1

The applicants, Tempus Energy Ltd, formerly known as Alectrona Grid Services Ltd, and Tempus Energy Technology Ltd (together, ‘Tempus’) sell electricity consumption management technology, also known as ‘demand-side response’ (‘DSR’), to individuals and professionals and hold a licence to operate as an electricity supply business in the United Kingdom.

2

The service that Tempus provides to its clients is intended to create cost efficiencies for its clients in the electricity supply chain by combining DSR technology with the services provided by an electricity supplier. Tempus sells electricity and helps its customers to move their non-time critical electricity usage to periods when wholesale prices are low, either because demand is low or because power from renewable sources is plentiful and therefore cheaper.

3

It is apparent from the documents before the Court that it is customary for demand-side response operators (‘DSR operators’) to enter into contracts with energy customers, which are often industrial and commercial customers or small- to medium-sized enterprises, under which the customer agrees to be flexible in the consumption of their electricity at a particular time period. The DSR operator calculates the total capacity available from all of the flexible customers at any one time and can then offer that capacity to the electricity network operator — National Grid (‘NG’) in the present case — in exchange for a payment, which it passes back to the flexible customer, whilst retaining a small profit margin for itself.

4

By its action, Tempus seeks the annulment of Commission Decision C(2014) 5083 final of 23 July 2014 not to raise objections to the aid scheme for the capacity market in the United Kingdom, on the ground that that scheme is compatible with the internal market pursuant to Article 107(3)(c) TFEU (State aid 2014/N-2) (OJ 2014 C 348, p. 5) (‘the contested decision’).

5

Through the aid scheme that is the subject of the contested decision (‘the measure at issue’), the United Kingdom of Great Britain and Northern Ireland establishes a capacity market involving centrally-managed auctions to procure the level of capacity required to ensure capacity adequacy. That aid scheme involves remunerating electricity capacity providers in exchange for their commitment to provide electricity or reduce or delay their electricity consumption during times of system stress (recital 4 of the contested decision).

6

The legal basis for the measure at issue is the UK Energy Act 2013 and the regulatory acts adopted on the basis of that Act, in particular the Electricity Capacity Regulations 2014 and the Capacity Market Rules 2014.

7

The capacity market works as follows: the amount of capacity required is decided centrally and the appropriate price for the supply of that amount is determined by the market through auctions in which all eligible capacity providers compete (recital 145 of the contested decision). The United Kingdom Government decides the amount of capacity required by taking into account, inter alia, recommendations from the electricity network operator, NG (recital 11 of the contested decision).

8

With regard to the auctions, the measure at issue provides as follows: each year, during the main auction, the capacity to be delivered four years later is auctioned (‘the T‑4 auction’); the capacity that was auctioned in 2014, for example, was to be delivered in 2018/2019, the delivery period running from 1 October 2018 to 30 September 2019 (recital 43 of the contested decision). Another auction takes place during the year prior to the delivery year of the main auction (‘the T‑1 auction’).

9

Some capacity will be systematically held back from the T‑4 auction and ‘reserved’ for the T‑1 auction; the amount of reserved capacity is to be based on an estimate of the amount of the ‘cost-effective’ DSR that could participate in the T‑1 auction and will be made public when the demand curve for the T‑4 auction is published (recital 45 of the contested decision). If demand falls between the T‑4 auction and the T‑1 auction, the amount of capacity auctioned in the T‑1 auction will be reduced. However, the contested decision notes that, because the T‑1 auctions provide a better route for DSR operators to access the market, the United Kingdom Government commits to procure in the T‑1 auctions at least 50% of the capacity ‘reserved’ four years earlier. The contested decision adds that flexibility will be retained to remove this guarantee if DSR does not prove cost-effective in the long run or if the DSR industry is considered sufficiently mature (recital 46 of the contested decision). The T‑4 auction and the T‑1 auction form the enduring regime.

10

With the temporary exception of interconnectors and foreign capacity providers, existing and new generators, DSR operators and storage operators are allowed to participate in the enduring auctions (recitals 4 and 149 of the contested decision).

11

In addition to the enduring regime, there is a transitional regime. Prior to the delivery period 2018/2019, ‘transitional’ auctions are scheduled, principally aimed at DSR operators. The first transitional auction was scheduled for 2015 for a delivery period running from October 2016 to September 2017; the second transitional auction was scheduled for 2016 for a delivery period running from October 2017 to September 2018 (recital 51 of the contested decision).

12

Generating and DSR resources participating in the capacity market are called ‘Capacity Market Units’ (‘CMUs’). Generating CMUs may participate individually as a CMU or in the aggregate with other eligible generating units subject to certain conditions. One of those conditions is that the aggregate capacity of all the units must be between the 2 megawatt (MW) de minimis threshold and 50 MW (recital 16 of the contested decision).

13

DSR CMUs are defined with reference to a commitment to reduce demand. Such commitment requires the DSR operator to cause its customer to reduce the import of electricity (as measured by half hourly meters) and/or export electricity generated by on-site generating units. Each component of a DSR CMU must be connected to a half-hourly meter and the provider’s total DSR capacity must be between 2 MW and 50 MW (recital 17 of the contested decision).

14

Eligible capacity providers have to undergo a pre-qualification process, which, in addition to basic administrative details, includes specific requirements depending on whether the potential participant is an existing or prospective operator (recital 26 of the contested decision). New capacity-generating providers and unproven DSR capacity providers (as opposed to proven DSR capacity providers, the declared capacity of which has been proved by the provider in a test) are also to be required to submit a bid bond of GBP 5000 (around EUR 5650) per megawatt for T‑4 auctions and T‑1 auctions and of GBP 500 (around EUR 565) per megawatt for transitional auctions.

15

The electricity network operator, NG, is responsible for organising the auctions in order to obtain the capacity level required to ensure electrical capacity adequacy.

16

Each auction will be a descending-clock, pay-as-clear auction in which all successful participants are paid the last-accepted bid. A high price is announced at the beginning of the auction, and participants submit bids to indicate how much capacity they are willing to supply at that price. This process is repeated in successive rounds according to a predetermined schedule until the auction discovers the lowest price at which demand equals supply, being the clearing price. All successful participants are paid the same clearing price (pay-as-clear model) (recital 49 of the contested decision).

17

Therefore, if successful at the auction, capacity providers are awarded a capacity agreement at the clearing price. The capacity agreements bid for by the applicants differ in terms of their length. Most existing capacity providers are to have access to one year agreements. Capacity providers undertaking capital expenditure above a GBP 125 per kilowatt (kW) (around EUR 141) threshold (plants to be refurbished) are to be eligible for capacity agreements of up to a maximum of 3 years. Capacity providers undertaking capital expenditure above a GBP 250 per kW (around EUR 282) threshold (new plants) are to be eligible for capacity agreements of up to a maximum of 15 years (recital 57 of the contested decision). Agreements longer than one year will only be available to participants in T‑4 auctions. In each case, the terms of the capacity agreement, including the capacity price, apply throughout the length of the contract.

18

Successful participants secure a steady payment during the term of their capacity agreement, which is financed through a levy on electricity providers. In return, they commit to deliver electricity at times of system stress. Financial penalties apply if successful participants do not deliver the amount of energy according to their capacity obligation set out in their agreement (recital 4 of the contested decision). Further, the capacity market is subject to a review process (recital 6 of the contested decision).

19

The costs incurred to fund capacity payments are to be paid by all licensed electricity suppliers (‘the cost recovery method’). Electricity suppliers’ charges are to be determined based on their forecast market share and calculated based on demand measured between 16.00 and 19.00 on all weekdays from November to February in order to incentivise suppliers to reduce their customers’ electricity demand at those times when demand is typically highest. According to the contested decision, this should reduce the amount of capacity that is needed, thereby also reducing the cost of the capacity market (recital 69 of the contested decision).

20

The gross capacity revenues paid to capacity providers have been modelled to be between GBP 0.9 billion and GBP 2.6 billion (around EUR 1.02 billion and EUR 2.94 billion) per year, that is between GBP 8.1 billion and GBP 23.4 billion (around EUR 9.14 billion and EUR 26.4 billion) in the period between 2018 and 2024, with that amount varying according to the level of ‘new build capacity’ required (recital 7 of the contested decision).

21

It should be borne in mind that, in the specific field of State aid, the European Commission may adopt guidelines on the exercise of its powers of assessment and, in so far as those guidelines do not contradict Treaty rules, the policy rules that they contain are to be followed by that institution (judgment of 13 June 2002, Netherlands v Commission, C‑382/99, EU:C:2002:363, paragraph 24 and the case-law cited).

22

Further, in its answers to the written questions put by the Court and again during the hearing, Tempus expressly confirmed that it did not contest the establishment of a capacity market as such, but only, first, the assessment of the elements provided by the United Kingdom regarding the impact of DSR and, second, the procedures envisaged to allow DSR operators to participate in the capacity market.

23

Among the various provisions of the Guidelines on State aid for environmental protection and energy 2014-2020 (OJ 2014 C 200, p. 1; ‘the Guidelines’) adopted by the Commission on 9 April 2014 and which entered into force on 1 July 2014, when considering a situation in which the principle of aid to procure the level of capacity required to ensure capacity adequacy is accepted (see paragraph 22 above), the relevant provisions for the purposes of the outcome of the dispute are the following.

24

First, in paragraph 224(b) under point 3.9.2, ‘Need for State intervention’, the Guidelines provide that, in its assessment of the elements provided by the relevant State, the Commission is to take account of, inter alia, the ‘assessment of the impact of demand-side participation, including a description of measures to encourage demand side management’.

25

Second, in paragraph 226, under point 3.9.3, ‘Appropriateness’, the Guidelines state as follows:

‘The measure should be open and provide adequate incentives to both existing and future generators and to operators using substitutable technologies, such as demand-side response or storage solutions. The aid should therefore be delivered through a mechanism which allows for potentially different lead times, corresponding to the time needed to realise new investments by new generators using different technologies. The measure should also take into account to what extent interconnection capacity could remedy any possible problem of generation adequacy.’

26

Third, as stated in paragraph 229, under point 3.9.5, ‘Appropriateness’, the Guidelines note that a ‘competitive bidding process on the basis of clear, transparent and non-discriminatory criteria, effectively targeting the defined objective, will be considered as leading to reasonable rates of return under normal circumstances’.

27

Fourth, in paragraph 232(a), under point 3.9.6, ‘Avoidance of undue negative effects on competition and trade’, the Guidelines point out, inter alia, that ‘the measure should be designed in a way so as to make it possible for any capacity which can effectively contribute to addressing the generation adequacy problem to participate in the measure, in particular, taking into account ... the participation of generators using different technologies and of operators offering measures with equivalent technical performance, for example, demand side management, interconnectors and storage’. That provision also observes that restriction on participation ‘can only be justified on the basis of insufficient technical performance required to address the generation adequacy problem.’

28

In the contested decision, with regard to Article 107(1) TFEU, the Commission took the view that the measure at issue did constitute State aid (recitals 109 to 115 of the contested decision).

29

Concerning the compatibility of the measure at issue with the internal market, the Commission noted that it had based its assessment on the conditions established under point 3.9 of the Guidelines, which sets specific conditions for procuring the level of capacity required to ensure capacity adequacy.

30

Under point 3.3.1, ‘Objective of common interest and necessity of the aid’, of the contested decision, the Commission states as follows:

...

...

31

Under point 3.3.2, ‘Appropriateness of the aid’, of the contested decision, the Commission states as follows:

...

...

32

Under point 3.3.5, ‘Avoidance of negative effects on competition and trade’, of the contested decision, the Commission notes as follows:

33

By application lodged at the Court Registry on 4 December 2014, Tempus brought the present action.

34

By document lodged at the Court Registry on 15 April 2015, the United Kingdom applied for leave to intervene in the present case in support of the form of order sought by the Commission. By order of 30 June 2015, the President of the Eighth Chamber of the Court granted leave to intervene. The United Kingdom lodged a statement in intervention and the main parties submitted their observations on that statement within the prescribed periods.

35

Following a change in the composition of the Court, the case was assigned to a different Judge-Rapporteur on 27 April 2016.

36

Following a change in the composition of the Chambers of the Court, the Judge-Rapporteur was assigned to the Third Chamber, to which the present case was therefore assigned.

37

By a measure of organisation of procedure of 20 December 2016, the Court invited the Commission to submit to it the notification of the measure at issue that had been sent to it by the United Kingdom.

38

By document lodged at the Court Registry on 13 February 2017, the Commission presented to the Court a non-confidential version of the notification that had been sent to it by the United Kingdom. It maintained that the notification at issue was a confidential document protected by Article 339 TFEU, and therefore could not be sent to Tempus. It stated that it could not send that document to the Court on the basis of a measure of organisation of procedure pursuant to Article 90 of the Rules of Procedure of the Court. The Commission indicated that, if the Court ordered it to produce that document on the basis of Article 91 of the Rules of Procedure, it would immediately comply with that measure of inquiry.

39

By order of 9 March 2017, relating to a measure of inquiry concerning the production of documents, the Court instructed the Commission, pursuant to Article 91(b), Article 92(1) and Article 103(1) of the Rules of Procedure, to produce the notification of the measure at issue that had been sent to it by the United Kingdom. The Commission complied with that request on 16 March 2017.

40

By measure of organisation of procedure of 3 April 2017, the Court communicated to the Commission and the United Kingdom its intention to include a number of extracts from the notification in the case-file. The Court invited them to submit observations regarding the disclosure of those extracts. The United Kingdom and the Commission submitted their respective observations in respect of the disclosure of those extracts on 19 April 2017 and 24 April 2017.

41

On 5 May 2017, some of the extracts from the notification, which were assessed as being relevant and non-confidential, were added to the case-file and sent to Tempus (‘the notification’).

42

By a measure of organisation of procedure of 5 May 2017, the Court invited all of the parties to provide replies to a series of questions. In essence, the Court’s questions to the parties related to certain paragraphs of the Guidelines and certain concepts that they refer to. The questions to the parties concerned, inter alia, the information available when the contested decision was adopted regarding the assessment of DSR and its possible technology developments with respect to security of supply in the United Kingdom. Further, the Commission was invited to outline its position with regard to certain statements in the contested decision and the elements available to it when it assessed the measure at issue. In addition, the Court invited Tempus to comment on the Commission’s statements in its rejoinder relating to the cost recovery method. The Commission was also invited to indicate the elements on which it had based its assessment in respect of the financing needs of the DSR operators, and Tempus was invited to indicate what evidence of their financing needs they had provided to the United Kingdom and the Commission.

43

On 22 May 2017, the parties complied with that measure.

44

By documents lodged at the Court Registry on 17 May 2017 and 9 June 2017, Tempus submitted observations on the report for the hearing.

45

By document lodged at the Court Registry on 29 June 2017, Tempus requested, on the basis of the principle of the equality of arms, that the Court re-examine Annex B to the notification, on which the Commission and the United Kingdom relied in their replies to the measure of organisation of procedure of 5 May 2017, and decide whether other parts of that annex that had not yet been disclosed had to be sent to the Court.

46

By order of 3 July 2017, relating to a measure of inquiry concerning the production of documents, the Court instructed the Commission, pursuant to Article 91(b), Article 92(1) and Article 103(1) of the Rules of Procedure, to produce Annex B to the notification. The Commission complied with that request on 4 July 2017. On 6 July 2017, that annex, which the Court found to be relevant and non-confidential, was added to the case-file and sent to Tempus.

47

Tempus claims that the Court should:

48

The Commission contends that the Court should:

49

The United Kingdom supports the form of order sought by the Commission and seeks the dismissal of the action.

50

The Commission notes, in its defence, that, in order for it to be admissible, the action of an interested party must be intended to safeguard the procedural rights available to it under Article 108(2) TFEU and Article 6(1) of Council Regulation (EC) No 659/1999 of 22 March 1999 laying down detailed rules for the application of Article [108 TFEU] (OJ 1999 L 83, p. 1). Therefore, Tempus’s arguments in respect of its first plea in law should be taken into account only to the extent that they seek to show that the Commission had not succeeded in overcoming the serious doubts that it must have encountered during the preliminary examination phase and that this affected Tempus’s procedural rights.

51

In that regard, it should be noted that the aim of the Commission’s argument is not to contest the admissibility of the action in its entirety, but only the part of the application that does not relate to the defence of Tempus’s procedural rights. During the hearing, the Commission accepted that the parties could agree that it was admissible for Tempus to defend its procedural rights. Moreover, it must be noted that the Commission has not specified the arguments that, in its view, should not be taken into account on the ground that they do not relate to Tempus’s procedural rights.

52

Having regard to the foregoing, the Commission’s argument relating to the admissibility of the first plea in law cannot be upheld.

53

As an interested party and in order to safeguard the procedural rights available to it under Article 108(2) TFEU and Article 6(1) of Regulation No 659/1999, Tempus submits two pleas in law in support of its action alleging, first, infringement of Article 108(2) TFEU and of several other rules and, second, a failure to state grounds.

54

As the United Kingdom and the Commission state, respectively, in the notification and the contested decision, there is a risk that the electricity available in the United Kingdom will be insufficient in the near future for the purposes of satisfying high-demand periods. The oldest generating plants will shortly be closing, and the electricity market is at risk of failing sufficiently to encourage generators to develop new generation capacities to make up for those closures. Equally, the electricity market does not offer sufficient encouragement to consumers to reduce their demand in order to remedy the situation. In order to guarantee the security of electricity supply, the United Kingdom therefore concluded that it was necessary to establish a capacity market.

55

The fundamental objective of the capacity market was to encourage capacity providers, that is to say, principally, both electricity generators (power plants, including those using fossil fuels) and DSR operators, who offer a service whereby consumption is rescheduled or reduced, to take into account the difficulties that may arise during high-demand periods. Thus, even if they intervene at different levels — supply for generation, demand for consumption — generators and DSR operators are essential elements for the structure and functioning of the capacity market envisaged by the United Kingdom.

56

Tempus takes the view that the Commission could not conclude, following a preliminary examination and in the light of the information available at the time when the contested decision was adopted, that the capacity market envisaged did not raise doubts as to its compatibility with the internal market. In that respect, Tempus highlights, as a DSR operator, seven aspects of the capacity market that do give rise to such doubts within the meaning given to that concept in Article 4 of Regulation No 659/1999. In essence, it submits that the capacity market privileges generation over DSR in a discriminatory and disproportionate manner that goes beyond what is necessary to achieve the objectives of the aid scheme and satisfy the relevant rules arising from EU law.

57

As is apparent from Article 108(3) TFEU, if, after having been informed of a plan to grant aid, the Commission considers that any such plan is not compatible with the internal market, it is to initiate without delay the procedure provided for in Article 108(2) TFEU. In accordance with the procedure envisaged by that provision, the Commission is then obliged to give the parties concerned notice to present their observations.

58

In that regard, Article 4 of Regulation No 659/1999 notes that, in so far as the plan notified by the relevant Member State does actually constitute a grant of aid, it is the presence or absence of doubts as to the compatibility of that plan with the internal market that allows the Commission to decide whether or not to initiate a formal investigation procedure after a preliminary examination, and provides as follows:

‘1.   The Commission shall examine the notification as soon as it is received. ...

2.   Where the Commission, after a preliminary examination, finds that the notified measure does not constitute aid, it shall record that finding by way of a decision.

3.   Where the Commission, after a preliminary examination, finds that no doubts are raised as to the compatibility with the [internal] market of a notified measure, in so far as it falls within the scope of Article [107](1) [TFEU], it shall decide that the measure is compatible with the [internal] market ([“the] decision not to raise objections”). The decision shall specify which exception under the Treaty has been applied.

4.   Where the Commission, after a preliminary examination, finds that doubts are raised as to the compatibility with the [internal] market of a notified measure, it shall decide to initiate proceedings pursuant to Article [108](2) TFEU ([“the] decision to initiate the formal investigation procedure”). ...’

59

Article 6 of Regulation No 659/1999 then sets out the detailed rules for the formal investigation procedure:

‘1.   The decision to initiate the formal investigation procedure shall summarise the relevant issues of fact and law, shall include a preliminary assessment of the Commission as to the aid character of the proposed measure and shall set out the doubts as to its compatibility with the [internal] market. The decision shall call upon the Member State concerned and upon other interested parties to submit comments within a prescribed period which shall normally not exceed one month. ...

2.   The comments received shall be submitted to the Member State concerned. ... The Member State concerned may reply to the comments submitted ... .’

60

Under the procedure for reviewing State aid, it is therefore necessary to distinguish between the preliminary stage of the procedure for reviewing aid under Article 108(3) TFEU, which is governed, inter alia, by Article 4 of Regulation No 659/1999 and is intended merely to allow the Commission to form a prima facie opinion on the partial or complete conformity of the aid in question, and the investigation stage envisaged by Article 108(2) TFEU, which is governed, inter alia, by Article 6 of Regulation No 659/1999 and is designed to enable the Commission to be fully informed of all the facts of the case (see judgment of 15 April 2008, Nuova Agricast, C‑390/06, EU:C:2008:224, paragraphs 57 and the case-law cited).

61

The Commission may restrict itself to the preliminary stage under Article 108(3) TFEU in order to take a decision favourable to an aid measure if it can satisfy itself, after having undertaken a preliminary investigation, that the project concerned is compatible with the FEU Treaty. However, where that preliminary investigation has led the Commission to the opposite conclusion or if such an investigation does not permit all the difficulties involved in determining whether the aid is compatible with the internal market to be overcome, the Commission is under a duty to carry out all the requisite consultations and for that purpose to initiate the procedure under Article 108(2) TFEU (see judgment of 15 April 2008, Nuova Agricast, C‑390/06, EU:C:2008:224, paragraphs 58 and 59 and the case-law cited).

62

With regard to the concept of ‘doubts’ as to the compatibility of the notified measure with the internal market, referred to in Article 4(3) and (4) of Regulation No 659/1999, three requirements have been established by case-law to act as a framework for the Commission’s assessment.

63

First, that concept is exclusive. Thus, the Commission may not decline to initiate the formal investigation procedure in reliance on other circumstances, such as third-party interests, considerations of economy of procedure or any other ground of administrative or political convenience (see, to that effect, judgments of 10 February 2009, Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraph 90 and the case-law cited, and of 10 July 2012, Smurfit Kappa Group v Commission, T‑304/08, EU:T:2012:351, paragraph 78).

64

Second, it follows, inter alia, from Article 4(4) of Regulation No 659/1999 that, when the Commission does not succeed in eliminating all doubt within the meaning of that provision, it is obliged to initiate the formal investigation procedure. It has no discretion in that regard (see, to that effect, judgments of 22 December 2008, British Aggregates v Commission, C‑487/06 P, EU:C:2008:757, paragraph 113 and the case-law cited, and of 10 July 2012, Smurfit Kappa Group v Commission, T‑304/08, EU:T:2012:351, paragraph 79 and the case-law cited).

65

Third, the concept of doubts referred to in Article 4(3) and (4) of Regulation No 659/1999 is an objective one. Whether or not such doubts exist requires investigation of both the circumstances under which the contested measure was adopted and its content. That investigation must be conducted objectively, comparing the grounds of the decision with the information available to the Commission when it took a decision on the compatibility of the disputed aid with the internal market. It follows that judicial review by the Court of the existence of serious doubts will, by nature, go beyond consideration of whether or not there has been a manifest error of assessment (see, to that effect, judgments of 2 April 2009, Bouygues and Bouygues Télécom v Commission, C‑431/07 P, EU:C:2009:223, paragraph 63, and of 10 July 2012, Smurfit Kappa Group v Commission, T‑304/08, EU:T:2012:351, paragraph 80 and the case-law cited).

66

In that context, in terms of a review of the legality of the Commission’s decision not to raise objections, it is for the Court to examine the argument put forward by Tempus to establish that, after a preliminary examination, doubts are raised as to the compatibility with the internal market of the notified measure within the meaning of Article 4(3) and (4) of Regulation No 659/1999, which should have led to the adoption of a decision to initiate a formal investigation procedure. Such an investigation is all the more important for Tempus given that those arguments correspond, in essence, to those that it could have put forward as an interested party exercising its procedural rights during the formal investigation procedure under Article 108(2) TFEU if such procedure had been initiated; it claims that, in the present case, those procedural rights have been infringed.

67

To that end, Tempus bears the burden of proof, which it can discharge by reference to a body of consistent indications, concerning, first, the circumstances and the length of the preliminary examination procedure and, second, the content of the contested decision (see, to that effect, judgment of 10 February 2009, Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraph 93). In particular, the insufficient or incomplete nature of the examination carried out by the Commission during the preliminary examination procedure constitutes an indication of the existence of doubts within the meaning of Article 4 of Regulation No 659/1999 (see judgment of 10 July 2012, Smurfit Kappa Group v Commission, T‑304/08, EU:T:2012:351, paragraph 81 and the case-law cited). As an interested party, Tempus has neither powers of investigation nor, in principle, investigatory capabilities that are comparable to those enjoyed by the Commission, which can, where necessary, request the cooperation of a Member State in order to complete its examination of the notified measure.

68

Accordingly, in the present case and as is apparent from paragraphs 79 and 82 below, at a stage of the procedure when the parties concerned have not yet been given notice to present their observations by a decision to initiate a formal investigation procedure, it is sufficient that Tempus set out its reasons for concluding, in the light of the contested decision, that the Commission should have had doubts as to the compatibility of the notified measure with the internal market. Therefore, it is not necessary for it to present sufficient evidence to establish the incompatibility of the aid scheme notified.

69

In that regard, it should be noted that, in order to be in a position to carry out a sufficient examination for the purposes of the rules that apply to State aid, the Commission is not obliged to limit its analysis to the information contained in the notification of the measure at issue. It can and, where necessary, must seek relevant information so that, when it adopts the contested decision, it has at its disposal assessment factors that can reasonably be considered to be sufficient and clear for the purposes of its assessment. By way of illustration, it has already been held that the Commission had carried out an ‘active and thorough’ examination of the compatibility of an aid measure as it had questioned the substance of the arguments put forward by the Member State (see, to that effect, judgment of 10 December 2008, Kronoply and Kronotex v Commission, T‑388/02, not published, EU:T:2008:556, paragraph 127), whereas an examination was held to be ‘insufficient’ on the basis that the Commission had failed to obtain information that would have allowed it to assess a measure (see, to that effect, judgment of 10 February 2009, Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraphs 109 and 110).

70

Therefore, in order to establish the existence of doubts within the meaning of Article 4(4) of Regulation No 659/1999, it is sufficient that Tempus show that the Commission has not researched and examined, thoroughly and impartially, all of the relevant information for the purposes of that analysis or that it has failed duly to take them into account in such a way as to eliminate all doubt as to the compatibility of the notified measure with the internal market.

71

In addition, it is established case-law that the lawfulness of a decision concerning State aid is to be assessed in the light of the information available to the Commission when the decision was adopted (see, to that effect, judgment of 15 April 2008, Nuova Agricast, C‑390/06, EU:C:2008:224, paragraphs 54 and 55 and the case-law cited). The information ‘available’ to the Commission includes that which appeared to be relevant for the assessment that the Commission had to carry out pursuant to Article 108(2) and (3) TFEU and Article 4(3) or (4) of Regulation No 659/1999.

72

In order to establish doubts as to the compatibility of the aid with the internal market, Tempus may therefore rely on all relevant information that was or could have been available to the Commission on the date when it adopted the contested decision. Equally, for the purposes of the review of lawfulness that the Court is required to carry out in that regard, the Court can take into consideration any information included in the contested decision in support of the assessment made by the Commission therein.

73

In the present case, it is therefore appropriate to review whether, in the light of the information available to the Commission when the contested decision was adopted, the elements put before the Court by Tempus were capable of raising doubts as to the partial or total compatibility of the notified measure with the internal market, therefore requiring the Commission to initiate the formal investigation procedure, without prejudice to the subsequent exercise by the Commission of its power to assess the compatibility of that measure with the internal market following the initiation of that procedure.

74

Tempus notes the significance of the questions that the Commission was confronted with in this case, which was the first investigation procedure relating to a capacity market and the first application of the Guidelines. In the present case, Tempus is of the view that the length of the discussions between the United Kingdom and the Commission prior to the notification went beyond what is required for the adoption of a decision not to raise objections and constituted an indication that may establish that there were doubts. Tempus also notes that it is paradoxical that the Commission takes the view, on the one hand, that it is not necessary to initiate a formal investigation procedure in respect of the first capacity market notified and, on the other hand, that it is appropriate subsequently to carry out the first sector inquiry ever initiated on the topic of State aid in order ‘to better understand these measures and ensure compliance with EU ... Rules’ and to reach ‘stakeholders who are not heard in ordinary State aid procedures’ (see Commission Decision C(2015) 2814 final of 29 April 2015 initiating an inquiry on capacity mechanisms in the electricity sector pursuant to Article 20a of Regulation No 659/1999 and associated press release).

75

The Commission contends that, in the absence of a complaint, it is the date of the notification alone that matters for determining whether the preliminary assessment procedure is of such a length as to raise doubts. As that procedure lasted less than two months, it is not of sufficient length to establish that there were doubts; rather, it implies that doubts did not arise. In any case, as the capacity market was designed in the context of a national public consultation that ended shortly before the notification was made, any contact the United Kingdom may have had with the Commission prior to notification is necessarily irrelevant for the purposes of knowing whether the Commission may have had doubts as to the measure that was ultimately notified. With regard to the paradox referred to by Tempus, the Commission claims that the lawfulness of the contested decision is to be assessed in the light of the information available when it came to its decision, without taking account of decisions adopted subsequently, which are included for their ‘psychological effect’. In any event, the ultimate decision to initiate a sector inquiry on capacity mechanisms in Member States other than the United Kingdom had the objective of ‘[gaining] a better understanding of the existence and functioning of [such] mechanisms’. With regard to the UK capacity market, such understanding had already been gained through its individual analysis of the notified measure.

76

The United Kingdom, in turn, submits that the length of discussion preceding notification does not give rise to doubts.

77

Under Article 108(3) TFEU, ‘the Commission shall be informed, in sufficient time to enable it to submit its comments, of any plans to grant or alter aid’. That obligation to notify allows the Commission to carry out a prior review of any proposed aid. Article 4(1) of Regulation No 659/1999 states that ‘the Commission shall examine the notification as soon as it is received’. If the notified measure raises doubts as to its compatibility with the internal market, Article 4(5) of Regulation No 659/1999 also specifies that the Commission has a period of ‘two months’, ‘[beginning] on the day following the receipt of a complete notification’, to take a decision in respect of a preliminary examination of the notification provided for in Article 108(3) TFEU, in order, where relevant, to initiate the formal procedure provided for in Article 108(2) TFEU.

78

In that regard, it must be recalled that the scale of the area of investigation covered by the Commission during the preliminary examination and the complexity of the matter under consideration may indicate that the procedure at issue considerably exceeded what is normally required for an initial examination carried out pursuant to the provisions of Article 108(3) TFEU. That circumstance constitutes probative evidence of the existence of doubts within the meaning of Article 4(3) and (4) of Regulation No 659/1999 (see judgment of 7 November 2012, CBI v Commission, T‑137/10, EU:T:2012:584, paragraph 285 and the case-law cited).

79

In the present case, it must be noted that the notified measure is significant, complex and novel.

80

First, in the contested decision, the Commission authorised for a period of ten years the implementation of an aid scheme intended to span several years (see recitals 6 and 162 of the contested decision). The amounts involved in the scheme are particularly high, namely between GBP 0.9 billion and GBP 2.6 billion per year (see recital 7 of the contested decision), that is between GBP 8.1 billion and GBP 23.4 billion over a period of 10 years, as was confirmed by the Commission during the hearing. Thus, for the period 2018-2019, the first delivery year envisaged by the capacity market, the United Kingdom had made known its intention to auction a capacity of 53.3 gigawatts (GW).

81

Second, both the definition and the implementation of that aid scheme are proving to be complex. The situations described to justify its existence are numerous and are based on probabilities (see recitals 79 to 82 of the contested decision). Different categories of operator are affected according to the modalities, which may vary from one category to another and which require certain eligibility criteria to be satisfied (see recitals 15 to 27 of the contested decision). Various auction mechanisms are described, which involve a number of stages (see, inter alia, recitals 28 to 51 of the contested decision). The measure is capable of producing significant, long-term effects, potentially continuing far beyond the authorised 10-year period, given that it allows certain capacity providers to enter into a contract of up to 15 years at each annual T‑4 auction (see recital 57 of the contested decision). In particular, those who will be impacted by such effects are existing and new generators and DSR operators, those effects being long term and experienced directly as well as indirectly.

82

Third, as Tempus states and as the Vice-President of the Commission responsible for competition explained when he summarised the contents of the contested decision in a press release, it is the first time the Commission has assessed a capacity market in the light of the Guidelines. The measure at issue is therefore novel in terms of both its subject matter and its implications for the future.

83

Other factors to be pointed out are the initiation of a sector inquiry relating to capacity markets and the subsequent initiation, in certain cases and for reasons specific to each case, of formal investigation procedures relating to capacity markets envisaged by other Member States, such factors having been pointed out by Tempus with the sole purpose of emphasising the complexity and novelty of the questions with which the Commission was confronted in the present case. However, as those events all occurred after the Commission adopted the contested decision, they cannot have any effect on the Court’s assessment of the lawfulness of the contested decision.

84

Yet, in the present case, it is apparent that, on 23 July 2014, following a preliminary examination lasting a month, the Commission concluded that the measure notified by the United Kingdom on 23 June 2014 did not give rise to doubts as to its compatibility with the internal market and that it could issue a decision not to raise objections in respect of the measure. The various capacity providers were not invited to submit their observations during that procedure, as the Commission took the view that it was not necessary.

85

Contrary to what is maintained by the Commission, given the circumstances of the present case, the fact that the preliminary examination lasted only a month is not, however, a reliable indication allowing it to be established that doubts had not arisen by the end of the first investigation of the measure at issue. As Tempus points out, it is also important to take into account the length and the content of the contact between the United Kingdom and the Commission before the measure was notified. While such contact is not unheard of, and is in fact encouraged by the Commission, in the present case, it does nonetheless have certain characteristics that highlight the complexity and novelty of the measure at issue.

86

As a preliminary point, it should be noted that the pre-notification phase, which arose as a result of the Commission’s experience in conducting State aid control proceedings, was formally established in the Best Practices Code on the conduct of State aid control proceedings, adopted by the Commission on 16 June 2009 (OJ 2009 C 136, p. 13; ‘the Best Practices Code’). Even though, as is stated in paragraph 8 of the Best Practices Code, the code does not alter any rights or obligations set out in the FEU Treaty and in the various regulations providing a framework for the procedures relating to State aid, the Best Practices Code does make clear the purpose, the length and the detailed rules of such informal contact.

87

According to paragraph 10 of the Best Practices Code, the pre-notification phase provides the Commission and the Member State with the possibility to discuss the legal and economic aspects of a proposed aid project informally prior to notification, and thereby enhance the quality and completeness of notifications. This phase thus paves the way for a more speedy treatment of notifications, once formally submitted to the Commission.

88

As the Best Practices Code points out, the purpose of the pre-notification phase is to facilitate the notification of the proposed aid in order to allow the Commission to carry out its preliminary examination optimally as soon as the notification is received. According to Article 2(2) of Regulation No 659/1999, the notification is to provide all necessary information in order to enable the Commission to take a decision pursuant to Articles 4 (decisions on preliminary examination of the notification) and 7 (decisions to close the formal investigation procedure) of the regulation.

89

The essential objective of the pre-notification phase is, therefore, to reduce the risk that the notification is found to be incomplete, thereby delaying the examination procedure of the notified project. In accordance with Article 5(1) of Regulation No 659/1999, the Commission can request all necessary additional information during the preliminary examination phase if it concludes that the information provided is incomplete.

90

By contrast, the purpose of the pre-notification phase is not to assess the compatibility of the notified measure with the internal market, especially in cases that are particularly novel or complex. Indeed, the Commission notes in paragraph 16 of the Best Practices Code that, in such a case, its services will not, in principle, provide at the end of the pre-notification phase an ‘informal preliminary assessment’ of the conformity of the draft notification and the prima facie compatibility of the planned project with the internal market. In any event, such guidance from the Commission in the form of informal, advance contact during the pre-notification phase cannot be considered to be the official position of the Commission, which is adopted after examining the notification.

91

As is stated in Article 4(1) of Regulation No 659/1999, it is only once the notification is received that the Commission is to start its examination. A prior statement on the completeness of the notification of the planned project cannot, in the light of the applicable rules, therefore be considered to be equivalent to an assessment carried out in the context of the procedures in place to monitor State aid laid down by the FEU Treaty and by Regulation No 659/1999. The Commission must not confuse the — possibly prior — phase of the preparation of the notification with the phase for the examination of the notification, which initially occurs as a preliminary examination and, where necessary, subsequently takes the form of a formal investigation, if it proves necessary in order to allow it to gather all the information needed to assess the compatibility of the aid and collect, to that end, the observations of the interested parties.

92

In that context, in the first place, it is apparent from the notification that the pre-notification phase, which includes all contact between the Commission and the relevant Member State prior to the notification, was significantly longer than the two-month period that is envisaged, as a general rule, by the Best Practices Code.

93

Preparations by the United Kingdom Government for establishing a capacity market started in 2010. In December 2012, that is to say around 18 months before the Commission commenced its preliminary examination of the notification pursuant to Article 4 of Regulation No 659/1999, the United Kingdom informed the Commission of the content of the planned measure.

94

As part of that informal contact, the Commission sent the United Kingdom a first set of questions regarding that measure, which the United Kingdom answered in July 2013.

95

It is also apparent from the notification that, on 25 March 2014, the Commission sent a second set of questions to the United Kingdom. In those 47 questions, the Commission asked the United Kingdom, inter alia, to provide information on the role of interconnection, the possibility of DSR participating in the capacity market and the various lengths of contract proposed to generators. By way of example, in question 32, the Commission asked the United Kingdom whether DSR operators were to be allowed to participate in the capacity market, and how they were to be treated. Further, in question 33, the Commission wishes to know the extent to which interconnected capacity was to be taken into account. On 31 March 2014, the United Kingdom answered those questions and sent the Commission an updated version of the planned measure.

96

On 17 June 2014, according to a letter from the United Kingdom of 27 June 2014 produced by the Commission before the Court, the Commission informed the United Kingdom that following a ‘preliminary assessment’ — carried out outside of the formal framework laid down in Article 4 of Regulation No 659/1999, as the preliminary examination starts only upon receipt of the notification — it concluded that the capacity market was prima facie compatible with Article 107(3)(c) TFEU in the light of the Guidelines.

97

Also on 17 June 2014, the Commission sent to the United Kingdom a third set of questions relating, inter alia, to the incentive effect of the planned measure, its proportionality and potential discrimination between capacity providers. The United Kingdom answered those questions directly in the notification in, inter alia, Annex H thereto, and provided a summary of those answers in its letter of 27 June 2014 referred to above.

98

However, in accordance with the Guidelines (see, inter alia, paragraphs 3.9.4, 3.9.5 and 3.9.6), questions relating to incentive effect, proportionality and the existence of potential discrimination are matters that must necessarily be assessed in the context of the examination of the compatibility of an aid measure that seeks to remedy problems of capacity adequacy, such as the measure at issue. In other words, contrary to the Best Practices Code for cases that are particularly novel or complex and despite the fact that such a non-binding, ‘informal preliminary assessment’ was not envisaged in paragraph 16 of that code, the Commission already concluded during that phase, even before its preliminary examination pursuant to Regulation No 659/1999, that the planned measure was prima facie compatible with the applicable rules, despite considering it necessary simultaneously to request information on key aspects of such an examination.

99

In the second place, in parallel to the discussions referred to above between the United Kingdom and the Commission, the United Kingdom organised a national public consultation from 10 October to 24 December 2013 relating to the planned capacity market. However, that consultation did not relate to the matter of compatibility of that measure with the applicable rules on State aid. It merely alluded to the requirement of authorisation from the Commission prior to the implementation of the planned measure.

100

In that regard, it cannot be held, as is suggested at times by the arguments submitted by the United Kingdom and the Commission, that a national consultation can be treated in the same way as a procedure allowing the interested parties to submit their observations, as would have been the case if the Commission had initiated the formal investigation procedure. In the context of State aid control proceedings, the relevant Member State providing the aid cannot substitute itself for the Commission, which must, as the guardian of the Treaties and in accordance with Article 108 TFEU, examine all projects intending to establish schemes of aid. It is for the Commission, rather than the Member State, where relevant and in the context of the procedure envisaged to that end, to gather all information necessary to allow it to assess the compatibility of the aid. Further, it is to the Commission, rather than to the Member State intending to provide the aid, that the interested parties must submit their observations, if they consider it necessary, in order to allow the Commission to come to a decision with full knowledge of the facts.

101

In the third place, even if the Commission takes the view that it did not receive any complaints (see paragraph 75 above), it is apparent from the notification and the contested decision that, in the light of the information available to them when they intervened, three types of operator wished to communicate directly and spontaneously to the Commission their observations on the compatibility of the aid.

102

Thus, on 30 May 2014 and 26 June 2014, the Commission received observations from a provider of balancing services asserting that the aid was not compatible with the Guidelines (see recitals 96 and 97 of the contested decision).

103

Further, on 9 June 2014, the Commission received observations from the UK Demand Response Association (‘UKDRA’), which also stated that the planned measure infringed the Guidelines. In particular, UKDRA referred to the different contract lengths available to DSR operators and generators, the mutual exclusivity between participation in the transitional auctions and the enduring auctions, the capacity market cost recovery method, and the failure to take into account the fact that DSR allows savings in electricity transmission and distribution costs. UKDRA also noted that, subsequent to the national consultation carried out by the United Kingdom, major changes had been made to the planned project, in order, inter alia, to support the generation of electricity from fossil fuels (see recitals 101 and 102 of the contested decision).

104

Finally, on 25 June 2014 and 3 July 2014, the Commission received letters from an operator that had acquired existing power plants. That operator claimed that the difference in treatment between existing and new plants — restricting existing plants, inter alia, to one year capacity contracts — infringed EU law (see recitals 98 to 100 of the contested decision).

105

The observations of the United Kingdom in response to those interventions were submitted to the Commission in Annex I to the notification. The United Kingdom noted, in that regard, that it responded to the statements contained in the two ‘complaints’ that had been annexed by the Commission to the third set of questions sent on 17 June 2014 — namely, those submitted on 30 May 2014 and 9 June 2014 — as well as to the statements made in ‘another complaint’ which, according to the United Kingdom, was to be sent to the Commission shortly afterwards by an operator that had acquired existing power plants (see recitals 103 to 107 of the contested decision, headed ‘Observations by the UK’). Further, in an email of 28 June 2014 sent to the Commission, the United Kingdom submitted, first, amongst information relating to contract length, operating reserve management and nuclear power, a table summarising difference in treatment between capacity providers and, second, noted that ‘for each of the complaints there is additional information within the notification, primarily in Annex I’.

106

It is apparent from the foregoing that the length of the pre-notification phase was significantly longer than the two-month period that is envisaged, as a general rule, in the Best Practices Code.

107

It is also apparent that, in addition to giving rise, from the very beginning, to a number of difficulties pointed out by the Commission in terms of the pre-notification phase, the planned measure also continued to give rise to such difficulties until the end of that phase. After more than a year of informal contact, the Commission asked the United Kingdom about the role of interconnection, the status of DSR and the difference in treatment between generators. Further, at the end of that informal procedure, the Commission asked the United Kingdom for further details on a number of points referred to by the Guidelines, which served as the framework for the analysis of the planned measure that was to be brought into force.

108

However, at that stage of the procedure, although the Commission was on the verge of starting its preliminary examination of the notification and three different types of operator had intervened by submitting to it their concerns as to the compatibility of the planned measure, it concluded that it did not have any doubts and therefore that it was not necessary to request observations from the interested parties on the various choices made by the United Kingdom in the notification.

109

In such circumstances, the length and the circumstances of the pre-notification phase, which attest to the difficulties caused by the need to gather relevant information to allow the Commission to examine the notification of the aid scheme in the light of the relevant provisions of the FEU Treaty, Regulation No 659/1999 and the Guidelines, as well as by the multiplicity of observations submitted in respect of that aid scheme by three different types of operator, are not such as to allow it to be concluded that the brevity of the preliminary examination procedure is an indication that there were no doubts as to the compatibility of that scheme with the internal market, and are, on the contrary, likely to constitute an indication that such doubts did exist.

110

In addition, it is not apparent from the documents before the Court that, during the month-long period of the preliminary examination of the notification, the Commission carried out a specific investigation into the role of DSR within the capacity market. The only document provided by the Commission to the Court in that regard is the email of the United Kingdom of 28 June 2014 (see paragraph 105 above). In that email, the United Kingdom sets out, inter alia, a summary, in the form of a table, of differences in treatment between capacity providers in response to concerns raised in that regard, most likely by the operator that had acquired existing power plants. The contested decision does not formally refer to that exchange and does not allow it to be concluded that, after it received that table or other information contained in the email, the Commission considered their content, insofar as concerns the role of DSR within the capacity market, either, for a second time, from the point of view of the United Kingdom or from the point of view of the interested parties. In any event, that table simply summarises the differences within the capacity market proposed in the notification between the following three types of operator: first, existing and refurbished power plants; second, new power plants; and, third, DSR operators. That document does not provide evidence allowing it to be established that the Commission carried out an independent assessment with regard to the criteria set out in the Guidelines (see, on that point, paragraph 117 et seq. below).

111

Consequently, with regard to the length of the discussions between the Member State and the Commission and to the circumstances surrounding the adoption of the contested decision, it is apparent from the foregoing that the notified measure is significant, complex and novel, and that it gave rise to a long pre-notification phase, during which the Commission asked the United Kingdom a number of questions in order to clarify important matters, particularly with regard to the assessment of the notified measure in the light of the Guidelines. It is also apparent that the notified measure was challenged in three respects by various operators who were supposed to benefit from it.

112

Similarly, it is not apparent from the documents before the Court that, under those circumstances, the Commission carried out a specific investigation during its preliminary examination concerning the information sent by the United Kingdom with regard to the role of DSR within the capacity market.

113

Taking into account the characteristics of the aid scheme at issue and the particular characteristics of its pre-notification phase, the Commission was not in a situation where it could simply rely on the information provided by the relevant Member State without carrying out its own investigation in order to examine and, if necessary, seek relevant information from, where appropriate, interested parties for the purposes of its assessment (see paragraph 69 above).

114

Given that the Commission has provided no evidence showing that such an examination was carried out, it can therefore be assumed that, in the present case, the Commission simply requested and reproduced the information submitted by the relevant Member State without carrying out its own analysis.

115

As Tempus claims, such circumstances are an indication that is capable of establishing that there were doubts as to the compatibility of the notified measure with the internal market.

116

Therefore, it is appropriate to examine whether elements relating to the content of the contested decision may, taking into account the observations submitted in that regard by Tempus in the present case, also indicate that the Commission should have had doubts following the preliminary examination of the notified measure.

117

Tempus claims that the Commission failed properly to assess the role of DSR within the capacity market with regard to, inter alia, the Guidelines, the objective of which is to ‘facilitate’ and ‘encourage’ DSR. Tempus also refers to the Commission documents mentioning the need to recognise ‘the potential for demand side to participate in the system’ to avoid ‘stranded investments in generation’. Thus, rather than simply approving the United Kingdom’s position, which is that ‘holding the first auction in December 2014 will be key to revealing information about DSR and DSR potential’ and that the transitional auctions should be sufficient (see recital 122 of the contested decision), the Commission should have required or carried out itself an appropriate assessment of the potential of DSR. The effect of that analysis would have been to reduce the ‘amount of generation capacity’ available from the first T‑4 auctions and avoid wasting public resources, which is detrimental to the consumer. The fact that DSR operators could participate in the T‑1 auctions was insufficient to compensate for that situation, due to the low capacity reserved for those auctions and the United Kingdom’s commitment to reduce, if possible, the ‘amount of capacity to procure in the year-ahead auction planned in 2017’. In such circumstances, generation from fossil fuels is ‘locked in’ from the outset, which risks blocking for years capacity that might have been provided by DSR.

118

Moreover, the other measures referred to by the Commission in its defence cannot be treated as measures supporting DSR within the aid scheme or measures specifically designed to deal with a capacity adequacy problem. Similarly, contrary to what is claimed by the Commission in its defence, the entry threshold of 2 MW which applies exclusively to DSR operators wanting to participate in the enduring regime is not ‘low’, as shown by a number of examples from other systems, such as the Pennsylvania Jersey Maryland (‘PJM’) capacity market and the New York Independent System Operator (‘NYISO’), where the entry threshold of 100 kW is 20 times lower.

119

The Commission observes that DSR would be insufficient in itself to prevent the risk of capacity shortages. Further, it maintains that Tempus has failed to establish that the Commission incorrectly analysed the potential role of DSR and that, as a result, it should have had doubts, obliging it to initiate a formal investigation procedure. Thus, Tempus failed to refer to the general measures supporting DSR implemented in the United Kingdom outside of the capacity market. Further, Tempus failed to take into account the characteristics of the UK capacity market intended to encourage DSR, such as the ‘low ... threshold of 2 MW’ required to participate in enduring auctions and the aggregation procedure for operators whose CMU falls below that threshold. Lastly, the Commission claims that DSR is not excluded from T‑4 auctions and DSR operators can bid for the contracts of the same length as existing generators. The risk of any ‘lock-in’ of capacity has been minimised and any remaining risk is outweighed by the positive effects of incentivising new investment as a prerequisite to address the capacity adequacy challenge.

120

The United Kingdom takes the view that Tempus complains that the United Kingdom ought to have done more to encourage and facilitate DSR without, however, establishing that there were doubts as to the capacity market’s compatibility with the internal market. It maintains that DSR is insufficient in itself to ensure the security of supply. Thus, while it is important to ensure that DSR operators can fully participate in the capacity market, the market was not designed specifically to encourage DSR operators above all else. The United Kingdom contests that awarding capacity contracts to existing or new generators is a waste of resources, given the importance of such contracts in achieving the objective of security of supply.

121

It is apparent from the replies to the written questions asked by the Court on 5 May 2017 that the Commission recognises that, for the purposes, inter alia, of the implementation of paragraph 232(a) of the Guidelines, both the capacity offered by generators and that offered by DSR operators are capable of contributing to solving the capacity adequacy problem identified by the United Kingdom. When questioned on that point, the Commission observed that ‘in principle, both capacities are capable of contributing, by their technical performance, to address the residual market failures identified by the UK’ (including the ‘missing money’ problem referred to in recital 85 of the contested decision).

122

The Commission also agrees with Tempus that the concept of technical equivalence does not, however, mean that the two are identical.

123

In its replies, the Commission stated as follows:

‘It should however be noted that both types of capacity have their specific capabilities and benefits. (New) generation capacity (incl. renewable energies) is required as a basis to have electricity fed into the grid, whereas peak shaving and peak shifting by demand response can reduce the need for generation capacity which would only be required in a limited amount of hours during the year (and constructing new generation capacities would therefore be overly expensive for that purpose).’

124

In that regard, Tempus noted the following:

‘DSR has a successful track record of exceeding the minimum technical performance requirement to deliver capacity, especially when legislators take the time to understand the nature of DSR and design the regulatory framework to best facilitate and encourage it. DSR is not the same as generation, nor should it be. “Equivalent technical performance” does not mean “the same” performance, it only means being technically capable of achieving the same result/outcome, ... A secure and reliable energy system requires diversity of sources, with different strengths. In addition to being a flexible and cost-effective source of capacity, DSR lowers network cost constraints by easing pressure on distribution and transmission infrastructure and reducing the need for expensive infrastructure upgrades, improves network reliability, encourages a smart and engaged customer side, boosts supply market innovation and competition, improves the value and viability of intermittent renewable generation by matching supply patterns without the need for a doubling up of generation through “back-up” fossil fuel plant, and overall plays a unique and significant role in ensuring security of supply at the lowest possible cost to customers.’

125

It is apparent from the foregoing that although generation and DSR can both contribute to remedying the capacity adequacy problem identified by the United Kingdom, each has its own advantage. Generation capacity supplies the network with electricity, while DSR capacity allows for peak shaving and peak shifting, thereby reducing the need to generate capacity which would only be required for a limited amount of hours during the year; that therefore avoids new generation capacities being built which, in such circumstances, would be too expensive.

126

Consequently, pursuant to paragraph 232(a) of the Guidelines, it was for the Commission to satisfy itself that the aid scheme was designed to allow DSR to participate alongside generation, because their respective capacities provide an effective solution to the capacity adequacy problem.

127

In that context, as follows, inter alia, from paragraph 226 of the Guidelines, the aid measures should be open and provide adequate incentives to the relevant operators.

128

The main question in the present case therefore is not whether DSR can play a role in ensuring the functioning of the capacity market, which it can, or, as the Commission and the United Kingdom claim, whether it is capable of dealing, as such and in the short term, with the capacity adequacy problem, which is not Tempus’s position. Rather, the key question is whether the role that DSR is capable of playing was assessed adequately having regard to the principles set out in the Guidelines.

129

The Court will examine, in the light of the foregoing, Tempus’s arguments criticising the Commission for authorising the measure at issue, following a one-month preliminary examination, without initiating the formal investigation procedure envisaged to allow the Commission to gather all the information needed to evaluate the compatibility of the aid and to allow the interested parties to submit their observations.

130

As follows from the Commission document of 5 November 2013 headed ‘Commission Staff Working Document, Generation Adequacy in the internal electricity market — guidance on public interventions’ (SWD(2013) 438 final), quoted by the United Kingdom in paragraph 219 of the notification: ‘capacity mechanisms should be designed fully taking into account the particular characteristics of demand response rather than defining products on the assumption that it will be filled by new generation’.

131

In that regard, in paragraph 220 of the notification, the United Kingdom observed that ‘demand side response ... [had] the potential to offer reliable capacity and a more cost effective way of delivering security of supply’. In that same paragraph, the United Kingdom presented DSR as an alternative to investment in generation capacity which was cheaper for consumers. For the United Kingdom, further development of the DSR is an important step towards a more efficient electricity market, in which participators respond better to price signals by reducing demand when electricity is scarce and prices are high and by consuming electricity when generation exceeds demand.

132

Further, the United Kingdom observed in paragraphs 61 and 188 of the notification that the participation of DSR operators in the capacity market was beneficial for competition in the capacity market, but also more widely in the energy market.

133

Likewise, in its notification the United Kingdom stresses that the bigger DSR becomes, the smaller the need for a capacity market will become. The United Kingdom notes that the capacity market will be withdrawn when it is no longer necessary, which should occur as DSR gradually develops (see pages 7, 8, 45 and 46 of the notification).

134

In essence, that analysis is identical to that set out by Tempus in the application, in which it maintains that DSR technology enables customers physically to move non-time-critical needs, thereby shedding demand peaks. Thus, usual consumption is replaced by highly reactive, optimised consumption. In that context, DSR, which allows peak demand to be reduced or enables reduction of demand during periods where production is scarce, is an alternative to generation, in particular to the type of generation that is most costly for the environment.

135

These statements from both the United Kingdom and Tempus show the real potential of DSR within the capacity market. Thus, the more quickly and comprehensively the United Kingdom implements this particular element of the mix of technologies, the smaller its reliance on generation capacity will have to be to solve the capacity adequacy problem.

136

It should be noted that in Annex B to the notification, headed ‘UK Checklist against the Commission Staff Working Document on Generation Adequacy in the internal electricity market — guidance on public interventions’, the United Kingdom gave the following reply to a question asking whether ‘the potential for demand side management and a realistic time horizon for it to materialise [had been] integrated into the analysis’:

‘DSR is expected to benefit from direct participation in the Capacity Market, as well as participation in the Transitional Arrangements. ... Each year, National Grid will assess the amount of DSR that currently exists and will estimate the potential DSR that could come forward over the period — as part of the analysis that will inform decisions on the amount of capacity to procure in the capacity auctions. The PJM Capacity Market in the US has brought forward about 15 GW of DSR over 10 years [that figure having been reached, according to the table, in delivery period 2015/2016]. However, the PJM market includes approximately triple the demand of the GB market. National Grid estimates that DSR could provide around 3 GW of capacity in 2018/19. Estimates of DSR development are very uncertain and difficult to obtain. However, the transitional arrangements will support the development of the sector and encourage greater DSR participation in the main Capacity Market.’

137

In paragraph 221 of the notification, the United Kingdom also refers to the results observed in the United States and notes that ‘evidence from the US has demonstrated that central auction capacity markets have been extremely successful at bringing forward DSR’. According to the graph submitted by the United Kingdom in the notification, from the period 2012/2013, the capacity provided by DSR in the annual PJM capacity market auctions was higher than 5 GW and reached 15 GW for contracts awarded in 2015/2016.

138

In addition, when questioned on the various points of information available on the evaluation of DSR’s potential, Tempus refers, inter alia, to reports compiled by Sustainability First in September 2012 and January 2014, commissioned by significant stakeholders in the electricity market, including NG, and claims that, in those reports, it was mentioned that 4 to 5 GW of capacity provided by DSR could presently be shifted from clients in the industrial sector. Tempus also submits other information from a report by Element Energy — De Montfort University Leicester headed ‘Demand side response in the non-domestic sector’ and dated July 2012, which specifies that between 1.2 GW and 4.4 GW can be shifted on a peak winter day in 2012 from non-industrial and non-residential buildings. For Tempus, that means that at least 5 GW of DSR capacity could have been available and taken into account on the capacity market, and the actual figure could have been even greater.

139

Finally, upon reading the contested decision, it appears that, when the Commission took its decision on the aid scheme notified, it was aware of the report compiled by a panel of technical experts (‘the PTE’), who were commissioned by the United Kingdom to examine NG’s recommendations concerning the capacity to be auctioned on the capacity market in December 2014, which was published on 30 June 2014 by the UK Department of Energy and Climate Change (‘DECC’). The Commission quotes the conclusions of the PTE report in recital 120 of the contested decision.

140

That report, which sets out the work carried out by the PTE from February 2014, contains a number of observations that could be made on the upcoming implementation of the first T‑4 auction in December 2014 for delivery period 2018/2019.

141

First, when it sets out its approach to examining NG’s findings, the PTE notes the following in paragraph 19 of its report in order to point out, inter alia, how useful the experience acquired by other capacity markets can be, in particular from the PJM capacity market, with regard to the integration of DSR:

‘The Panel has also drawn from the experience of its members in other Capacity Mechanism markets, such as PJM and New England, as well as its experience in other key areas where the need to procure capacity includes the demand side. The Panel has been somewhat reassured that DECC has drawn on the PJM experience, but the Panel remains concerned that not enough evidence has been provided on the potential contribution that the demand side might make, particularly the extent to which embedded generation might become available, with some retrofitting and aggregation, and the extent to which [combined heat and power] can deliver additional power to the system over and above its own demand in stress periods.’

142

Second, when it examined the contribution that could be made to the capacity market by DSR, the PTE observed as follows in paragraphs 96 to 106 of that report:

...

143

Both Tempus’s application and the analysis above highlight the urgent need to identify adequate incentives to allow DSR to participate effectively in the capacity market, taking into account its full potential. The conclusion of such an analysis is therefore that the capacity market’s modalities should not ‘spoil’ the potential of DSR and replace it with new generation capacity. In that regard, the PTE notes that it is regrettable that currently no organisation is even collecting the data needed to understand and gather information on the various aspects of the potential of DSR, despite some already being available.

144

Although NG and the UK DECC are criticised in that respect by the PTE, it can be concluded that, through its formal investigation procedure, the Commission does have sufficient means to request and obtain the relevant information to assess the situation in order, inter alia, to identify the potential need and, where relevant, the level of incentive necessary to allow the real potential of DSR to be exploited in response to the capacity adequacy problem identified by the United Kingdom.

145

Third, as part of its conclusions and recommendations, the PTE stated the following in paragraph 6(c) and in paragraphs 119 and 132 of its report, outlining its concerns on the lack of information and understanding of DSR in the United Kingdom:

‘The Panel raised concerns regarding the lack of information and understanding regarding Demand Side Reduction (DSR). The Panel prefers the term Distributed Energy Resources (DER) which imports the full range of contribution that could come from sources other than conventional generation whereas the term DSR appears to constrain demand-side awareness to mere reductions in demand and embedded generation. Noting the importance of building a strong institutional knowledge of DER amongst DECC and NG, the Panel recommended a programme to investigate this area further so that opportunities are captured in the future.

... NG’s overall Scenario and model-based approach is in principle sound, and it has sought to take account of evidence and stake-holders’ views. However, the Panel’s consensus view is that NG tended to take an overly conservative view on a few key assumptions, most notably interconnector flows, and by treating that (and weather) as sensitivities rather than including interconnector capacities based on their estimated availability probabilities (as with generation plant), it exaggerated the amount to procure. If instead the expected net interconnector flows are 2.25 GW (described as 75% imports) then the range of capacity to be procured would correspondingly decrease and, with a little more effort to procure DSR and accelerate interconnector commissioning, as well as expecting more coal plant to be offered into the auction, could be enough to avoid new [combined cycle gas turbines].

... Recommendation 9: ... a programme to research the full potential of DER should be instituted as soon as possible to inform future auctions with particular focus on the full range of peak demand mitigation resources that are referred to in this report.’

146

At this stage of the Court’s assessment, it appears that, when the Commission carried out its preliminary examination, it was in a position to analyse elements allowing it not only to envisage the current role of DSR — a technology that was found to be reliable and cost-effective by the UK authorities, the usefulness and effectiveness of which was already shown by US examples — but also to envisage the real potential of DSR, as illustrated, inter alia, by NG’s estimate, quoted by the United Kingdom in its notification, according to which DSR could provide around 3 GW of capacity in 2018/2019.

147

Similarly, the Commission was aware of the difficulties referred to by the PTE regarding the appreciation of the potential of DSR in the capacity market. As in the case of the appreciation of the potential of interconnection, in respect of which the concerns of the PTE were described as ‘serious’ by the Commission in recital 124 of the contested decision, and as is apparent from paragraphs 141 to 145 above, there was a risk that the capacity market envisaged would fail sufficiently to take into account the potential of DSR or, more widely, all potential that may reduce the need to resort to generation capacity in response to the capacity adequacy problem.

148

In that regard, it must be recalled that it is apparent from paragraph 224 of the Guidelines that the assessment of the impact of DSR operator participation is one of the elements of the examination to be carried out by the Commission when it takes a decision on the need for State intervention.

149

However, in that context, it is apparent from the contested decision that the Commission considered that, for the purposes of assessing the actual appreciation of DSR — and in order no longer to be in a situation where it could have doubts in that respect as to the compatibility of the aid scheme with the internal market — it was sufficient to accept the modalities envisaged by the United Kingdom in that regard.

150

According to the Commission’s statement in recital 122 of the contested decision in respect of its assessment of the compatibility of the aid and the need for State intervention and with regard to the contribution of DSR, it is sufficient to note as follows:

151

The Commission also observed, in recital 128 of the contested decision, that, even if the aid scheme ‘may [have resulted] in support to fossil fuel generation’, it was able to establish that the evaluation of the capacity adequacy problem, which was carried out annually, took into consideration all types of operator, including DSR operators (see also recitals 134 and 149 of the contested decision at other stages of the analysis). It therefore concluded, in recital 129 of the contested decision, that ‘the measure is technology neutral’ and therefore does not strengthen the position of fossil fuel generation operators.

152

Taking into account the elements available to the Commission, referred to above, and the elements that could have been available to it through the means it could use pursuant to Regulation No 659/1999, and given the size of the role that could be played by DSR within the capacity market in order, inter alia, best to decide whether State intervention is needed and to limit aid for fossil fuel generation to the appropriate amount, such assessments are insufficient to allow the Commission to dispel doubts emerging from the elements that were already in its possession or that could have been available to it when it adopted the contested decision.

153

In particular, as is apparent from paragraphs 226 and 232(a) of the Guidelines, according to which the aid measures ‘should be open’, but should also ‘provide adequate incentives to both existing and future generators and to operators using substitutable technologies, such as demand-side response’, and ‘should be designed in a way so as to make it possible for any capacity which can effectively contribute to addressing the generation adequacy problem to participate in the measure’, it is particularly important for the Commission to ensure that the capacity market at issue allows all solutions genuinely and effectively to participate in solving the generation adequacy problem, as each solution has its advantages and disadvantages.

154

Contrary to what is claimed by the Commission in the contested decision, given the elements available and taking into account the role of DSR, in the present case, the Commission could not be satisfied merely by the ‘openness’ of the measure and conclude, consequently, that it was technology neutral, without examining in greater detail the reality and the effectiveness of the appreciation of that technological solution in the capacity market.

155

Although it is apparent from the contested decision that the Commission examined the elements referred to by the United Kingdom with regard to the appreciation of generation capacity in the capacity market and, as a result, inter alia, of the concerns expressed by the PTE report, the elements provided subsequently concerning the appreciation of the capacity provided by interconnection, no element referred to in the contested decision proves that the Commission carried out its own examination concerning the actual appreciation of DSR, the potential of which was recognised and expedient, for the purposes of that market. By way of example, in the contested decision there is no reference to the NG’s 3 GW estimate. By failing to examine the potential role and capacity of DSR within the capacity market, the Commission accepted the United Kingdom’s information and assumptions (see paragraphs 149 to 151 above), despite their influence on the level of capacity auctioned and the amount of aid necessary for the capacity market.

156

Thus, with regard to the outcome of the first T‑4 auction in December 2014, given that that auction is being carried out in accordance with modalities that appear to fail sufficiently to take into account the potential of DSR, there is a risk — which is referred to in the PTE report — that generation capacity is granted a larger role than is strictly necessary. Therefore, it cannot be excluded that if the Commission had carried out its own examination of the potential of DSR, in particular with regard to the modalities of the appreciation of the NG’s estimates and of other sources and with regard to the reasons for the success of the US examples, the detailed rules for the participation of DSR operators would have been different.

157

Further, with regard to the T‑1 auction, it must be recalled that, according to recitals 45 to 47 of the contested decision, some capacity will be held back from the T‑4 auction and ‘reserved’ for the T‑1 auction, based on an assessment of the amount of the ‘cost-effective’ DSR that could participate in an auction. If demand falls between the T‑4 auction and the T‑1 auction, the amount of capacity auctioned in the T‑1 auction could also be reduced. As was confirmed during the hearing, it is apparent from those recitals that the United Kingdom Government committed to auction at least 50% of the ‘reserved’ capacity in the T‑1 auction, while retaining the flexibility to resile from that commitment if DSR did not prove to be ‘cost effective in the long run’ or if the DSR sector was considered sufficiently mature. Regulation 10 of the Electricity Capacity Regulations 2014, headed ‘Determining whether capacity auction is to be held’, provides, for example, in paragraph 3 that the relevant Secretary of State may determine that a T‑1 auction is not to be held if the forecasts show that no DSR providers will apply to bid in such an auction. Therefore, as the T‑1 auction requires, inter alia, an assessment of the ‘cost-effective’ DSR capacity that will participate in the auction, the potential capacity of DSR and the assessment thereof should have been analysed by the Commission in the contested decision.

158

In such circumstances, taking into account, inter alia, the importance of the role that could be played by DSR within the capacity market, the available elements concerning the potential of DSR are such as to give an indication that there were doubts as to the compatibility of that scheme with the internal market, which, upon reading the contested decision, cannot be held to have been allayed following the Commission’s preliminary examination.

159

Tempus maintains, in essence, that the assessment of the measure at issue should have given rise to doubts as to its compatibility with the internal market due to several infringements of the principles of equal treatment, protection of legitimate expectations and proportionality to the detriment of, inter alia, DSR operators. In particular, Tempus complains of the treatment of DSR operators concerning the length of capacity contracts, the capacity market’s cost recovery method, the conditions of participation in the auctions and the lack of additional remuneration for DSR operators in respect of the savings in the amount of electricity lost during transmission and distribution.

160

Tempus submits, in essence, that the Commission should have had doubts as to the compatibility of the measure at issue with the internal market as it limits the possibility of granting capacity contracts of longer than one year to generators who have exceeded a certain threshold of capital expenditure, thereby violating the principle of equal treatment to the detriment of DSR operators. Such discriminatory treatment is contrary to the objective that the measure at issue be technology neutral, gives a competitive advantage to generators and locks in a significant part of demand which could have been avoided through DSR.

161

Tempus claims that that inequality in treatment is not justified by the higher investment costs borne by generators who refurbish existing plants or build new ones. DSR operators and their customers also incur investment costs. Further, the customers of DSR operators want a guarantee that they will receive income over a number of years before they make such an investment. In that regard, Tempus submits that information had been sent to the United Kingdom and that the latter could not simply assert in the notification that the DSR sector had failed to provide ‘quantitative’ evidence that contracts with a longer term are necessary to support the participation of DSR operators in the auction.

162

The Commission contends, in essence, that it properly examined the measure at issue with regard to the treatment of DSR operators compared with that of other capacity providers and found that that measure allowed them to be competitive, taking into account their specific characteristics. The different contract lengths offered are justified by the central objective of the measure at issue, which is securing sufficient capacity in the future, including by incentivising investments in new plants. DSR operators are not permitted to have contracts that are longer than one year because they do not have the same financial needs as generators refurbishing an existing plant or building a new one. In that regard, the Commission argues that, during the administrative procedure, neither Tempus nor UKDRA submitted a substantiated argument that DSR operators need investment at a level comparable to projects for the refurbishment of existing plants or the building of new plants and provided no quantitative evidence whatsoever to that effect.

163

In its replies to the written questions of the Court, the Commission claims that, given the high level of capital costs, if plants to be refurbished and new plants were offered only one-year contracts, they would probably not participate at all or would exit from the auction as their bids would be above the price cap; even if they could bid, this would likely be at a very high price, which would become the clearing price for the entire auction. The Commission argues that not only would this produce significant windfall profits for existing generation capacity and DSR operators, but it would involve higher costs for all final customers and, consequently, a disproportionate amount of aid.

164

In that regard, according to settled case-law, the general principle of equal treatment, as a general principle of EU law, requires that comparable situations must not be treated differently and different situations must not be treated in the same way unless such treatment is objectively justified. The comparability of different situations must be assessed with regard to all the elements which characterise them. These elements must in particular be determined and assessed in the light of the subject matter and purpose of the European Union act which makes the distinction in question. The principles and objectives of the field to which the act relates must also be taken into account (see judgment of 12 December 2014, Banco Privado Português and Massa Insolvente do Banco Privado Português v Commission, T‑487/11, EU:T:2014:1077, paragraph 139 and the case-law cited).

165

In the present case, as an initial point, it must first be noted that the measure at issue gives DSR operators no opportunity to be granted capacity contracts of longer than one year.

166

The contested decision states that only capacity providers undertaking capital expenditure above a GBP 125/kW threshold (plants to be refurbished) will be eligible for capacity agreements of up to a maximum of 3 years and only capacity providers undertaking capital expenditure above GBP 250/kW (new plants) will be eligible for capacity agreements up to a maximum of 15 years (recital 57 of the contested decision).

167

Although recital 57 of the contested decision uses the technology neutral term ‘capacity providers’, it is apparent from the notification and from a reading of Regulation 2(1) of the Electricity Capacity Regulations 2014 in conjunction with Regulations 4, 5 and 11(3) thereto that capacity contracts of up to a maximum of 3 years and 15 years are expressly limited to generating CMUs undertaking capital expenditure exceeding thresholds established by the Secretary of State to the exclusion of DSR CMUs.

168

Consequently, first, DSR operators are ineligible to be issued contracts of up to 3 or 15 years, even if they establish that they have undertaken capital expenditure above the thresholds set for generating CMUs in the measure at issue. Second, the measure at issue contains no capital expenditure threshold for DSR CMUs that gives DSR operators the possibility of being issued capacity contracts of longer than one year.

169

In the contested decision, the Commission approved the United Kingdom’s position that it was sufficient to offer capacity contracts of up to 3 or 15 years to generating CMUs with a capital expenditure exceeding a certain threshold set by the Secretary of State and it was justified to offer to DSR CMUs only one-year contracts (recitals 106, 129 and 145 of the contested decision). Thus, it decided that DSR operators needed lower capital expenditure than generating CMUs building or refurbishing plants. On that basis, the Commission accepted that new-build and refurbishment generation capacities involving high investment costs are eligible for longer capacity contracts in order to allow operators to obtain necessary financing and concluded that offering shorter capacity contracts to DSR operators did not disadvantage them vis-à-vis operators of plants to be refurbished or new plants. Consequently, it concluded that the measure at issue was technology neutral and did not strengthen the position of fossil fuel generation operators.

170

It is therefore appropriate to verify whether the Commission could approve the difference in treatment between DSR operators and generators without initiating a formal investigation procedure or whether the fact that it was impossible for DSR operators to be issued capacity contracts of longer than one year should have led the Commission to have doubts as to the compatibility of the measure at issue with the internal market.

171

In that regard, in the first place, it should be noted that the measure at issue seeks to be technology neutral, in accordance with the requirements in paragraph 226 of the Guidelines.

172

It is apparent from the contested decision that the measure at issue aims to procure the necessary amount of capacity to ensure the security of the UK’s electricity supply (recitals 4 and 126 of the contested decision).

173

In more detail, it is apparent from the notification that the main aim of the measure at issue is to enable the provision of adequate reliable capacity in the UK electricity market at minimum cost to consumers in a way that minimises unintended consequences and risks, and supports the delivery of wider government objectives, such as enabling the decarbonisation of the electricity system, the development of a more responsive demand side and further integration of the internal energy market (paragraph 139 of the notification). Further, in order to achieve the objective of guaranteeing the UK’s electricity supply, the notification specifies that the measure at issue seeks to incentivise sufficient investment in generation and non-generation capacity, including specific support for DSR (paragraph 140 of the notification). Those objectives correspond to those referred to in paragraphs 216 to 221 of the Guidelines as being lawful for aid measures promoting capacity adequacy.

174

With regard to the main objective of guaranteeing security of electricity supply, as referred to previously in paragraph 121 above and as the Commission expressly recognised in its replies to the Court’s written questions of 5 May 2017, the parties agree that both generation capacity and DSR capacity can, in principle, contribute to solving the generation capacity adequacy problem.

175

Similarly, with regard to the secondary objective of incentivising sufficient investment in new capacity, it should be noted that the objective of the measure at issue is aimed at both generation capacity and other capacity, such as DSR. The notification also expressly points out in paragraph 355 that the capacity market does not seek to procure specific volumes of capacity on the basis of technology type. On the contrary, the measure at issue seeks to allow the market to determine what is the optimum quantity of each type of capacity (new generation capacity, generation capacity from plants to be refurbished, generation capacity from existing plants, existing DSR capacity, unproven DSR capacity) in order to reach the level of security of supply determined by the United Kingdom.

176

It follows that DSR operators must be considered to be in an equivalent situation to that of generators with regard to the objectives of security of electricity supply pursued by the measure at issue, which are technology neutral. That is particularly true, given that the Guidelines require that aid measures implementing a capacity market be open and provide adequate incentives to both existing and future generators and to operators using substitutable technologies, such as DSR or storage solutions (paragraph 226 of the Guidelines). It was therefore for the Commission to check whether the limitation on capacity contracts longer than one year to only generation capacity would nonetheless allow the capacity market to be technology neutral without distorting competition between generators and DSR operators.

177

In the second place, it is apparent from the contested decision that the fact that capacity contracts of longer than one year are offered to certain capacity providers is justified by their high capital expenditure and by their difficulties in securing financing.

178

Thus, according to the contested decision, the rationale for longer term contracts for new entrants is to help promote competitive new entry into the market. Allowing new entrants to receive a long term contract enables new entrants to secure lower-cost financing for their investment. This can help mitigate barriers to entry for independent firms who cannot finance investment in new capacity on the back of revenues from other plants in their portfolio. By encouraging competition in the market, longer term contracts can therefore help to lower costs for consumers in both the energy and capacity markets. Longer term contracts should also reduce the risk that participants with high investment or refurbishment costs load all of these costs into a single year agreement (recital 59 of the contested decision).

179

It is apparent from the contested decision that the use of longer capacity contracts seeks to achieve the technology-neutral objectives, recalled in paragraph 173 above, of guaranteeing the security of electricity supply and incentivising sufficient investment in capacity. Further, although the contested decision stresses that it is necessary to encourage new entrants onto the market, it has to be noted that granting contracts of longer than one year pursues a wider goal insomuch as operators refurbishing existing plants are also eligible for capacity contracts for a maximum of three years. It follows that the main reason for granting capacity contracts of longer than one year is to mitigate the financing difficulties of certain operators, due to the size of their capital expenditure, by guaranteeing them an income over a number of years and giving them the means to make a competitive bid during the auction, by allowing them to recoup their costs over a number of years.

180

It should therefore be noted that the decisive criterion referred to by the measure at issue to decide which operators are eligible for capacity contracts of longer than one year is the level of capital expenditure and the financing difficulties that could prevent those operators from participating in the capacity market.

181

As longer contracts were found to be needed in order to create a level playing field, it was necessary to consider what length was required to allow each category of capacity provider fully to participate in the capacity market, having regard to their investment costs and their financing difficulties, in order to comply with the obligation to provide sufficient incentives to all operators. It was therefore for the Commission to investigate whether reserving capacity contracts of longer than one year to certain technologies was discriminatory and was contrary to the objective of establishing a technology neutral capacity market, thereby contravening the requirements under the Guidelines.

182

In the third place, it should be noted that the Commission approved the United Kingdom’s position that it was not necessary to offer capacity contracts of longer than one year to DSR operators without examining whether their capital expenditure and their financing difficulties could require the possibility of their obtaining such contracts in order to allow them to participate in the auction, while avoiding an excessively high clearing price.

183

It is apparent from the contested decision that the Commission examined in detail the financing needs of capacity providers engaged in building new plants or refurbishing existing ones. The Commission expressly requested that the United Kingdom provide it with additional information in support of its choices relating to the length of capacity contracts, as longer contacts were in principle more problematic and had to be justified meticulously (table 16, page 161 of the notification). The Commission was therefore able to rely on detailed information provided by the United Kingdom regarding the capital expenditure and the financing difficulties associated with the building of new plants in order to determine the optimum length of capacity contracts and the capital expenditure thresholds that had to be a condition to being granted one of those contracts. This was a question of helping those operators to obtain the necessary financing and avoiding that their participation in the capacity market lead to an excessively high clearing price, while not allowing them to recover all of their fixed investment costs simply on the basis of income from the capacity market. The information provided by the United Kingdom included, inter alia, a number of detailed case studies analysing different scenarios as well as model business plans of various operators and various plants (paragraphs 4.3.1, 4.6.5 and C.4.3 of the notification).

184

On the other hand, it is apparent from the contested decision that the Commission did not attempt to analyse in detail the capital expenditure and financing needs of DSR operators. Indeed, after the Commission received UKDRA’s letter of 9 June 2014, by which it drew the Commission’s attention to the difference in treatment between generators and DSR operators, the Commission did ask the United Kingdom to respond. However, the Commission then merely took formal note of the United Kingdom’s response in which it simply stated, first, that DSR operators did not have the same capital needs as operators of new plants and, second, that UKDRA had not provided a single piece of quantitative evidence to support the suggestion that longer capacity contracts are necessary to support DSR participation in the auction (paragraph 511 of the notification).

185

Firstly, it should be noted that the United Kingdom has provided no detailed analysis supporting its position, in striking contrast to the information relating to the financing needs of generators. Indeed, the lack of information on United Kingdom DSR has also been pointed out by the PTE in its report (see, inter alia, paragraphs 19, 96 and 101 of the report quoted in paragraphs 141 and 142 above). Although the Commission quotes some of the conclusions from the PTE’s report in recital 120 of the contested decision, it did not deem it appropriate to gather more information itself on DSR to make up for the lack of information provided by the United Kingdom. Therefore, it must be held that neither the contested decision nor the notification contains a detailed examination of the capital needs of DSR operators.

186

Secondly, it must be pointed out that equal treatment, with regard to the length of the capacity contracts for which the various capacity providers could bid, was the DSR operators’ main claim against both the United Kingdom Government and the Commission. It is apparent from UKDRA’s contribution to the public consultation, to which the Commission could have had access when it adopted the contested decision, that UKDRA contested not only the fact that the capacity contracts of longer than one year were reserved for generating CMUs with capital expenditure exceeding a certain threshold, but also the capital expenditure threshold selected. In particular, UKDRA expressly invited the United Kingdom to create a model of the financing needs of the various technology types and to review the amount of the thresholds on that basis. In addition, by its letter of 9 June 2014, UKDRA communicated its doubts to the Commission as to the compatibility of 15 year contracts with the Guidelines and, independently of the matter of the length of the contracts granted, reiterated the desire of DSR operators to be able to bid for capacity contracts of the same length as those offered to generators.

187

Thirdly, both UKDRA and Tempus accept that new DSR operators do not necessarily have the same capital expenditure as generators building new plants. Nevertheless, they submit that, much like new generating CMUs, new DSR CMUs have capital expenditure and financing difficulties that justify them being granted capacity contracts of longer than one year in order to allow them to participate fully in the capacity market. In particular, Tempus claims that DSR operators must develop a sufficiently wide portfolio of clients to enable them to cover open-ended capacity events. Further, the investment necessary to allow the electrical consumption of each of those clients to be flexible over time, the financing of which DSR operators may be required to contribute, can be significant and can require longer capacity contracts. Similarly, the PTE observes expressly in paragraph 99 of its report that certain investments relating to DSR ‘[need] the certainty of a long term capacity contract to secure finance’ and concludes that it is essential to identify the potential DSR resources which are not likely to respond to capacity market signals because the existing incentives do not meet their needs. Indeed, DSR operators were not the first to claim that all capacity providers should be able to bid for contracts of the same length, regardless of the technology used, in order to create a level playing field. Indeed, it is apparent from the notification that a number of vertically integrated electricity companies had pointed out that, in order to enable a fair capacity price comparison, all capacity providers should have access to contracts with the same length and that each capacity provider should have the freedom to choose the length of the contract that allows it to be competitive in the auction (page 79 of the notification).

188

Fourthly, contrary to what is claimed by the Commission, Tempus and UKDRA cannot be criticised for failing to present more detailed information during the administrative procedure. According to case-law, where appropriate, the Commission must seek relevant information (see paragraph 69 above) so that, when it adopts the contested decision, it has at its disposal elements that can reasonably be considered to be sufficient and clear for the purposes of its assessment (see, to that effect, judgments of 10 December 2008, Kronoply and Kronotex v Commission, T‑388/02, not published, EU:T:2008:556, paragraph 127, and of 10 February 2009, Deutsche Post and DHL International v Commission, T‑388/03, EU:T:2009:30, paragraph 109). Therefore, in order to establish doubts within the meaning of Article 4(4) of Regulation No 659/1999, it is sufficient that Tempus show that the Commission has neither researched nor examined, thoroughly and impartially, all of the relevant information for the purposes of that analysis or that it has failed duly to take them into account in such a way as to eliminate all doubt as to the compatibility of the notified measure with the internal market.

189

In the present case, having regard to the considerations set out in paragraphs 171 to 176 above and given the lack of information on the capital needs of DSR operators, it is for the Commission to inform itself further on the matter by, for example, accepting UKDRA’s offer in its letter of 9 June 2014 to provide it with further information to determine whether granting contracts of a different length to DSR operators and other capacity providers was compatible with the principle of equal treatment. That is particularly so given that the DSR sector is highly diversified and it is established that UK authorities had little knowledge of that sector when compared with the generation sector (see paragraphs 141 to 145 above).

190

Finally, it should be pointed out that granting contracts of up to 3 or 15 years to certain generators has an effect on competition throughout the length of that contract. As the Commission concluded in recital 131 of the contested decision that the DSR sector is still in its infancy in the United Kingdom, it was for the Commission to examine whether the fact that it was impossible for DSR operators to bid for contracts of the same length as generators did not reduce their opportunities of contributing to solving the UK capacity adequacy problem when the sector was more developed. The granting of contracts of 3 or 15 years reduces in the future the capacity auctioned on the capacity market.

191

Fifthly, in paragraph 102 of its report, the PTE described the risk associated with the implementation of the UK capacity market in the absence of sufficient information relating to DSR as follows:

‘For all these reasons, we fully appreciate that NG were unable to carry out analysis on the demand side with the rigour and distinction that has been the hallmark of much of their other work. Nevertheless, this implies an urgent need to create a systematic process for ensuring that the resources of the demand side are not wasted only for new generation to step into the inefficiency gap.’

192

Therefore, it must be noted that the Commission approved the United Kingdom position that it was not necessary to offer contracts of longer than one year to DSR operators without examining how high the capital expenditure associated with new DSR CMUs was or investigating whether it was necessary to set specific thresholds for DSR CMUs having regard to their financial needs and the objectives pursued by the measure.

193

In the light of the foregoing, it is appropriate to conclude that the difference in the length of the capacity contracts offered to DSR operators and that of those offered to generators indicates that there were doubts as to the compatibility of the measure at issue with the internal market. It was for the Commission to examine the level of capital expenditure and the financial needs of DSR operators for the purposes of establishing that there was no infringement of the principle of equal treatment between generating CMUs and DSR CMUs, despite it being impossible for DSR operators to be granted capacity contracts of longer than one year. Having regard to the technology neutral objectives pursued by the measure at issue and the criteria established by the measure at issue, it was necessary to carry out such an examination prior to reaching a conclusion that the measure was compatible with the internal market. Therefore, the fact that the Commission did not have all the information with regard to the United Kingdom’s decision not to allow DSR operators to bid for contracts of the same lengths as those of other technologies, in the context of the preliminary examination procedure, indicates that there were doubts.

194

Tempus submits, in essence, that the Commission should have had doubts as to whether the measure at issue was proportionate and, consequently, as to whether it was compatible with the internal market, due to the cost recovery method selected, which fails sufficiently to incentivise consumers to reduce their consumption during demand peaks and therefore does not allow the total amount of aid to be limited to the minimum amount necessary.

195

Thus, Tempus claims that the cost recovery method adopted, namely cost recovery based on electricity consumption between 16.00 and 19.00 each weekday in winter, rather than consumption during the three highest annual demand peaks (‘the triad’), disadvantages DSR operators and infringes the principle of proportionality by increasing the amount of aid granted. Such a method makes it more difficult for consumers not to contribute to capacity market costs by reducing their consumption, that is to say their demand, at the relevant time, taking into account the fact that that consumption is inevitable for businesses and families. That is particularly the case given that small businesses and residential consumers could not avoid capacity market costs through DSR due to the fact that, in the United Kingdom, they would be categorised according to their profile and not according to the settlement of their consumption, which is divided up into half-hour periods.

196

Further, by failing sufficiently to incentivise consumers to reduce their electricity demand precisely when demand is highest and capacity is weakest, Tempus claims that the method adopted increases the amount of aid granted, by obliging the United Kingdom to procure more capacity than is necessary. According to Tempus, the United Kingdom does not contest that a clearer price signal is likely to lead to a decrease in the amount of aid, given that, originally, it had opted for a method based on the triad, before subsequently changing its mind after the close of the national public consultation.

197

In addition, Tempus states that the change in cost recovery method was specifically requested by vertically integrated suppliers, which benefited from the change. On 26 June 2014, one month before the adoption of the contested decision, the United Kingdom Office of Gas and Electricity Management (‘Ofgem’) decided formally to request that the United Kingdom Competition and Markets Authority carry out an investigation into the market for the supply of gas and electricity to individuals and small businesses due to, inter alia, concerns caused by the strong position of vertically integrated suppliers, in particular with regard to access to a market in which competition is already weak (Annex E3 to the observations on the statement in intervention, paragraphs 3.16 to 3.18, and Annex E4 to the observations on the statement in intervention, paragraph 1.39).

198

The Commission responds that the cost recovery method belongs to the financing side of the capacity market, which is not (at least not directly) relevant for the compatibility assessment of the measure at issue. The cost recovery method used in the present instance is an attempt to balance the interest in maintaining a demand reduction incentive with the need to reduce uncertainty for electricity suppliers as to their share of costs, the analysis in the contested decision being based on explanations provided by the United Kingdom Government in the notification in response to UKDRA’s letter. The Commission submits that the cost recovery method benefits DSR operators and that, in any event, it adds an additional layer of peak demand charging, which would not be the case with other methods such as flat charges or general taxation. The United Kingdom adds that, during the national public consultation, participants were specifically asked to suggest alternatives to the cost recovery method based on the triad.

199

In that regard, the Guidelines state that an aid measure is considered to be proportionate only if its amount is limited to the minimum needed to reach the objective aimed for (paragraphs 27(e) and 69 of the Guidelines). Further, with regard to aid measures promoting capacity adequacy, they must be constructed so as to ensure that the price paid for capacity availability automatically tends to zero when the level of capacity supplied is expected to be adequate to meet the level of capacity demanded (paragraph 231 of the Guidelines).

200

In the present case, first of all it should be noted that the method of recovering costs incurred in financing the capacity contracts adopted as part of the measure at issue consists of a charge levied on all licensed electricity suppliers, which is calculated on the basis of their market share of the electricity demand registered between 16.00 and 19.00 on weekdays from November until February (recital 69 of the contested decision).

201

Next, it should be stated that, in the contested decision, the Commission approved the cost recovery method adopted in the measure at issue. Thus, it concluded that the cost recovery method was an incentive to reduce electricity demand during peak times, while being predictable for electricity suppliers (recital 129 of the contested decision).

202

Therefore, it is appropriate to determine whether the Commission was able to approve the cost recovery method adopted in the measure at issue without initiating a formal investigation procedure or whether that method should have led the Commission to have doubts as to whether the measure at issue was compatible with the internal market.

203

In that regard, in the first place, contrary to the submissions of the Commission, it should be noted that the cost recovery method is relevant when assessing whether the measure at issue is compatible with the internal market and, in particular, whether it is proportionate.

204

Firstly, the amount of aid depends on the volume of capacity purchased through the capacity market and the auction clearing price. On the one hand, the volume of the capacity auctioned by the United Kingdom is determined by an electricity demand recommendation in combination with the available capacity and the application of a reliability standard seeking to achieve the adequacy level desired during demand peaks (recital 32 of the contested decision). The necessary capacity volume is therefore directly linked to the level of capacity consumed during demand peaks. The lower the demand peaks, the less necessary it is for the United Kingdom to purchase capacity to reach the desired level of security of electricity supply. On the other hand, a reduction in the capacity volume auctioned may also lead to a reduction in the clearing price insofar as it leads to a greater number of capacity providers competing for the same capacity volume. As the United Kingdom points out in its notification, greater competition leads to a lower auction clearing price (point 2, ‘Box 2 — When will it be possible to withdraw the Capacity Market ?’ of the notification). Indeed, the notification acknowledges that a reduction in electricity consumption during demand peaks leads, in due course, to an exit from the capacity market (see figures 8 and 9 on page 47 of the notification, which show that the ‘missing money’ problem will be reduced as DSR progresses).

205

Secondly, the United Kingdom accepted that the capacity market cost recovery method influences the volume of capacity auctioned. The United Kingdom explains that linking the charges intended to finance the recovery of the costs of the capacity market to the consumption of electricity during demand peaks is a clear incentive for the parties concerned to reduce their consumption during demand peaks, which reduces the amount of capacity needing to be purchased in order to reach the desired level of security of supply and, consequently, also reduces costs for consumers (point 624 of the capacity market project that was the subject of public consultation).

206

In the second place, it is apparent from the notification that the United Kingdom amended the cost recovery method after the public consultation. It was initially envisaged that the amount of the charges would be calculated on the basis of the electricity suppliers’ market share in the electricity demand registered during the so-called ‘triad’ periods, that is to say the three half-hour periods registering the highest annual electricity consumption in the United Kingdom during the period from November to February (paragraphs 521 and 522 of the notification). It was only after the public consultation that the United Kingdom decided to amend the cost recovery method to adopt one based on electricity consumption between 16.00 and 19.00 during winter weekdays, as described in paragraph 200 above.

207

In the third place, it should be noted that the Commission approved the United Kingdom’s position without examining the consequences of that amendment on the total amount of aid and, consequently, on whether the measure at issue was proportionate.

208

In its letter of 9 June 2014, UKDRA notified the Commission of its concerns following the amendment of the cost recovery method. UKDRA took the view that that method would blunt the price signal that should be sent to consumers during the highest demand peaks to reduce their demand. Further, the Commission’s attention was also drawn to the fact that residential consumers were categorised according to pre-established profiles and could not avoid capacity market costs by changing their consumption between 16.00 and 19.00.

209

Indeed, following UKDRA’s letter of 9 June 2014, the Commission asked the United Kingdom to respond. However, the Commission then simply noted the United Kingdom’s response, in which it merely affirmed that the cost recovery method ultimately adopted retained an incentive to reduce electricity consumption at peak times, while being predictable for suppliers (recital 129 of the contested decision). In particular, according to the United Kingdom, as the triad periods were identified after the event, using them as a reference period for calculating the charges created uncertainty for suppliers regarding the amount of their contribution to the financing of the system, which could incentivise them to make consumers pay a higher price. The United Kingdom also maintained that taking a predictable and defined period for the calculation of the charges was also intended to encourage the development of time-of-use tariffs, an approach that could also benefit residential consumers, who would then be able to respond and reduce their consumption between 16.00 and 19.00 on winter weekdays (paragraph 522 of the notification).

210

Despite the acknowledgment that the cost recovery method influences the volume of capacity needing to be procured on the capacity market, it is apparent from the contested decision that the Commission did not investigate whether the new cost recovery method effectively maintained an equivalent incentive to reduce electricity consumption during demand peaks by, inter alia, encouraging the development of DSR.

211

The Commission also did not investigate whether the cost recovery method adopted affected, inter alia, DSR operators’ access to the market, in particular by increasing the barriers to entry and expansion resulting from the strong position of vertically integrated suppliers. An aid measure may also have distortive effects on competition by strengthening or maintaining substantial market power of the beneficiary. Even where aid does not strengthen substantial market power directly, it may do so indirectly, by discouraging the expansion of existing competitors or inducing their exit or discouraging the entry of new competitors (paragraph 92 of the Guidelines).

212

With regard to the United Kingdom’s argument that the method initially proposed could incentivise electricity suppliers to impose higher premiums on end consumers, the Commission did not explain how that risk had any effect on the total amount of aid. Further, the Commission did not examine whether, with regard to consumers, such a premium could be offset by reducing the volume of capacity purchased on the capacity market, or even the clearing price, for the reasons set out in paragraphs 204 and 205 above.

213

In the light of the foregoing, it must be concluded that it was for the Commission to examine the potential effect of the change to the cost recovery method on whether the measure at issue is proportionate and, consequently, whether it is compatible with the internal market. Therefore, the fact that the Commission did not have all the information with regard to the consequences of changing the cost recovery method, in the context of the preliminary examination procedure, was another indication that there were doubts.

214

Tempus submits, in essence, that the Commission should have had doubts as to whether the measure at issue was compatible with the internal market, insofar as it infringed the Guidelines, and, in particular, the obligation to encourage and provide adequate incentives to DSR operators, due to the conditions of participation in the capacity market to which DSR operators were subject and which made it difficult for them to participate in the capacity market.

215

Firstly, Tempus submits that the interplay between the transitional auctions and the enduring auctions encourages DSR operators to participate in the transitional auctions due to their more favourable participation conditions, which leads them to be de facto excluded from the first T‑4 auctions. Instead of encouraging DSR operators to participate in the capacity market by giving them additional opportunities to bid, the transitional auctions thus actually lead to their participation in the enduring actions being restricted.

216

Secondly, Tempus submits that the measure at issue disadvantages DSR operators by obliging all capacity market participants to guarantee open-ended capacity events, whereas the majority of capacity events are time bound. By doing so, the measure at issue failed sufficiently to take into account the specific nature of DSR operators and discouraged them from participating in the capacity market.

217

Thirdly, Tempus submits that imposing the same bid bond requirement on participants in the capacity market may cause a market entry problem for DSR operators, given that the sector is still in its infancy. The problem is aggravated by the requirement to bid to cover open-ended capacity events. Indeed, the United Kingdom had initially intended to require a bid bond from new DSR operators that was lower than new generators. The measure at issue therefore discourages DSR operators from participating in the capacity market.

218

Fourthly, in response to the Commission’s argument in its defence that a 2 MW de minimis threshold for participation in the enduring auction was low and encouraged DSR operators to participate, Tempus submits that the threshold was actually rather high, in particular in the light of the threshold adopted in the US capacity market examples, and worsened the problems associated with the amount of the bid bond.

219

The Commission, supported by the United Kingdom, submits, firstly, that the measure at issue does not require DSR operators to choose between the transitional auctions and the enduring auctions, but, on the contrary, offers an additional chance to DSR operators who were not able to participate in the first T‑4 auctions or who participated in them unsuccessfully, with a view to support growth in the sector. Since the transitional auctions do not aim to provide additional support to DSR operators that are capable of winning enduring auctions, the exclusion from transitional auctions of DSR operators that have successfully participated in enduring auctions is entirely justified.

220

Secondly, the Commission submits that, in the light of the objective pursued by the measure at issue, namely security of supply regardless of the actual duration of each stress event on the system, the fact that it was not possible to bid for time-bound capacity events does not constitute discrimination against DSR operators insofar as providing for such an eventuality would limit the reliability of DSR operators vis-à-vis other capacity providers, would make the auctions more complicated and could also force the United Kingdom to procure more capacity. The United Kingdom adds that although the possibility of bidding for time-bound capacity events may be appropriate in the DSR sector, as such bidding is more familiar there and allows DSR operators to bid more accurately, it is not justified in the enduring regime insofar as it risks compromising the objective pursued or making it more expensive.

221

Thirdly, the Commission concludes that the bid bond conditions are reasonable. It observes that such a bond is intended to demonstrate the seriousness of the intention of new operators to participate and encourage them to provide the capacity necessary to reach the objective of security of supply. It notes that no discrimination issue was raised in that regard during the administrative procedure. The Commission also points out that the difference between the amounts of the bid bond required in the context of the enduring regime and those required in the context of the transitional regime is due to the fact that the latter regime was specifically designed to encourage new DSR operators. The United Kingdom adds that the majority of responses to the public consultation supported the inclusion of a collateral requirement for DSR operators and considered that the proposed level was appropriate.

222

Lastly, in its rejoinder, the Commission submits that Tempus’s arguments relating to the 2 MW de minimis threshold are inadmissible as they were only made at the reply stage. Further, it submits that the 2 MW de minimis threshold is low.

223

In that regard, it should be noted that paragraph 226 of the Guidelines states that the measure should be open and provide adequate incentives to both existing and future generators and to operators using substitutable technologies, such as DSR or storage solutions.

224

In the present case, it should first be noted that the notification indicates that the measure at issue contains some measures designed to encourage the development of DSR.

225

Firstly, the notification states that the organisation of transitional auctions is expressly intended to support the growth of DSR and provide the best possible chance that DSR operators are able to compete successfully in the enduring regime subsequently. In addition to their existence, the transitional auctions have certain features designed to encourage the development of DSR. The bid bond required to participate in the transitional auctions is set at 10% of the level required to participate in the enduring auctions. Further, the transitional auctions allow bids to cover time-bound capacity events, whereas the enduring auctions require participants to commit to cover open-ended capacity events (paragraphs 222 and 223 of the notification).

226

Secondly, the notification states that DSR is encouraged to develop by the organisation of the T‑1 auctions and, in particular, the guarantee from the United Kingdom to procure during those auctions at least 50% of the volume of its capacity initially reserved for such auctions, regardless of how need has developed between the date when the T‑4 auctions were organised and the date when the T‑1 auctions are organised (paragraphs 224 to 226 of the notification).

227

Thirdly, the notification states that DSR is also encouraged to develop by some of the conditions of participation in the enduring auctions. In particular, those conditions include the setting of a 2 MW de minimis threshold, the possibility of aggregating and the possibility for DSR operators of influencing the clearing price (paragraph 224 of the notification).

228

In the contested decision, the Commission approved the United Kingdom’s position. It states expressly in recital 131 of the contested decision that the measure at issue supports the development of DSR and that it includes measures specifically designed to help to develop the sector, which is still in its infancy. It is apparent from the contested decision that those measures include, inter alia, the fact that the transitional auctions are ‘limited’ to DSR operators and are specifically designed to encourage the development of DSR, by helping DSR operators that are not yet mature enough to be competitive in the enduring auctions (recitals 51 and 107 of the contested decision). They also include the guarantee that T‑1 auctions, which are ‘a better route to market’ for DSR operators than the T‑4 auctions, are organised and that the United Kingdom ‘commits’ to procure at least 50% of the capacity reserved in such auctions, while retaining some flexibility in the long term (recital 46 of the contested decision).

229

Therefore, it is appropriate to determine whether the Commission was able to confirm that the measure at issue provided adequate incentives to DSR without initiating a formal investigation procedure or whether the Commission should have had doubts as to whether the measure at issue was compatible with the internal market.

230

Tempus submits, in essence, that the more favourable participation conditions will lead DSR operators to favour participating in transitional auctions. According to Tempus, that would lead to a de facto exclusion of DSR operators from the first T‑4 auction. That would also lead to long-term capacity allocated to generators during the auctions being locked into the market.

231

In that regard, firstly, it should be noted that the measure at issue does not exclude DSR operators from the enduring T‑4 and T‑1 auctions, provided that they satisfy the participation conditions.

232

Secondly, contrary to what is claimed by Tempus, it should be pointed out that there is no genuine mutual exclusion between participation in the transitional auctions and participation in the enduring auctions. DSR operators whose bids have not been successful in the first T‑4 auction are still able to participate in the transitional auction. DSR operators that have bid successfully in the transitional auction are still able to participate in the subsequent T‑4 and T‑1 auctions. Therefore, the measure at issue does not require DSR operators to choose between participating in the transitional auction or the enduring auction.

233

Indeed, DSR operators that have obtained a capacity contract after the first T‑4 auction are not eligible to participate in the transitional auction. However, contrary to what is claimed by Tempus, that limitation is not tantamount to excluding DSR operators from the first T‑4 auction. The transitional auction is intended only to help DSR operators that are not sufficiently mature to participate successfully in the first enduring auction to develop by giving them an additional chance to receive a payment for capacity from 2015 and 2016 in order to be more competitive in the subsequent enduring auctions. In that regard, as the Commission rightly claims, the fact that the DSR sector is still in its infancy does not mean that some DSR operators have not already reached a sufficient level of maturity to be able to participate competitively in the enduring auction from the first T‑4 auction.

234

Thirdly, with regard to the risk of market lock-in due to the lower level of participation of DSR operators in the T‑4 auctions and due to the successive granting of an excessive amount of long capacity contracts to generators, that argument will be examined with those relating to the interaction between the T‑4 auctions and the T‑1 auctions. The risk of market lock-in referred to by Tempus assumes that the capacity volume reserved for T‑1 auctions does not allow DSR to develop.

235

It must be concluded that the interaction between the transitional auctions and the enduring auctions does not lead DSR operators to be excluded from the enduring auctions.

236

However, it must also be held that, by definition, the transitional auctions are not part of the enduring regime. Further, contrary to what is claimed by the Commission in recital 51 of the contested decision, it is apparent from the documents before the Court that the transitional auctions are not reserved solely for DSR operators, but are also open to small CMUs, as is observed in Regulation 29 of the Electricity Capacity Regulations 2014. In such circumstances, it is appropriate also to examine whether the enduring auctions provide DSR operators with adequate incentives.

237

Tempus maintains, in essence, that the enduring auctions do not provide DSR operators with adequate incentives because, first, T‑4 auctions have not been designed to consider the lead time of DSR operators and, second, the capacity volume reserved for the T‑1 auctions is limited.

238

In that regard, at the outset, it should be noted that the T‑1 auctions are particularly important for DSR operators.

239

The parties are in agreement that the T‑1 auctions may be better adapted for DSR operators than the T‑4 auctions due to those operators’ lead times. Thus, according to the contested decision, the T‑1 auctions are a ‘better route to market’ for DSR operators (recital 46 of the contested decision). Similarly, in its application, Tempus claims that it may be difficult for DSR operators to participate in the T‑4 auctions as those auctions require participants to bid and make investments immediately to provide capacity four years later and also receive payment only four years later (paragraph 75 of the application).

240

Indeed, the volume of capacity reserved for the T‑1 auction is calculated in the measure at issue on the basis of an assessment of the amount of cost-effective DSR that could participate in the auction (recital 45 of the contested decision).

241

However, firstly, it should be noted that the volume of capacity reserved in the T‑1 auctions is limited when compared with the volume of capacity auctioned during the T‑4 auctions. Further, the T‑1 auctions are in no way reserved for DSR operators and a part of the volume of capacity auctioned in the T‑1 auctions must therefore be attributed to capacity providers other than DSR operators.

242

Secondly, contrary to what is claimed by the Commission in recital 46 of the contested decision, it is apparent from the documents before the Court that there is no guarantee that the United Kingdom will organise a T‑1 auction if a T‑4 auction is organised or that it will procure through the T‑1 auction at least 50% of the volume initially reserved for that auction. While Regulations 7(4)(b), 10 and 26 of the Electricity Capacity Regulations 2014, read together, mean that the Secretary of State may decide not to organise T‑1 auctions, the text is silent on the commitment to auction at least 50% of the volume of capacity initially reserved for those auctions. At the hearing, when answering the Court’s questions, the representatives of the Commission and of the United Kingdom were also unable to point to the legal provision confirming the existence of such a guarantee, aside from the United Kingdom’s political statements.

243

In the light of the foregoing, it is appropriate to conclude that, although the organisation of T‑1 auctions may genuinely encourage the development of DSR, the Commission should have had doubts as to the size of the incentive in the present case, having regard to the limited volume of capacity reserved to the T‑1 auction and the absence of an express legal provision guaranteeing that the United Kingdom would procure at last 50% of the volume reserved for those auctions.

244

Tempus claims, in essence, that the conditions of participation in the enduring auctions fail adequately to incentivise DSR operators. According to Tempus, in practice, it is unlikely, having regard to some of the conditions of participation, that DSR operators are in a position to participate in the T‑4 auctions. Tempus relies, inter alia, on that fact that it is impossible for DSR operators to bid for a time-bound capacity obligation and on the amount of the bid bond.

245

In the first place, with regard to the duration of the capacity events, Tempus claims that the measure at issue discriminates against DSR operators by treating all participants in the enduring auctions equally and by requiring that everyone, including DSR operators, submit bids in respect of open-ended capacity events.

246

In that regard, as Tempus points out, the United Kingdom decided, in respect of the enduring regime, to require all operators to be capable of responding to open-ended capacity events. Conversely, under the transitional auctions, Article 29(3) of the Electricity Capacity Regulations 2014 allows DSR operators to elect whether to bid for either a time-bound or an open-ended capacity obligation. Further, as the United Kingdom acknowledges in its written pleadings before the Court, time-bound capacity obligations are more familiar to DSR operators and can help them accurately to quantify their risk exposure, meaning they can make more accurate bids in the enduring regime.

247

However, the Commission rightly claims that bids restricted to covering time-bound capacity events provide a lower level of security of supply vis-à-vis bids covering open-ended capacity events, and therefore do not allow the desired level of security of supply to be reached as easily. Requiring all capacity providers to cover open-ended capacity events, thereby making DSR operators responsible for the risk of default during ongoing capacity events, is therefore insufficient to give rise to doubts as to the compatibility of the measure at issue, insofar as that measure takes into account the financing needs relevant to each technology in order to allow all capacity providers to participate effectively in the capacity market. As is set out in paragraphs 182 to 192 above, it does not appear, however, that the Commission sought to verify whether the measure at issue took into account the financing needs of DSR operators.

248

In the second place, with regard to the bid bond, Tempus submits that imposing the same bid bond requirement on all participants in the capacity market may cause a market entry problem for DSR operators, given that the sector is still in its infancy.

249

In that regard, firstly, it should be noted that the United Kingdom accepted that the bid bond could be a barrier to new DSR operators accessing the market. It is apparent from the documents before the Court that the United Kingdom had initially envisaged reducing the amount of the bid bond for unproven DSR CMUs to avoid that bid bond being a barrier to entry for DSR operators new to the market (point 565 of the capacity market project that was the subject of public consultation). Further, during the public consultation, some DSR operators had also stated that the amount of the bid bond was a barrier to entry for new DSR operators. The amount of the bid bond may constitute a barrier to entry for new DSR operators in particular as all participants in the capacity market had to commit to covering open-ended capacity events while DSR operators may have more difficulty than generators in covering an ongoing capacity event. As DSR operators risk potentially being perceived as more likely to default, they may therefore have more difficulties in financing the amount of the bid bond.

250

Secondly, it should be observed that, following observations from generators and distributors in response to the public consultation, the United Kingdom decided, in the measure at issue, to align the bid bond imposed on unproven DSR CMUs with that imposed on new generating CMUs that are not yet operational. Therefore, the measure at issue is less favourable to DSR operators than the system initially envisaged in order to respond to the financing difficulties encountered by those operators.

251

However, as the Commission points out in its defence, while the United Kingdom had initially intended that the bid bond would be entirely forfeited in the event of a supply default, it should be noted that the measure at issue provides that the bid bond will be forfeited only pro rata to the volume of capacity that was not actually supplied by the DSR operators, provided that they provide at least 90% of the volume of capacity that they had committed to. Therefore, the measure at issue contains a measure specifically designed to make up for the fact that DSR operators lost the advantage that they had in the reduced bid bond amount when the amount of the bid bond required from unproven DSR CMUs was aligned with that of new generating CMUs.

252

Consequently, in the light of the goal pursued by the bid bond, aligning the amount of the bid bond required from unproven DSR CMUs with that of new generating CMUs is insufficient in itself to give rise to doubts as to the compatibility of the measure at issue, insofar as the measure takes into account the financing needs relevant to each technology in order to allow all capacity providers to participate effectively in the capacity market. As is set out in paragraphs 182 to 192 above, it does not appear, however, that the Commission sought to verify whether the measure at issue took into account the financing needs of DSR operators.

253

In the third place, in response to the arguments put forward by the Commission in its defence, Tempus claims that the setting of a 2 MW de minimis threshold is a barrier to entry to the capacity market for new DSR operators.

254

At the outset, it should be stated that Tempus’s arguments with regard to that threshold relate to the line of argument in the application concerning the discriminatory or disadvantageous treatment of DSR operators within the capacity market, as pointed out by Tempus at the hearing. Further, this argument responds to the Commission’s claims in its defence that the threshold was low and favourable to DSR. Thus, in the present case, the argument in question has not only a close connection with the application, but results, further, from the normal evolution of debate in proceedings before the Court (see, to that effect, judgment of 26 November 2013, Groupe Gascogne v Commission, C‑58/12 P, EU:C:2013:770, paragraph 31). Contrary to the Commission’s claims, that argument must therefore be considered merely to amplify a plea set out in the application.

255

Then, it should be observed that the notification presents the 2 MW de minimis participation threshold as low having regard to the participation threshold adopted by NG in the context of other measures and, consequently, as incentivising DSR operators to participate in the capacity market (paragraph 224 of the notification).

256

However, firstly, it should be noted that the participation threshold for the PJM capacity market was only 100 kW, being 20 times lower than the threshold adopted by the measure at issue. The PJM capacity market is expressly taken as a reference by the United Kingdom in the notification in support of its statement that the measure at issue allows the DSR sector to develop (paragraph 221 of the notification).

257

Secondly, while it is true that it is indeed possible for DSR operators to aggregate several sites in order to reach the 2 MW de minimis threshold, it should be noted that they are liable to pay a bid bond on the whole of the 2 MW, if even a tiny proportion of that volume is unproven DSR capacity. For the reasons set out in paragraphs 249 to 252 above, the amount of the bid bond may constitute a barrier to entry for new DSR operators.

258

Consequently, the Commission should have had doubts as to the statement that the setting of a 2 MW de minimis participation threshold was a measure encouraging DSR.

259

It is apparent from the entirety of the foregoing that the interplay between the T‑4 and the T‑1 auctions and some of the conditions of participation in the capacity market applicable to DSR operators should have led the Commission to have doubts as to, first, the capacity of the measure at issue to reach the objectives claimed by the United Kingdom in terms of development of DSR and, second, its compatibility with the requirements of the Guidelines in terms of adequate incentives for DSR operators and, consequently, as to the compatibility of the measure at issue with the internal market.

260

Tempus claims that the measure at issue gives rise to doubts as to its compatibility with the internal market in that it does not remunerate DSR operators for the savings in the amount of electricity lost during transmission and distribution. According to Tempus, the capacity provided by DSR operators reduces not only the overall amount of capacity required and circulating in the capacity market, but also the amount of the capacity lost in transmission and distribution of the electricity by around 7-8%. It takes the view that those savings should be incorporated into the remuneration of DSR operators in order to incentivise improvements to grid efficiency.

261

The Commission, supported by the United Kingdom, takes the view that the lack of additional remuneration for savings in the amount of electricity lost during transmission and distribution was analysed in the notification and was examined in the contested decision. Thus, the Commission adopted the explanation of the United Kingdom Government, according to which the capacity market has the sole objective of guaranteeing the availability of sufficient capacity on the system, not to reward all other benefits provided by each type of technology.

262

In essence, Tempus claims that the measure at issue gives rise to doubts as to its compatibility with the internal market in that it does not remunerate DSR operators for the savings in the amount of electricity lost during transmission and distribution.

263

In the contested decision, the Commission concluded that the measure at issue remunerates providers solely for making capacity available to the exclusion of all other services, such as delivering energy (recital 132 of the contested decision). It then took the view that, in the light of the objective pursued by the measure at issue, namely ensuring capacity adequacy in order to reach the desired level of security of supply, the lack of additional remuneration in respect of the savings in the amount of electricity lost during transmission and distribution was justified (recital 140 of the contested decision).

264

In that regard, it must be held that the measure at issue essentially establishes a capacity market that seeks to remedy the UK’s capacity adequacy problem.

265

The Guidelines expressly provide that the appropriateness of aid measures such as the one at issue is conditional on ‘the aid [remunerating] solely the service of pure availability provided by the generator, that is to say, the commitment of being available to deliver electricity and the corresponding compensation for it, for example, in terms of remuneration per MW of capacity being made available’ and that ‘the aid should not include any remuneration for the sale of electricity, that is to say, remuneration per MWh sold’.

266

Having regard to those considerations, it must be held that the lack of additional remuneration in respect of the savings in the amount of electricity lost during transmission and distribution did not give rise to doubts within the meaning of Article 4(3) and (4) of Regulation No 659/1999 that should have led the Commission to initiate the formal investigation procedure referred to in Article 108(2) TFEU. Consequently, Tempus’s arguments in that regard must be rejected.

267

It is apparent from examination of the first plea that there is a body of objective and consistent indications, based (i) on the length and circumstances of the pre-notification phase and (ii) on the incomplete and insufficient content of the contested decision owing to the lack of appropriate investigation by the Commission at the preliminary examination stage with regard to some aspects of the capacity market, that demonstrates that the Commission adopted the contested decision despite the existence of doubts. Without needing to adjudicate on Tempus’s other arguments, the Court concludes that the assessment of the compatibility of the measure notified with the internal market gave rise to doubts within the meaning of Article 4 of Regulation No 659/1999, which should have led the Commission to initiate the procedure referred to in Article 108(2) TFEU.

268

The contested decision must therefore be annulled.

269

In view of the annulment of the contested decision, which is necessary in the light of the first plea in law, there is no need to examine the second plea in law.

270

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. Since the Commission has been unsuccessful, it must be ordered to bear its own costs and to pay the costs incurred by Tempus Energy Ltd and Tempus Energy Technology Ltd, in accordance with the form of order sought by them.

271

Under Article 138(1) of the Rules of Procedure, the Member States which have intervened in the proceedings are to bear their own costs. Consequently, the United Kingdom must be ordered to bear its own costs.

On those grounds,

THE GENERAL COURT (Third Chamber, Extended Composition),

hereby:

I. Background to the dispute

A. Applicants and subject matter of the dispute

B. The contested measure

C. Relevant guideline provisions

D. The contested decision

II. Procedure and forms of order sought

III. Law

A. Admissibility

B. Substance

1. The first plea in law, alleging infringement of Article 108(2) TFEU, infringement of the principles of non-discrimination, proportionality and protection of legitimate expectations and incorrect assessment of the facts

(a) Preliminary observations

(b) Concept of doubts and the Commission’s decision whether or not to initiate a formal investigation procedure

(c) Length of the discussions between the Member State and the Commission and the circumstances surrounding the adoption of the contested decision

(d) Assessment by the Commission at the preliminary examination stage of the role of DSR within the capacity market in the light of the available information

(1) Equivalence and advantages of generation and DSR

(2) Positive role of DSR

(3) Available information on the potential of DSR

(e) Presumed discriminatory or disadvantageous treatment of DSR within the capacity market

(1) Length of capacity contracts

(2) Cost recovery method

(3) Conditions of participation in the capacity market

(i) Transitional auctions

(ii) T‑1 auctions and their interaction with T‑4 auctions

(iii) Conditions of participation in the enduring auctions

(iv) Conclusion

(f) Lack of additional remuneration for DSR operators in respect of the savings in the amount of electricity lost during transmission and distribution

(g) Conclusion

2. The second plea in law, alleging a failure to state reasons

Costs