 50 )

86.

Accordingly, I take the view that the Court’s test should be regarded as granting standing to local or regional entities when a direct cause and effect relationship between the contested EU act and the exercise of a specific autonomous legal power of a local entity can be identified. That relationship exists when the EU act is the determining factor which either precludes the local or regional authorities from using their powers at all, or compels them to act, while significantly altering the manner in which they can lawfully make use of those powers.

87.

If that is so, do the respondents meet that test?

88.

Although it is admittedly a close call, in the light of all the legal and factual elements in the case file, I would, on balance, be inclined to give a positive answer to that question. The contested regulation indeed has an impact on the way in which the respondents are able lawfully to exercise their specific powers in the area of environmental protection and the protection of public health when tackling pollution and ensuring the appropriate standards of air quality, as they are legally obliged to do.

89.

The fact that the respondents enjoy, at the constitutional level, specific powers in the fields of environmental protection and the regulation of traffic has already been established by the General Court and does not appear to be contested by the appellants. In that regard, I would add that, within their legal systems, the respondents have even corresponding legal obligations in that regard. Those are important elements in so far as they distinguish the respondents from other national authorities in the Union which may be affected by the contested regulation only indirectly, to the extent that those other national authorities are charged with the task of looking after the general well-being of their citizens. A fortiori, those elements differentiate the respondents from any physical persons living in the Union that – for the mere fact of breathing polluted air – may feel affected by the contested regulation.

90.

The key issue thus becomes whether the exercise of those powers by the respondents, in order to meet their legal obligations, is significantly affected by the contested regulation.

91.

First, the parties agree that, under national law, local entities such as the respondents are responsible for the adoption of a variety of measures to combat pollution and ensure certain standards of air quality in order to preserve the health and safety of citizens. Those measures may be necessary – something which is crucial in this context – in order to comply with obligations flowing from EU law, such as those stemming from the provisions of Directive 2008/50. As mentioned in point 59 above, that directive requires Member States to comply with European air quality standards (‘AQS’) and, to that end, adopt air quality plans (including short-term action plans) (‘AQPs’) for specific zones and agglomerations where pollutants exceed the limits.

92.

Second, it is also uncontested that large European cities, such as the respondents, are plagued by pollution deriving, in large part, from vehicle emissions. That is especially true with regard to NOx emissions in large cities, produced mainly by local road traffic, and as such are often well beyond the relevant legal limits. ( 51 )

93.

Third, tests carried out after the ‘dieselgate’ scandal revealed that almost all Euro 5 and Euro 6 light-duty diesel vehicles significantly exceeded applicable NOx limits. ( 52 ) Therefore, the introduction of those standards has not substantially improved the situation with regard to NOx emissions so far. ( 53 )

94.

Fourth, it is also hardly disputable that, if the EU institutions do not strive to ensure – as the respondents argue – compliance with the limits set out in ‘their own’ legislation regarding polluting emissions from vehicles, the burden to adopt measures to ensure air quality logically shifts onto the authorities of the Member States. That is especially true with regard to national authorities responsible for the regulation of traffic, particularly those acting in the most polluted areas. Those authorities will obviously be required to act sooner rather than later, considering the normal life-cycle of light-duty vehicles. The Commission is also – and one cannot fail to note a certain irony in that – calling on national authorities to act swiftly and vigorously, having recently made it a priority to start infringement proceedings against Member States that do not comply with the AQS. ( 54 )

95.

To that end, national authorities such as the respondents are therefore required to adopt certain measures, including within the framework of the AQPs, that, had the Union prohibited the registration and sale of vehicles failing to meet its own emission standards, would most likely have been less restrictive or perhaps even unnecessary. Thus, the choice as to the measures to be adopted by the respondents, and the manner in which those measures will have to be implemented, is bound to be reduced as a direct consequence of the contested regulation. The room for manoeuvre in meeting their legal obligations under both EU and national law is considerably narrowed, thereby obliging them to perform the tasks required to that end differently.

96.

Certainly, as has been suggested above, ( 55 ) the cities or other regional entities still technically retain their powers under national law to regulate local traffic for environmental or public health reasons, irrespective of whether a ‘true’ or ‘false’ Euro 6 standard is fulfilled, in the manner they see fit. Should such authorities wish to do so, they can adopt bans on Euro 6 vehicles, or even introduce their own Euro 7 or Euro 8 standards. They are still capable of deciding that only certain cars may enter the city-centre, or that car traffic is only allowed at certain times or on certain days, or is in fact banned altogether.

97.

The fact remains, however, that the exercise of their legal powers and obligations has been made infinitely more difficult. Reasonable regulation of local traffic on environmental and public health grounds must rely on some broader, standardised criteria. No one is likely to seriously suggest that each major city or region in Europe should now start establishing its own list of vehicles which will be admitted to its city centres, with emissions measuring stations at the city gates. However, the inability to rely on (for this very purpose) established EU categories and standards means that the cities have to choose their own criteria, which are bound to generate further contention in terms of social or financial costs for local residents or certain groups of residents, eventually exposing the cities to possible legal challenges based on discrimination and/or arbitrary limitation of access or free movement. ( 56 )

98.

The unenviable task of dealing with all these (admittedly difficult) decisions – and the potential liability associated therewith – is simply passed on to the (Member States and their) local authorities. Those authorities may thus end up being forced to adopt draconian and unpopular measures in order to ensure compliance with EU law, for the very reason that the Commission opted to tolerate other breaches of EU law.

99.

To suggest, in such a situation, that the position of the local entities is affected only in fact and not in law by the contested regulation is extremely formalistic, ( 57 ) if not outright cynical. The reality is that an alleged failure of the Commission to ensure compliance with the NOx emissions limits is bound to have a direct and significant impact on the manner in which entities such as the respondents have to exercise their autonomous legal powers in the fields of environmental protection and the regulation of traffic. ( 58 )

100.

Within the latter dimension, the impact upon the position of the respondents is legal, not merely factual. I readily acknowledge that drawing a line between what is legal and what is factual in scenarios such as the present one is far from straightforward. Of course, just like anybody else, the respondents are also factually concerned: they must accept the EU legislation as a fact and the starting point for further action. However, in addition, there is also an element of legal concern. The contested regulation has a direct impact on the way in which the respondents will exercise their legal rights and obligations.

101.

A simple mental experiment might help to illustrate that point. Assuming that the respondents were to be qualified as being concerned merely in fact, not in law, who then could ever be legally concerned by such legislation? The Hungarian Government has suggested that only vehicle manufacturers and the national authorities tasked with assessing compliance with the provisions of the contested regulation could be legally concerned. ( 59 ) However, how much more legally concerned is a vehicle manufacturer by legislation with which it must comply if it wishes to market its products that it is yet to produce? ( 60 ) A producer has no legal right whatsoever to have certain legislation adopted, containing certain standards or values. As a result, its legal position (for the purposes of existing rights and obligations) cannot by definition be affected by any new legislation decreeing future rules since it has no (prospective) legal right in that regard in the first place. The same would be true, a fortiori, of national authorities charged with the task of checking compliance with the emission values provided for in the EU legislation. What would be the vested legal rights or obligations (independent of the Member State of which it is likely to be a simple organ) of a central national institution tasked with the type-approval of vehicles? For both these types of actors, new values with which they will have to comply in the future (one by producing vehicles to that specification if it wishes for them to be approved, the other by measuring that compliance with those specification) are simply the factual points of departure, but do not really have any direct and immediate impact on their legal position.

102.

I am not claiming that I subscribe to such a reductionist view. The argument made above is simply to illustrate that, if an expansionist vision of a merely factual concern to the exclusion of any legal concern were to be embraced, no one would ever be legally concerned by any EU law measure.

103.

Finally, there are three broader, systemic reasons for embracing a balanced interpretation of the fourth paragraph of Article 263 TFEU when the applicants are the regional entities of Member States. As they largely overlap with the arguments already expressed in my Opinion in Région de Bruxelles-Capitale on this point, ( 61 ) I can be concise.

104.

First, an overly restrictive application of Article 263 TFEU vis-à-vis local entities would, arguably, hardly be consistent with some basic constitutional principles, in particular those enshrined in Article 4(2) and (3) TEU.

105.

On the one hand, an approach that considers – for the purposes of judicial control at EU level – the constitutional roles played, at national level, by regional and local entities to be irrelevant, would sit uneasily with Article 4(2) TEU which requires the Union to respect, inter alia, the Member States’ national identities, ‘inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-government’.

106.

On the other hand, a more open access to the EU judicial forum by infra-State entities appears to be the logical (and almost inevitable) counterpart of the numerous and far-reaching obligations that EU law imposes on those entities. These are not only the overall obligation to adopt all measures that are necessary to implement EU law and to ensure its application, and to abstain from adopting any measure that could render EU law ineffective, but also the very concrete and specific obligations in the individual policy areas, as also illustrated in the present case. The principle of sincere cooperation, enshrined in Article 4(3) TEU, must obviously cut both ways. ( 62 )

107.

Second, denying entities such as the respondents the ability to challenge EU acts which significantly affect the exercise of certain powers that they enjoy at the constitutional level could lead to what I cannot but call a ‘constitutionally unhealthy situation’ in terms of incentives and control. Indeed, it is not infrequent that EU delegated or implementing acts are the result of some type of (legally foreseen) agreement reached between an EU institution and the relevant industry or stakeholders. In those cases, a restrictive approach to the concept of ‘direct concern’ may mean that the only category of applicants that will (perhaps) have standing before the EU Courts are the very members of that industry. However, that is unlikely to provide the best incentive for the negotiations with the respective stakeholders and a balanced legislative outcome, if the effective absence of any challenge to that outcome is conditional upon producing legislation that is to the liking of the relevant industry.

108.

Third, at an institutional level, I remain perplexed by a judicial structure, and an ongoing judicial policy which petrifies that structure, that, by restricting direct access to the EU Courts, channels cases in which the validity of EU acts is at stake before the national courts, in order for those cases to eventually arrive many years later before the Court of Justice via the preliminary ruling procedure. ( 63 ) This complex and time-consuming combination of procedures is simply unreasonable in view of the latest reform to the judicial structure of the Union which gave the General Court the means required to carry out its constitutional role as the EU first-instance adjudicator which, within the ambit of its jurisdiction, can exercise full judicial review by examining both the relevant law and facts. ( 64 )

109.

All of the foregoing applies, a fortiori, in cases relating to complex, technical fields, for the adjudication of which the distribution of tasks within the preliminary ruling procedure might be sub-optimal in practice. On the one hand, a national court, sometimes even a first-instance court seised with a somewhat different matter, is unlikely to carry out a preliminary detailed, factual and technical assessment of the issues relevant for the interpretation of EU law to be carried out by this Court. ( 65 ) On the other hand, it is not for the Court of Justice when giving preliminary rulings to assess facts. In this manner, there is a risk that rather complex technical issues that are bound to influence the outcome of the interpretation or validity of EU law will not be examined in either set of proceedings. ( 66 )

110.

In the light of the foregoing, and despite the error of law incurred in that respect by the General Court, I suggest that the Court of Justice dismiss the second part of Hungary’s first ground of appeal, as well as Germany’s first and second grounds of appeal. The respondents are directly concerned.

111.

The first part of Hungary’s first ground of appeal concerns the General Court’s finding, made in paragraph 40 of the judgment under appeal, that the contested regulation did not entail implementing measures vis-à-vis the respondents. In essence, Hungary takes the view that the contested regulation, in and of itself, produces no legal effect and a variety of acts must be adopted at national level before any legal effect may be deployed.

112.

This plea should, in my view, be dismissed.

113.

It follows from case-law that the question of whether a regulatory act entails implementing measures should be assessed by reference to the position of the person bringing the action and the subject matter of that action. ( 67 ) Therefore, it is immaterial whether the act in question entails implementing measures with regard to other persons, ( 68 ) and whether other parts of the challenged act, which are not contested by the applicant, entail implementing measures. ( 69 ) What is crucial, in that context, is whether the specific legal effects which alter the position of the applicant materialise vis-à-vis that person as a result of the EU act challenged, or of any other act adopted by the Union or the Member State in question. ( 70 )

114.

Against that background, and in view of the position of the respondents and of the subject matter of the present proceedings, it seems to me that the General Court did not err in finding that, with regard to the specific situation of the respondents, the contested regulation did not entail implementing measures for the purposes of the third limb of the fourth paragraph of Article 263 TFEU.

115.

If the contested regulation is said to allow ‘false Euro 6’ vehicles to receive type-approval, making it possible for those vehicles to be marketed and thus for them to circulate, the effect on the respondents’ legal ability to exercise their powers as they see fit is direct and immediate. No other EU or national act is required to (metaphorically speaking) tie the hands of the respondents. There is, in other words, a cause and effect relationship between the EU rules contested by the respondents and the alteration of their legal status.

116.

It would be unreasonable and artificial to claim that infra-State entities should wait for the first type-approval of any ‘false Euro 6’ vehicle in order to challenge that approval before a national court and, in the context of those proceedings, plead the invalidity of the EU legislation being implemented. No such implementing acts would be – manifestly – acts adopted ‘vis-à-vis’ the respondents. With regard to the legal effects contested by the respondents (the implied or surreptitious amendment of Euro 6 standards), the EU act challenged thus clearly is, borrowing the words of Advocate General Cruz Villalón, ‘fully and autonomously operational’. ( 71 )

117.

Accordingly, the first part of Hungary’s first ground of appeal should be rejected.

118.

By its third ground of appeal, Germany claims that the General Court misinterpreted Article 5(3) of Regulation No 715/2007 in order to come to the conclusion that the Commission did not have competence to adopt the contested regulation. Germany maintains that the Commission enjoys a wide margin of discretion both in relation to the qualification of a provision as ‘non-essential’, and in relation to the practical effects of a provision. When that margin of discretion is duly taken into account, the introduction of RDE tests as such, of a compliance factor for those tests, and of transitional rules, should be considered as non-essential elements within the meaning of Article 5(3) of that regulation.

119.

The Commission’s single ground of appeal also concerns the General Court’s findings on its alleged lack of competence to adopt the contested regulation. The Commission does not contest that, as the General Court stated in paragraph 118 of the judgment under appeal, the NOx emission limits are an essential element of Regulation No 715/2007. However, the Commission considers that that element has not been amended. It is revealing, in that regard, that the General Court refers to a ‘de facto’ amendment of the limits (in paragraphs 128, 137 and 144 of the judgment under appeal), while acknowledging that legally those limits have not been modified (in paragraph 150 of that judgment).

120.

The Commission points out that Regulation No 715/2007 did not include specific provisions on how the emissions have to be checked. In the Commission’s view, the General Court thus confuses the method to verify the limit with the amendment of that limit. In so far as the old laboratory tests, the NEDC, and the new RDE tests apply cumulatively, a vehicle having thus to satisfy both to be type-approved, the contested regulation cannot lead to any deterioration of the verification methods and, by way of a consequence, to any de facto amendment of that limit.

121.

The Commission also claims that the General Court reversed the burden of proof by requiring the Commission to justify the methods chosen, and by annulling the challenged act because it found the Commission’s arguments to be unconvincing. It is for the applicants to prove the unlawfulness of an EU act, which is, in principle, presumed to be valid.

122.

In this context, the ACEA emphasises that the coefficients introduced by the Commission are necessary so as to allow for different sets of data to be compared in the light of the technical and statistical uncertainties relating to the new tests.

123.

First, I am not convinced by the arguments relating to the Commission’s alleged room for manoeuvre to determine which elements of a piece of legislation are essential and which are not. It is, in the first place, for the EU legislature (in casu, the Council and the Parliament) to determine the aspects on which they wish to assign delegated or implementing powers to the Commission and those on which they do not. ( 72 ) In the second place, ‘essential’ and ‘non-essential’ elements are legal categories and it is thus for the Court to ultimately interpret them and verify their correct application in a specific case. ( 73 )

124.

Second, in similar vein, I do not consider the burden of proof to have been reversed in any way. The General Court simply applied, in a legally and factually complex case, the normal principles on the burden on proof in direct actions. To the extent that the respondents had made a prima facie case, it was incumbent on the Commission to contradict and to rebut, in substance and in detail, the information produced by the respondents and the inferences to be drawn from that information. ( 74 ) However, having found the Commission’s counter-arguments to be unconvincing, the General Court came to the conclusion that the Commission could not, without de facto altering the NOx emissions limit set out in Annex I of Regulation No 715/2007: (i) establish NTE values for NOx emissions that, due to the CF pollutants, are higher than that limit and, in any event, (ii) adopt factors of a magnitude such as that of the CF pollutants.

125.

Third, having made those observations, the key point raised by these grounds of appeal is whether the General Court erred in law by holding that, through the contested regulation, the Commission de facto amended an essential element of the main legislation.

126.

In the light of the arguments put forward by the appellants, I fail to see any such error.

127.

First, the NOx emission limit set out in Regulation No 715/2007 constitutes, in all evidence, an essential element of that piece of legislation. ( 75 ) Similarly to the General Court, in view of the wording, context, and, above all, purpose of that EU measure, I fail to see how one could come to a different conclusion on that point. Consequently, the Commission would not have been entitled to amend it via the adoption of implementing legislation such as the contested regulation. The parties appear to agree on this point.

128.

Second, it can indeed be said that, in principle, elements such as the NTE values or the CF pollutant are – especially when each is taken in isolation – non-essential elements of the regulation. Those are technical elements of a merely functional and instrumental nature. It follows rather clearly from recital 26, Article 5(3) and Article 14(3) of Regulation No 715/2007 that the procedures and tests for type approval constitute non-essential elements of that regulation, which can thus be adopted by the Commission following the regulatory procedure with scrutiny.

129.

Third, in view of the above, it does not, however, follow that, by fixing certain parameters for the RDE tests (such as the NTE values and/or the CF pollutant), the Commission could never affect what is, in itself, an essential element (the NOx emission limit). It stands to reason that, if the verification process chosen and the specific values adopted therein are too lenient and ineffective, the limit established by the EU legislature is either made impossible to check or de facto surpassed and, as a consequence, it will often not be complied with in practice. I thus find no fault, at least in theory, with the reasoning of the General Court according to which the adoption of excessively high parameters could, de facto, lead to an indirect or surreptitious amendment of the NOx emission limit.

130.

Whether or not the specific parameters adopted by the Commission, in the light of the specific circumstances of this case, led to a de facto amendment of the NOx emission limit seems to me to be a complex assessment of fact. As such, that assessment is not subject to review by the Court on appeal, unless the appellants claim (and prove) a distortion of the facts or the evidence referred to by the first-instance court. ( 76 ) According to settled case-law, the distortion must be obvious from the documents in the case file without there being any need to carry out a new assessment of the facts and the evidence. ( 77 )

131.

In the present case, I do not see any such distortion.

132.

None of the appellants or interveners have put forward any convincing argument to demonstrate that the General Court has fundamentally misunderstood the facts or misread the evidence submitted by the parties. On the contrary, the General Court came to a conclusion that is not only possible, but also appears to find support in various official documents that are relevant and which were referred to by the parties. For instance, the European Parliament’s ‘Report on the inquiry into emission measurements in the automotive sector’ concluded that the conformity factor ‘in practice, weakens the emissions standards currently in force’, is ‘not justifiable from a technical perspective and [does] not reflect an obvious need to develop new technology’, and could be considered ‘a de facto blanket derogation from the applicable emissions limits for a considerable amount of time’. ( 78 ) Similarly, a recent briefing paper from the European Court of Auditors reports the view of some researchers according to whom the NOx limits set for RDE tests appear to weaken the effectiveness of those tests, and notes the much lower limits that are in force in the United States. ( 79 )

133.

That being said, in my view, an argument which ought to be considered – since it constitutes a criticism of the legal reasoning followed by the General Court, and not of its assessment of the relevant elements of fact – is that relating to the fact that the General Court allegedly overlooked the cumulative nature of the two tests. Since both the old tests (laboratory testing – NEDC) and the new tests (real driving conditions – RDE) must be satisfied for a vehicle to be type-approved, the appellants argue that the contested regulation does not alter the prior situation. Accordingly, there could not be any de facto amendment of the main legislation.

134.

That begs the question: what is the relevant ‘prior situation’ here? The appellants seem to be correct in arguing that the contested regulation has not, and could not, lead to an alteration in peius of the prior situation, if that prior situation is understood as referring to the actual state of affairs back in 2016. If my understanding of the testing system is correct, that regulation should logically have indeed improved the verification of compliance with the Euro 6 standard.

135.

However, the fact that the situation, from a factual point of view, has not changed (or possibly only improved), cannot call into question the conclusion reached by the General Court with regard to the unlawful alteration of the legal situation, that logically ought to be the point of departure for that assessment.

136.

It is true that, action by the Commission in this matter was most probably necessary. Pursuant to recital 15 and Article 14(3) of Regulation No 715/2007, the Commission is obliged to keep under review the procedures, tests and requirements referred to in Article 5(3) of the same regulation, as well as the test cycles used to measure emissions. Those provisions state that, if the review finds that those are no longer adequate or no longer reflect real world emissions, they must be adapted so as to adequately reflect the emissions generated by real driving on the road.

137.

However, that does not detract from the fact that, as the General Court found, the Commission has enacted legislation that, for a given period, still tolerated deviations from the NOx emission limit set out in the main legislation.

138.

In other words, the point to which, following the discovery of a breach, the situation is normally supposed to revert in law is the renewed upholding of the legal requirement that existed all along. The argument that the Commission’s contested regulation did not ‘make things worse’ might well be correct with regard to the factual situation that existed at that time and came to light after the ‘dieselgate’ scandal. Indeed, if a number of producers and vehicle types on the market at that time were in fact non-compliant, some of them even to a remarkable degree, the legislative effort seeking to gradually ensure at least some compliance under the actual conditions prevailing at that moment cannot but be seen to be acting in the best interest of the system as such and its future function.

139.

However, the fact remains that, as part of that process, as established by the General Court, the Commission allowed for further deviations from the point to which legally the situation ought to have immediately reverted: to the full compliance with the already existing standard as previously enacted by the EU legislature which ought to have been upheld and enforced all along. That is indeed the relevant ‘prior situation’ from the legal point of view.

140.

Therefore, I see nothing wrong per se with the conclusion of the General Court that, with the adoption of the contested regulation, the Commission did alter the legal situation by expressly tolerating emissions above the limits decided by the EU legislature, with those specific limits being an essential element that the Commission was not entitled to effectively alter by mere implementing legislation.

141.

Accordingly, it seems to me that, on this point too, the judgment under appeal should be upheld.

142.

By its fourth ground of appeal, Germany claims that the General Court erred in finding that point 2 of Annex II to the contested regulation was severable from the rest of the instrument. Germany argues that, according to the reasoning followed in the judgment under appeal, the General Court should have also annulled Article 1, point 2, of the contested regulation which amends Article 3(10) of Regulation No 692/2008 in so far as it impliedly determines – as the judgment under appeal states in paragraph 156 – the point from which the RDE tests are no longer conducted simply for monitoring purposes.

143.

The City of Brussels raises a plea of inadmissibility on this ground of appeal, arguing that Germany has no interest in a full annulment of the contested regulation.

144.

Indeed, it is not immediately obvious what purpose further annulment would serve if a key component of the structure has already been removed. However, in general, the German Government could perhaps take the view that, rather than having a lopsided system (old system minus the CF pollutant, or with a lower CF decided by the Commission), it would be more advantageous or reasonable to have an overhauled system. If that is indeed so, it may have an interest in the annulment of the regulation in its entirety. For that reason, the plea of inadmissibility should be dismissed.

145.

However, and in any event, as regard the merits of the argument, it does not seem to me to be well founded.

146.

According to settled case-law of the Court, partial annulment of an EU act is possible when the elements the annulment of which is sought may be severed from the remainder of the act. The Court has also held that that requirement of severability is not satisfied where the partial annulment of an act would have the effect of altering its substance. ( 80 )

147.

In the present case, the problematic element identified by the General Court in the contested regulation (the CF) is very specific. That coefficient will continue to apply until the EU legislature repeals it with one which better reflects reality, or decides to simply do away with it.

148.

The German Government failed to explain why the repeal or replacement of that coefficient would necessarily affect the correct functioning of the legal framework which the contested regulation amends or implements. Indeed, the General Court has found no fundamental error in the design of the tests, but rather only in the NTE values established to that end (because of the CF).

149.

Nor has that government explained what, precisely, would have changed had the General Court also annulled Article 1 point 2 of the contested regulation.

150.

Accordingly, I am of the view that this ground of appeal should also be rejected.

151.

By its second ground of appeal, Hungary claims that the General Court erred in limiting the temporal effects of its annulment to a reasonable period not exceeding 12 months. In its view, that maximum period is too short because in practice it is impossible for the EU legislature to enact a new legislation within that time frame.

152.

Very similar arguments are put forward in Germany’s fifth ground of appeal. The German Government refers to the estimates of the German automobile industry with regard to how many cars from German and other manufacturers throughout Europe could no longer be sold, further suggesting that the potential legal void allegedly created by the judgment of the General Court could possibly result in a complete halt in the production, approval, and registration of new cars due to the ensuing legal uncertainty.

153.

I am unconvinced by these grounds of appeal.

154.

In simple terms, the question that the appellants ask the Court to address is the following: is the 12-month period long enough for the EU institutions to enact new measures which, by remedying the error identified, ensure compliance with the standards set out in the main legislation?

155.

It seems to me that, in the light of the arguments raised by the appellants, to answer that question, the Court of Justice would have to carry out an ex novo factual examination. Indeed, the appellants do not invoke an error in the methods and criteria used by the General Court to carry out that assessment (such as, the wrong standards applied, certain pertinent elements left out of the equation, and so on). The appellants simply contend that the General Court was mistaken in its overall evaluation. In that regard and for that reason, I doubt that these grounds of appeal are even admissible.

156.

In any event, the appellants do not point to any specific factual or legal element that supports the view that a 12-month period is insufficient for the Commission to intervene, on the basis of its implementing powers, to rectify the problems identified in the contested regulation.

157.

Consequently, in my view, the present grounds of appeal are not admissible, and in any event, are unfounded.

158.

Pursuant to Article 138(1) of the Rules of Procedure of the Court of Justice, applicable to appeal proceedings pursuant to Article 184(1) thereof, the unsuccessful party shall be ordered to pay the costs if they have been applied for in the successful party’s pleadings.

159.

The respondents having applied for the costs, those costs should thus be borne by the appellants.

160.

Pursuant to Article 140(1) and (3) of the Rules of Procedure of the Court of Justice, all interveners shall bear their own costs.

161.

I suggest that the Court of Justice should: