Language of document : ECLI:EU:T:2021:645

JUDGMENT OF THE GENERAL COURT (Third Chamber)

6 October 2021 (*) (1)

(State aid – Polish electricity market – Capacity mechanism – Decision not to raise any objections – Aid scheme – Article 108(2) and (3) TFEU – Concept of doubts – Article 4(3) and (4) of Regulation (EU) 2015/1589 – Serious difficulties – Article 107(3)(c) TFEU – Guidelines on State aid for environmental protection and energy 2014-2020 – Failure to initiate the formal investigation procedure – Procedural rights of the interested parties – Obligation to state reasons)

In Case T‑167/19,

Tempus Energy Germany GmbH, established in Berlin (Germany),

T Energy Sweden AB, established in Gothenburg (Sweden),

represented by D. Fouquet and J. Derenne, lawyers,

applicants,

v

European Commission, represented by K. Herrmann and P. Němečková, acting as Agents,

defendant,

supported by

Republic of Poland, represented by B. Majczyna, acting as Agent,

by

PGE Polska Grupa Energetyczna S.A., established in Warsaw (Poland), represented by A. Ryan and A. Klosok, Solicitors, and by T. Janssens and K. Bojarojć-Bartnicka, lawyers,

by

Enel X Polska z o.o., established in Warsaw, represented by V. Cannizzaro, S. Ventura and L. Caroli, lawyers,

and by

Enspirion sp. z o.o., established in Gdańsk (Poland), represented by A. Czech, lawyer,

interveners,

APPLICATION under Article 263 TFEU for annulment of Commission Decision C(2018) 601 final of 7 February 2018 not to raise objections to the aid scheme for the capacity mechanism in Poland, on the ground that that scheme is compatible with the internal market pursuant to Article 107(3)(c) TFEU (State aid SA.46100 (2017/N)),

THE GENERAL COURT (Third Chamber),

composed of A.M. Collins, President, V. Kreuschitz (Rapporteur) and G. Steinfatt, Judges,

Registrar: E. Coulon,

gives the following

Judgment

I.      Background to the dispute

A.      The applicants

1        The applicants, the companies Tempus Energy Germany GmbH and T Energy Sweden AB (together, ‘Tempus’), sell electricity consumption management technology, also known as ‘demand-side response’ (‘DSR’), to individuals and professionals, inter alia in the electricity markets of the Federal Republic of Germany and the Kingdom of Sweden.

2        The service provided by Tempus to its clients is intended to create cost efficiencies in the electricity supply chain by combining DSR technology with the services provided by an electricity provider. Tempus sells electricity and helps its customers to move their non-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.

B.      The administrative procedure and the contested decision

3        By its action, Tempus seeks the annulment of Decision C(2018) 601 final of the European Commission of 7 February 2018 not to raise objections to the aid scheme for the capacity mechanism in Poland (‘the aid scheme’), which provides for the annual payment to capacity providers of 4 billion Polish zlotys (PLN), spread over a period of 10 years, on the ground that that scheme is compatible with the internal market, pursuant to Article 107(3)(c) TFEU (State aid SA.46100 (2017/N)) (‘the contested decision’).

4        On 16 November 2016, the Polish authorities had, pursuant to the Code of Best Practice for the conduct of State aid control procedures (OJ 2009 C 136, p. 13; ‘the Code of Best Practice’), pre-notified to the European Commission a proposed aid scheme to support capacity providers on the electricity market in Poland (the ‘Polish capacity market’ or the ‘Polish capacity mechanism’). On 27 January 2017, the Commission sent a request for information to the Polish authorities, to which they responded on 31 March 2017. A number of meetings and telephone conversations between those authorities and the Commission took place between 10 February 2017 and 18 January 2018 (recital 1 of the contested decision).

5        On 6 December 2017, the Polish authorities notified the Commission of the aid scheme (recital 2 of the contested decision).

6        On 7 February 2018, the Commission adopted the contested decision.

C.      The Polish electricity market

7        Four principal vertically integrated operators are active on the Polish electricity market, three of which, namely Enea S.A., Energa S.A. and PGE Polska Grupa Energetyczna S.A. (PGE), are majority-controlled by the Polish State through the public undertaking Polskie Sieci Elektroenergetyczne (PSE), whereas that State holds only approximately 30% of the shares of the fourth operator, Tauron Polska Energia S.A. In 2015, those operators together were responsible for 65% of Polish electricity generation and a significant part of the distribution and retail market. The national transmission grid, unbundled from the distribution network, is owned and operated by PSE. The fifth-largest operator active on the Polish electricity market is RWE Polska S.A., a subsidiary of RWE AG and renamed in 2016 Innogy Polska S.A., which is exclusively under private control.

8        The Polish electricity market remains dominated by hard coal and lignite generation. A significant share of the often ageing power plants is set to be decommissioned in the near future. The resulting gap in generation capacity poses a challenge for Poland, but also provides an opportunity to modernise, diversify and decarbonise its energy mix. Poland has an installed capacity of 41 gigawatts (GW) and a total demand of about 165 terawatts (TW) per year. Hard coal and lignite represent 40% and 20% of capacity, respectively, while renewable energy sources (‘RES’), in particular wind energy, provide a combined 19.5% of the installed capacity. However, coal still represents about 75% of actual energy generation. Although the peak demand of about 25 GW occurs in winter, electricity security is more at risk during the summer months, since a number of the combined heat and power plants stop operating in the warmer months and low water levels may restrict generation of thermal plants.

9        Like other wholesale electricity markets in the European Union, electricity in Poland is traded on the Polish power exchange or ‘over-the-counter’. Although both the wholesale and the retail markets are highly concentrated, electricity prices are within the average European range. The Polish balancing market is controlled by PSE as transmission system operator (‘TSO’). In addition to capital and operational expenditure, transmission charges also cover net balancing costs. Despite a drop in prices in recent years, Poland is a net exporter of electricity. It is envisaged to supplement the existing border-link interconnectors with Germany, the Czech Republic, Lithuania, Sweden and Slovakia with another border-crossing interconnector with Denmark. Electricity transmission to and from Sweden is controlled through a market coupling mechanism. Transmission with Germany, Slovakia and the Czech Republic is determined via auction between the TSOs. To mitigate the problem of transmission network congestion between Germany and Poland, 50Hertz Transmission GmbH, a German TSO, and PSE have recently agreed to build phase shifters at the interconnections. Splitting the common German-Austrian bidding zone has further decreased congestion.

D.      The aid scheme

10      The aid scheme establishes a capacity market mechanism or market which is intended to fill expected gaps between demand and capacity and, in so doing, ensure security of supply in a sustainable manner on the Polish electricity market, having regard to the estimate by the Polish authorities that that market will reach critical levels of resource adequacy or generation capacities in 2020. More specifically, according to those authorities, due to the extensive programme of phasing-out and mothballing of old power units by 2020, capacity shortages which cannot be met by market forces alone are expected, which is described as the ‘missing money problem’ (recitals 6 to 8 of the contested decision). In order to demonstrate that market failure, the Polish authorities relied, inter alia, on PSE’s data and medium-term forecasts on capacity adequacy (‘PSE’s adequacy assessment’), anticipating, in essence, for 2020, 2025 and 2030, a loss of load higher than the reliability standard expressed in terms of a Loss of Load Expectation (‘LoLE’) of 3 hours per annum (‘the reliability standard at issue’), which translates to a system security level of 99.97% (recital 31 of the contested decision). The data underlying that assessment were transmitted to the European Network of Transmission System Operators for Electricity (‘ENTSO-E’) for the purposes of the preparation of its 2017 report, entitled ‘Mid-term Adequacy Forecast (MAF) 2017’ (‘the 2017 MAF’), and PSE’s forecasts were reviewed by an external consulting firm (recitals 9 to 13 of the contested decision). As is apparent from recitals 15 and 16 of the contested decision, the Polish authorities committed to improving price signals during times of scarcity by a series of measures, including the assurance that, by 1 January 2021, DSR operators would be eligible to participate in the wholesale electricity and balancing markets and would be treated in a similar way as other market participants (recital 16(f) of the contested decision).

11      PSE is responsible for managing the capacity market, and one of its main tasks is to organise centrally managed auctions to procure the level of capacity required. Those auctions are in principle open to existing and new generators, DSR and storage operators, located in Poland or in the control area of neighbouring EU TSOs (recital 4 of the contested decision). Successful providers are to receive a steady payment for the duration of the capacity agreement granted (‘capacity payments’) in return for a commitment to deliver capacity at times of system stress when called on by PSE (the ‘capacity obligation’). If they fail to deliver the volume of energy corresponding to their capacity obligation, successful providers will incur financial penalties. Capacity payments are financed through a levy on electricity supplies (the ‘capacity charge’), collected from final consumers, on the basis of annual electricity consumption or of electricity consumed in the ‘selected hours of the day’.

12      In recitals 18 and 19 of the contested decision, it is however stated that, first, the Polish capacity market excludes capacity providers in receipt of operating aid, inter alia, aid granted under the Polish aid scheme aimed at encouraging the deployment of production from RES, as approved by Commission Decision C(2017) 8334 final of 13 December 2017 not to raise objections to the Polish support scheme for RES and relief for energy-intensive users (Ustawa o odnawialnych źródłach energii – aukcyjny system wsparcia OZE oraz ulgi w opłacie OZE dla przedsiębiorstw energochłonnych) (State aid SA.43697 (2015/N)) (‘the RES support scheme’), and, second, in order to avoid any overcompensation, capacity providers have the investment subsidies they receive deducted from their capacity payments, inter alia, in the context of the European Union’s Emissions Trading System (see Commission Decision C(2013) 6648 final of 22 January 2014 not to raise objections to the aid scheme relating to the derogation of Poland from Article 10c of Directive 2003/87/EC on emission trading – Free allowances to power generators (State aid SA.34674 (2013/N))).

13      The Polish capacity mechanism was introduced by the Ustawa o rynku mocy (Polish Capacity Market Act of 8 December 2017; ‘the Act’) (Dz. U. of 2018, item 9), which entered into force on 18 January 2018. Pursuant to Article 34 of the Act, the Polish Minister of Energy adopted executive regulations which detail the provisions governing the functioning of the capacity market. On 30 March 2018, the President of the Urzęd Regulacji Energetyki (Energy Regulatory Office, Poland) approved those executive regulations. On 24 August 2018, the Polish Minister of Energy adopted the executive ordinance on the parameters of main auctions for the delivery period between 2021 and 2023.

14      The purpose of the Act is to provide medium- and long-term security of electricity supply to consumers, in a cost-effective, non-discriminatory and sustainable manner (Article 1(2)). The aim of the Polish capacity market is to create and negotiate capacity obligations, that is to say, the obligation of an operator to ensure the provision of capacity during delivery periods and its actual provision during emergency periods. That capacity can be made available either by generating and providing electricity or, in the case of DSR, by reducing demand at times of system stress. The auctions conferring capacity obligations are preceded by a registration and certification procedure (Articles 11 to 28 of the Act; recitals 20 to 26 of the contested decision), with characteristics concerning the certification of DSR operators (recitals 27 and 28 of the contested decision). Capacity providers participate in the auctions and in the Polish capacity market in the form of Capacity Market Units (‘CMUs’) which may, inter alia, be generating or DSR CMUs and consist of one or more physical units aggregated into a single group for the purposes of bidding (Article 16 of the Act; recital 17 of the contested decision). In order to be eligible as CMUs, physical generating or DSR units, including foreign ones, must achieve a minimum (net) capacity threshold of 2 megawatts (MW) (Article 16(1) (1), (2), (5) and (6) of the Act). For groups of physical generating or DSR units, including foreign ones, maximum capacity is to be 50 MW and each physical unit may not exceed the maximum (net) capacity of 10 MW (Article 16(1) (3), (4), (7) and (8) of the Act).

15      It is apparent from recitals 42 and 43 of the contested decision that the length of the capacity agreements to be granted to CMUs is determined, in principle, on the basis of the investment costs (‘CAPEX’) which are expected to be incurred in the five years before the start of the delivery year, with the exception of the first main auction, in which it is the CAPEX incurred since January 2014 that is relevant. The CAPEX threshold is stipulated in the Minister of Energy executive regulation for each auction. The Polish authorities have offered commitments to the Commission whereby CAPEX thresholds can be lowered only in the future and whereby it will notify it of any lower adjustment below 20% of current thresholds. Moreover, they are required to ensure that the CAPEX thresholds for obtaining 5- or 15-year capacity agreements are technologically neutral. CMUs which do not undertake significant CAPEX, in particular existing generating CMUs, may conclude only one-year capacity agreements. New generating CMUs, modernised generating CMUs and DSR CMUs incurring CAPEX equal to or higher than PLN 500 000/kilowatt (kW) are eligible for capacity agreements for a maximum period of five years. New generating CMUs the CAPEX of which is greater than or equal to PLN 3 million/kW may conclude capacity agreements for a maximum period of 15 years. Capacity agreements longer than one year are granted only in the main auctions. According to Article 25(5) of the Act, CMUs may receive a ‘Green Bonus’ allowing the duration of a multi-year capacity agreement to be extended by two years either where the CMU’s carbon dioxide (CO2) emission performance standard is less than or equal to 450 kilograms per megawatt hour (kg/MWh) of energy generated or, in case of cogenerating units, where the CMU delivers at least half of the heat generated to a heating network using hot water as the heat carrier (recital 49 of the contested decision).

16      Article 29 of the Act provides for two types of capacity auction, namely main auctions and supplementary auctions. The main or ‘N‑5’ auction must be organised during the fifth year prior to the delivery period, with the exception of the first three main auctions for the delivery period between 2021 and 2023, which were already organised in 2018. Supplementary or ‘N‑1’ auctions intended for short delivery periods for each quarter of the calendar year must be organised at the same time in the year preceding the delivery year (recitals 37 and 38 of the contested decision).

17      According to recitals 51 to 73 of the contested decision, the participation of foreign capacity in auctions is permitted within the limits set by two models, which are referred to by the Commission as the target and transitory solutions (Article 6(1) of the Act). According to that institution, under the target solution, which is intended to be implemented in the long term, PSE preselects eligible capacity providers by organising three separate pre-auctions, under Article 4(1) of the Act, for each of Poland’s borders. To that end, the area covering part of Germany, the Czech Republic and Slovakia is treated as one border because of its integration into a synchronised frequency area, whereas the borders with Lithuania and Sweden are treated as separate borders (Article 6(1)(2) of the Act, read in conjunction with Article 6(6) thereof; recital 54 of the contested decision). Bidders winning a pre-auction are subject to a certification procedure as foreign CMUs, registered and eligible for the award of one-year capacity agreements in the main auctions in which they take part in a ‘passive’ way. This means that their exit offers are automatically equal to the offers submitted during the pre-auctions. (recitals 59, 62 and 63 of the contested decision). In recitals 66 and 67 of the contested decision, it is stated that such participation of foreign CMUs requires the conclusion of cooperation agreements between PSE and its neighbouring TSOs (Article 6(2) of the Act), which is why, according to the Commission, only the transitory solution is applied at that stage. Under the latter solution (Article 6(1)(1) of the Act), instead of foreign capacity providers, the (five) interconnectors for each of Poland’s borders, represented by the neighbouring TSOs as co-owners of the interconnecting assets, participate in the main auctions (recital 68 of the contested decision). The implementation of that solution also requires the conclusion of inter-TSO agreements (Article 6(2) of the Act, read in conjunction with Article 8(1) thereof), although the Commission describes these as simpler than those required in the target solution (recital 73 of the contested decision).

II.    Procedure and forms of order sought

18      By application lodged at the Court Registry on 14 March 2019, Tempus brought the present action.

19      By document lodged at the Court Registry on 3 June 2019, PGE applied to intervene in the present case in support of the form of order sought by the Commission. By order of 20 September 2019, the President of the Third Chamber of the General Court granted it leave to intervene. PGE lodged its statement in intervention and the main parties lodged their observations on that statement within the periods prescribed.

20      By document lodged at the Court Registry on 7 June 2019, the Republic of Poland applied to intervene in the present proceedings in support of the form of order sought by the Commission. By order of 23 September 2019, the President of the Third Chamber of the General Court granted it leave to intervene. The Republic of Poland lodged its statement in intervention and the main parties lodged their observations on that statement within the periods prescribed.

21      By document lodged at the Court Registry on 17 June 2019, Enspirion sp. z o.o., a DSR operator active on the Polish electricity market since 2012, applied to intervene in the present proceedings in support of the form of order sought by the Commission. By order of 20 September 2019, the President of the Third Chamber of the General Court granted it leave to intervene. Enspirion lodged a statement in intervention and the main parties submitted their observations on that statement within the periods prescribed.

22      By document lodged at the Court Registry on 27 June 2019, Enel X Polska z o.o. (‘Enel X’), a DSR operator active on the Polish electricity market since 2014 with a share of around 70% of DSR capacity in Poland, applied to intervene in the present proceedings in support of the form of order sought by the Commission. By order of 20 September 2019, the President of the Third Chamber of the General Court granted it leave to intervene. Enel X lodged a statement in intervention and the main parties submitted their observations on that statement within the periods prescribed.

23      Following a change in the composition of the Chambers of the General Court, pursuant to Article 27(5) of the Rules of Procedure of the General Court, the Judge-Rapporteur was assigned to the new composition of the Third Chamber, to which the present case was accordingly allocated.

24      No request for a hearing was made by the main parties, under Article 106 of the Rules of Procedure, within the period prescribed.

25      On a proposal from the Judge-Rapporteur, the General Court (Third Chamber) decided, pursuant to Article 106(3) of the Rules of Procedure, to rule on the action without an oral part of the procedure and, by way of measures of organisation of procedure provided for in Article 89 of the Rules of Procedure, put written questions to the Commission and to the interveners, to which they replied within the period prescribed. Tempus submitted its observations on the replies of the Commission and of the interveners within the period prescribed.

26      Tempus claims that the General Court should:

–        annul the contested decision;

–        adopt measures of organisation of procedure, pursuant to Article 89(3)(d) of the Rules of Procedure, ‘ordering’ the Commission to produce the following documents, referred to in recital 1 of that decision, namely:

–        the request for information addressed by the Commission to the Polish authorities of 27 January 2017 (see paragraph 4 above);

–        the records of the ‘several meeting or phone calls’ which took place between the Polish authorities and the Commission between 10 February and 20 November 2017;

–        the full text, with confidential information redacted where appropriate, of the notification of the aid scheme made by the Polish authorities on 6 December 2017 (see paragraph 5 above);

–        order the Commission to pay the costs.

27      The Commission and the interveners contend that the General Court should:

–        dismiss the action;

–        order Tempus to pay the costs.

III. Law

A.      Admissibility

28      The Commission does not call into question the admissibility of the action as such.

29      By contrast, the Republic of Poland, PGE and Enspirion contend that the action is inadmissible for Tempus’s lack of standing to bring proceedings, which the General Court can examine of its own motion. PGE states that the case-law permits interveners to put forward arguments different from those of the Commission, on the condition that those arguments do not alter the framework of the dispute and that the intervention is still intended to support the form of order sought by that party. In addition, the Republic of Poland and PGE recall that, in the rejoinder, the Commission itself contends that Tempus is not seeking to preserve its procedural rights, but rather challenges the contested decision on the merits without having standing to do so.

30      The Republic of Poland, PGE and Enspirion are of the view that Tempus does not have the status of interested party within the meaning of Article 108(2) TFEU and Article 1(h) of Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the application of Article 108 [TFEU] (OJ 2015 L 248, p. 9). According to the Republic of Poland, a distinction must be made in that regard between the standing of the two applicant companies. First, it is apparent from Tempus Energy Germany’s memorandum of association that it was not created until 26 July 2018, that is to say, over five months after the adoption of the contested decision. That company therefore cannot claim to be seeking to defend, by the present action, its right to be heard in a formal investigation procedure. If, instead of adopting the contested decision, the Commission had decided to initiate that procedure, that company would not have been able to submit observations. Second, even though T Energy Sweden was created in 2017, at no stage of the national consultation did it submit observations or attempt to influence the final structure of the Polish capacity market, for example by contacting the Polish authorities or the Commission. Moreover, although the Commission published the contested decision on its website as early as 18 April 2018, that company did not bring the action until 14 March 2019, such that its assertion as to its willingness to defend its procedural rights is not credible. Enspirion states that Tempus offers its customers only technical solutions enabling the provision of DSR services by other entities, meaning that it is not a direct beneficiary of the capacity market in the strict sense and that it can benefit from it only indirectly through its contracting parties. Likewise, PGE observes that Tempus has merely noted, in a speculative and unsubstantiated manner, its plans to participate in that market via various interconnectors ‘in the next three years’. Tempus, however, has not incorporated a company in Poland or taken part in the general certification process or in the auctions held under the Polish capacity mechanism. On the contrary, 21 months have elapsed since the adoption of the contested decision and there is no concrete sign of any meaningful ‘plans’ on Tempus’s part showing its interest in becoming active on that market.

31      PGE concludes from this that Tempus has no interest in the annulment of the contested decision, for the purposes of the concept of interest in bringing proceedings, and that the action must be declared inadmissible for that same reason. It states that, without participating in the Polish capacity mechanism, Tempus cannot be considered a ‘direct competitor’ of the other beneficiaries of the aid scheme. According to PGE, in essence, in the absence of evidence of its alleged ‘plans’ to participate in the Polish capacity market, Tempus’s interest is purely general and hypothetical. At the very most, it could be affected only by those parts of the contested decision which relate to Swedish DSR operators – either those pertaining to the transitory and target solutions which provide for different roles for interconnectors – but the application fails to make a single relevant point in that regard. In any event, given that the target solution has not yet been finalised and so concerns a future legal situation, it contends that Tempus has not demonstrated that the prejudice to that situation is certain. Nor did Tempus make any submissions during the public consultations in July 2016 on the initial concept for the Polish capacity market, in contrast to numerous interested parties, including DSR operators. The same applies to the public consultations conducted by the Polish authorities in December 2016, during which the other DSR operators did not raise any concerns with regard to the Polish capacity market.

32      According to the Republic of Poland and PGE, Tempus’s situation is very different from that of the applicants in Case T‑793/14, which were present on the United Kingdom market and were actively involved in shaping the United Kingdom capacity market, such that their standing to bring proceedings was beyond doubt. In the case at hand, however, neither of the two applicant companies operates on the Polish electricity market. Not being an interested party, Tempus must therefore prove that the contested decision concerns it directly and individually within the meaning of Article 263(4) TFEU, which it has not done. PGE submits that there is no ‘direct causal link’ between Tempus’s interests and the contested decision, Tempus neither being present nor operating on the Polish capacity market and merely alleging unsubstantiated ‘plans’ to enter that market. That Tempus is not individually concerned by that decision is also confirmed by its failure to participate in the consultation procedure at domestic level and in the pre-notification and notification procedures at EU level. The concept of individual concern would have no legal significance and that remedy would be transformed into an ‘actio popularis’ if, in order to recognise standing to bring proceedings, it were sufficient that an applicant, like Tempus in this case, was only indirectly and potentially affected by an aid scheme and thus was only marginally concerned by the act at issue.

33      As far as admissibility is concerned, Tempus merely states in the application that, in accordance with established case-law, competitors have standing to challenge a failure to initiate the formal investigation procedure under Article 108(2) TFEU in breach of their procedural rights. Tempus is a ‘party concerned’ within the meaning of Article 108(2) TFEU and an ‘interested party’ within the meaning of Article 1(h) of Regulation 2015/1589 which can challenge a decision adopted under Article 4(3) of the same regulation by invoking a breach of its procedural rights under Article 108(2) TFEU and Article 6(1) of that regulation. That status of interested party is sufficient to base its standing to bring proceedings under the fourth paragraph of Article 263 TFEU, provided that it claims that its interests could be affected by the grant of the aid at issue and that it demonstrates, to the requisite legal standard, that the aid is likely to have a material impact on its situation.

34      Tempus states, in essence, that it is a potential beneficiary of the aid scheme and a direct and indirect competitor of the other – more privileged – beneficiaries of that scheme. The two applicant companies have plans to enter the Polish market in the next three years and will become a direct competitor in Poland. As a DSR operator active in Germany and Sweden, Tempus is already a potential beneficiary and a competitor of those beneficiaries and the grant of aid has a material impact on its situation, distorting the competitive relationship in question.

35      As a preliminary point, it should be recalled that, according to the fourth paragraph of Article 40 of the Statute of the Court of Justice of the European Union, applicable to the procedure before the General Court by virtue of Article 53 thereof, an application to intervene is limited to supporting the form of order sought by one of the principal parties to the dispute. In addition, under Article 142(3) of the Rules of Procedure, the intervener must accept the case as he finds it at the time of his intervention. Consequently, although those provisions do not preclude the intervener from advancing arguments which are new or differ from those of the party which he supports, those arguments must not alter the framework of the dispute (see, to that effect, judgment of 20 March 2013, Andersen v Commission, T‑92/11, not published, EU:T:2013:143, paragraph 31 and the case-law cited; see, also, to that effect, judgment of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraph 36 and the case-law cited).

36      It is true that the question of whether an intervener is entitled to plead the inadmissibility of the action where the main party has not done so and whether or not such a plea falls outside the scope of the dispute as determined by the form of order sought by the main party, has not yet been decided by the Court of Justice (judgments of 10 November 2016, DTS Distribuidora de Televisión Digital v Commission, C‑449/14 P, EU:C:2016:848, paragraph 121, and of 4 June 2020, Hungary v Commission, C‑456/18 P, EU:C:2020:421, paragraphs 22 to 24). However, given that admissibility is one of the absolute bars to proceedings, however, the General Court is, in any event, required to examine it of its own motion (see, to that effect, judgments of 9 June 2016, Magic Mountain Kletterhallen and Others v Commission, T‑162/13, not published, EU:T:2016:341, paragraph 38 and the case-law cited, and of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraph 36 and the case-law cited).

37      As regards standing to bring proceedings, within the meaning of the fourth paragraph of Article 263 TFEU, it must be stated that, contrary to what is argued by the Republic of Poland, PGE and Enspirion, Tempus is a ‘party concerned’ within the meaning of Article 108(2) TFEU or an ‘interested party’ within the meaning of Article 1(h) of Regulation 2015/1589. Thus, its action, including all the pleas and complaints put forward in support of it to demonstrate that the Commission should have harboured doubts or experienced serious difficulties obliging it to initiate the formal investigation procedure provided for in Article 108(2) TFEU, is admissible in that it seeks to safeguard the procedural rights which Tempus would have enjoyed under that provision (see, to that effect, judgments of 24 May 2011, Commission v Kronoply and Kronotex, C‑83/09 P, EU:C:2011:341, paragraphs 59 and 63 to 65 and the case-law cited; of 3 September 2020, Vereniging tot Behoud van Natuurmonumenten in Nederland and Others v Commission, C‑817/18 P, EU:C:2020:637, paragraph 81; and of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraphs 45, 46 and 49).

38      Article 1(h) of Regulation 2015/1589, after all, defines the concept of ‘interested party’, synonymous with that of ‘party concerned’ within the meaning of Article 108(2) TFEU, as relating, inter alia, to ‘any person, undertaking or association of undertakings whose interests might be affected by the granting of aid, in particular the beneficiary of the aid, competing undertakings and trade associations’. The use of the expression ‘in particular’ establishes that that provision contains merely a non-exhaustive list of persons that could be categorised as ‘interested parties’, with the result that that term covers an indeterminate group of persons (see, to that effect, judgments of 14 November 1984, Intermills v Commission, 323/82, EU:C:1984:345, paragraph 16; of 24 May 2011, Commission v Kronoply and Kronotex, C‑83/09 P, EU:C:2011:341, paragraph 63; and of 13 June 2019, Copebi, C‑505/18, EU:C:2019:500, paragraph 34).

39      Having regard to that definition, the EU Courts have interpreted ‘interested party’ broadly. Thus, it is apparent from the case-law that Article 1(h) of Regulation 2015/1589 does not rule out the possibility that an undertaking which is not a direct competitor of the beneficiary of the aid can be categorised as an interested party, provided that it demonstrates that its interests could be adversely affected by the grant of the aid, and that, for that purpose, it is sufficient that that undertaking establishes, to the requisite legal standard, that the aid is likely to have a specific effect on its situation (see, to that effect, judgment of 24 May 2011, Commission v Kronoply and Kronotex, C‑83/09 P, EU:C:2011:341, paragraphs 63 to 65 and the case-law cited). Similarly, a trade union can be categorised as ‘concerned’ within the meaning of Article 108(2) TFEU if it shows that its interests or those of its members might be affected by the granting of aid, provided that that trade union shows, to the requisite legal standard, that the aid is likely to have a specific effect on its situation or that of the members it represents (see, to that effect, judgment of 9 July 2009, 3F v Commission, C‑319/07 P, EU:C:2009:435, paragraph 33).

40      Accordingly, the argument of PGE and the Republic of Poland according to which Tempus does not warrant the status of interested party on the ground that it is not a ‘direct competitor’ present on the Polish capacity market or has not made sufficiently concrete plans for entering that market cannot succeed. Tempus has demonstrated, to the requisite legal standard, that its interests are liable to be affected by the aid scheme and that the grant both of the agreements and the capacity payments is likely to have a material impact on its situation. It has thus plausibly explained that it is at least a potential competitor on the Polish capacity market, in that it has the firm intention and an inherent ability to enter it in the near future and that the aid scheme raises barriers making that entry more difficult (see, as regards the concept of ‘potential competition’, judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraphs 36 to 58). In addition, Tempus’s status as an interested party is borne out by its status as an operator active on the adjacent German and Swedish electricity markets, which enables it, through interconnectors – or, in the case of the Swedish market, by means of a market coupling mechanism (see paragraph 9 above) – to participate in the Polish capacity market. That assessment is not invalidated by the fact that Tempus intervened neither in the national consultation procedures nor in the pre-notification and preliminary examination procedures before the Commission; in the context of the latter it does not have, in any event, as the Commission has rightly noted, its own procedural rights that would have enabled it to submit observations.

41      In any event, the mere fact that Tempus Energy Germany’s memorandum of association dates from 26 July 2018 – more than five months after the adoption of the contested decision, meaning that it could not necessarily have participated in a formal investigation procedure following a decision to initiate the procedure adopted on the same date – does not affect the admissibility of its action. Given that T Energy Sweden, with which it jointly brought the present action, holds the status of interested party and has standing to bring proceedings within the meaning of the case-law cited in paragraph 39 above, there is no need to examine Tempus Energy Germany’s standing to bring proceedings separately (see, to that effect, judgments of 9 June 2016, Magic Mountain Kletterhallen and Others v Commission, T‑162/13, not published, EU:T:2016:341, paragraphs 40 and 41 and the case-law cited, and of 20 September 2019, Le Port de Bruxelles and Région de Bruxelles-Capitale v Commission, T‑674/17, not published, EU:T:2019:651, paragraph 36).

42      Consequently, the action must be declared admissible.

B.      Substance

1.      Subject matter of the dispute and review of substantive legality

43      In support of its action, Tempus puts forward two pleas for annulment.

44      The first plea alleges a failure by the Commission to fulfil its obligation to initiate the formal investigation procedure and, therefore, breach of Tempus’s procedural rights, as an interested party, under Article 108(2) TFEU and Article 6(1) of Regulation 2015/1589. This plea is subdivided into two main parts, the second of which in particular contains several sections, subsections and complaints aimed at demonstrating the existence of serious difficulties, within the meaning of the case-law, or of doubts, within the meaning of Article 4(3) and (4) of Regulation 2015/1589, which the Commission ought to have had during its preliminary examination.

45      The second plea alleges breach by the Commission of its obligation to state reasons under the second paragraph of Article 296 TFEU.

46      As regards the first plea and the scope of the review of legality which the General Court is called upon to carry out in that regard, it should be recalled that Article 108(3) TFEU and Article 4 of Regulation 2015/1589 establish a preliminary examination stage for notified aid measures. On completion of that stage, the Commission is to make a finding either that the measure does not constitute aid or that it falls within the scope of Article 107(1) TFEU. In the latter case, it may be that the measure does not raise doubts as to its compatibility with the internal market; on the other hand, it is also possible that the measure may raise such doubts (see, to that effect, judgment of 24 May 2011, Commission v Kronoply and Kronotex, C‑83/09 P, EU:C:2011:341, paragraph 43).

47      When, after the preliminary examination stage, the Commission adopts a decision, as in the present case, whereby it finds that a State measure does not constitute aid that is incompatible with the internal market, it implicitly refuses to initiate the formal investigation procedure. That principle applies both where the decision is taken on the ground that the Commission considers that the aid is compatible with the internal market, under Article 4(3) of Regulation 2015/1589, namely ‘a decision not to raise objections’, and where it is of the opinion that the measure does not come within the scope of Article 107(1) TFEU and thus does not constitute State aid pursuant to Article 4(2) of the same regulation (see, to that effect, judgment of 19 June 2019, Ja zum Nürburgring v Commission, T‑373/15, EU:T:2019:432, paragraph 111 and the case-law cited; see, also, to that effect, judgment of 16 March 2021, Commission v Poland, C‑562/19 P, EU:C:2021:201, paragraph 50 and the case-law cited).

48      On the other hand, it is apparent from settled case-law that, if the Commission is unable to conclude, following an initial examination in the context of the procedure under Article 108(3) TFEU, that a State aid measure either is not ‘aid’ within the meaning of Article 107(1) TFEU or, if classified as aid, is compatible with the FEU Treaty, or where that procedure does not make it possible for it to overcome the serious difficulties involved in determining whether the measure in question is compatible with the common market, the Commission is under a duty to initiate the formal investigation procedure under Article 108(2) TFEU, without having any discretion in that regard. That obligation corresponds to the one enshrined in Article 4(4) of Regulation 2015/1589, pursuant to which the Commission is required to initiate the procedure under Article 108(2) TFEU where the measure at issue raises doubts as to its compatibility with the internal market (see, to that effect, judgment of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraph 57 and the case-law cited).

49      The concept of serious difficulties coincides with that of doubts (see judgment of 9 September 2020, Kerkosand v Commission, T‑745/17, EU:T:2020:400, paragraph 106 and the case-law cited) and is objective in nature. The existence of such difficulties must be sought not only in the circumstances in which the Commission’s decision was adopted at the end of the preliminary examination but also in the assessments upon which it has relied. It follows that the legality of a decision not to raise objections, based on Article 4(3) of Regulation 2015/1589, depends on the question whether the assessment of the information and elements which the Commission had or should have had at its disposal, during the preliminary examination stage of the notified measure, should have objectively raised doubts as to the compatibility of that measure with the internal market, given that such doubts must give rise to the initiation of a formal investigation procedure in which the interested parties referred to in Article 1(h) of the same regulation may participate (see, to that effect, judgments of 3 September 2020, Vereniging tot Behoud van Natuurmonumenten in Nederland and Others v Commission, C‑817/18 P, EU:C:2020:637, paragraphs 79 and 80 and the case-law cited; see, also, to that effect, judgment of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraph 58 and the case-law cited).

50      Indeed, the case-law has further made clear in that regard that the legality of such a decision is to be assessed in the light of the information available to the Commission when the decision was adopted, it being understood that the information ‘available’ to the Commission includes that which appeared to be relevant for the assessment to be carried out and which could have been obtained, upon request by the Commission, during the preliminary examination stage (see, to that effect, judgment of 20 September 2017, Commission v Frucona Košice, C‑300/16 P, EU:C:2017:706, paragraphs 70 and 71).

51      It is for the applicant to prove the existence of serious difficulties or doubts. It may do so by reference to a body of consistent evidence, in particular, by claiming and establishing that the examination carried out by the Commission during the preliminary examination procedure was insufficient or incomplete (see, to that effect, judgments of 3 September 2020, Vereniging tot Behoud van Natuurmonumenten in Nederland and Others v Commission, C‑817/18 P, EU:C:2020:637, paragraph 82 and the case-law cited, and of 20 June 2019, a&o hostel and hotel Berlin v Commission, T‑578/17, not published, EU:T:2019:437, paragraphs 59 and 60 and the case-law cited).

52      It is in the light of those principles of case-law that the merits of the first plea must be examined.

2.      First plea: failure by the Commission to fulfil its obligation to initiate the formal investigation procedure in accordance with Article 108(2) TFEU

53      According to Tempus, the Commission should have had doubts or serious difficulties when assessing the compatibility of the aid scheme with the internal market and should therefore have initiated the formal investigation procedure under Article 108(2) TFEU. In not doing so, it breached the procedural rights enjoyed by Tempus under that provision, by depriving it of the opportunity to submit comments that would have been liable to change its assessment of the Polish capacity market. Tempus maintains that those doubts arise in view of the existence of a body of consistent indications concerning, first, the conduct and the length of the administrative procedure (first part) and, second, the content of the contested decision (second part).

(a)    First part of the first plea: existence of doubts as to the conduct and length of the procedure

(1)    Summary of the main arguments of the parties

54      As a preliminary point, having regard to the fact that the majority of the contacts between the Commission and the Polish authorities took place during the pre-notification phase, Tempus requests the General Court to ‘order’ the Commission to produce a copy of the documents or minutes of the telephone calls mentioned in order to verify the exact subject matter of those exchanges and to ensure that they were limited to the measures necessary for the notification and did not extend to an assessment of the compatibility of the aid scheme with the internal market anticipating the preliminary examination to be carried out following the notification of the aid scheme of 6 December 2017. The Act was enacted on 8 December 2017 and entered into force on 18 January 2018 and the Commission adopted the contested decision on 7 February 2018, exactly two months after notification of the aid scheme.

55      According to Tempus, the length and conduct of the pre-notification discussions between the Polish authorities and the Commission, together with a body of consistent evidence, are an indication of the existence of doubts and exceed what is normally required for the adoption of a decision under Article 4(3) of Regulation 2015/1589. In particular, the purpose of the pre-notification phase is not to assess the compatibility with the internal market of a significant, complex and novel measure, such as that in the case at hand. In such a case, the procedure could not be completed after a preliminary examination, but required a thorough assessment, inter alia, with the aid of comments from interested parties in the context of the formal investigation procedure which the Commission circumvented in the present case.

56      First, Tempus considers that the size of the capacity volumes auctioned and the expected capacity payments – in the 2021 delivery year, a volume of more than 20 million kW for an expected annual payment of EUR 1 020 million – highlights the importance of the Polish capacity mechanism.

57      Second, the length of the pre-notification phase is in itself an indication of the complexity of the case and contrary to the requirements of paragraph 14 of the Code of Best Practice, applicable in the present case, and to paragraph 16 of the Code of Best Practices for the conduct of State aid control procedures, adopted by the Commission on 19 July 2018 (OJ 2018 C 253, p. 14). That pre-notification phase lasted more than a year, which indicates a particular degree of complexity demonstrating, together with other consistent evidence, the existence of doubts. The assessment of the characteristics of the aid scheme, in particular the mechanisms intended to facilitate cross-border access, to allow access for capacity providers and DSR providers, including foreign ones, and to set up an effective cost recovery methodology, requires information from the Member State concerned and from interested parties, given that those elements have a significant impact on the volume of State aid and a lasting effect on the opportunities of competitors with different business models. According to Tempus, the numerous amendments made by the Polish authorities to the draft act during the pre-notification phase not only are an indication that the Commission’s questions or comments cast doubt on the design of the Polish capacity market and its compatibility with the internal market, but also show the complexity of the case.

58      Tempus also recalls that, during pre-notification and the preliminary examination procedure, the Commission was still involved in proceedings before the General Court concerning the capacity market in the United Kingdom, which gave rise to the judgment of 15 November 2018, Tempus Energy and Tempus Energy Technology v Commission (T‑793/14, EU:T:2018:790). Thus, the Commission should have taken account of the fact that the Polish capacity market was designed on the basis of the United Kingdom model, such that the elements that were being questioned before the General Court, such as the possibility for DSR effectively to participate in that market, should have been subject to particular control by it in the present case. In addition, the General Court upheld the complaint against the United Kingdom capacity mechanism concerning the excessive length of the pre-notification phase and its misuse. For those reasons, the Commission should have had doubts and considered more carefully the compatibility of the Polish capacity market with the internal market. The likelihood of committing a misjudgement of the complexity of the Polish capacity market was all the greater given that the Commission was conducting parallel examinations of six different European capacity markets.

59      Third, according to Tempus, the high complexity of the aid scheme and the large amount of aid involved are sufficient to justify the necessity of initiating the formal investigation procedure. Several features of the Polish capacity market justify its being qualified as novel, notwithstanding the fact that the Commission had already launched a sector inquiry into capacity mechanisms in 2015 which also related to Poland (‘the sector inquiry’) and that it had prepared a preliminary report in that regard on 13 April 2016 (C(2016) 2107 final)), followed by a final report of 30 November 2016 (C(2016) 752 final)). The numerous adjustments made to the Polish capacity market shows that new difficulties had arisen which could not be taken into account in the context of the sector inquiry, given that that market was not yet sufficiently developed at that stage.

60      Fourth, that sector inquiry underlines the complexity of the examination of the Polish capacity market and the resulting risks for the internal energy market. However, it does not substitute a proper assessment by the Commission of capacity markets, including the Polish one, under Article 108(2) TFEU, an argument which the General Court already rejected in the judgment of 15 November 2018, Tempus Energy and Tempus Energy Technology v Commission (T‑793/14, EU:T:2018:790, paragraphs 99 and 100). Furthermore, like the situation which gave rise to that judgment, the pre-notification phase was relatively long in the present case. Nevertheless, following the notification of the aid scheme, the Commission adopted the contested decision precisely within the two-month period prescribed in Article 4 of Regulation 2015/1589. As the General Court held in paragraphs 92 and 108 of that judgment, the pre-notification phase must not be disproportionate to the time required for the adoption of the subsequent decision. Tempus recalls that, in paragraph 84 of the aforementioned judgment, the Court added that the various capacity providers had not been invited to submit their observations during that phase, which is also the case here. It is, however, unacceptable for pre-notification contacts to be used for assessing the compatibility of the aid scheme with the internal market.

61      In order to substantiate further its complaint alleging misuse of the pre-notification phase, Tempus requests the Court to order the Commission to provide the full text – redacted of confidential information as appropriate – of the notification submitted by the Polish authorities on 6 December 2017. In that regard, it recalls the similar approach in Case T‑793/14, in which the General Court had ordered a measure of inquiry in accordance with Article 92 of the Rules of Procedure, which allowed for the appropriate confidential treatment of the relevant documents, in particular in the case of Article 91(b) and Article 92(3) of the Rules of Procedure.

62      The Commission and the interveners dispute Tempus’s line of argument. Similarly, the Commission contends that the request for a measure of organisation of procedure pursuant to Article 89(3)(d) of the Rules of Procedure, ‘ordering’ it to produce certain documents, should be rejected.

(2)    Findings of the General Court

63      As a preliminary point, it should be noted that the line of argument developed by Tempus in the first part of its first plea is largely inspired by the considerations set out in paragraphs 78 to 115 of the judgment of 15 November 2018, Tempus Energy and Tempus Energy Technology v Commission (T‑793/14, EU:T:2018:790). It is true that, in paragraphs 90 and 91 of that judgment, the General Court found that the purpose of the pre-notification was not to assess the compatibility with the internal market of a significant, complex and novel measure and that the Commission must not confuse that – 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 proved necessary in order to allow it to gather all the information it had needed to assess the compatibility of the aid and collect, to that end, the observations of the interested parties.

64      It is not apparent from the foregoing, however, that the Commission is generally expected to refrain from any assessment, even a provisional one, of the compatibility of proposed aid in the pre-notification phase. That finding corresponds to the content of paragraphs 11, 12 and 16 of the Code of Best Practice, by the setting and publication of which the Commission imposed a limit on the exercise of its discretion as regards the organisation of its procedures (see, to that effect, judgment of 3 September 2020, Vereniging tot Behoud van Natuurmonumenten in Nederland and Others v Commission, C‑817/18 P, EU:C:2020:637, paragraph 100 and the case-law cited), according to which, in essence, the pre-notification phase is specifically intended to enable the services of the Commission and the Member State concerned, including in particularly novel or complex cases, to address key competition problems, carry out economic analysis and, where appropriate, obtain the external expertise required to demonstrate ‘the compatibility of a planned project with the internal market’. In that regard, the Republic of Poland and Enel X are right to submit that, during that phase, the Commission must necessarily be able to assess that information in order to determine whether, following formal notification, it is sufficient to enable it to carry out a full examination of the compatibility of the proposed aid with the internal market. This is all the more true since, otherwise, contrary to paragraph 16 of that code, at the end of the pre-notification phase, the Commission would not even be in a position to provide the national authorities with an informal and non-binding provisional assessment of the said proposed aid for that purpose.

65      Tempus acknowledges that it does not criticise the Commission for having carried out an excessively long examination of the notified aid scheme in the preliminary examination procedure initiated following its complete notification by the Polish authorities, which lasted only two months, as provided for in Article 4(5) of Regulation 2015/1589, but considers that the allegedly excessive length of the pre-notification phase is indicative of the existence of doubts or serious difficulties. It must be pointed out, however, that, even if the Commission cannot misuse the pre-notification phase in order to avoid the constraints, in particular temporal, of the preliminary examination procedure – or even circumvent it (see the case-law cited in paragraph 63 above, concerning a case in which the draft decision was already ready at the time of the notification) – in complex cases, exceptionally, pre-notification contacts are capable of lasting several months (see paragraph 14 of the Code of Best Practice).

66      The General Court considers that, in the case at hand, the duration of a pre-notification phase of approximately one year cannot be considered indicative of the existence of such misuse or of a circumvention, nor can that duration constitute an indication of doubts, even if it is a complex case. Paradoxically, Tempus itself considers that the aid scheme was of such complexity that it necessitated a thorough examination in the context of a formal investigation procedure. However, in accordance with paragraph 14 of the Code of Best Practice, in such a complex case, the Commission could legitimately exceed the indicative period of two months and continue the pre-notification phase for ‘several months’ in order to ensure that the Member State submits a complete notification in order to enable it to carry out its preliminary examination in full knowledge of the facts. In the present case, as the Commission notes, when it was pre-notified, the draft law was not yet final and still had to pass several stages in the legislative process in Poland, including a public consultation, to be finally adopted by the Polish Parliament on 8 December 2017, that is to say, only two days after its formal notification. Thus, in order to comply with the State aid rules and following the recommendations set out in paragraphs 10 to 18 of the Code of Best Practice, the Polish authorities had initiated pre-notification contacts with the Commission at an early stage of the internal decision-making process in order to be able to take account of the provisional positions of that institution throughout that process and to ensure that the project ultimately notified was likely to meet the criteria for compatibility under Article 107(3)(c) TFEU and the Guidelines on State aid for environmental protection and energy 2014-2020 (OJ 2014 C 200, p. 1; ‘the Guidelines’). Therefore, without other elements indicating misuse, which are absent in this case (see paragraphs 67 and 68 below), such a collaborative approach on the part of the Polish authorities and the Commission, based on the principle of sincere cooperation, under Article 4(3) TEU, cannot in itself be regarded as giving rise to doubts or serious difficulties.

67      In addition, the Commission and the interveners rightly submit that, unlike the procedure relating to the United Kingdom capacity market that was the subject of the judgment of 15 November 2018, Tempus Energy and Tempus Energy Technology v Commission (T‑793/14, EU:T:2018:790, paragraphs 101 to 105), in the present case there is no evidence that, during the procedures both at national level and before the Commission, in particular during the public consultation of the draft law initiated by the Polish authorities following its pre-notification, the proposed aid scheme was challenged by interested parties and, in particular, by DSR operators. Tempus itself does not claim to have participated in that consultation or to have submitted observations or complaints to the Commission, unlike the applicants’ approach in that other case concerning the United Kingdom capacity market, as members of the UK Demand Response Association (UKDRA).

68      In that regard, Tempus is not justified in invoking the absence of procedural guarantees on the part of the parties concerned during the pre-notification and preliminary examination stages, since a prudent and diligent operator wishing to enter a national electricity market which is subject to significant reform is supposed to take, like the applicants in the case concerning the United Kingdom capacity market, all the steps necessary to defend its commercial interests before the competent authorities. Thus, in its preliminary examination, the Commission was entitled to rely not only on the results of its sector inquiry into the capacity markets of 11 Member States which already included Poland, but also, in the absence of detailed objections, to rely on the results of the public consultation which had given rise to a multitude of comments from interested parties, including Polish DSR operators such as Enel X and Enspirion. In addition to this, Tempus had experience in assessing the United Kingdom capacity market, the policy design of which bears, according to Tempus’s own statements, certain similarities to the Polish capacity market.

69      Nor is Tempus justified in arguing that the long pre-notification phase, during which the Polish authorities made numerous amendments to the draft act at the Commission’s instigation, is in itself evidence of the complexity of the Polish capacity market and, therefore, an indication of doubts. That line of argument is not capable of invalidating the plausibility of the conclusion according to which those elements demonstrate, on the contrary, that, following those amendments, introduced partly on the basis of commitments offered by the Polish authorities to the Commission (see recitals 16, 43, 73, 79, 82 to 87, 138 and 165 of the contested decision), that institution no longer had doubts as to the compatibility of the aid scheme with the internal market (see the conclusion under Section 4 of that decision), and must therefore be rejected as ineffective. That assessment does not, however, prejudge the possibly complex nature of the examination of the various elements of the Polish capacity market which are the subject of the second part of the present plea.

70      The same applies to Tempus’s argument that the Commission’s sector inquiry into national capacity markets, finalised in November 2016 – long before the adoption of the contested decision – is not evidence of the lack of complexity of the aid scheme. Even assuming that, in the meantime, the policy design of the Polish capacity market had changed significantly, that argument is not in itself capable of demonstrating that, due solely to the duration of the administrative procedure and of its process, the Commission must have had doubts or serious difficulties when assessing its compatibility with the internal market. Furthermore, Tempus’s argument that the Commission should have been more careful in view of the proceedings pending before the General Court in Case T‑793/14 concerning its decision on the United Kingdom capacity market cannot succeed, since, as has been stated in paragraphs 67 and 68 above, the procedural situations in those cases were not comparable, in particular due to the explicit challenge to the policy design of that market by DSR operators.

71      Finally, contrary to what Tempus maintains, the scale or volume of the aid to be granted under the aid scheme cannot in itself be deemed to be indicative of doubts or serious difficulties. In that regard, the Commission rightly contends that, as is also apparent from paragraphs 10 to 18 of the Code of Best Practice, even proposed aid of a certain size, complexity or novelty must, in principle, be able to receive the same treatment as other, less significant aid proposals, the provisions of the FEU Treaty, Regulation 2015/1589 and that code making no distinction in that regard.

72      Consequently, the first part of the first plea, alleging the existence of doubts concerning the conduct and the length of the procedure, must be rejected as unfounded.

73      As regards Tempus’s requests for a measure of organisation of procedure or of inquiry, it is sufficient to state that the General Court considers itself to be sufficiently informed by the elements in the file to rule on the first part of the first plea and that there is therefore no need to allow Tempus to seek in the documents requested additional evidence which, in any event, is not likely to have an impact on the General Court’s assessment.

74      Those requests must therefore also be rejected.

(b)    Second part of the first plea: existence of doubts concerning the content of the contested decision

(1)    Preliminary observations

(i)    The alleged doubts or serious difficulties in the light of the provisions of the Guidelines

75      In the second part of the first plea, Tempus argues, in essence, that, as demonstrated by a comparison of the grounds of the contested decision with the information available on the Polish capacity market, the Commission should have had doubts or serious difficulties as to the compatibility of the aid scheme with the internal market in the light of Article 107(3) TFEU, read in conjunction with the relevant provisions, in particular Section 3.9 of the Guidelines. It did not, however, research and examine, thoroughly and impartially, all the relevant information so as to eliminate all those doubts. In particular, the Commission’s assessment is insufficient and incomplete with respect to (i) the objective of common interest and the need for State intervention (first section); (ii) the appropriateness of the aid scheme (second section); (iii) the incentive effect (third section); (iv) the proportionality of the aid (fourth section) and (v) the avoidance of undue negative effects on competition and trade between Member States (fifth section).

(ii) The legal nature of the Guidelines and the scope of the review of legality by the EU Courts in that regard

76      So far as concerns the legal nature of the Guidelines and the scope of the review of legality that the EU Courts are called upon to exercise in the light of their provisions, it should be recalled that, in adopting such rules of conduct and announcing them by publishing them that they will henceforth apply to the cases to which they relate, the Commission imposes a limit on the exercise of its aforementioned discretion and, in principle, cannot depart from those rules without being found, where appropriate, to be in breach of general principles of law, such as equal treatment or the protection of legitimate expectations (see judgment of 3 September 2020, Vereniging tot Behoud van Natuurmonumenten in Nederland and Others v Commission, C‑817/18 P, EU:C:2020:637, paragraph 100 and the case-law cited).

77      It is in the light of those principles of case-law that the various sections of the second part of the first plea must be examined.

(2)    First section: alleged incompleteness of the assessment of the objective of common interest and the need for State intervention

(i)    First subsection: the objective of common interest

78      Tempus submits that a number of aspects should have given rise to doubts on the part of the Commission in the light of paragraph (220) of the Guidelines. It states that it does not contest the objective of common interest pursued by the Polish capacity mechanism, namely that of securing sufficient supplies of electricity to final consumers in Poland. However, the general objective of the Guidelines is ‘to ensure a competitive, sustainable and secure energy system in a well-functioning Union energy market’ (paragraph (30)) and they recognise that ‘aid for generation adequacy may contradict the objective of phasing out environmentally harmful subsidies including fossil fuels’ (paragraph (220)). It follows that the Member States are required not to view generation adequacy as an isolated objective of common interest, but as an objective in the greater context of a general aim to ‘support the shift towards a resource-efficient, competitive low-carbon economy’ (paragraph (30)). That reading also applies in the light of Article 194(1) TFEU, which lists the functioning of the energy market, energy efficiency, the development of new and renewable forms of energy and the promotion of interconnection as aims pursued by the European Union, alongside security of supply.

–       The first complaint

79      By its first complaint, Tempus submits that, in its brief assessment, set out in recitals 138 and 163 of the contested decision, of the potential of DSR in the Polish electricity market, the Commission applied the wrong test. It merely relied on the commitments offered by the Polish authorities to facilitate the development of DSR, first, by ensuring that DSR would be eligible to participate in the wholesale electricity markets and in the balancing market in 2021 and, second, by adjusting the rules for DSR in the draft law or after its implementation. However, the Commission failed to assess the real potential of DSR in the Polish energy-only market, as is required in paragraph (220) of the Guidelines, and whether the design of the Polish capacity market could prevent it from achieving that potential.

80      Thus, the Commission did not attempt to assess the expected growth of DSR in the light of the planned changes, the impact of the capacity market on it and, above all, the risk of its being excluded in practice from most of the revenues from that market. On that subject, it merely mentioned the measure referred to in recital 16(f) of the contested decision, intended to ensure ‘direct DSR customer access to [the] wholesale market from 2021’. However, the Commission failed to consider the impact of the ‘other measures’ on DSR potential, with or without the proposed capacity market. In particular, Tempus calls into question the basis for the statement in footnote 31 to the contested decision, according to which ‘the estimates show that the potential for DSR in the Polish [electricity] market is between [1 200 MW and 2 500 MW]’, having regard to the fact that the Commission did not assess, first, its actual potential given the full suite of energy-only market measures proposed plus other enabling changes that could have been proposed but were not, and, second, the extent to which the capacity market was likely to reduce or enhance that potential, based on the real world impact of its policy design. According to Tempus, the lack of a sufficiently profound analysis of DSR potential is confirmed by the fact that recital 138 of the contested decision devotes only around eight lines to it.

81      Tempus disputes that the focus of the Polish capacity mechanism must be on generation capacity. The Commission did not properly assess the role of DSR on the Polish energy-only market and its potential in contributing to the objective of security of supply. In order to secure sufficient energy supplies, it is necessary to quantify the adequacy issue. Since it did not follow the requirements of paragraph (220) of the Guidelines in its examination, the Commission could not accurately quantify the difference in generation capacity which that mechanism aimed to reduce. Tempus emphasises the expression ‘primarily’ used in that paragraph, which shows that the first task of a Member State is to assess properly the potential of DSR and increased interconnection capacity, before it is possible to assess accurately how much new generation capacity a market actually needs. In addition to that error, the Commission erred in considering that DSR was only ‘required for a limited number of hours during the year’, for example in stress events, which shows its misunderstanding of the very concept of DSR, which avoids not only the building of new generation capacity but also reduces the need for supplying electricity to the network by existing generating CMUs. It allows for a more efficient use of existing generation and network infrastructure, encouraging customers to change their demand habits better to match times of higher renewable energy generation. According to Tempus, especially when combined with storage, DSR could exponentially increase the volume of intermittent renewable generation and assist the fundamental changes in the energy supply system. Even the Commission acknowledged that DSR could make a valuable contribution to resource adequacy (Final Report of the Sector Inquiry, paragraphs 275, 276 and 282). It should have had doubts as to whether the Polish authorities had assessed all pathways of achieving generation adequacy – meaning DSR, interconnection capacity and increased generation capacity – and how those pathways interact to achieve resource adequacy as well as an adjustment of the energy system.

82      The Commission, supported by the interveners, contends, in essence, that it correctly identified the objective of common interest pursued by the Polish capacity market and complied with paragraph (220) of the Guidelines.

83      By the first complaint, Tempus disputes, inter alia, the considerations set out in recitals 138 and 163 of the contested decision.

84      It is apparent, in essence, from recitals 137 and 138 of the contested decision that, having regard to paragraph (220) of the Guidelines, the Act is intended to facilitate DSR and reinforce the grid, and that, currently, DSR is developed only to a minimum extent in Poland and therefore cannot meet the reliability standard at issue. The Commission nevertheless acknowledges that the Polish authorities committed to facilitating the development of DSR in Poland, as demonstrated by the commitment described in recital 16(f) of that decision and by the adjusted certification rules within the proposed capacity mechanism. It considers those steps to be likely to intensify the development of DSR up to its full potential, namely, according to the estimates, between 1 200 MW and 2 500 MW on the Polish electricity market. In addition, as regards the increase of interconnection capacity, the Commission takes note of PSE’s grid reinforcement programme, which runs until 2025 and concerns, in particular, the 400-kilovolt network in the south-west of Poland. The Polish authorities explained that that programme was intended to solve bottleneck issues and enable the net transfer capacity of interconnectors on Poland’s synchronous profile to be increased.

85      Furthermore, it is stated in recital 139 of the contested decision that those measures alone are not sufficient to achieve the reliability standard at issue, as is apparent from the conclusions of the adequacy assessment described in recitals 10 to 14 of that decision, such that, as is shown in Section 3.3.2 thereof, aid for generation adequacy appears necessary. The Commission notes that the Polish capacity mechanism is technologically neutral and open to all capacity providers and may therefore involve capacity payments to all those providers, including to conventional plants generating from fossil fuels such as coal. Against that background, it states, inter alia, that the Polish authorities have introduced a number of design features to that mechanism in order to ensure compliance with the objective of phasing out environmentally harmful subsidies. The Green Bonus, in particular, will enable CMUs emitting less than 450 kg of CO2/MWh to have access to cheaper financing and bid lower prices in capacity auctions.

86      In recital 140 of the contested decision, the Commission concludes from the foregoing that, bearing in mind that the Polish capacity market is a technology neutral scheme open to all potential capacity providers, including conventional generation based on fossil fuels such as coal to a great extent, several design features and actions aiming at alternative ways for achieving generation adequacy were provided by the Polish authorities.

87      In recital 163 of the contested decision, it is essentially recalled that, as regards technological neutrality, all types of capacity may participate in the Polish capacity market, including DSR. The Commission considers the eligibility rules in that respect to be appropriate for ensuring a level playing field between the various potential capacity providers in the capacity market. Regarding DSR in particular, it underlines that, as is described in Section 2.3.3, the Polish authorities adapted the certification rules to overcome the obstacles which DSR operators would have faced considering their specific characteristics and the fact that that industry is still in its infancy in Poland.

88      Tempus bases its challenge on the main premiss according to which the objective of common interest of generation adequacy, as set out in paragraph (220) of the Guidelines, is not an isolated objective, but an objective in the wider context of a general aim, namely that of supporting the ‘shift towards a resource-efficient, competitive low-carbon economy’, provided for in paragraph (30) of the Guidelines, which corresponds to the European Union’s objectives referred to in Article 194(1) TFEU, including the functioning of the energy market, energy efficiency, the development of new and renewable forms of energy and the promotion of interconnection, which should go alongside security of energy supply.

89      In that regard, Tempus does argue, rightly, that the objectives of common interest which environmental aid is meant to pursue are apparent from a combined reading of paragraphs (30) and (220) of the Guidelines.

90      After all, in the ‘general compatibility provisions’, under the subheading ‘General conditions’ of the heading ‘Contribution to an objective of common interest’, paragraph (30) of the Guidelines recognises a ‘general objective’ of environmental aid, including in the energy sector, which is to ‘increase the level of environmental protection compared to the level that would be achieved in the absence of the aid’. In that regard, reference is made to the Europe 2020 strategy, which sets ‘targets and objectives for sustainable growth to support the shift towards a resource-efficient, competitive low-carbon economy’. It specifies the primary objective of aid in the energy sector, which is to ensure ‘a competitive, sustainable and secure energy system in a well-functioning Union energy market’. Under paragraph (31) of the Guidelines, Member States intending to grant environmental or energy aid have to ‘define precisely the objective pursued and explain what is the expected contribution of the measure towards th[at] objective’. That obligation to specify objectives is recalled in paragraph (221) of the Guidelines as regards aid for generation adequacy.

91      Having regard to the more general definitions set out above, paragraphs (219) and (220) of the Guidelines specify the content of the objective of common interest that aid for generation adequacy, such as that in the present case, is intended to pursue. Paragraph| (219) recognises that such aid ‘can pursue different objectives’ and may aim at addressing ‘short-term concerns brought about by the lack of flexible generation capacity to meet sudden swings in variable wind and solar production’ or the definition of ‘a target for generation adequacy, which Member States may wish to ensure regardless of short-term considerations’. That in itself indicates that the Member States have a certain discretion in defining those sub-objectives that are deemed to equate to an objective of common interest.

92      That discretion of the Member States in the defining those sub-objectives and in balancing them is confirmed in paragraph (220) of the Guidelines, which recognises that aid for generation adequacy ‘may contradict the objective of phasing out environmentally harmful subsidies including for fossil fuels’ and therefore derogate from the general objective of common interest referred to in paragraph (30) of the Guidelines to ‘increase the level of environmental protection’. Its exercise is however limited by the recommendation, also mentioned in paragraph (220) of the Guidelines, according to which ‘Member States should therefore primarily consider alternative ways of achieving generation adequacy which do not have a negative impact on the objective of phasing out environmentally or economically harmful subsidies, such as facilitating demand side management and increasing interconnection capacity’.

93      This means that Member States are supposed to balance the potentially conflicting objectives of security of energy supply against environmental protection, all the while observing the principle of proportionality in the strict sense, with the aim of reducing the environmental impact of aid to the strict, necessary and acceptable minimum. Moreover, that requirement to balance these objectives is perfectly compatible with, on the one hand, the objectives – also potentially divergent – set out in Article 194(1) TFEU, which aim both to ensure the functioning of the energy market and security of energy supply in the European Union and to promote energy efficiency, energy saving, the development of new and renewable energy sources and the interconnection of energy networks, and, on the other hand, the requirements of the principle of proportionality, as set out in Article 5(4) TEU. Thus, that requirement is recalled, under the heading ‘Appropriateness of the aid’, in paragraphs (42) and (43) of the Guidelines, according to which, inter alia, ‘a measure addressing a generation adequacy problem needs to be balanced with the environmental objective of phasing out environmentally or economically harmful subsidies, including for fossil fuels’.

94      It follows that, in principle, Tempus is correct to note that, under paragraph (220) of the Guidelines, when a Member State introduces a capacity mechanism, it is supposed to take account of the objective of common interest of environmental protection by refraining from frustrating that objective by unilaterally favouring generation capacities using fossil fuels, and by promoting it, inter alia, by ‘facilitating demand side management’.

95      It must however be stated that, regard being had to the discretion of the Member State recalled in paragraphs 91 and 92 above, including concerning its choice between various energy sources and the general structure of its energy supply in order to guarantee its security (see, to that effect, judgment of 22 September 2020, Austria v Commission, C‑594/18 P, EU:C:2020:742, paragraph 48 and the case-law cited), a clear and precise obligation as to the ways in which DSR potential should be assessed or promoted, however, follows neither for that Member State (see, to that effect and by analogy, judgment of 19 July 2016, Kotnik and Others, C‑526/14, EU:C:2016:570, paragraph 44) nor for the Commission. Similarly, as the Commission and the interveners note, that provision cannot be interpreted as prohibiting aid measures for conventional power plants, including fossil fuel plants, where these prove necessary to guarantee generation adequacy and therefore the security of energy supply, or as requiring them to give absolute priority to alternative techniques, such as DSR or interconnection capacities.

96      Tempus does not, however, claim that, in the present case, by enacting the Act, the Polish legislature failed to exercise that discretion or balance the potentially divergent objectives recalled in paragraphs 93 and 94 above, but merely claims that the assessment set out in recitals 138 and 163 of the contested decision is based on the ‘worst test’ in that it failed to assess, as is allegedly required by paragraph (220) of the Guidelines, the actual potential of DSR in the ‘energy-only’ Polish market, that is to say, in the absence of the capacity market.

97      However, such a requirement of a counterfactual examination of DSR potential cannot be inferred either from that paragraph or from any other relevant provision of the Guidelines. Even though paragraph (30) thereof states that environmental aid is supposed to ‘increase the level of environmental protection compared to the level that would be achieved in the absence of the aid’, it does not follow that there is a requirement to quantify precisely the level of protection, inter alia, by DSR as a technique for economic and efficient use of energy and, therefore, environmental protection. Nor, contrary to what Tempus argues, is such a requirement apparent from the wording of the second sentence of paragraph (220) of the Guidelines, which provides that ‘Member States should … primarily consider alternative ways of achieving generation adequacy which do not have a negative impact on the objective of phasing out environmentally or economically harmful subsidies’. That sentence contains only a requirement addressed to the Member States to balance potentially divergent objectives, recalled in paragraphs 93 and 94 above, in the context of which it is recommended that such subsidies should no longer be promoted and that instead recourse should be had to support measures intended, inter alia, for facilitating DSR and increasing interconnection capacity. Indeed, it is only in the context of the examination of the proportionality of the aid as such that paragraphs (69) and (70) of the Guidelines provide for a counterfactual scenario, such as that invoked by Tempus.

98      Therefore, in the case at hand, it was sufficient for the Commission to assess the question of whether the aid scheme was likely to facilitate DSR on the basis of the information which, at the time of the adoption of the contested decision, it had in its possession relating to the factual and legal situation of DSR operators, their potential for development and their prospective development on the Polish electricity or capacity market. It is on that basis that the Commission was entitled to take into consideration DSR potential in order to assess its likely growth prospects and to make sure that it was not discriminated against compared to other conventional capacity providers (see the examination of the second complaint below), without it having to entertain any doubts in that regard.

99      Consequently, the first complaint must be rejected as unfounded.

–       The second complaint

100    In the second complaint, Tempus complains that the Commission committed an error of assessment in assuming that the Polish capacity market facilitated the development of DSR. In support of that complaint, it calls into question the considerations, including the reference to the commitments offered by the Polish authorities (set out in recital 16(f) of the contested decision), and the estimates of growth of the DSR sector between 1 200 MW and 2 500 MW, as set out in recital 138 and in footnote 31 to the contested decision.

101    In that regard, Tempus claims, in essence, that, over the last five years, the Polish electricity market has given rise every year to approximately 95 ‘extreme’ prices compared to the short run marginal costs of power plants. Given a rise in electricity demand of 9% in Poland since 2014 with an upward trend, there is thus a clear incentive for DSR. According to two alternative scenarios for 2020, based on the Power2Sim model designed by Energy Brainpool, known as ‘BAU’ (Business as Usual) and ‘+ 1 GW’, the effect of the introduction of the capacity market would be that the number of extreme prices fell to 82 and 46, respectively, which would therefore strongly affect the incentive for DSR. Tempus also criticises the Commission for not requiring the Polish authorities to increase interconnection capacity.

102    The Commission and the interveners contend, in essence, that DSR potential must be assessed in relation to the reliability standard at issue, that is to say in relation to the need to ensure a level of system security of 99.97%. PSE’s adequacy assessment anticipated a generating capacity shortfall of 2 750 MW in 2020 and 8 068 MW in 2025 (recitals 12 and 147 of the contested decision). Having regard to the planned decommissioning of generation capacity, namely 2 960 MW for coal and 413 MW for lignite in 2020, and 5 822 MW for coal and 600 MW for lignite in 2025, which cannot be offset either by import capacity or by DSR, and to the general lack of revenue (recitals 8 and 9 of the contested decision), the Polish capacity market should also promote generation capacity. The Commission relies, in particular, on PSE’s adequacy assessment, as audited by a consulting firm, which it says is the basis for its own assessment set out in recitals 138 and 139 of the contested decision, pointing to very high LoLE values greatly exceeding the reliability standard at issue in 2020, 2025 and 2030 in four different scenarios. In its capacity adequacy assessment, after having launched public tenders for two DSR programmes in 2016, PSE therefore rightly estimated the DSR potential, before the implementation of the Polish capacity market, of providing capacity at 200 MW yearly during the 2017-2040 period, as well as its maximum potential beyond 2020 as being between 1 200 and 2 500 MW. According to PSE, new DSR capacity would enter the market only if the net revenues for new entrants exceeded at least EUR 30/kW, that is to say if the fixed costs of new DSR were recovered by the net energy revenue. Similarly, that consulting firm’s modelling showed that further DSR development would not be triggered, as the revenue level would not be sufficient for new DSR providers to be profitable on the market. The interveners Enspirion and Enel X, two Polish DSR operators, share that line of argument and highlight the nature of those operators as ‘price makers’ as well as the fact that, in optimal conditions, the capacity reserves provided by DSR can eventually reach around 10% of Poland’s peak demand, that is to say around 2 600 MW. The Commission, supported by Enel X, Enspirion and PGE, disputes the argument of Tempus which is based on ‘extreme prices’ in the Polish electricity market, those prices not having led to a greater development of DSR in the past, which is illustrated by the results of the auctions in which Enel X and Enspirion had participated. Last, the Commission, supported by the Republic of Poland, recalls that PSE’s adequacy assessment also concluded that there was insufficient interconnection capacity, which was unable, either alone or in combination with DSR, to cover the supply gap, compared with the estimates for the decommissioned capacity of old hard coal and lignite installations (3 373 MW in 2020, 6 422 MW in 2025 and 11 653 MW in 2030), and to alleviate significantly resource adequacy concerns in Poland. It states that the succinct nature of the reasons set out in that regard in the contested decision in fact shows that there were no doubts in that regard, since DSR and interconnection clearly could not provide the required capacity.

103    The General Court considers that it is without having any doubts or serious difficulties that the Commission was able to rely on estimates of the growth of the DSR sector between 1 200 MW and 2 500 MW to conclude that the measures proposed by the Polish authorities were such as to facilitate the development of DSR in Poland, in accordance with the objective referred to in paragraph (220) of the Guidelines.

104    Indeed, the very detailed facts and evidence submitted by the Commission and by the interveners – including the two largest Polish DSR operators – establish to the requisite legal standard the solid evidentiary basis underpinning the considerations set out in recitals 138 and 139 of the contested decision, such as to dispel any doubt as to the true development potential of DSR on the Polish electricity or capacity market. Those elements are based, inter alia, on PSE’s adequacy assessment, as audited independently by a consulting firm, and on the experience and expertise which those DSR operators were able to gain on the Polish electricity market before and after the entry into force of the Act. In particular, the demonstrated growth of those operators and the energy capacity acquired by them in the past, in auctions of the Polish capacity mechanism and in previous programmes, credibly demonstrate the potential of that sector, which is precisely limited by the figures on which the Commission had relied, namely around 200 MW per year.

105    Aside from the fact that Tempus relies mainly on methodological aspects from the Energy Brainpool study, it does not dispute the detailed figures for the capacity lots that Enspirion and Enel X were able to obtain, before and after the adoption of the contested decision, during the auctions for the relevant supply periods, or the fact that DSR alone is not capable of contributing substantially to the objective of achieving the reliability standard at issue. Nor does Tempus dispute the relevance of that reliability standard, or the extent of the loss of capacity until 2020 and 2025, respectively, which will be due to the decommissioning of fossil fuel plants. It merely calls into question the estimates of DSR potential of between 1 200 MW and 2 500 MW that PSE had made, starting from the 200 MW volume allocated to DSR operators in 2016, as part of its adequacy assessment on which the Commission relied in footnote 31 to recital 138 of the contested decision. When drafting the application, however, Tempus clearly still did not know that those figures were the result of the application of the standardised and widely recognised ENTSO-E methodology – including in the Commission’s final report of the sector inquiry on capacity mechanisms of 30 November 2016 (pp. 9 and 10) – and that those figures had been audited, independently, and had been broadly endorsed by a consulting firm. Thus, in the reply, Tempus has not been able to challenge, in a substantiated manner, those plausible elements, even on the basis of the methodology proposed in Energy Brainpool’s study.

106    Therefore, in merely proposing an alternative modelling, used in the Energy Brainpool study, which simply extrapolates ‘extreme prices’, linked to an estimate of an increase in electricity demand, Tempus has failed to establish that the Commission should have entertained doubts or serious difficulties as to the true potential of DSR in Poland. Even assuming that that modelling were capable of demonstrating, indirectly, an incentive effect for the use of DSR, it cannot undermine the plausibility of the figures on which the contested decision is based, which are supposed to represent immediately the potential of DSR to meet electricity demand, or the probative value of the methodology applied to arrive at those figures.

107    Similarly, it is in an unsubstantiated manner that Tempus criticises the Commission for not having required the Polish authorities to take measures to promote interconnection capacity. In that regard, it must be recalled that, in recital 138 of the contested decision, the Commission took note of PSE’s grid reinforcement programme running until 2025, which is intended to solve bottleneck issues and enable the net transfer capacity of interconnectors on Poland’s synchronous profile. In those circumstances, Tempus has failed to put forward sufficiently precise and concrete evidence capable of establishing that the Commission should have had doubts, in the light of paragraph (220) of the Guidelines, as to whether the objective of increasing interconnection capacity had been achieved.

108    Consequently, the second complaint must also be rejected as unfounded.

–       The third complaint

109    By the third complaint, Tempus criticises the Commission, in essence, for having accepted, contrary to the requirements of paragraphs (3), (220) and paragraph (233)(e) of the Guidelines, that the Polish capacity mechanism mainly subsidises fossil fuel power plants and carbon-intensive technologies at the expense of more environmentally friendly generation and supply technologies. In support of its complaint, it presents the results of the auctions that granted substantial capacities to such power plants and technologies. In addition, it disputes that the Green Bonus is sufficient to promote low-carbon technologies and to contribute to the elimination of coal and lignite generation capacity. On the contrary, that bonus favours, inter alia, CMUs burning natural gas, a fossil fuel harmful to the environment due to its greenhouse gas emissions, and discriminates against DSR operators.

110    The Commission and the interveners contend, in essence, that the results of auctions subsequent to the adoption of the contested decision cannot be taken into account for the review of the legality of that decision. In any event, Tempus has not established that the awarding of capacity agreements to generating CMUs using fossil fuels is to the detriment of DSR operators showing growth in the same auctions, namely, for Enel X, the allocation of capacity volumes of 614.6 MW (2.74%), 761 MW (7.19%) and 791 MW (7.44%), respectively, and for Enspirion, the allocation of capacity volumes of 152 MW, 202 MW and 237 MW, respectively, which corresponds, inter alia, to the development forecasts in PSE’s adequacy assessment. In that regard, the Commission and the interveners recall that the Guidelines do not prohibit support for conventional power plants, which continue to play an important role in the Polish energy mix, as long as their generation capacity is required to ensure capacity adequacy and security of supply in accordance with the reliability standard at issue. They also deny that the Green Bonus discriminates against DSR, as that bonus creates specific economic incentives for low-emission generation capacity, which cannot be transferred to DSR operators.

111    It is appropriate to recall the considerations set out in paragraphs 92 to 95 above, according to which paragraph (220) of the Guidelines means that a Member State which introduces a capacity mechanism must take account of the objective of common interest of environmental protection and promote it, inter alia, ‘by facilitating demand side management’, which presupposes, a contrario, that it will refrain from frustrating that objective by unilaterally favouring generation capacities using fossil fuels, in accordance with the objective of phasing out environmentally harmful subsidies. Contrary to what Tempus maintains and as is set out in paragraph 95 above, however, those objectives cannot be understood as prohibiting aid measures for fossil fuel power plants, where those measures prove to be necessary to the guarantee of generation adequacy and therefore of security of energy supply, or as requiring the Member State and the Commission to give absolute priority to alternative and more environmentally friendly techniques, such as DSR. Furthermore, as is apparent from the positions taken by Enel X and Enspirion (see paragraph 110 above), since the introduction of the Polish capacity mechanism, Polish DSR operators have shown a net growth in capacity obtention which corresponds to the development forecasts underlying both PSE’s adequacy assessment as audited and the contested decision (see paragraphs 104 and 105 above), which tends to confirm that that mechanism enables them to exhaust their potential.

112    Thus, the fact that the Polish capacity mechanism entails the awarding to fossil fuel-based generating CMUs of capacity agreements and payments, even of substantial volume, cannot in itself be characterised as contrary to the objectives referred to in paragraph (220) of the Guidelines, as long as the award of that capacity appears necessary in the light of the requirements identified in PSE’s adequacy assessment and the reliability standard at issue, as a precise and clearly defined objective within the meaning of paragraph (221) of the Guidelines. That latter aspect has not been specifically called into question by Tempus in that precise context, however, and will be assessed in the assessment of the complaints relating to the necessity and proportionality of the Polish capacity mechanism (see paragraphs 235 et seq. and 287 et seq. below).

113    With regard to the Green Bonus for generating CMUs meeting the relevant low-carbon criteria, it is sufficient to note, first, that eligible CMUs are not necessarily inefficient fossil fuel power plants, but also new or modern less polluting plants requiring high CAPEX, and that an incentive to reduce those emissions appears to be in line with the objective of common interest of environmental protection, as advocated by the Guidelines, in particular in paragraph (233)(e) thereof, relating to aid measures which should ‘give preference to low-carbon generators’. Tempus has not, however, established that, inter alia, the operation of modern natural gas-based generating CMUs was as harmful to the environment as that of generating CMUs using coal or lignite. Second, as the Commission and the interveners have noted in detail, Tempus has not demonstrated that the application of that bonus gave rise to unequal or even discriminatory treatment of DSR operators, whose internal or external generation capacity could also be based on a polluting or fossil fuel energy source, such as diesel engines, and whose emissions were difficult to quantify (see, also, paragraph 161 et seq. below with regard to the complaints relating to allegedly discriminatory treatment of DSR operators).

114    Therefore, the third complaint must also be rejected as unfounded, as must, accordingly, the first subsection in its entirety.

(ii) Second subsection: the need for intervention by the Polish State

–       The first complaint

115    By the first complaint in the second subsection aimed at demonstrating that the Commission should have entertained doubts as to the need for intervention by the Polish State, within the meaning of paragraphs (222) to (224) of the Guidelines, Tempus submits, in essence, that the Commission carried out an insufficient examination of the alleged capacity adequacy problem in Poland. Contrary to the requirements set out in paragraph (221) of the Guidelines, the data underlying the adequacy assessment, including the 2017 MAF, were incomplete and did not allow valid conclusions to be drawn until 2030, or until 2028, when the aid scheme will end, but at most until 2025. Moreover, contrary to what is required by paragraph (224) of the Guidelines, the contested decision contains only an insufficient description of the impact of the participation of DSR operators and of the measures intended to encourage DSR. According to Tempus, as the consulting firm confirmed and contrary to what is stated in recital 150 of the contested decision, the adequacy assessment mentions DSR only briefly in its conclusion, stating that price caps would be removed and that DSR programmes had been announced.

116    The Commission, supported by the interveners, counters, in essence, that it had sufficiently robust information at its disposal to identify and assess fully generation adequacy concerns in Poland throughout the duration of the Polish capacity market, in particular on the basis of PSE’s adequacy assessment, which it had asked that it be submitted to an independent audit by a consulting firm. That consulting firm confirmed those adequacy concerns and concluded that, in the absence of a capacity market, the Polish electricity network would not be able to meet the reliability standard at issue. As is stated in recital 12 of the contested decision, the audited adequacy assessment, relying on the base scenario, namely the decommissioning of up to 3.5 GW thermal generation units in 2020 and up to 6.5 GW by 2025, anticipated that the LoLE would reach 176.4 hours per annum in 2020 and 101.7 hours in 2025. In addition, the projections for 2030 are in line with those concerning the previous period, made, inter alia, applying ENTSO-E’s methodology. As is apparent from recital 13 of the contested decision, for 2030, the least conservative scenario resulted in 12.56 hours of LoLE, while other, more realistic, modelled scenarios showed significantly higher LoLE values, namely 1 165.30 or 210 hours. In any event, Tempus’s criticism of the methodology for calculating capacity adequacy risks established for 2030 is ineffective given that the aid scheme is approved only until 2028 and the last main auction is to take place in 2025.

117    It should be recalled that it is apparent from paragraphs (222) to (224) of the Guidelines, grouped under Section 3.9.2, entitled ‘Need for State intervention’, that the Member State concerned is supposed, on the one hand, to analyse and quantify properly ‘the nature and causes of the generation adequacy problem, and therefore the need for State intervention to ensure generation adequacy’ (paragraph (222)), and, on the other hand, ‘clearly demonstrate the reasons why the market cannot be expected to deliver adequate capacity in the absence of intervention, by taking account of on-going market and technology developments’ (paragraph (223)). The Commission, for its part, is required to ensure in particular that that Member State has assessed ‘the impact of demand-side participation’ and described ‘measures to encourage demand side management’ (paragraph (224)).

118    In addition, it should be noted that, at the stage of the adoption of the contested decision, the Commission had at its disposal a body of information and expert reports enabling it to dispel any doubts as to the need to establish the Polish capacity market, in accordance with those provisions. In particular, it could rely on PSE’s adequacy assessment of 24 February 2017, which it had expressly asked to be submitted to an independent audit by a consulting firm which delivered its report on 17 August 2017, several months before the notification by the Polish authorities of the aid scheme, in order for them and the Commission to be able to take into account the results and recommendations set out in that report.

119    On that basis, the Commission explained in detail, in recitals 5 to 14 of the contested decision, the generation adequacy issue in Poland which had to be resolved and which, in its view, included the following forecasts:

–        the mothballing/declassification of thermal generation units based on coal and lignite in 2020 up to approximately 3.5 GW and until 2025 for approximately 6.5 GW (recitals 6 and 11 of the contested decision);

–        the missing money problem as a failure of the Polish electricity market due to its inability to compensate for the capacity shortfall which is linked to that declassification (recitals 7 to 9 of the contested decision);

–        the resulting inability of the Polish electricity market to achieve the reliability standard at issue, namely three hours of LoLE per annum (comparable to that used and recognised for the United Kingdom and French electricity markets), as defined by PSE in its adequacy assessment, in accordance with the probabilistic methodology, known as a ‘Monte Carlo simulation’, applied by ENTSO-E, inter alia, in its 2017 MAF, on the basis of a number of forecast scenarios, and as independently confirmed by a consulting firm (recitals 9 to 11 of the contested decision);

–        the resulting capacity shortfalls or LoLE for 2020 and 2025 being 176.4 and 101.7 hours per annum respectively, and corresponding, in the least conservative scenario, in accordance with the 2017 MAF assumptions, to a LoLE still above the reliability standard at issue with 14.2 hours in 2020 and 32.8 hours in 2025; according to the consulting firm’s calculations, the volume of additional capacity thus needed would be 2 750 MW in 2020 and 8 068 MW in 2025 (recital 12 of the contested decision);

–        according to PSE’s adequacy assessment for 2030, that is to say beyond the period covered by the 2017 MAF assumptions, the least conservative scenario results in a LoLE of 12.56 hours, while the other modelled scenarios show significantly higher rates of LoLE, up to 1 165.3 hours; the consulting firm confirmed in that regard that PSE’s methodology was consistent with similar adequacy studies of ENTSO-E (recital 13 of the contested decision).

120    In the light of those audited forecasts, the Commission concluded that the Polish authorities had demonstrated that the identified capacity shortfalls gave rise to the need for intervention by the Polish State in the light of paragraphs (222) to (224) of the Guidelines (recitals 144 to 151 of the contested decision). It inter alia underlined that such shortfalls were expected to arise in 2020 and, according to PSE’s adequacy assessment, amount to 2 750 MW if the Polish market had to rely on energy-only market revenues (recital 146 of the contested decision). Moreover, the Commission noted, in essence, that the quantity of capacity that the Polish authorities intended to procure and remunerate was directly based on that assessment, which was based on an objective reliability standard, aimed at reaching a level of system security of 99.97% (recital 147 of the contested decision).

121    The Commission also found that the way in which the Polish authorities identified and substantiated the existence of various market failures, as is described in Section 2.2.1 of the contested decision, was in line with paragraph (223) of the Guidelines. Those authorities convincingly demonstrated, by way of a probabilistic adequacy assessment (see recitals 9 to 12 of the contested decision), that capacity in the Polish energy-only market was expected to suffer from the missing money problem in case it had to rely on revenues from that market only (recital 148 of the contested decision).

122    Finally, in the light of the criteria set out in paragraph (224) of the Guidelines, the Commission noted that PSE’s adequacy assessment was based on the simulation of different scenarios, each one relying on different assumptions regarding all four of the elements mentioned in that paragraph. In all modelled scenarios, capacity shortfalls were expected to arise as compared to the reliability standard at issue (recital 150 of the contested decision).

123    The foregoing considerations suffice to find that, first, the Commission could not have had any doubts as to the various forecasts that required and justified the intervention of the Polish State in establishing a capacity mechanism and that, second, it set out its assessment in that regard in a sufficiently detailed and intelligible manner in the contested decision.

124    In that regard, it should be noted that Tempus does not dispute the base scenario on which PSE’s adequacy assessment and the independent audit of the consulting firm are founded. That scenario is based on the planned decommissioning of thermal generation units up to 2020 for approximately 3.5 GW and up to 2025 for approximately 6.5 GW, respectively, on the missing money problem affecting the Polish electricity market and on the expected capacity shortfall for those years, taking into account the reliability standard at issue of which neither the calculation methodology, in accordance with that set out in ENTSO-E’s 2017 MAF, nor the appropriateness is called into question by Tempus. Nor does Tempus call into question, in a substantiated manner, the resulting market failure, including the expected capacity shortfalls or LoLE for 2020 and 2025 (176.4 and 101.7 hours, respectively), or the simulated scenarios and assumptions made by PSE, as independently audited by a consulting firm, in accordance with ENTSO-E’s recognised probabilistic methodology, which led to the conclusion that this LoLE was above the reliability standard at issue and required an additional 2 750 MW of generating capacity in 2020 and 8 068 MW in 2025. In those circumstances, it cannot be considered that the Commission should have entertained doubts in that regard.

125    The only real challenge by Tempus in that context concerns the validity of extrapolating scenarios and probabilistic assumptions until 2030 and the fact that the Commission had concluded that there were no doubts as to the compatibility of the aid scheme with the internal market beyond 2025. Indeed, it is limited to criticising the incompleteness of the data underlying the adequacy assessment, including the 2017 MAF, to infer that the Polish authorities and the Commission were not entitled to draw valid conclusions from it until 2028, when the aid scheme will end, or even until 2030.

126    However, Tempus has not been able to undermine the plausibility of the extrapolation method used by PSE, based on the relevant figures for the previous period, and validated, in general, by a consulting firm, which confirmed that this methodology corresponded to the one applied by ENTSO-E in the context of other similar simulations. That is even less the case given that Tempus proposes no alternative methodology capable of calling into question the various probabilistic simulations made by PSE, for which concrete data were clearly lacking, as is also shown by the 2017 MAF, which covers forecasts only until 2025.

127    Consequently, that argument must be rejected, without there being any need to rule on its possible inoperative nature on the purely formal ground that the year 2030 is not included within the duration of the aid scheme (until 2028, with the last main auction in 2025), as is argued by the Commission.

128    Moreover, in so far as Tempus complains that the Commission accepted PSE’s adequacy assessment as being sufficient, even though that assessment did not include, contrary to the requirements of paragraph (224) of the Guidelines, an estimate of DSR potential and of its possible contribution to generation adequacy, it is admittedly true that recital 150 of the contested decision does not contain any specific reasoning in that regard. Nevertheless, that issue was sufficiently addressed during the pre-notification phase in the consulting firm’s report, which expressly recommended that the adequacy assessment be completed on those points, without however concluding that that had a significant impact on the results of its analysis of DSR potential (see pages 6 and 43, paragraph 5.18, and page 48, paragraph 6.8, of that report). Furthermore, the Commission having explicitly mandated that that report be available both at the stage of notification of the aid scheme and at the time of adoption of the contested decision – which is based, inter alia, on that analysis – in the absence of any indication to the contrary, it must be concluded that it necessarily took account of those elements, including the impact of DSR potential, without its having to entertain doubts in that regard.

129    In those circumstances, the first complaint must be rejected as unfounded in its entirety.

–       The second complaint

130    By its second complaint, Tempus alleges, in essence, that the Commission should have entertained doubts as to the necessity of the Polish capacity market on account of the Republic of Poland’s systematic and unlawful refusal, contrary to Article 4(3) TEU, to promote capacity from RES and to take into account the objectives of increasing capacity by 15% until 2020, in accordance with the requirements set out in Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC (OJ 2009 L 140, p. 16). Recitals 134 to 143 of the contested decision, and recital 137 in particular, make no reference to measures taken by the Republic of Poland, in the context of other legislation or schemes aimed at mitigating its greenhouse gas emissions and thus at balancing its support for coal production ‘to a great extent’ (recital 140 of the contested decision). In the light of the requirements of paragraph (224) of the Guidelines, the Commission should have questioned, first, whether similar capacity problems would have arisen if the Polish authorities had emphasised the attainment of the objective laid down by Directive 2009/28 and, second, whether the aid for generation adequacy impeded the installation of generation capacity from renewable sources, contrary to the objective of phasing out environmentally harmful subsidies. Additionally, Tempus complains that the Commission failed to examine whether the well-established intensive use of co-firing of biomass, in addition to natural gas, in current Polish coal power plants, as an integral part of the national renewable energy action plan, would be affected and further hindered by the market reform which was supposed to enable the Republic of Poland to phase out the capacity market. It should have verified whether the Polish authorities intended to introduce, as part of that reform, a mechanism to compensate for a lower co-firing share replacing old coal power plants, which is not the case. The Commission should further have had doubts as to the use of the heat wave constraints in Poland in 2015 to justify the necessity of establishing a capacity market. However, in its working document of 30 November 2016, it underlined that the heat waves of 2015 – which did not affect only Poland – ‘were sufficiently addressed at national level’.

131    The Commission and PGE counter, in essence, that, in the context of the examination in the light of the Guidelines, any shortcomings on the part of the Republic of Poland in achieving the objective of Directive 2009/28 have no bearing on the existence of doubts as to the necessity of the capacity market. Tempus does not provide any evidence to support the existence of a link between the degree of deployment of RES and generation adequacy concerns or its general statement that aid for generation adequacy would hamper the instalment of those sources in Poland. The Republic of Poland highlights, in essence, its efforts, including on the basis of the aid schemes at issue, to develop the operators of renewable energy installations, whose aid ensuring a full return on investment is much more economically advantageous to investors than the capacity market.

132    It should be pointed out that paragraphs (222) to (224) of the Guidelines, under the heading ‘Need for State intervention’, do not lay down any specific requirement for Member States to promote the production of energy from renewable sources, but require only demonstration of the need to introduce an aid scheme for generation adequacy in the light, in particular, of the impact of certain technologies, such as DSR and the actual or potential existence of interconnectors (paragraph (224)(b) and (c) of the Guidelines). Nor does that alleged requirement follow from the ‘objective of common interest’, as described in paragraphs (219) and (220) of the Guidelines, in the light of which Tempus again attempts to call into question recitals 134 to 143 of the contested decision (see, also, the first complaint). While those provisions do set out the objective of phasing out environmentally harmful subsidies, including for fossil fuels, they do not contain any precise objective, in terms of capacity volumes, of promoting RES in return, as is enshrined in Directive 2009/28. The same is true of the vague and unsubstantiated argument which Tempus draws from the promotion of the technology of co-firing biomass in Polish coal power plants. That does not, however, prejudge the question of whether those factors are liable to have an impact on other criteria of the Guidelines, such as the appropriateness of the aid, within the meaning of paragraph (225) of the Guidelines (see paragraph 235 et seq. below).

133    As regards Tempus’s argument relating to the heat waves that occurred in Poland in 2015 to justify the need for a capacity market, it is sufficient to note that, even assuming that the Commission should have doubted the impact of those heat waves on the availability of generation capacity during the summer, that aspect alone cannot call into question the fact that, in general, the Polish authorities sufficiently demonstrated that need by relying on a set of relevant factors to dispel all doubts on the part of the Commission, within the meaning of Article 4(3) of Regulation 2015/1589. That argument is therefore ineffective.

134    Consequently, the second complaint must be rejected in its entirety.

–       The third complaint

135    By its third complaint, Tempus claims that the contested decision fails to take sufficient account of the fact that, since 2014, Poland has already had a capacity market, in the form of a strategic reserve of 830 MW (a so-called ‘cold contingency reserve’) for a period of two years from 2016 and extendable by two years, in the context of which tenders have been launched in order to attract DSR services, four of which, organised between 2013 and 2015, resulted in 200 MW of DSR capacity being contracted. The Commission, however, failed to examine thoroughly the effect of those previous mechanisms – which have since been abolished – on the Polish electricity market, even though such a comparative analysis was essential for determining the necessity, appropriateness and proportionality of the new capacity market. Tempus also reiterates, in essence, its complaint alleging an incomplete assessment in the light of the potential development of the Polish energy-only market, as is required by paragraph (220) of the Guidelines (see the first complaint of the first subsection). The Polish capacity market forms an obstacle to the phasing out of environmentally harmful subsidies for fossil fuels, which the Commission failed to take properly into account, in particular in recital 138 of the contested decision. Finally, Tempus recalls that, in its examination, the Commission overlooked the Republic of Poland’s repeated failures to fulfil its obligation to transpose EU energy rules, including Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe (OJ 2008 L 152, p. 1) and Directive 2009/28, and that it had been notorious for granting State aid to old and polluting coal power plants.

136    The Commission, supported by PGE, contends, in essence, that it was not under an obligation to analyse the impact of an earlier measure intended to ensure generation capacity in Poland which came to an end prior to the implementation of the Polish capacity market. It states that, as a tool for managing a situation of overcapacity and risks of premature exit of existing plants, a strategic reserve is aimed at fixing short-term security of supply problems, including unforeseen situations of scarcity, but that it cannot contribute to the solving of long-term capacity problems. Unlike a capacity mechanism, capacity in a strategic reserve is held in reserve and does not participate in the market or encourage investment in new capacity. Moreover, having regard to the fact that the adequacy assessment had confirmed that the Polish energy-only market could not secure a sufficient volume of capacity in order to comply with the reliability standard at issue and, therefore, to ensure security of electricity supply in Poland, the procedures concerning, inter alia, the implementation of Directives 2008/50 and 2009/28 cannot give rise to doubts as to the need for the aid scheme, the objective of which is not to improve the quality of the environment but to provide necessary capacity for electricity supplies in Poland.

137    To the extent that the third complaint overlaps with the first complaint of the first subsection, it must be rejected as unfounded at the outset for the reasons set out in paragraphs 95 to 99 above.

138    As regards the alleged lack of need for the Polish capacity mechanism due to the existence of an earlier ‘strategic reserve’ scheme, suffice it to note, as the Commission and PGE contend, that, from the point of view of both its objectives and its instruments, that scheme was not comparable to that mechanism. In addition, notwithstanding the fact that that ‘strategic reserve’ scheme was limited in time until 2016, as the case may be on the basis of previous probabilistic forecasts, Tempus does not put forward any plausible evidence capable of explaining how, even following a change of situation in 2016, that scheme would have been sufficient to deal with the problems mentioned in paragraph 119 above, as outlined in recitals 5 to 13 of the contested decision, namely, in particular, to guarantee long-term security of electricity supply using the reliability standard at issue. Moreover, in recital 16(g) of the contested decision, the Commission took account of the fact that, before the first delivery year of the capacity market, namely 2021, the ‘cold contingency reserve’, inter alia, would be terminated. That argument cannot therefore be accepted.

139    With regard to Tempus’s arguments alleging that the Republic of Poland failed to comply with its alleged obligations to implement a more environmentally friendly energy policy – in particular Directives 2008/50 and 2009/28 – it is sufficient to recall the considerations set out in paragraph 132 above, which apply mutatis mutandis.

140    In those circumstances, the third complaint must be rejected as unfounded in its entirety.

–       The fourth complaint

141    By its fourth complaint, Tempus criticises the Commission for not having sufficiently assessed, in its examination of the objective of common interest, the effects of the Polish capacity mechanism on the electricity markets of neighbouring countries and on electricity exchange. Such an assessment was necessary, however, in view of the numerous applications for capacity mechanisms, six of which were approved by the Commission on the same day, of its conclusions in its final report of the sector inquiry and of its proposal for a regulation on a new electricity market design. At the instigation of the Republic of Poland, which had carried out a campaign to that effect, Article 18(a) of the proposed regulation allowed Member States to opt for the new ‘sunset clause’ allowing old coal power capacity also with emitting quality of 550 g CO2/kWh to participate in all capacity auctions, provided that they had been accepted beforehand under the conditions of the Polish capacity market until the end of 2019 in a previous national capacity mechanism. Aware of that campaign by the Polish authorities to maintain old and polluting coal power capacity, the Commission should have had doubts as to the reliability of the Polish capacity market and its validity under the objective of common interest. In addition, in view of Article 23(4) of the proposed regulation, like its approach in the case on the French capacity market (Commission Decision (EU) 2017/503 of 8 November 2016 on State aid scheme SA.39621 2015/C (ex 2015/NN) (OJ 2017 L 83, p. 116)), the Commission should have had doubts as to the consequences of the Polish capacity market and its effect on neighbouring Member States. The Act contains a specific and detailed chapter entitled ‘Participation of Foreign Capacities in the Capacity Market’, but the contested decision does not even refer to any negotiations or exchanges with the neighbouring Member States involved. Thus, the conclusion set out in recital 185 of the contested decision, according to which the Commission was ‘satisfied’ that, ‘due to its design, the negative effects of the [Polish] capacity market on competition and trade in the internal electricity market’ were sufficiently limited cannot conceal the incomplete and erroneous assessment of the relevant facts and, therefore, the existence of doubts as to the compatibility of the aid scheme, having regard to the objective of common interest and the need for State intervention.

142    The Commission considers that it did carry out a sufficient assessment of the impact of the Polish capacity market on neighbouring electricity markets and on electricity exchange. First, the discussions within the legislative procedure at issue on a definition of the conditions for capacity markets as well as a mere proposal for a regulation have no bearing on its assessment of such a market under the State aid rules, nor can they call into question the validity of that assessment in the contested decision. The Republic of Poland contends that Tempus’s argument is ineffective, given that the contested decision was adopted on 7 February 2018, more than 4 months before the start of informal trilogues, more than 10 months before the political agreement concluding the negotiations and more than 16 months before the publication of Regulation (EU) 2019/943 of the European Parliament and of the Council of 5 June 2019 on the internal market for electricity (OJ 2019 L 158, p. 54) in the Official Journal of the European Union. Second, the Commission takes the view that the allegation that there are doubts as to the effects of the aid scheme on neighbouring markets is not substantiated. In recital 150 of the contested decision, it assessed the impact of variable generation, including that originating from neighbouring systems, as required by paragraph (224)(a) of the Guidelines, declaring, after having analysed different scenarios presented in the adequacy assessment, that capacity shortfalls were expected to arise as compared to the volume needed to achieve the reliability standard at issue. In addition, in Section 2.5, entitled ‘Participation of foreign capacity’, as well as in recital 165 of the contested decision, it rightly concluded that the rules of the aid scheme were appropriate in that regard. Furthermore, the Polish authorities committed, with regard to the opening of the Polish capacity market for the capacities from neighbouring Member States, to re-notify the interconnectors’ de-rating methodology by the end of March 2022 (recital 86 of the contested decision). Last, the reference to the French and German (Commission Decision (EU) 2018/860 of 7 February 2018 on the Aid Scheme SA.45852 – 2017/C (ex 2017/N) – Germany – Creation of a capacity reserve (OJ 2018 L 153, p. 143)) cases cannot establish the existence of doubts regarding the rules of the aid scheme relating to foreign capacity participation. After all, the French capacity market, as initially notified, excluded explicit participation of foreign capacity and allowed only for implicit participation, in that the total capacity demanded was adjusted to account for expected imports, without that foreign capacity being remunerated for its contribution to security of supply. The German capacity reserve, as initially notified, excluded foreign capacity participation completely. By contrast, the Polish capacity market enabled the explicit participation of foreign capacity and contained important commitments as regards the opening of the borders.

143    As a preliminary point, it should be noted that Tempus cannot validly base an argument solely on the fact that the Commission has dealt with a number of similar cases and adopted several decisions in parallel, such an approach rather being beneficial to consistency between the different national capacity markets within the European Union, with a view to the creation of an internal market for electricity. In any event, it must be held that the Republic of Poland is right to argue that subsequent events, including the alleged campaign by the Polish authorities concerning the creation of such a market on the basis of Regulation 2019/943, which took place after the adoption of the contested decision and which was not available to the Commission, are not such as to call into question the legality of that decision (see the case-law cited in paragraph 50 above). Therefore, all of Tempus’s arguments relating to the genesis of that regulation must be rejected as ineffective and as incapable of creating doubts on the part of the Commission at the time of the adoption of that decision.

144    Furthermore, as the Commission submits, it is apparent from recital 150 of the contested decision, which is admittedly succinct, that the Commission did indeed take account of the requirement laid down in paragraph (224)(a) and (c) of the Guidelines which consists in ascertaining whether the Member State concerned has assessed, inter alia, the impact of variable generation, including that originating from neighbouring systems, and the actual or potential existence of interconnectors, in order to determine whether its intervention, linked to the introduction of a capacity mechanism, was necessary. Tempus, however, does not put forward any specific evidence capable of establishing the claim that the Commission did not, in fact, take those aspects into account, even though the various scenarios used in PSE’s adequacy assessment and audited by the consulting firm clearly included those cross-border aspects (see, also, recital 165 of the contested decision and paragraph 209 et seq. below).

145    Furthermore, as the Commission contends, it is clear that the cases and decisions relating to the French and German markets are neither comparable to the present case nor such as to support the existence of doubts as to the necessity of introducing the Polish capacity market. In the case concerning the French capacity market, unlike the present case, the Commission had initially criticised the absence of any rule providing for the participation of foreign capacity that the French authorities ultimately granted (see recitals 119, 238, 239, 192, 193 and 293 to 295 of Decision 2017/503). Furthermore, even though such an absence had also been criticised in the decision to initiate the procedure in the German case, it turned out that the relevant market was not a capacity market in the strict sense, but that the relevant rules provided for the creation of a capacity reserve outside the electricity market, such that the exclusion of foreign capacity from participating could not give rise to a distortion of cross-border competition (see recitals 42, 59, 80, 92, 123 to 126 of Decision 2018/860).

146    It follows that the fourth complaint and, accordingly, the second subsection, alleging that there was no need for intervention by the Polish State, must be rejected.

147    Consequently, the first section, alleging that the assessment of the objective of common interest objective was incomplete and that State intervention was necessary, must be rejected in its entirety.

(3)    Second section: alleged incompleteness of the assessment of the appropriateness of the aid scheme

(i)    First subsection: error of assessment of the relevance of the Polish capacity mechanism in a highly concentrated and State-dominated electricity market

148    By the first subsection, Tempus criticises the Commission for not having properly examined the appropriateness of the aid scheme in the light of the high degree of concentration of the Polish electricity market, dominated by the Polish State, and of the requirement, under paragraph (233)(d) of the Guidelines, that a capacity mechanism ‘not unduly strengthen market dominance’. Similarly, the Commission failed to make the necessary comparison with the previously used capacity mechanisms that had been introduced by the Republic of Poland from 2014, even though current market conditions should have been fully taken into account in determining whether the measures for new and existing capacity were sufficient to address market failures. In view of the barriers identified for new entrants and for DSR in particular in the Polish capacity market, the Commission should therefore have had doubts as to the risk that it might strengthen the concentration of the Polish electricity market and as to its appropriateness.

149    The Commission, supported by the interveners, contends that the Guidelines do not preclude the introduction of a measure intended to ensure generation adequacy in a concentrated electricity market. The Polish capacity market does not create barriers to entry for generating or DSR CMUs, but provides a source of stable income for new entrants easing the start of their business in Poland. In addition, paragraph (233)(d) of the Guidelines does not require the Commission to take due account of the structure of the national energy market when assessing the appropriateness of the notified aid measure (recital 183 of the contested decision). In any event, Tempus does not establish that the approval of the Polish capacity market would unduly strengthen the market dominance of a particular capacity provider or undertaking on that market.

150    It must be considered, in line with the Commission, that paragraphs (225), (226) and paragraph (233)(d) of the Guidelines do not prevent the approval of a capacity market on account of its inappropriateness merely because the electricity market concerned has a certain degree of concentration or is still dominated by the State, which is a recurrent phenomenon in a sector that has undergone successive liberalisation at European level only in recent times. It is precisely for that reason that paragraph (233)(d) of the Guidelines has the sole objective of preventing pre-existing market dominance from being unduly strengthened and no other provision of the Guidelines addresses that phenomenon from the perspective of the appropriateness or proportionality of aid for generation adequacy. Therefore, the argument of Tempus which alleges the existence of barriers to entry is ineffective in this context (see, also, paragraph 161 et seq. below on the complaint based on the alleged discrimination of DSR). As regards the alleged need for the Commission to assess the appropriateness of the introduction of the Polish capacity market in the light, inter alia, of the pre-existing market mechanisms and the current conditions in which the Polish electricity market operates, it is sufficient to note that this complaint overlaps with that alleging that there was no need for the intervention of the Polish State and must be rejected for the reasons set out in paragraph 138 above, Tempus not having succeeded in calling into question the plausibility of the forecasts for the development of the said market by 2020, 2025 and 2030.

151    Accordingly, the first subsection must be rejected as unfounded.

(ii) Second subsection: discrimination of DSR

–       Preliminary observations

152    In the second subsection, Tempus complains that the Commission dispelled doubts, without having conducted a more detailed assessment in that regard, of discrimination on two counts, contrary to paragraph (226) of the Guidelines, to the detriment of DSR operators, pertaining both to the length of the capacity agreements and to the CAPEX eligibility criteria. First, it submits that foreign DSR operators are discriminated against in comparison with Polish DSR operators, when the former can enter into capacity agreements of only one year’s duration, whereas the latter are eligible for capacity agreements of between one and five years depending on their CAPEX. Second, it claims that DSR operators as a whole are discriminated against as compared with generating CMUs.

153    In view of the fact that the first aspect is barely developed by Tempus in this context and is more related to possible discrimination of foreign capacity relating to the conditions governing its access to the Polish capacity market, the General Court considers it necessary to deal, first, with the second aspect, which concerns the link between the CAPEX criteria and the length of the capacity contracts, as well as the alleged discrimination between generating and DSR CMUs. The first aspect will be assessed in the answer to the fourth subsection (see paragraph 209 et seq. below).

154    It should also be noted that Tempus’s complaints are based, to a large extent, on the considerations set out in the study of the Polish capacity market prepared by the energy consulting firm Energy Brainpool, carried out at the request of Tempus for the purposes of the present proceedings.

–       The first complaint

155    In the first complaint, Tempus submits, in essence, that DSR CMUs are treated in a discriminatory manner as regards both the length of the capacity agreements and the eligibility criteria. DSR CMUs as a whole are not eligible for 15-year capacity agreements, which are reserved to generating CMUs incurring CAPEX above PLN 3 million/MW. The Commission, however, failed to assess properly the CAPEX system proposed by the Polish authorities, its appropriateness for meeting the objective of common interest pursued and its discriminatory effect.

156    According to Tempus, as is apparent from Table 1 in recital 42 of the contested decision, the CAPEX criteria for determining the length of capacity agreements wrongly include that of retrofitting to best available technology (‘BAT’) and BAT modernisation. Those CAPEX criteria are incompatible with the objective of security of supply on the Polish capacity market, in that retrofitting to BAT admittedly results in a reduction of greenhouse gas emission levels, but does not increase the capacity of the relevant CMU. In addition, the grant of capacity payments – up to 17 years if the Green Bonus has been awarded – to the CMUs concerned based on investments made to retrofit to BAT appears to be unlawful and probably incompatible State aid for adaptation to EU standards, within the meaning of paragraphs (53) to (55) of the Guidelines. CAPEX incurred by reason of retrofitting to BAT should not be used in the Polish capacity mechanism. After all, retrofitting an existing coal plant to BAT standards by means of a cost-intensive upgrade is not required either for increasing its capacity, the level of which remains the same, or for capacity availability, but rather for keeping it in operation. That contradicts the objective of guaranteeing security of supply by ensuring technological neutrality and promoting new technologies and energy generating capacity allowing for a reduction of market prices in both off-peak (more low cost capacity) and peak periods (more total capacity). Tempus also recalls that the Polish investment plan already provides aid for the modernisation of existing coal-fired power plants during the period of transitional allocation of free greenhouse gas emission allowances between 2013 and 2019, in accordance with Article 10c of Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC (OJ 2003 L 275, p. 32), which has been approved by the Commission (Commission Decision C(2013) final of 22 January 2014, State aid SA.34674 (2013/N) – Poland – Derogation from Article 10c of Directive 2003/87/EC on emission trading – Free allowances to power generators).

157    Moreover, the shorter length of capacity agreements envisaged for DSR CMUs places them at a disadvantage in general compared to the inefficient, high-emission generating CMUs using fossil fuels such as coal. According to Tempus, even if the CAPEX of generating CMUs for enhancing their capacity is considerably higher than that required by DSR CMUs, they cannot satisfy themselves with capacity agreements of up to 5 years on a capacity market where established, fossil fuel-based generating CMUs receive agreements lasting up to 17 years. Furthermore, only a minority of DSR operators can receive five-year capacity agreements (recital 43 of the contested decision). Thus, as a foreign DSR operator, Tempus could obtain only a one-year capacity agreement, irrespective of its CAPEX and its other financing needs, including operational costs (OPEX), which vary widely depending on the type of customer and the flexible asset used. It specifies that, even though Polish DSR operators are in theory permitted to access five-year capacity agreements, the majority of the capacity concerned and almost all of the capacity of new entrants is expected to be procured in ‘N-1’ auctions offering only one-year agreements rather than in ‘N‑5’ auctions. Longer agreement lengths, however, enable the bidder to bid a lower auction price and thereby obtain a competitive advantage by spreading the revenue stream over a longer period, which is the intended benefit of the Green Bonus (recital 50 of the contested decision). This results at least in indirect discrimination or implicit exclusion of DSR, the possibility of which was recognised in the Commission’s working document on the sector inquiry. However, the Commission failed to examine whether the eligibility criteria linked to the length of the capacity agreements were discriminatory, inter alia, towards foreign DSR operators (recital 62 of the contested decision).

158    The Commission and the interveners deny that the aid scheme discriminates against DSR operators. The Guidelines, in particular paragraph (226) thereof, do not require identical treatment of those operators and other generation capacity providers, but only to open the capacity market to DSR and provide it with adequate incentives. As regards the length of the capacity agreements, they submit that the difference in treatment is justified while providing such adequate incentives. The inclusion of the criterion of retrofitting to BAT standards is essential, since, otherwise, pursuant to Article 14 of Directive 2010/75/EU of the European Parliament and of the Council of 24 November 2010 on industrial emissions (integrated pollution prevention and control) (OJ 2010 L 334, p. 17), a generation unit might not be authorised to operate and thus keep the capacity available or to pass the certification process in order to be qualified as a CMU (recital 25(f) and (g) of the contested decision). Furthermore, capacity payments are aimed at rewarding the availability of capacity, not at covering investment costs. The retrofitting of a CMU to BAT standards is not intended to increase its capacity, but to extend its lifetime and improve its reliability, both of which increase the availability level and thus contribute to security of supply. Moreover, the modernisation of an existing unit will not lead to the grant of a 17-year capacity agreement, since the CAPEX is well below the threshold. The inclusion of the costs of the operation permit in the CAPEX/MWh calculation reflecting the retrofitting to BAT therefore does not amount to distinct aid for investment to finance the retrofitting to BAT standards of old existing generation units, but is inherent to the aid for capacity availability of the modernised CMUs. Tempus also fails to have regard to the fact that the investment subsidies granted on the basis of the free allocation of greenhouse gas allowances, classified as investment aid, are deducted from the capacity payments in order to avoid any overcompensation (recital 19 of the contested decision).

159    The Commission, supported by the interveners, observes that generating CMUs and DSR units are in different factual situations, the level of their CAPEX requirements being different, which is an objective criterion for determining the duration of the capacity agreements. In addition, the claim that DSR CMUs cannot be satisfied with capacity agreements of up to five years’ duration is not substantiated. On the contrary, the significantly lower costs of those CMUs should enable them to submit correspondingly lower bids in the auctions and thus increase their chances of winning them. On the other hand, long-term capacity agreements for generating CMUs are necessary in order to create a level playing field, as, in contrast to existing capacity, new capacity – or refurbishing one – is likely to need to secure financing for CAPEX, which is more difficult and more expensive without the relative stability provided by multi-year agreements. Thus, the possibility of concluding capacity agreements for more than one year is intended to ensure predictable and stable income over a long period, which is not necessary to the same extent for developing DSR, following large, risky capital investments that would not have been undertaken were it not for such a guarantee.

160    The Commission disputes that DSR operators participate mainly in ‘N‑1’ auctions and not in ‘N‑5’ auctions, since there is no legally binding rule in the aid scheme providing for such a limitation. That is confirmed by the high participation of those operators in the first three main auctions organised in Poland. In view of the uncontested information provided by the Polish authorities and its long experience with the CAPEX criteria establishing the length of capacity agreements, the Commission was able to take the view, without having any doubts in that respect, that the duration of capacity agreements as foreseen in the Polish capacity market was appropriate. The Commission disputes, moreover, that that market tends to favour the modernisation of existing CMUs over new power generation or technologies and, in particular, new and refurbishing fossil fuel generation projects.

161    It should be recalled that, in accordance with settled case-law, applicable in matters of State aid, 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 (judgment of 15 April 2008, Nuova Agricast, C‑390/06, EU:C:2008:224, paragraph 66). 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 (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).

162    Moreover, it is important to note that, regardless of whether the contested decision is a decision not to raise objections that is subject to a review of legality aimed at assessing the existence of doubts or serious difficulties (see paragraphs 48 to 51 above), the observance by the Commission of the principle of equal treatment is a question of law involving no discretion on its part and is therefore amenable to a full review by the EU Courts (see, to that effect, judgments of 11 September 2007, Lindorfer v Council, C‑227/04 P, EU:C:2007:490, paragraph 63 and the case-law cited, and of 17 September 2009, Commission v Koninklijke FrieslandCampina, C‑519/07 P, EU:C:2009:556, paragraphs 100 to 103 and the case-law cited). It is in the light of that premiss that it is appropriate to examine whether the Commission should have had doubts as to observance of, inter alia, the principle of equal treatment which it was required to observe also when applying Article 107(3)(c) TFEU (see, to that effect, judgment of 15 June 1993, Matra v Commission, C‑225/91, EU:C:1993:239, paragraph 41), and the rules of the Guidelines which are intended to implement that provision.

163    Thus, in the case at hand, observance of the principle of equal treatment must be assessed in the light of the objectives recognised by the Guidelines (see Section 3.9.1, entitled ‘Objective of common interest’), which are intended to implement, inter alia, Article 107(3)(c) TFEU, read in conjunction with Article 194(1) TFEU (see paragraphs 88 and 93 above).

164    Those objectives include, in particular, the objective of generation adequacy defined by the Member State irrespective of any short-term concerns (paragraph (219) of the Guidelines). In that regard, while it is true that Member States may be authorised to grant aid which may ‘contradict the objective of phasing out environmentally harmful subsidies including for fossil fuels’, they ‘should … primarily consider alternative ways of achieving generation adequacy which do not have a negative impact on th[at] objective …, such as facilitating demand side management and increasing interconnection capacity’ (paragraph (220) of the Guidelines). Furthermore, in the light of those objectives, coupled with the objective of the aid remunerating ‘solely the service of pure availability provided by the generator …’ and not for the sale of electricity (see Section 3.9.3, entitled ‘Appropriateness’), such aid measures ‘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’ (paragraphs (225) and (226) of the Guidelines).

165    It must be held that DSR CMUs and generating CMUs, in particular new ones and ones to be modernised, are, in the light of the objectives pursued by the Polish capacity market, in different factual and legal situations as regards the application of the CAPEX criteria determining the length of the capacity agreements, which is not as such disputed by Tempus.

166    First, it is common ground that the CAPEX criteria are closely linked to the differentiation of the duration of capacity agreements, ranging from 1 to 15 years, in so far as that duration takes account of the period that a generating CMU may need to recoup significant investments. Guaranteeing capacity payments for a certain period of time and limiting the associated economic risk create a significant incentive to make the necessary investment for modernising an already-established power plant or the entry of an operator into the Polish electricity market through the establishment of a new power plant and thus to ensure, in accordance with the capacity market’s main objective of guaranteeing security of supply, the long-term availability of the associated generating capacities (see recitals 42 to 44 of the contested decision). It should be specified that, under the first subsection of the third section of the second part of the first plea (see paragraph 248 below), Tempus does not dispute that the CAPEX criteria create such an incentive effect, in that they encourage operators to invest in the creation or modernisation of generation capacity.

167    It follows that that differentiation of the duration of the capacity agreements according to the CAPEX criteria does not, a priori, constitute unequal treatment of operators, since they are not all obliged to make such significant investments and, therefore, are in different factual and legal situations.

168    Second, Tempus acknowledges that DSR CMUs generally incur lower costs than generating CMUs investing in modernisation or in the construction of a new power plant. In addition, without precise quantification and more detailed information, Tempus argues to no avail that those operators also incur substantial costs justifying granting them capacity agreements longer than five years, those costs not satisfying, according to its own admission, the relevant CAPEX criteria. Furthermore, Tempus’s claims are contradicted, inter alia, by Enspirion, which, as a Polish DSR operator, explains in a plausible manner the reasons for which such operators, from both an economic and a legal perspective, are not interested in offering a long-term commitment to providing capacity. That assessment is borne out by the finding – not disputed as such by Tempus – according to which, unlike generating CMUs classified as price takers, DSR operators, where they are not yet fully integrated into the energy market, are price makers which need the capacity payments to recoup their generally low fixed costs and at least part of their generally high variable costs (recital 47 of the contested decision; see, also, recital 62 of that decision). Taking into account their classification as price makers, however, Tempus has not substantiated the reasons for which it is necessary to grant DSR operators capacity agreements longer than five years to enable them to bid a lower auction price and thus obtain a competitive advantage by spreading the revenue stream over a longer period, as is provided for large investments in generating capacity.

169    In that regard, Tempus’s attempt to call into question the inclusion of the criterion of adaptation to BAT standards in the determination of CAPEX cannot succeed, either. As the Commission notes, adaptation to those standards is a precondition not only for the continued operation of a generating CMU under Article 14 of Directive 2010/75 and the corresponding Polish law, and thus for the maintenance of the related generating capacities, but also for certification as a CMU in order to be eligible to participate in the capacity auctions (recital 25(f) and (g) of the contested decision; see, also, Article 19(2)(2)(b) of the Act concerning the conditions governing the making of a certification application). It is appropriate also to recall that, by virtue of the objective of the Polish capacity market of ensuring security of supply and its operating rules (see paragraph 11 above), capacity payments are intended primarily to reward the availability of capacity and, only secondarily, to cover investment costs as such. The Commission is therefore right to contend that the adaptation of a CMU to BAT standards is not necessarily aimed at increasing capacity, but, primarily, at maintaining its availability and ensuring its reliability. The same is true where such an adaptation involves the costly modernisation of an old and polluting fossil fuel plant, which also has the objective of keeping that plant in operation and making its capacity available. It follows that the taking into account of the costs of adaptation to BAT standards in the calculation of CAPEX cannot be classified as distinct aid for investment favouring old existing and polluting power plants, as Tempus claims, but constitutes an integral part of the aid scheme ensuring the proper functioning of the Polish capacity mechanism. Finally, even assuming that it is distinct aid for investment for the purposes of adaptation to BAT standards, such aid, like aid under the RES support scheme, should be deducted from the capacity payments to avoid any overcompensation (see recital 19 of the contested decision and paragraph 11 above).

170    Consequently, the argument alleging that such CAPEX criteria were wrongly recognised in the Polish capacity mechanism must be rejected.

171    Third, Tempus fails to make a clear and precise distinction between the modernisation of an established generating CMU, on the one hand, and the creation of a new generating CMU, which is much costlier, on the other. In the first case, a CMU is not likely to be granted capacity agreements lasting up to 17 years, in view of the Green Bonus, given that, in the case of a modernisation, the CAPEX is generally well below the threshold and does not normally give rise to the grant of that bonus (see, inter alia, Table 1, second line, contained in recital 42 of the contested decision). In accordance with the foregoing, PGE has argued, without being contradicted by Tempus, that the majority of the costs of adapting to BAT standards are well below the CAPEX threshold for capacity agreements of five years’ duration. Moreover, without prejudice to the examination of the second complaint (see paragraphs 174 to 182 below), it is important to note that the fact that the Green Bonus allows a generating CMU, in accordance with the incentives described in paragraph 166 above, to bid a lower price and thus obtain a competitive advantage by spreading the revenue stream over a longer period (recital 50 of the contested decision), fits perfectly into the system of CAPEX criteria linked to the duration of the capacity agreements while promoting a more environmentally friendly technology. Thus, it cannot be held that there is unequal treatment to the detriment of other capacity providers in a different factual and legal situation, including DSR operators which have much lower costs and are generally not under the obligation to improve the emissions of their generation capacity.

172    It follows from the foregoing considerations that Tempus has neither succeeded in demonstrating that generating and DSR CMUs were in at least comparable factual and legal situations, nor has it put forward sufficiently detailed and plausible evidence to justify the conclusion that the Commission should have entertained doubts as to possible discrimination, at least indirectly or implicitly, in that regard, on the ground that the duration of the capacity agreements to be granted was differentiated according to CAPEX. This complaint must therefore be rejected.

173    The same applies to the ineffective argument that only a minority of DSR operators can access five-year capacity agreements.

–       The second complaint

174    By the second complaint, Tempus submits, in continuation of the first complaint, in essence, that it is in a discriminatory manner that DSR CMUs are not eligible for the Green Bonus. That exclusion results in an additional unjustified advantage for generating CMUs in bidding processes, in terms of potential cost reduction, and is contrary to paragraphs (43) and (221) of the Guidelines. Eligibility for that bonus would enable DSR CMUs to provide increased incentives for large industrial and commercial customers to switch to genuine DSR load-shifting rather than using polluting on-site oil or diesel generators. By contrast, extending by two years the capacity agreements of generating CMUs using conventional energy sources has the opposite effect, in that they may emit less than the threshold of 450 kg CO2/MWh but still receive higher support than CMUs using lower-emission technologies, such as DSR. The Commission did not, however, examine that inconsistency or that violation of paragraphs (43) and (221) of the Guidelines. Nor did it assess the criteria for eligibility for the Green Bonus against paragraph (233)(e) of the Guidelines. According to Tempus, under Article 25(5) of the Act, eligibility for the Green Bonus is based solely on the level of emissions of the relevant generating CMUs and, as their business model is, by its nature, low on emissions, there is no technical or economic parameter preventing DSR CMUs from being eligible for it. In accordance with the estimates of the Polish authorities, the latter CMUs could also reduce their offer price by 10% (recital 50 of the contested decision), without placing them at a disadvantage compared with fossil fuel producers, including from natural gas, in violation of the abovementioned provisions.

175    The Commission and the interveners counter, in essence, that the objective of the Green Bonus is to incentivise the development of low-emission generating CMUs. In the light of the generation shortfall identified in the adequacy assessment, clear incentives had to be provided to those CMUs to stimulate costly investment in new, more environmentally friendly capacity. Therefore, the fact that only generating CMUs are eligible for the Green Bonus is in line with the objective of common interest of the Polish capacity market and does not constitute discrimination against DSR CMUs. By contrast, Tempus’s argument that DSR is ‘less emissive’ and should therefore be awarded the Green Bonus cannot succeed, given the difficulty – or even impossibility – of determining in advance the actual emissions resulting from or linked to DSR. Enspirion states, in essence, that no DSR operator is interested in long-term capacity agreements, given that DSR is regarded as an ancillary activity, that power plants generate profit through production, not by stopping it, and that those operators are not able to determine the possibility of taking over the capacity obligation five years in advance.

176    It must be pointed out that Tempus fails to recognise that the Green Bonus is merely a ‘bonus’ aimed at incentivising only generating CMUs, which are allocated 15-year capacity agreements due to their substantial investment needs exceeding the relevant CAPEX thresholds (see paragraphs 166 to 171 above), to take measures to reduce their emissions in order to improve environmental protection. Tempus has not been able to demonstrate why DSR CMUs should also be awarded such long-term capacity agreements, so it has not established any justification for applying the Green Bonus to them, either. The one-off difference in treatment relating to that bonus between generating and DSR CMUs under Article 25(5) of the Act, read in conjunction with Article 25(4)(2) thereof, cannot therefore lead the General Court to find that there has been discrimination.

177    Tempus’s line of argument seeks to dissociate artificially the environmental protection aspect underlying the Green Bonus from the objectives, structure and functioning of the Polish capacity mechanism as a whole into which it is integrated, including the economic incentives for generating CMUs to invest in the capacity required to ensure security of supply in the long term. It follows that the argument that, in so far as they use an environmentally friendly technology, DSR CMUs must necessarily be eligible for the Green Bonus is based on a reductionist and erroneous premiss. To that end, Tempus merely takes into account the objective of environmental protection while disregarding that of security of supply, which is, however, a priority and the reason behind granting generating CMUs, including conventional ones, 15-year capacity agreements. It has therefore not established, in the light of all those objectives, that generating and DSR CMUs were in at least comparable factual and legal situations requiring that the Green Bonus be applied to them in the same way, even outside the 15‑year capacity agreements for which DSR CMUs are ineligible.

178    Furthermore, as Enspirion and PGE rightly note, it has not been established that the capacity provided by DSR operators is capable of fulfilling the eligibility criteria for the Green Bonus, given the impossibility of quantifying the related emissions. In that regard, Tempus’s argument that eligibility for that bonus would enable DSR CMUs to provide increased incentives for large industrial and commercial customers to switch to genuine DSR load-shifting rather than using polluting on-site oil or diesel, tends rather to confirm, at present, that those CMUs do indeed depend, in part, on such polluting generation technologies.

179    It follows that Tempus has also failed to demonstrate that, in its assessment set out in recitals 49 and 50 of the contested decision concerning the Green Bonus, the Commission should have entertained doubts in the light of paragraphs (43), (221) and paragraph (233)(e) of the Guidelines.

180    In so far as Tempus relies on paragraph (43) of the Guidelines, it should be recalled that that provision takes account of the possible need to balance ‘a measure addressing a generation adequacy problem … with the environmental objective of phasing out environmentally or economically harmful subsidies, including for fossil fuels’. Thus, it does not follow that the environmental objective which is inherent in the Green Bonus is an absolute priority as such, in the sense that that bonus should be generally available to any operator using low CO2-emitting technologies, such as DSR. Likewise, paragraph (221) of the Guidelines merely sets out the requirement for the Member State concerned to define clearly the precise objective at which the aid measures for generation adequacy are aimed, which is the case here as regards both the objective of common interest (see paragraph 89 et seq. above) and the Green Bonus (see paragraph 171 above). Finally, the requirement to give preference to low-carbon producers through those measures, in case of equivalent technical and economic parameters (paragraph (233)(e) of the Guidelines), should not be understood as being absolute, either, but is also subject to the aforementioned balancing requirement.

181    In the light of the foregoing considerations, Tempus has not succeeded in establishing that DSR operators, which, in the light of the objectives pursued by the Polish capacity mechanism and by the Guidelines, are in a different factual and legal situation from that of generating CMUs with significant CAPEX, suffer unequal treatment in that they are not eligible for the Green Bonus. For those reasons, the Commission could validly accept, without having any doubts, in recital 50 of the contested decision, that the CMU eligible for that bonus was placed in a position to bid a lower auction price and thus obtain a competitive advantage by spreading the revenue stream over a longer period, that incentive fitting perfectly into the system of CAPEX criteria linked to the duration of the capacity agreements while promoting a more environmentally friendly technology (see paragraph 171 above).

182    The second complaint must therefore be rejected as unfounded.

–       The third complaint

183    In the third complaint, Tempus claims that DSR CMUs are discriminated against in the light of the rules governing the termination of capacity agreements, those rules being much stricter for them than for new generating CMUs in case of any delays in the fulfilment of formal obligations, which the Commission neither took into consideration in its assessment of the appropriateness of the aid scheme nor disputed in the course of the proceedings.

184    The Commission, supported by the interveners, points out that a new generating CMU is obliged, prior to the commencement of the first delivery period, to demonstrate its ability to deliver, for at least one uninterrupted hour, 95% of total contracted capacity, multiplied by the corrective availability factor. By contrast, a DSR CMU unconfirmed for auctions – currently all such CMUs – has to pass a demand reduction ability test at the latest one month before the first delivery period, which is already the case where that CMU can provide at least 80% of the contracted capacity and, if not, where, at its request, it demonstrates within three working days its ability to provide at least 50% of that capacity (see Article 53(4) and (5) of the Act). Its capacity agreement will be amended accordingly and its capacity obligation, as well as the value of the security lodged, will be reduced proportionately. That solution enables DSR CMUs to reduce their business risk significantly compared to that of generating CMUs.

185    The Republic of Poland states that the construction of a new generating CMU is a long and complex investment process, involving many risks that may materialise at various stages of the works, such that it was necessary to mitigate the risk of loss of the capacity agreement at an advanced stage of construction. Postponing the termination of that agreement to no later than the end of the third delivery year envisaged is an additional incentive to ensure the timely completion of investments. No remuneration is paid for each month of delay, and the electricity supplier pays a penalty of 15% of the value of the unfulfilled capacity obligation, which is calculated on the basis of the highest price at the capacity auction for a given delivery year. By contrast, a DSR CMU certified as confirmed not incurring such large financial outlays does not have to provide any collateral before the conclusion of the capacity agreement and, between its conclusion and commencement of its enforcement, no additional measures are required which, if not completed, could result in its termination. In addition, unlike generating CMUs, unconfirmed and planned DSR CMUs are required to indicate the facilities which will achieve the reductions no later than three months before the delivery period.

186    PGE adds that, as part of the general certification process, DSR CMUs must adhere to much lower information and formal requirements than generating CMUs, the latter having the obligation to provide detailed technical parameters, production plans for the next five years and expected planned outages for each month. DSR CMUs are required to provide only simplified operating plans, which include, for example, their estimated maximum and minimum physical capacity. Similarly, in the certification process for the main and supplementary auctions, generating CMUs are required to provide an electricity generating licence, without there being any equivalent requirement for DSR CMUs.

187    Suffice it to note that the present complaint is particularly lacking in detail and does not take account of the reasons that gave rise to the differentiation of the rules on termination of the capacity agreements of the CMUs concerned, which the Commission and the interveners rightly point out (see Articles 46 and 47 of the Act). Those authorities have explained, in a plausible manner and without being contradicted by Tempus, that, by analogy with the reasons set out in paragraphs 165 to 172 above concerning the differentiation of the duration of those agreements, generating CMUs must be given substantial economic incentives either to modernise old plants or to build new plants. Those incentives are strengthened by the specific and favourable termination rules for new and modernised generating CMUs by relieving them of part of the economic risk which is linked to the long-term capacity commitment (Article 46(4) of the Act). As with its line of argument relating to the differentiation of the duration of capacity agreements, Tempus has not been able to establish the reasons why DSR CMUs are in a factual and legal situation at least comparable to that of those generating CMUs and should therefore be covered by the same termination rules. Furthermore, Tempus does not take into account either the obligations to which those latter CMUs are subject in return for the alleged advantage linked to the possibility of postponing the date of termination, namely that of the payment of a penalty of 15% of the value of the unfulfilled capacity obligation (Article 47(2) of the Act), the specific exemptions for generating CMUs which, inter alia, for the purposes of obtaining a capacity agreement, can be satisfied with a relatively easy to pass demand reduction ability test at the latest one month before the first delivery period (Article 53(4) and (5) of the Act), or the fact that, in the event of non-performance, those CMUs merely have their capacity agreement reduced to one year, as is the case for modernised generating CMUs (Article 46(2) and (3) of the Act). Similarly, Tempus has failed to challenge the Republic of Poland’s argument that DSR CMUs are not required to provide collateral before the conclusion of the capacity agreement.

188    Consequently, Tempus has not established that the Commission should have had doubts as to possible discrimination of DSR CMUs in the light of the conditions governing the termination of the capacity agreements of new generating CMUs (see, also, recital 45 of the contested decision).

189    The third complaint must therefore be rejected as unfounded.

–       The fourth complaint

190    By the fourth complaint, Tempus submits that unconfirmed DSR CMUs suffer discrimination on account of the relatively high collateral, namely EUR 10/kW, payable by them and by new generating CMUs, which the Commission failed to analyse properly. The combination of the 2 MW threshold for participation in auctions and that collateral means that there is a minimum bid bond of EUR 20 000 per entry, which represents a significant barrier for a new market entrant subject to the automatic requirement to pay full collateral rather than only in proportion to the unconfirmed DSR. That amount is twice as high as that applied under the United Kingdom capacity mechanism and means that the revenue stream available to DSR operators, spread over one year, is not sufficient to raise capital to cover the cost of participating in the bidding process. Tempus states, in essence, that alternatives to paying the collateral, such as an investment grade financial rating or a parent company guarantee, do not remove those barriers to entry, but benefit only vertically integrated incumbents, and should have caused the Commission doubts.

191    The Commission, supported by the interveners, counters, in essence, that the introduction of collateral to the Polish capacity market has been driven by the uncertainty of offered capacity. That collateral mitigates the risk, for PSE, that contracted capacity will not be delivered during the delivery year, which could lead to a failure to meet the reliability standard at issue and the involuntary disconnection of consumers. There is no barrier to entry for unconfirmed DSR CMUs, provisionally certified (see recital 28(a) of the contested decision), in view of their minimal size and the required bid bond. The aid scheme enables the aggregation of DSR CMUs, thereby easing the reaching of the threshold (recital 28 of the contested decision). Moreover, there are several alternatives to the payment of the bid bond, which were specifically defined for unconfirmed DSR CMUs, namely the possibility of showing an investment grade financial rating, a cash deposit, a bank or insurance guarantee, or a guarantee from the parent company, provided that it is rated investment grade (recital 28 of the contested decision). Furthermore, aggregated DSR CMUs can be provisionally certified without having to provide the detailed information required for the main certification with respect to each single physical unit and metering point, that information being required only at the time of testing (see recital 28(c) of the contested decision). The amount of collateral will be reduced in proportion to the percentage of physical metering points provided to the operator (see recital 28(d) of the contested decision). The Republic of Poland adds that, where the same DSR CMU participates in multiple auctions, no collateral is required for the second and subsequent auctions if the collateral established for the purposes of the first auction remains valid. That solution does not apply to generating CMUs, which are required to establish separate collateral for each capacity auction in which they participate. In any event, according to the Commission, the collateral required of new generating CMUs is of the same amount (EUR 10/kW or PLN 43/kW), such that no discrimination could be observed (recital 45 of the contested decision). The Republic of Poland states that the amount of collateral is the same for all sources which do not yet exist or which have not yet proved their ability to meet the capacity obligation, including new generating CMUs and unconfirmed DSR CMUs. However, no collateral is required for confirmed DSR CMUs. In addition, the required level of collateral is comparable to other cases where collateral is required in the Polish electricity market, such as advance fees for connecting new generation sources (PLN 30/kW), for the participation of existing sources (PLN 30/kW) or for the participation of new sources in RES auctions (PLN 60/kW). Finally, the Commission questions whether, in the specific case of the Polish capacity market, a threshold of 2 MW represents an insurmountable barrier to entry for small DSR operators and points out that, even if that were the case, the mechanism enables aggregation to overcome the threshold. PGE and the Republic of Poland add, in essence, that, at the time of the adoption of the contested decision, the 2 MW threshold did not prevent effective DSR participation in the Polish capacity mechanism. By the time the aid scheme was notified, the Commission had already assessed significantly higher minimum entry thresholds.

192    In the present complaint, Tempus fails to call into question, in a sufficiently substantiated manner, the very detailed considerations set out in recitals 28 and 45 of the contested decision relating to the criteria governing CMU certification, including that of DSR CMUs, to which the requirement to provide bid bonds is linked. In view of the fact, which is not disputed as such by Tempus, that, on the one hand, that requirement is a key element in ensuring the proper functioning of the Polish capacity mechanism and, on the other hand, that collateral in the amount of EUR 10/kW is the same for both new generating CMUs and DSR CMUs, Tempus cannot claim unequal treatment to its detriment. In particular, it has not established that DSR operators, as the case may be, following aggregation as DSR CMUs, would not be able to meet the alternative options for providing the required collateral, which are specifically designed to facilitate their participation (recital 28, first sentence, and recital 28(a) of the contested decision), including proof of an investment grade financial rating, a bank guarantee, an insurance guarantee or a guarantee from the parent company, provided that it is rated investment grade (recital 28(b) of the contested decision). Furthermore, in so far as Tempus claims that that collateral is excessive, in particular, in comparison with that required under the United Kingdom capacity market, and that its combination with the horizontally applicable minimum capacity requirement of 2 MW equates to a barrier to entry which is difficult to overcome for DSR operators alone, those arguments are unsubstantiated and ineffective in the context of the complaint alleging possible discrimination, but rather relate to the possible inappropriate or disproportionate nature of the aid scheme (see paragraph 287 et seq. below).

193    Accordingly, the fourth complaint is not capable of establishing that the Commission should have entertained doubts as to possible discrimination against DSR CMUs in relation to the criteria governing collateral and must therefore be rejected as unfounded.

–       The fifth complaint

194    In the fifth complaint, Tempus complains that the Commission erred in its assessment of the obligation to open the Polish capacity market to the largest variety of capacity providers of all technologies, in accordance with the objectives set out in paragraphs (226) and (232)(a) of the Guidelines. In the light of the ‘corrective availability factor’, which is determined based on historical data from the last five years, the Polish capacity mechanism disadvantages wind and solar generating CMUs, the intermittent nature of those energy sources being particularly well suited to flexible demand management and storage. According to Tempus, by virtue of Article 18 of the Act, the maximum number of capacity obligations that a CMU may bid in a certification application must be specified taking into account the said factor indicated for particular groups of technology, which must be specified annually by ordinance of the Polish Minister of Energy. Those factors laid down in the first ordinance, preceding the certification process for the first three auctions, were only 10.94% for onshore wind energy and 2.07% for photovoltaic energy, whereas no specific factor was provided for offshore wind energy, which thus means, in theory, that a ‘general’ factor of 87.76% should apply. In addition, Article 18(4) of the Act provides that, if a CMU is composed of a group of physical units belonging to different groups of power supply technology, the corrective availability factor applied to that CMU shall be equal to the lowest factor applied to the physical units of which it is composed. Therefore, in case of a CMU aggregating photovoltaic installations and battery storage, that factor would be 2.07%, even though the same factor for battery alone is as high as 96.11%. Tempus concludes from this that the Commission should have had doubts as to the appropriateness of the Act with respect to the principle of technology neutrality and the obvious discrimination of DSR.

195    The Commission, supported by the interveners, counters, in essence, that, according to common practice in capacity mechanisms, the ‘de-rating’ or ‘corrective availability factor’ is essential to enable a fair selection between many different types of capacity capable of contributing to the security of supply objective. Without de-rating, 1 MW of installed wind or solar capacity would be considered equivalent to 1 MW of installed gas or biomass capacity, which is not the case. Gas and biomass power plants can usually generate as long as they are supplied with fuel, which is reflected in a de-rating factor exceeding 80%. By contrast, wind and solar capacity can generate only if the wind or sun is sufficiently strong at the point in time that generation is required, such that they are usually allocated a de-rating factor higher than 15%. That differentiation stems from the need to recognise the contribution to security of supply of many different capacity sources or types and to enable them to participate in capacity mechanisms. The Republic of Poland and PGE state that the lower coefficients assigned to intermittent RES have no impact on the operation of DSR providers, since the coefficient value for DSR CMUs is the highest possible, namely 1 (100%), and similar to that previously calculated by the Commission under the Italian capacity mechanism.

196    It is apparent from paragraph (226) of the Guidelines, inter alia, that ‘the [aid] 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’ and that 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’. Similarly, under the heading ‘Avoidance of undue negative effects on competition and trade’, paragraph (232)(a) of the Guidelines states that that 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’.

197    As the Commission and the interveners submit, those provisions are an expression of the principle of technological neutrality, which requires that a capacity mechanism not unilaterally favour a particular energy supply or generation technology, including one based on fossil fuel or RES (see paragraph 90 et seq. above and paragraph 205 below). It is in that spirit of technological neutrality that, in order to address a capacity adequacy problem, those provisions require the creation of appropriate incentives to make greater use of technologies that are substitutable for equivalent technical qualities, such as DSR, interconnections and storage solutions.

198    Tempus has not demonstrated, however, that the Commission should have entertained doubts as to the compatibility of the Polish capacity market with those criteria.

199    In so far as Tempus criticises the Commission for accepting the content of Article 18 of the Act in that it recognises the application of a ‘corrective availability factor’ to certain technologies, it has not been able to call into question the plausibility of the Commission’s explanation that the application of such a factor is widely recognised and customary in the energy sector, essentially on the ground that the generation of intermittent energy from wind or solar installations does not have the same degree of reliability in terms of ensuring security of supply on a continuous basis as does generation capacity based on gas or biomass. Therefore, the fact that, for that reason, the Polish authorities apply separate ‘corrective availability factors’ to those different energy production technologies and that those factors are lower for wind or photovoltaic generation capacity than for other conventional generation capacities cannot in itself be regarded as contrary to the principle of technological neutrality, as is expressed in paragraphs (226) and (232)(a) of the Guidelines, as long as it is established that all those generation capacities may contribute to the main objective of the capacity mechanism of ensuring security of supply.

200    Furthermore, even assuming that DSR CMUs primarily use RES generation capacity, the ‘corrective availability factor’ of which is relatively low – which is not established – Tempus does not explain the reasons for which the application of that factor would adversely affect their fair participation in the Polish capacity market compared to other generating CMUs. As the Republic of Poland submits, without being contradicted by Tempus on that point, Tempus disregards the fact that the capacity provided by DSR CMUs is given the higher factor of 100%, irrespective of the energy sources they use. In those circumstances, it is also not substantiated that Article 18 of the Act, in providing for the application of the ‘corrective availability factor’, implies unequal treatment to the detriment of DSR operators or that the Commission should have entertained doubts in that respect.

201    Consequently, the fifth complaint must be rejected, as must, accordingly, the second subsection in its entirety.

(iii) Third subsection: incorrect assessment of the sufficiency of the incentives for new market participants

202    By the third subsection, Tempus essentially criticises the Commission for having failed to assess the impossibility of triggering any incentives and cost-effective results in view of the fact that, in particular, coal generating CMUs can obtain capacity agreements for a maximum duration of 17 years, even though their investment decision dates back to 2014. The CAPEX limits referred to in recital 45 of the contested decision will induce fossil fuel-based generating CMUs with such multi-year agreements to reach those limits in order to avoid penalties, thereby discouraging them from being cost-effective since they will be able to access better refinancing terms later on. That is contrary to the principle that the amount of aid granted must be kept to a minimum, increases the financial burden on consumers, and has a discriminatory and penalising effect on cost-effective, lower carbon solutions. According to Tempus, the objective of competitive price discovery, which underpins paragraph (229) of the Guidelines, means that the most cost-efficient technologies are at an advantage in order to ensure the participation of resources providing the best solution at the lowest cost to customers. On the other hand, the long-term capacity agreements cannot be described as adequate incentives for new entrants without due consideration of the unintended consequences for price discovery and competition, which casts doubt on the appropriateness of the aid scheme within the meaning of paragraph (226) of the Guidelines. Tempus states that the reverse engineering of the main auction process so that the most expensive technologies can spread their bids over periods of up to 15 years in order to bid on an annual basis against another, more cost-efficient technology with a one-year revenue stream does not contribute to competitive pricing, but amounts to rigging competition. A truly competitive process would place more expensive solutions at a disadvantage so that the needs identified are met, as far as possible, by cheaper – and, in the case of DSR, greener – resources, resorting to more expensive and polluting options where necessary to top up to the required capacity volume. However, the Polish capacity mechanism, based on the auction principle ‘pay as clear’ with one price, combined with the discriminatory length of capacity agreements, prevents this.

203    The Commission, supported by the interveners, counters that the introduction of the CAPEX milestones is intended to ensure that new generating CMUs do not delay the construction of the units in order to meet the delivery year planned for each specific unit, and not to penalise CMUs which do not reach the planned level of CAPEX. The Polish capacity market provides adequate incentives for new entrants by making available to them long-term capacity agreements – up to 17 years – which enable them to compensate part of their CAPEX with capacity payments and to secure more favourable financing on the market, without however guaranteeing them a complete cost advantage in capacity auctions compared to technologies that require only a low level of investment, such as the majority of DSR CMUs. The Commission stresses that paragraph (226) of the Guidelines does not require equal treatment of all kinds of capacity providers but only open access and adequate incentives, which the Polish capacity market provides to DSR CMUs, even if those incentives are not identical to those provided for generating CMUs. Moreover, Tempus has failed to establish that the competitive bidding process and the capacity agreement lengths available were not capable of achieving the reliability standard at issue.

204    It must be stated that the line of argument under the present complaint is not only vague and ambiguous, but also overlaps substantially with other complaints that have already been rejected, in particular those relating to the CAPEX criteria and to the differentiation of the length of capacity agreements (see paragraph 165 et seq. above), without however specifying them or elaborating on them. In essence, Tempus vaguely alleges that the CAPEX criteria are ineffective and contrary to paragraphs (226) and (229) of the Guidelines, in that they offer neither a sufficient incentive for the participation of CMUs operating from alternative, more cost-efficient and less polluting technologies, nor a level playing field for those technologies in the bidding process, the results of which are necessarily skewed in favour of fossil fuel-based generating CMUs and to the detriment of the consumer who must pay more for energy.

205    As has already been found in paragraph 197 above, paragraph (226) of the Guidelines is an expression of the principle of technological neutrality. In addition, it is necessary to recall the considerations set out in paragraph 165 et seq. above, according to which Tempus has not succeeded in demonstrating that the CAPEX criteria governing the length of capacity agreements were liable to lead to unequal treatment of, inter alia, DSR operators. The Commission is right to observe that those criteria are intended to encourage investment in the quicker construction of new generation capacity or in the modernisation of old capacity required in order to achieve the objective of security of supply pursued by the Polish capacity market. As such, the CAPEX criteria cannot therefore be deemed contrary to the principle of technological neutrality, provided that it is common ground that other technologies, including DSR, are offered adequate incentives and opportunities to participate in the said market, which is the case, and that their potential to provide capacity is not sufficient to achieve the said objective (see paragraphs 103 to 107 and paragraph 168 et seq. above). In that regard, Tempus cannot call into question the fact that generating CMUs attempt to make their investment profitable or at least offset the costs of producing a new power plant, given that that aspect is an integral part of the economic incentives underlying the CAPEX criteria and the differentiation of the length of capacity agreements.

206    In addition, Tempus does not dispute the fact that DSR operators are allowed to participate in capacity auctions, albeit only for the grant of capacity agreements of a maximum duration of five years, taking into account their lower CAPEX, which is not contrary to the principle of equal treatment (see paragraph 165 et seq. above). Nor is Tempus justified in arguing, therefore, that those criteria and those governing access to auctions and their conduct are contrary to paragraph (229) of the Guidelines, according to which ‘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’. After all, in doing so, it merely extrapolates its unfounded reasoning in the context of the ‘proportionality’ of the aid scheme (see the title of Section 3.9.5 preceding paragraphs (228) and (229) of the Guidelines), without however adding anything capable of calling into question the considerations set out in paragraph 290 et seq. below. Lastly, Tempus makes only vague claims that increasing the participation in the Polish capacity market of other technologies, such as DSR, would produce more competitive, efficient and cost-efficient results, but fails to produce any evidence to support these claims, the plausibility of which is seriously called into question in light of the limited prospects of that technology to achieve the reliability standard at issue (see paragraph 103 et seq. above) and, therefore, to exert sufficient competitive pressure on generating CMUs.

207    Accordingly, Tempus has not established that the Commission should have entertained doubts as to the compatibility of the aid scheme in the light of paragraphs (228) and (229) of the Guidelines.

208    Consequently, the third subsection must be rejected as unfounded.

(iv) Fourth subsection: insufficient participation of foreign capacity

209    In support of the fourth subsection, first, in the light of paragraph (232)(b) and paragraph (233)(a) of the Guidelines, Tempus criticises the Commission for not sufficiently assessing the additional hurdles and limitations on the admission of foreign DSR operators to auctions (see recitals 51 to 73 and 165 of the contested decision). It states that, contrary to what is outlined in recital 190 of that decision, namely that foreign CMUs can participate directly in main auctions only as from 2020 (for delivery in 2025) and in additional auctions as from 2024, such a requirement does not appear in the Act. Moreover, foreign CMUs can obtain only one-year capacity agreements, for which they compete with Polish CMUs, and receive only the price of the highest bid at their respective border. Furthermore, foreign CMUs, including DSR CMUs, have to participate in pre-auctions in order to be eligible for main auctions and pay a collateral of EUR 10/kW, whereas Polish DSR CMUs are exempted from that if they can show an investment grade financial rating (recital 28(b) of the contested decision). Tempus adds that, if a Polish CMU is unsuccessful in the auction, the collateral is paid back. Successful bidders have to pass the main certification procedure, failing which they lose the collateral proportional to the capacity volume that has not received that certification, while foreign DSR CMUs have to pass the ‘DSR-test’ or pay an additional collateral of up to EUR 10/kW until their certification is proved. Tempus concludes from this that the Polish capacity mechanism disadvantages foreign generating and DSR CMUs.

210    Second, Tempus takes the view that the Commission failed to carry out an adequate assessment and should have had doubts as to the appropriateness of the methodology used to open an interim pathway towards the participation of foreign CMUs in the Polish capacity market. According to Tempus, until the target solution providing for the direct participation of those CMUs in auctions is implemented, which is dependent on the conclusion of agreements between the TSOs (recital 51 of the contested decision), in the context of the transitory solution, TSOs on neighbouring markets act as gatekeeper of CMUs. Thus, German DSR operators are forced to bid via the German TSO. This creates a serious conflict of interest, since that TSO is incentivised to protect German resource adequacy first and foremost and not to reach a timely agreement with PSE, which would cause it to lose its role as gatekeeper of German CMUs participating in the auctions of the Polish electricity market. In particular, a DSR CMU may not be the preferred choice of such a TSO, since this reduces grid constraints and, therefore, the business case for transmission capacity expansion.

211    The Commission neither discussed potential conflicts of interest in respect of the role of TSOs in the transitory solution and how the Act provides for means to overcome them, nor called into question their legal validity. By provisionally authorising that solution, it allowed a breach of the unbundling rules for TSOs under the new EU electricity market design rules. Thus, it endorsed the provision of the Act pursuant to which a foreign TSO or interconnector, such as 50 Hertz from Germany, whose ownership structures have been unbundled, is entitled under the Polish capacity mechanism to perform the role of generator and that of DSR operator simultaneously. A TSO which operates an interconnector can thus actively participate in the Polish auctions, which is contrary to EU rules under which a TSO or interconnector can be responsible only for safeguarding balance, ensuring non-discriminatory access to the grid, enforcing balancing positions and organising markets so as to provide energy in order to maintain that balance. By contrast, such a TSO cannot act as an energy provider or as a DSR operator, even in an interim phase until at least 2022. Similarly, the Commission authorised the rule that, during the first phase, in the absence of possibilities for foreign DSR to participate in auctions, the foreign TSO or interconnector is invited to play the role of DSR under the Polish capacity mechanism, which is a violation of the strict unbundling rules under EU law, such as Article 9 of Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC (OJ 2009 L 211, p. 55).

212    The Commission and the interveners counter that neither paragraph (226) of the Guidelines nor its other provisions under Section 3.9 require that the conditions for participation in the capacity market be identical for domestic and foreign CMUs (see, in particular, paragraph (232)(b) of the Guidelines). The participation of foreign CMUs in the Polish capacity market in accordance with those provisions is ensured by the target and transitory solutions (see recitals 73 and 165 of the contested decision). Like its approach in other cases and in the sector inquiry, in essence, the transitory solution is necessary to address the complex problem of sufficient participation of foreign capacity to address the competition distortions caused by capacity mechanisms while avoiding the creation of further distortions, in particular as regards the signals for cross-border trade and fair use of cross-border transmission.

213    PGE notes that participation in the Polish capacity market by foreign capacity via the target solution cannot be implemented immediately due to the need to overcome objective administrative burdens, including the conclusion of necessary agreements between the respective TSOs. The transitory solution, the first additional auction of which will be held in 2020 and the last in 2023, allows for cross-border participation much sooner (recital 51 of the contested decision) and, in any case, before the target solution is scheduled to be applied in 2025 (the first delivery year) on the basis of the main auction conducted in 2020 (recitals 68 to 73 of the contested decision). As a result of the transitory solution, the first delivery year of foreign capacity via interconnectors will be the same as that of domestic capacity – namely, 2021 – and there is therefore no delay in the integration of foreign capacity into the aid scheme. That solution therefore not only allows foreign operators to gain access to the Polish electricity market, but also ensures that Poland has enough time to implement a comprehensive target solution.

214    The Commission, supported by the interveners, disputes that TSOs are exposed to a conflict of interest. Since the transitory solution applies from the first delivery year, Tempus’s criticism must refer only to the target solution, in which TSOs are supposed to play the alleged role of ‘gatekeeper’. In any case, the German TSO has a huge interest in cooperating with PSE, the Polish TSO, to ensure a successful participation of German capacity in the Polish capacity market. First, such participation will provide them with additional revenue and will increase security of supply in Germany and thus align with the German TSO’s objective. Second, it will lead to revenues derived from congestion, which will accrue to the German TSO and to German consumers, representing the value of interconnection between Poland and Germany for Polish security of supply (recitals 65 and 67 of the contested decision). Furthermore, the participation of German capacity will not alter or affect market coupling, as required by paragraph (232)(d) of the Guidelines. The conditions of the Polish capacity market are actually aimed at achieving the opposite result, that is to say, the overall liberalisation of the market. The Republic of Poland and PGE state that foreign capacity providers continue to provide their services and supply electricity and capacity on the domestic market. They thereby affect generation adequacy in their power grid and support, at the same time, capacity adequacy in the Polish grid as a result of cross-border trade. In addition, Tempus is wrong to claim that the provision of DSR services during stress periods is disadvantageous for TSOs, because that limits the need to expand the grid infrastructure. DSR providers reduce demand for grid electricity only during periods of shortages which at the same time results in higher prices in the electricity market. In other periods, they run their customers normally, which requires an adequate availability of grid infrastructure. Indeed, the situation described by Tempus in which DSR contributes to the reduction of demand for grid infrastructure can occur only where users consistently and continuously use DSR methods in response to pricing signals from the electricity market. Moreover, it is impossible to separate completely the activities of electricity generators and DSR operators from those of TSOs, since the latter are actively involved in, and necessary for, many electricity market processes, including, in particular, cross-border trade. Finally, Article 68(7) and (8) and Article 73(e) of Commission Regulation (EU) 2015/1222 of 24 July 2015 establishing a guideline on capacity allocation and congestion management (OJ 2015 L 197, p. 24) specifically stipulate that TSOs may obtain congestion income for allocating capacities in the electricity market.

215    According to the Commission, the argument that DSR CMUs may not be the preferred choice for a TSO is ineffective in the transitory solution, which provides for the participation of interconnection capacity, and not foreign capacity. In any event, it is for PSE, the Polish TSO, to identify the foreign capacity that can participate in the auctions of the Polish capacity market, based on pre-auctions. Their selection is decided solely by the price, and successful bidders must be certified in the same way as domestic participants (recitals 53, 57 and 59 of the contested decision). The only additional information required is the neighbouring TSO’s commitment to provide availability information for each foreign capacity unit, which is essential in order to enable PSE to assess whether those units have delivered the required capacity. The Republic of Poland states, in essence, that the annual forecasts of maximum capacity obligation volumes for individual zones with a synchronous profile, including those covering Germany, the Czech Republic, Slovakia and separate zones assigned to the interconnectors with Lithuania and Sweden, are the most important parameter determining the share of foreign capacity in the Polish capacity market. Since those maximum capacity obligation volumes required may change over time, it was decided that units located outside the Polish power grid would be able to conclude capacity agreements for no more than one delivery year. Otherwise, the volume of foreign capacity contracted would exceed the maximum capacity obligation volumes for individual zones in subsequent years. As a result, on the one hand, other capacity suppliers from outside the Polish power grid would be deprived of the possibility of participating in capacity auctions, which would reduce competition, and, on the other hand, there would be a significant risk that capacity contracted outside the Polish power grid would not affect generation adequacy in that power grid. This would exacerbate the risk of capacity shortages in the Polish power grid or make it necessary to conclude agreements for the same capacity again, for instance within Poland or in another zone.

216    The Commission adds that the fact that interconnector CMUs are directly eligible to participate in the Polish capacity market is perfectly in line with paragraph (226) of the Guidelines. In any event, the capacity offered by interconnectors will consist only of foreign capacity which cannot participate directly in the capacity market during the transitory solution. Moreover, the Polish capacity market does not have any tools at its disposal to force foreign interconnectors to participate in its mechanism and ‘the same also applies to the [contested] decision’.

217    In response to the General Court’s written question referred to in paragraph 25 above, the Commission, Enel X, PGE and the Republic of Poland state, in essence, that Article 9(1) of Directive 2009/72 does not preclude the transitory solution, as assessed in recitals 68 to 73 of the contested decision, in the context of which TSOs merely carry out tasks, which they are required to do also pursuant to Article 12 of that directive, including that of managing electricity flows on the system, taking into account exchanges with other interconnected systems, that of ensuring, to that end, a secure, reliable and efficient electricity system, that of granting and managing third-party access to the system, of ensuring non-discrimination between system users, and that of collecting congestion charges and payments under the operator compensation mechanism. As with those tasks, in the transitory solution, in order to facilitate access to the TSO or interconnector of a Member State, such as PSE in the present case, foreign TSOs, after having estimated the volume of capacity that can be supplied through the interconnector, participate in the auctions of the Polish capacity mechanism by submitting a corresponding offer and ensure, subsequently, non-discriminatory access for generating and DSR providers to the capacity won (recital 68 of the contested decision). In addition, in accordance with Article 16(6) of Regulation (EC) No 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 (OJ 2009 L 211, p. 15), any revenues generated by the sale of capacity volumes thus won are used by those TSOs to ensure the availability of allocated transmission capacity and to increase interconnection capacity (see recital 65 of the contested decision). That assessment is confirmed a posteriori by Regulation 2019/943 and by Directive (EU) 2019/944 of the European Parliament and of the Council of 5 June 2019 on common rules for the internal market for electricity and amending Directive 2012/27/EU (OJ 2019 L 158, p. 125).

218    As a preliminary point, it should be borne in mind that the first part of the present subsection overlaps with the first complaint of the second subsection in so far as Tempus alleges discrimination between Polish and foreign DSR operators (see paragraph 152 above).

219    The question of the fair and non-discriminatory access of foreign capacity, including DSR, to a national capacity market is regulated, specifically, in Section 3.9.6, under the heading ‘Avoidance of undue negative effects on competition and trade’, and, in particular, in paragraph (232)(b) of the Guidelines, according to which ‘the [aid] 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 operators from other Member States where such participation is physically possible in particular in the regional context, that is to say, where the capacity can be physically provided to the Member State implementing the measure and the obligations set out in the measure can be enforced’. It follows that a Member State which introduces a capacity market is not supposed to open that market to foreign capacity immediately and treat it on an equal footing with domestic capacity, but is obliged only to make its access possible in so far as this is necessary in order to address a generation adequacy problem and ‘where such participation is physically possible’, in particular ‘where … the obligations set out in the [aid scheme] can be enforced’. Nor does paragraph (233)(a) of the Guidelines lay down a requirement to treat domestic and foreign capacity perfectly equally, but merely requires, in the negative, that the aid scheme not reduce ‘incentives to invest in interconnection capacity’.

220    It is in the light of those requirements for the gradual opening of the domestic capacity market that it should be examined whether Tempus’s challenge to the introduction by the Polish State of the transitory and target solutions (see Article 6 of the Act), which seek to implement those requirements, should have raised doubts on the part of the Commission.

221    The rules of the aid scheme permitting the participation of foreign capacity are set out in recitals 51 to 88 of the contested decision and are assessed in recitals 165 and 190 thereof.

222    The target solution is described in detail in recitals 53 to 67 of the contested decision. It is stated there, inter alia, that, according to the Polish authorities, its implementation assumes that the neighbouring TSOs will conclude cooperation agreements for each border for issues related to the certification of foreign capacity, which may take several years. For that reason, those authorities proposed to apply a transitory solution allowing cross-border participation via interconnectors, instead of foreign capacity providers, which would be allowed to bid in the capacity auctions (recitals 51, 66 to 68 and 165 of the contested decision).

223    As far as the transitory solution is concerned, recitals 68 to 73 of the contested decision state, inter alia, that the participation of interconnectors for each border (five in total) also requires the conclusion of inter-TSO agreements, although simpler ones than those provided for in the target solution. The Polish authorities committed to implementing the transitory solution as of the first delivery year, namely 2021, by opening an additional auction to the participation of interconnectors and by setting aside enough capacity for that additional auction, namely at least 1 160 MW (recitals 73 and 83 of the contested decision).

224    Regarding the first part of the fourth subsection, it follows from paragraphs (232)(b) and (233)(a) of the Guidelines that the complaint alleging discrimination between Polish and foreign CMUs, including Polish and foreign DSR operators, on account of the rules governing the target solution, is based on a false premiss and cannot succeed. In the light of the objectives of the Polish capacity mechanism and of those set out in the Guidelines, including that of provisionally restricting the access of foreign capacity within the limit of what is both necessary to ensure generation adequacy and physically possible, those CMUs and operators are in different factual and legal situations, such that the argument alleging their unequal treatment is unfounded.

225    That finding is sufficient to reject as unfounded the first part of the fourth subsection, as well as the first complaint of the second subsection, without it being necessary to assess more precisely the criteria governing the access of foreign CMUs to the Polish capacity market under the target solution, the implementation of which is in any event delayed by several years (see recitals 53 to 67 of the contested decision), or the detailed arguments of the Commission and of the interveners seeking to justify that solution. After all, in view of those different factual and legal situations, the Commission cannot be criticised for having accepted rules providing for different treatment of foreign CMUs in order to regulate their progressive access to that market within the limits of what appeared necessary to address a generation adequacy problem and what was physically possible, within the meaning of the abovementioned provisions. It is therefore not necessary to assess separately and anew each of the sub-complaints, some of which have already been rejected, alleging such unequal treatment, including those based on the lack of immediate participation of foreign CMUs in main auctions, on their obligation to participate in pre-auctions and to pay a collateral of EUR 10/kW, and on the fact that they can obtain only one-year capacity agreements (see paragraphs 165 et seq. and 192 above).

226    By the second part of the fourth subsection, Tempus criticises the Commission for having provisionally accepted the transitory solution, although it is both inappropriate and unlawful. First, as ‘gatekeepers’ of CMUs participating in the auctions of the Polish electricity market, neighbouring TSOs are faced with a serious conflict of interest, in that they are incentivised to protect the resource adequacy of their domestic grid or market first and foremost and not to reach a timely agreement with PSE, which would cause them to lose that role as gatekeeper. Second, a DSR CMU may not be the preferred choice of a TSO, since DSR reduces constraints on that grid and, therefore, the business case for capacity transfer expansion. Third, the Commission allowed a breach of the EU rules governing the unbundling of TSOs, in so far as an unbundled TSO cannot, even in an interim phase, simultaneously act as a producer and as a DSR operator actively participating in auctions of the Polish capacity mechanism.

227    As a preliminary point, it should be noted that Tempus has not disputed PGE’s plausible explanations according to which, in essence, the transitory solution, the first additional auction of which was to be held in 2020 and the last in 2023, allows for cross-border participation – albeit limited – of foreign capacity at a later stage than under the target solution, the implementation of which depends on the as yet uncertain conclusion of cooperation agreements between PSE and neighbouring TSOs, and thus ensures that the first year of delivery of foreign capacity via interconnectors is the same as that of national capacity, namely 2021 (see, also, recital 73 of the contested decision). Nor does Tempus dispute, in that regard, the need for neighbouring national TSOs to conclude cooperation agreements with each other, which is therefore not the sole responsibility of the Polish State or of PSE. Furthermore, Tempus has failed to call into question the validity of the de-rating methodology used for foreign participation and accepted by the Commission as part of the transitory solution, including of the commitments offered by the Polish authorities to that effect, such as their obligation to re-notify that methodology (see recitals 74 to 88 of the contested decision).

228    As regards the TSOs’ alleged role as ‘gatekeeper’ of the CMUs, it is sufficient to note that Tempus’s first argument, based on the TSOs’ alleged conflict of interest, is neither plausible nor substantiated, in view of the detailed explanations to the contrary provided by the Commission and the interveners, including in response to the General Court’s written question. Thus, it seems illogical, or even contradictory, to consider that a foreign TSO, in its capacity as interconnector subject to the obligations laid down in Article 12 of Directive 2009/72 and as a bidder participating in the auctions of the Polish capacity mechanism, does not seek to facilitate the participation of foreign capacity in the Polish capacity market. On the contrary, the volume of capacity obtained by that TSO in those auctions is specifically intended to be reserved for that capacity whose non-discriminatory access to that volume must be ensured (Article 12(d) and (f) of that directive, read in conjunction with Article 2(18) thereof). In addition, the revenues from such participation are liable to strengthen the market position of that foreign capacity in the interest both of security of supply on interconnected markets and of the secure, reliable and efficient management of electricity flows on the system, taking into account exchanges with other interconnected systems, in accordance with the objective referred to in Article 12(d) of Directive 2009/72. Similarly, by increasing cross-border trade which is liable to influence generation adequacy in interconnected systems in a positive manner, the participation of foreign capacity via a foreign TSO is in fact beneficial to the market coupling referred to in paragraph (232)(d) of the Guidelines and to the liberalisation of the internal market for electricity, and not the other way round. In any event, Tempus has not been able to establish that such participation would prevent or reduce the contribution of foreign capacities to security of supply on their own systems or on their domestic markets.

229    The first argument therefore cannot be accepted.

230    It must be held that Tempus’s vague and succinct second argument, according to which a DSR CMU may not be the preferred choice of a TSO since DSR reduces grid constraints and, therefore, the business case for capacity transfer expansion, is not convincing and cannot succeed. First, the argument relating to the saving of capacity on the grid, which is therefore available for export in cross-border areas, contradicts Tempus’s first argument, as rejected in paragraphs 228 and 229 above, according to which TSOs are incentivised first and foremost to protect the resource adequacy of their domestic grid or market. Second, discrimination between generation capacity and DSR operators as regards access to the capacity volume won by a foreign TSO at the auctions of the Polish capacity mechanism is expressly prohibited by Article 12(f) of Directive 2009/72, read in conjunction with Article 2(18) thereof. Third, the Republic of Poland and PGE have contended, in a plausible manner, that DSR does not affect the need to expand grid infrastructure since it reduces demand for grid electricity only in periods of shortages, which at the same time results in relatively high prices in the electricity market, and that, in other periods, DSR operators run their customers normally, for which an adequate availability of grid infrastructure is thus required. Indeed, Tempus does not allege that DSR contributes to the reduction of the demand for grid infrastructure in the event of stable and continuous use of its methods by customers, irrespective of pricing signals from the electricity market. Such a reduction in infrastructure needs does not appear to be the aim of their current business model, which could explain, in part, their limited potential on the Polish capacity market (see paragraph 103 et seq. above). Last, as those interveners correctly point out, it is not possible to separate completely electricity generators and DSR operators from the activities of TSOs, particularly in the context of cross-border exchanges.

231    The second argument must therefore also be rejected as unfounded.

232    As regards Tempus’s third argument, alleging breach of the EU rules governing the unbundling of TSOs and of the alleged prohibition on simultaneously acting as producer and DSR operator participating in auctions of the Polish capacity mechanism, it is apparent from the detailed observations of the Commission, Enel X, PGE and the Republic of Poland in response to the written question of the General Court, which Tempus has disputed only very vaguely and succinctly, that that argument is based on a false premiss.

233    The unbundling obligation laid down in Article 9(1) of Directive 2009/72, by virtue of which Member States are required to unbundle transmission systems and the operation of generation and supply activities in their national energy markets, is without prejudice to the tasks of the TSOs, as interconnectors, under Article 12 of the same directive, which correspond to those entrusted to PSE and to foreign TSOs under the transitory solution (see paragraph 228 above). Moreover, the Commission and those interveners have explained, convincingly, that, in that capacity, those TSOs merely act as intermediary and facilitator enabling foreign capacity to have non-discriminatory access to the volume of capacity won by a foreign TSO in the auctions of the Polish capacity mechanism. Therefore, contrary to what Tempus claims, for a TSO, there is no confusion between its status as grid operator and interconnector, on the one hand, and the activities of generation or supply of electricity of generation capacity and DSR capacity, on the other.

234    Consequently, the third argument of the second part of the fourth subsection must also be rejected as unfounded, as must that subsection in its entirety.

(v)    Fifth subsection: inappropriateness of the aid scheme for guaranteeing security of supply

235    In the fifth subsection, Tempus complains that the Commission failed to analyse the risk of a security of supply failure and the potential for risk reduction on the Polish electricity market that the Polish capacity mechanism is supposed to reduce. In that regard, it disputes PSE’s analysis of the risks of system deficiencies related to heat waves, cold waves and storms, as well as historical price peaks. According to Tempus, in reality, those price peaks do not demonstrate the existence of an ‘absolute’ market failure, even if demand peaks occur rarely and back-up capacities are used to a low degree, which may lead to the missing money problem (see recital 8 of the contested decision). A technology which can cover infrequent demand peaks in an economically feasible way should therefore have low fixed costs. Instead, the Polish capacity mechanism extends capacity agreements with CMUs with higher CAPEX. During summer heat waves, when demand is much lower than in winter, power plants equipped with steam turbines face severe outages due to the lack of cooling water. Nevertheless, generating CMUs with 15-year capacity agreements will receive reliable payment even if they are unavailable in times of extreme heat waves, which may occur every 5 years, thereby placing all technologies that can reduce that risk at a disadvantage.

236    The Commission replies, in essence, that Tempus is wrong to assume that capacity mechanisms aim to promote the construction of capacity solely in order to deal with consumption peaks. Such mechanisms should ensure that all providers capable of delivering the capacity necessary are able to participate in the auctions in order to limit the aid to the minimum necessary. The qualification criteria are therefore the same for all capacity providers, and any power plant or CMU that fulfils them can participate in the capacity mechanism.

237    As a preliminary point, it should be noted that the line of argument under the present subsection is difficult to understand, or even contradictory, in that it does not call into question the principle according to which the Polish electricity market suffers from a market failure in the form of a missing money problem, which is one of the main justifications for the introduction of the Polish capacity mechanism (see paragraph 119 et seq. above). In essence, Tempus seeks to argue that that failure is not as ‘absolute’ as PSE and the Commission claim and does not justify granting 15-year capacity agreements to certain generating CMUs, given their partial inefficiency during summer months in the event of a cooling water shortage. There is therefore no justification for covering the high fixed costs of those inefficient CMUs by means of capacity payments, when such capacity is not even available.

238    As the Commission contends, that line of argument is a further attempt by Tempus to dissociate a key element, namely the grant of long-term capacity agreements to CMUs with high CAPEX, from the structure and functioning of the Polish capacity mechanism as a whole, which, in view of its main objective of ensuring security of supply over 10 years, is supposed to provide operators with sufficient economic incentives to invest in new generation capacity or to modernise old generation capacity intended to maintain or increase capacity availability (see paragraph 119 et seq. above). However, neither the justification for that approach nor the proportionality of the capacity payments can be called into question on the sole ground that that capacity is, potentially every five years, unavailable for a certain period of the summer due to a cooling water shortage, which may moreover, by virtue of Article 67(1) and (4) of the Act, even give rise to full quarterly reimbursement of those payments. In any case, such an eventuality does not call into question the long-term availability of that capacity, the guarantee of which remains the raison d’être of the Polish capacity mechanism.

239    Therefore, that aspect cannot have given rise to doubts on the part of the Commission, with the result that the fifth subsection must be rejected.

240    In the light of all the foregoing considerations, the second section must therefore be rejected as unfounded in its entirety.

(4)    Third section: alleged incompleteness of the assessment and alleged violation of the incentive effect of the aid

(i)    First subsection: alleged retroactivity of the aid scheme

241    By the first subsection, first, Tempus submits, in essence, that the Commission should have had doubts as to the incentive effect of the aid scheme for existing generating CMUs which are entitled to conclude multi-year capacity agreements (recital 44 of the contested decision). It does not dispute that the taking into account of CAPEX, determined on the basis of the amount of investments envisaged during the five years preceding the delivery year, is, in principle, compatible with the incentive effect criterion, and that that rule is derogated from for the first auctions to be held in 2018 for delivery in 2021. However, the Commission was wrong to approve January 2014 as the starting date on the ground that it occurred ‘a few months after the announcement [in July 2013] by [the Polish authorities of their] plans to introduce a [capacity market]’, since the incentive effect criterion had to be assessed against the date on which the aid measure was granted and not on the date of that announcement. That first announcement cannot be considered a binding and reliable commitment on the part of the Polish authorities to introduce a capacity market in Poland, the characteristics of which were, moreover, redesigned several times until the end of 2016. According to Tempus, the early date of January 2014 was in fact crucial for awarding long-term capacity agreements for coal power plants already operating or under construction. Thus, 15-year capacity agreements that are supposed to be reserved for new entrants have been awarded to coal power plants which, at the date of entry into force of the Polish capacity mechanism, were already at advanced stages of construction or were already in operation. Under paragraph (49) of the Guidelines, however, the assessment of incentive effect requires an examination of whether the aid induces a change in behaviour in the beneficiary which it would not undertake without the aid. That is not the case for investment in new capacities arising from projects that were already under construction or operating before the entry into force of the aid scheme.

242    Second, according to Tempus, the choice of the date of 1 July 2017 as a criterion for differentiating new CMUs from existing CMUs (recital 44 of the contested decision) is arbitrary. The starting date of electricity generation is irrelevant, as long as CMUs deliver capacity pursuant to their agreement. Moreover, the appropriateness of the criterion of that date in relation to the CAPEX eligibility criterion is questionable, particularly in view of the fact that it applies only to the first auctions and only to CMUs eligible for 5-year and 15-year capacity agreements. Moreover, the notion of ‘generating electricity’, interpreted literally, excludes DSR operators, even those eligible for five-year capacity agreements. It is thus not clear in contested decision whether DSR CMUs that have incurred CAPEX since 2014 and that started operating after 1 July 2017 are eligible in the same manner as generating CMUs. If that is not the case, it would be another example of discrimination between generating CMUs and DSR CMUs. Tempus disputes that the date of 1 July 2017 is used in order to avoid potential new projects artificially delaying their development to take advantage of the long-term capacity agreements, as that date is intended specifically to benefit the Kozienice (Poland) plant, which already started operating in December 2017. Tempus recalls that the very point of the incentive effect criterion is that the aid measure incentivises the beneficiaries to undertake operations or investments they would not have undertaken in the absence of aid. The Member State should therefore ensure that a CMU does not undertake investments with a certain level of CAPEX only in view of obtaining capacity agreements. Although paragraph (219) of the Guidelines provides that aid for generation adequacy may be granted in the form of investment or operating aid, the contested decision does not mention which type of aid is concerned by the Polish capacity market. According to Tempus, if that market is deemed to involve operating aid, its retroactive nature does not appear to be consistent with the object of the measure.

243    First, the Commission, supported by PGE and the Republic of Poland, counters that the capacity payments under the capacity agreements are operating aid for rewarding and maintaining availability, and are not intended to cover the investment costs of generating CMUs, with the result that, in the present case, the condition set out in paragraph (50) of the Guidelines is not applicable. According to Section 3.3.4 of the contested decision, by enabling CMUs to cover their costs and penalising them in case of non-delivery, the aid scheme will incentivise existing and new capacities to remain in or enter the market, and to be available at times of scarcity. PGE states in that regard that capacity payments do not correlate with the level of CAPEX, but result from a competitive tender. Second, the choice of the date of 1 January 2014, far from being arbitrary, was closely related to the date of the public announcement of the capacity mechanism by the Polish authorities and was meant to ensure that the relevant CAPEX was determined on the basis of the investment costs that had been incurred five years before the date of the first auction. The postponement of that date by six months compared to the date of that announcement reflects the fact that no investment could have taken place at such short notice following that announcement. Third, the fact that the aid scheme had been redesigned several times since its announcement in 2013 has no effect on the incentive created by that political declaration and the fact that the Polish authorities had identified a problem of capacity adequacy on the Polish electricity market. The capacity investors concerned could not have known that that assessment would undergo modifications according to the rules in the Guidelines that did not even exist in 2013. Fourth, it was crucial to determine the CAPEX eligibility criteria as of 1 January 2014 in order to avoid potential new investments being artificially delayed until the formal introduction of the Polish capacity market granting access to long-term capacity agreements (see recital 44 of the contested decision). Such a delay of investment in new capacity would only increase the adequacy concerns and lead to higher costs. Finally, so far as concerns the date of 1 July 2017, the Commission stresses that there is no rule governing Polish capacity market that stipulates when the construction of a generation plant must start in order to qualify as a new plant. Thus, the contested decision states that the eligible CAPEX is determined on the basis of the investment costs which are expected to be incurred in the five years before the start of the delivery year, as the only decisive cut-off point without any conditions being set for the actual commencement of the investment.

244    PGE observes that, as early as July 2013, the Polish authorities clearly indicated that they were intending to introduce a capacity mechanism in Poland in 2014. The incentive effect of those announcements is evidenced by the fact that significant market players, including PGE, relied on them in making final investment decisions on the development of adequate capacity on the assumption of future capacity market revenues. In view of the first delivery planned for 2021 and the strong signals given by the Polish authorities regarding the start date of the Polish capacity mechanism, it was logical for the Commission to conclude that the CAPEX was to be deemed eligible as from 1 January 2014. According to the Republic of Poland, in essence, by agreeing that CAPEX incurred from 1 January 2014 could be taken into account within the framework of the Polish capacity market and that power plants which started production after 1 July 2017 could be considered new generating units, the Commission avoided different treatment of units that were objectively in the same factual situation, that is to say, generating units whose construction required significant CAPEX. Pursuant to Article 96 of the Act, CAPEX incurred by operators of generating CMU and DSR operators were treated in the same manner as regards their eligibility to bid for capacity agreements with terms exceeding one year.

245    It should be recalled, as a preliminary point, that, as regards the incentive effect of the aid, the provisions of Section 3.2.4 (in particular paragraphs (49) to (52)) of the Guidelines must be taken into consideration. In accordance with paragraph (49) of the Guidelines, the aid scheme can be found compatible with the internal market only if it has an incentive effect. This assumes that ‘the aid induces the beneficiary to change its behaviour to increase the level of environmental protection or to improve the functioning of a secure, affordable and sustainable energy market, a change in behaviour which it would not undertake without the aid’. In addition, it ‘must not subsidise the costs of an activity that an undertaking would anyhow incur and must not compensate for the normal business risk of an economic activity’. In accordance with paragraph (50) of the Guidelines, in essence, aid is considered not to have an incentive effect for the beneficiary – and, therefore, to be incompatible with the internal market – where that beneficiary submitted its aid application to the national authorities after the start of work on the project.

246    As regards paragraph (49) of the Guidelines, it is apparent from the passage ‘to increase the level of environmental protection or to improve the functioning of a secure, affordable and sustainable energy market’ that the incentive effect of aid for generation adequacy may be linked to one or other of those objectives, depending on the balancing exercise which the Member State is supposed to carry out in the light of the general criteria and objectives set out in paragraphs (30), (219) and (220) of those guidelines (see paragraph 89 et seq. above). This is confirmed by paragraph (69) of the Guidelines, which states that ‘environmental and energy aid is considered to be proportionate if the aid amount per beneficiary is limited to the minimum needed to achieve the environmental protection or energy objective aimed for’. Thus, where, as in the case at hand, the Member State pursues above all, by means of a capacity mechanism, the objective of ensuring security of supply and thus ‘improv[ing] the functioning of a secure, affordable and sustainable energy market’, that incentive effect is linked, primarily, to the incentives for operators to make available the generation capacities necessary for that purpose and, only secondarily, to the objective of environmental protection. As has been explained in paragraph 117 et seq. above, in the context of the Polish capacity mechanism, those incentives are based on the CAPEX criteria allowing operators to obtain multi-year capacity agreements, which gives them the necessary stability of revenue, also to finance or recoup their investments in the creation or modernisation of generation capacity.

247    In recital 44 of the contested decision, the merits of which are disputed by Tempus, it is stated, in essence, that the CAPEX eligibility criteria are determined on the basis of the investment costs which are expected in the 5 years before the start of the first delivery year, with the exception of the very first auction, in which CMUs are eligible to obtain 5- or 15-year capacity agreements on the basis of their CAPEX incurred since January 2014, that is to say, since a few months after the announcement by the Polish authorities, in July 2013, of their plans to introduce a capacity mechanism, provided that those CMUs did not start generating electricity before 1 July 2017. This aims at avoiding potential new projects artificially delaying their development to take advantage of long-term contracts.

248    By the present subsection, Tempus calls into question the incentive effect created by the three aforementioned dates. The date of 1 January 2014 is used in Article 96(1) of the Act and in recital 44 of the contested decision in connection with that of July 2013 (first complaint), while the date of 1 July 2017 is mentioned in Article 95 of the Act (second complaint). However, Tempus does not dispute that the CAPEX criteria, including the taking into account of previous investment expenditure over a five-year period, in themselves create such an incentive effect, in that they encourage operators to invest in the creation or modernisation of generation capacity.

249    Specifically, under the first complaint, Tempus merely calls into question the five‑year period before the first auction starting in January 2014, given that the Polish authorities announced their plans to introduce the capacity mechanism in July 2013. First, the announcement of July 2013 was vague and unreliable and cannot be confused with the date on which the aid measure was granted. Second, the date of January 2014 was arbitrarily chosen as being relevant to the grant of the aid, even though the aid scheme was not yet in force, in order to allow the financing of coal-fired power plants already in operation or under construction.

250    In so far as Tempus calls into question the justification for the choice of January 2014 as the starting point for the taking into account of eligible CAPEX, it must be pointed out that that complaint is based on a misreading of the grounds of recital 44 of the contested decision. Contrary to the impression which may emerge from the Commission’s written submissions, those grounds must not be understood as imputing the key rationale for the choice of the January 2014 date to the July 2013 announcement alone, but rather as describing the historical context in which that choice was made. In particular, PGE and the Republic of Poland have explained, in a sufficiently plausible manner, that the choice of January 2014 was due to the fact that, following the public announcement by the Polish authorities, in July 2013, of their plans to introduce a capacity mechanism, it was necessary to set a starting date for the taking into account of the CAPEX incurred over a five‑year period which, on the one hand, was early enough to allow the participation of modernised or new generating or DSR CMUs in the first regular auction, the organisation of which could not take place before the end of 2018 (see recital 38 of the contested decision and Article 96(1) of the Act, read in conjunction with Article 36(2) thereof), and, on the other hand, was sufficiently far in advance of the supply of capacity during the first delivery year, namely 2021. Thus, in view of the fact that that first regular auction was expected to take place in January 2019 at the earliest (see recital 38 of the contested decision), it was not arbitrary, but rather consistent, to set January 2014 as the starting date of the five‑year period for the purpose of the taking into account of CAPEX, which recital 44 of the contested decision takes into consideration.

251    In that regard, Tempus cannot argue that the aid scheme did not take a sufficiently concrete form either in July 2013 or in January 2014 to enable those dates to be characterised as the dates at which the aid measure was granted to create the required incentive effect.

252    First, that line of argument disregards the fact that the aid at issue, namely the capacity payments, could be granted only following the authorisation of the aid scheme by the Commission, which took place on 7 February 2018, the entry into force of the Act and the close of the very first auction, scheduled for 2018, involving the allocation of the first capacity agreements for the first delivery year in 2021. Thus, the date of grant of the aid cannot be before that of the close of the first auction the result of which confers on the successful bidder, pursuant to the Act, the legal right to receive capacity payments (see, to that effect and by analogy, judgments of 21 March 2013, Magdeburger Mühlenwerke, C‑129/12, EU:C:2013:200, paragraphs 40 and 41; of 6 July 2017, Nerea, C‑245/16, EU:C:2017:521, paragraphs 32 and 33; and of 28 October 2020, INAIL, C‑608/19, EU:C:2020:865, paragraphs 30 to 34). Moreover, even assuming that Tempus sought to argue that the incentive effect must be linked to the entry into force of the aid scheme, namely in December 2017, such a line of argument would contradict its own finding that the taking into account of previous investment expenditure over a five-year period was permissible and gave rise to such an incentive effect.

253    Second, in the light of the foregoing, it was equally consistent for the Commission to accept the justification that the taking into account of the January 2014 date was intended to avoid new investments in capacity development being artificially delayed (see recital 44 in fine of the contested decision). Furthermore, PGE has confirmed, in a credible manner, that market participants relied on the announcement of the introduction of a capacity market from 2014 to take investment decisions with a view to obtaining capacity agreements at a later stage. It follows that the taking into account of that date, in conjunction with that announcement, was able to create an incentive effect in accordance with paragraph (49) of the Guidelines. It is therefore irrelevant that, in July 2013 or January 2014, there was not yet, as the case may be, a ‘binding and reliable’ commitment on the part of the Polish authorities to introduce a capacity market in Poland and that its specific criteria were still being defined until the end of 2016.

254    Third, Tempus’s vague, even speculative, allegation that the choice of January 2014 was intended solely to advantage coal-fired generating CMUs already in operation or under construction is not such as to call into question the plausibility of that choice. In any event, the Commission cannot be criticised for having accepted the rule that CAPEX incurred in the five years preceding the first delivery year should be taken into account also for coal-fired generating CMUs in the process of being modernised or under construction, account being taken of the main objective of the Polish capacity market of creating incentives to generate and maintain in a timely manner the availability of sufficient capacity to ensure security of supply.

255    The first complaint must therefore be rejected as unfounded.

256    As regards the second complaint, it must be held that Tempus’s argument that the choice of the date of 1 July 2017 relating to the start of electricity generation seeks essentially to advantage coal-fired generating CMUs is neither substantiated nor convincing. It is certainly apparent from Article 95 of the Act, read in conjunction with Article 32(1)(4) thereof, that existing generating CMUs which commenced electricity generation after 1 July 2017 and which fulfil the CAPEX criteria referred to therein are regarded as new generating CMUs. However, as the Republic of Poland essentially contends, Tempus has not established the extent to which that deadline could give rise to unequal treatment in favour of CMUs compared to modernised or new generating CMUs commencing generation after the entry into force of the Act, or vice versa. On the one hand, all those generating CMUs which fulfil the CAPEX criteria incur substantial investment expenditure and, on the other hand, the existing CMUs which do not incur such expenditure are in a different factual situation in the light of the objectives of the Polish capacity market. In addition, it was consistent to set a uniform starting date of generation for all those (new) CMUs which was sufficiently close to the entry into force of the Act in December 2017. In that regard, the example of the Kozienice plant which began operating in December 2017 is ineffective, in view of the entry into force of the Act during the same month. Furthermore, in so far as Tempus invokes anew, in that context, discrimination against DSR CMUs compared to generating CMUs, that complaint faces the same objections as those set out in paragraph 153 et seq. above, as those CMUs are in very different factual and legal situations in the light of the objectives pursued.

257    In any event, even if the choice of the precise date of 1 July 2017 is not mandatory to distinguish existing and new generating CMUs, given the need to plan investments in the modernisation or construction of a new generating unit in the long term, that date is sufficiently close to the entry into force of the Act – in December 2017 – and to the planned organisation of the first auction, from January 2018. That choice cannot therefore be characterised as unreasonable, even arbitrary, or as depriving the capacity payments to be made to the CMUs concerned of the required incentive effect, within the meaning of paragraph (50) of the Guidelines. Moreover, in the light of the foregoing, there is no risk that certain CMUs may be led to make investments of a certain CAPEX level with the sole aim of obtaining capacity agreements rather than with the aim, in accordance with the main objective of the capacity mechanism, of making the necessary capacity available. Furthermore, as the Commission submits, the consideration set out in recital 44 in fine of the contested decision according to which the approach described therein aims at avoiding potential new projects artificially delaying their development refers not solely to the choice of the date of 1 July 2017, but to that approach as a whole (see paragraph 250 above).

258    Last, it should be noted that, as the Commission and the interveners submit, inter alia, in the case at hand, the aid for generation adequacy constitutes operating aid and not investment aid, given that the capacity payments are intended primarily to reward the availability of capacity and, only secondarily, to enable CMUs to recoup the investment costs necessary for that purpose (see, also, paragraph 169 above).

259    It follows that Tempus has not established that the Commission should have entertained doubts as to the alleged retroactive nature of the aid scheme and that the first subsection must be rejected in its entirety.

(ii) Second subsection: incomplete assessment of the incentives for new capacity

260    By the second subsection, having regard to recital 164 of the contested decision and the content of the interim report of the sector inquiry, Tempus submits, in essence, that the Commission failed to evaluate the incentives for the ‘already committed new generation’ and therefore should have had serious difficulties in identifying the potentially different needs of the aforementioned new capacity under the Polish capacity market.

261    Neither the Commission nor the interveners expressly address this subsection.

262    In recital 164 of the contested decision, it is explained, in essence, that, with regard to the incentives for new and existing generating CMUs, the auctioning process has been designed to consider different lead times for making capacity available, namely periods of one or five years for which CMUs can bid and which are supposed to cater for the potentially different needs of new, existing and refurbishing capacity. Moreover, the aid scheme includes different capacity agreement lengths, depending on the CAPEX incurred to put capacity on the market. The Commission considers that this contributes to creating a level playing field between new and existing CMUs as, in contrast to existing CMUs, new or refurbishing CMUs are likely to need to secure financing for CAPEX, which is more difficult and more expensive without the relative stability provided by multi‑year capacity agreements. Without such agreements, this financing issue would be particularly acute for independent CMUs.

263    In the light of those grounds, which correspond, in essence, to those which have already been assessed in paragraphs 117 et seq. and 246 above, by the second subsection, Tempus merely reiterates, in different words and from the perspective of the incentive effect of the aid, its challenge relating to the differentiation of the length of capacity agreements and to the alleged lack of sufficient or equitable incentive that results for all CMUs. In contrast, recital 164 of the contested decision, which must be read in conjunction with recitals 42 and 43 of the same decision, describes in a sufficiently intelligible and comprehensive manner the relevant elements justifying the view that the CAPEX criteria and the various durations of capacity agreements linked to them produce the requisite incentive effect for new CMUs, depending on their respective financing needs. In that regard, reliance on the interim report of the sector inquiry fails, since that inquiry concerned the situation of the Polish electricity market before the introduction of the capacity mechanism and, in particular, the effects of the contingency reserve that was replaced by that mechanism (see recital 16(g) of the contested decision).

264    Therefore, the second subsection must be rejected as unfounded.

(iii) Third subsection: insufficient assessment of the cost recovery methodology

265    Under the third subsection, Tempus claims that the cost recovery scheme of the Polish capacity mechanism would have required an in-depth analysis, especially regarding the rules on determining the capacity charge (see recitals 107 and 108 of the contested decision, and judgment of 15 November 2018, Tempus Energy and Tempus Energy Technology v Commission, T‑793/14, EU:T:2018:790, paragraph 203), in order to verify its proportionality. Member States should concentrate the costs of the capacity market on consumers consuming at peak times so as to increase the price incentive to shift away from those times, thereby reducing the resource adequacy problem. Under Articles 69 to 78 of the Act, the cost recovery methodology is based on the principle that household customers are charged for the capacity mechanism based on their annual consumption only, whereas non-household customers are invoiced according to the time of use, based on their electricity consumption during the ‘selected hours of the day’, which are not specified in the act, but are published annually, by decree of the President of the Energy Regulation Office, for each quarter of the year (Article 74(4) of the Act). In addition, customers from the energy-intensive industries (‘EII’) sector, defined in Article 71 of the Act, are entitled to a reduction of the capacity charge depending on their electricity consumption intensity factor, with the most energy-intensive customers receiving the greatest discount, and having to pay only 80%, 60% or 15% of the charge, determined in relation to consumption in the ‘selected hours of the day’ (Article 70(3) of the Act). According to Tempus, without any certainty as to the ‘selected hours of the day’ or as to the range within which those hours are likely to be, it is impossible for the Commission to assess whether or not that window constitutes an effective incentive, or even whether non-household customers are able to avoid capacity market costs by reducing their peak-demand usage. In addition, instead of maximising the price incentive for EII customers to use DSR to reduce their peak-time usage, such customers will be exempt from up to 85% of the capacity charge, which substantially reduces the incentive effect of a targeted cost recovery methodology. Thus, those customers who are most likely to have the technical capability to use DSR will have less incentive to avoid demand peaks, which increases the likelihood of their continuing to consume at those times and the increased cost of the scheme being borne by other electricity customers. Therefore, more capacity will have to be procured to cover those persisting peaks, which will unnecessarily increase the amount of the aid. Similarly, household customers will not be able to avoid capacity market costs, even by using DSR technology to avoid other peak time electricity costs. Tempus concludes from this that the aid scheme removes the household DSR market since it makes it possible to avoid capacity market costs, significantly impedes the EII market and provides absolutely no certainty as to the size of the non-EII commercial customer market from DSR. In that regard, it states, in essence, that the commitment to notify separately any reductions of the capacity charge (recital 108 of the contested decision) is insufficient, the Commission being required immediately to ensure that the aid scheme has an incentive effect, observes the principles of necessity and proportionality and facilitates DSR. In any event, it should have had doubts on those points, irrespective of the formal investigation procedure underway in Case SA.51502 (see Commission Decision C(2019) 2504 final of 15 April 2019, State aid SA.51502 (2019/C)/(2018/N) – Poland – Reductions from a capacity mechanism levy for [Energy-intensive users] in Poland). In the case of the Polish capacity mechanism, the cost recovery methodology is intrinsically linked to the aid scheme in that it provides the financing for the costs incurred in providing the capacity market.

266    The Commission counters that it carried out a complete assessment of the financing method of the Polish capacity market (recitals 107 and 114 to 117 of the contested decision), which is, moreover, the subject of a formal investigation procedure in Case SA.51502. PGE submits that Tempus mischaracterises the incentives flowing from that financing method in arguing that the Polish capacity mechanism should concentrate costs on those who consume at peak times rather than spreading them across all consumers equally. That argument is undermined by the fact that household customers agree to stable electricity prices in their electricity supply contracts and that they are incentivised to reduce their consumption more generally and thus the costs of the capacity market. The Commission and the Republic of Poland submit, in essence, that the decision to invoice households on the basis of their annual electricity consumption as a lump sum or on a monthly basis was based on the fact that energy meters in Poland are not equipped to provide sufficiently detailed data on their consumption profile and that it would not be possible to apply the same invoicing method to all households by reference to electricity consumption in peak hours, unlike customers other than households. The Commission adds that the financing method of State aid may impact the assessment of its compatibility with the internal market solely if it forms an integral part of it and entails a non-severable violation of EU law like, for instance, of Articles 30 and 110 TFEU (paragraph (29) of the Guidelines), which Tempus has not claimed. It takes the view that the capacity charge levied on consumers to recover the costs of the Polish capacity market is not affected by the capacity payments. The increase in revenue from that charge does not necessarily result in an increase in the amount of the capacity payments made, which depends solely on the outcome of the bidding process. The Republic of Poland states that reductions of charges for energy-intensive users were not assessed by the Commission (recital 108 of the contested decision), as they had not been notified together with the Polish capacity market and are a mechanism distinct from the capacity market, which can operate without those reductions. The material and procedural separation of those two mechanisms is evidenced primarily by Article 104 of the Act, which provides for separate suspension clauses which suspended the implementation of the capacity agreements and the application of reductions, respectively, until the adoption of compatibility decisions by the Commission. Moreover, that distinction is reflected in the Commission’s previous decision-making practice.

267    By the present subsection, Tempus disputes the incentive effect of the cost recovery methodology, which is defined in Articles 69 to 78 of the Act, in Chapter 3 on the ‘capacity charge’. In particular, it denies that the methodology set out in Article 70 of the Act for determining the capacity charges to be paid by different types of electricity customers is capable of creating such an incentive effect. On the contrary, it gives rise to inefficient management of the capacity market to the detriment of, in particular, DSR, by penalising household customers and preventing large industrial customers from making their consumption more flexible, which results in the need to make more capacity available.

268    As PGE in particular contends, Tempus starts from an erroneous premiss in claiming that, for the purposes of satisfying the incentive effect criterion, Member States are required to pass on the costs of the capacity market primarily to consumers consuming at peak times so as to increase the price incentive to shift away from those times, thereby reducing the resource adequacy problem. That line of argument disregards the fact that the rules relating to the capacity charge are primarily intended to ensure the financing of capacity payments (see Article 69(1) of the Act), and not to issue price signals encouraging consumers to make their consumption more flexible in order to reduce capacity needs. Furthermore, in the absence of an explicit objection by Tempus, the argument of the Commission and the Republic of Poland that the monthly invoicing of capacity charges to household customers on the basis of their annual consumption (see Article 70(1) of the Act) is due mainly to the fact that energy meters in Polish households are not yet equipped to provide sufficiently detailed data on their consumption profile and that it is therefore not possible to charge them fairly by reference to electricity consumption in peak hours is sufficiently credible.

269    As regards the invoicing of non-household customers according to the time of use, based on their electricity consumption during the ‘selected hours of the day’, under Article 70(1)(2) and (2) and (3) of the Act, read in conjunction with Article 71 thereof, having regard to the plausible explanations provided by the Commission, that method cannot be criticised for undermining the incentive effect of the aid scheme as a whole on the ground that it unduly favours large industrial customers. In that regard, Tempus again starts from an erroneous premiss in establishing a link between the amount of the capacity charge and that of the capacity payments, since the latter is determined exclusively by the outcome of the auction, namely according to the price offered by the bidder that wins the capacity obligation. It is therefore neither coherent nor possible to argue, as Tempus does, that the sum of the capacity charges collected in a given year is liable to influence the total amount of aid to be granted, namely that of the capacity payments made in that same year to reward compliance with capacity obligations. In the absence of an intrinsic link between the sum of the capacity charges, on the one hand, and the sum of the capacity payments, on the other, Tempus’s argument that the more favourable treatment of large industrial customers according to the intensiveness of their electricity consumption should necessarily have an impact on the amount of aid paid cannot be accepted. Accordingly, that argument must be rejected, without there being any need to examine further the merits of that alleged favourable treatment which, in any event, is the subject of the separate formal investigation procedure initiated by the Commission in Case SA.51502 (see, also, paragraph 271 below).

270    Furthermore, in so far as Tempus submits that the calculation method based on the ‘selected hours of the day’ is not capable of creating an incentive effect, suffice it to note that the amount of the capacity charge thus calculated makes no difference to the determination of the relevant incentive effect arising from the aid scheme and the Polish capacity mechanism, which is essentially linked to the capacity payments and to the length of the capacity agreements (see paragraph 165 et seq. above).

271    Finally, in so far as Tempus argues that, as a result of this method, those large industrial customers have no interest in making their electricity consumption more flexible, in particular by using DSR, thereby reducing the market’s capacity needs, it is sufficient to note that this aspect is in fact related to the question of whether and how the Polish capacity market is supposed to combine the incentive effect of making available the necessary generation capacity with that of protecting the environment (see paragraphs 245 and 246 above). Furthermore, that question is the subject of the formal investigation procedure in Case SA.51502, following the separate notification by the Republic of Poland of the system for reducing capacity charges for EIIs (see recitals 12, 19 to 22, 49, 57 and 60 of the decision to initiate the procedure in Case SA.51502), for which reason it was not assessed by the Commission in the present procedure (see recital 108 of the contested decision).

272    Accordingly, the Commission cannot be criticised for having accepted a different method of calculating capacity charges for different types of customers, without having doubts as to its incentive effect.

273    Consequently, the third subsection must be rejected as unfounded.

(iv) Fourth subsection: penalty regime

274    By the fourth subsection, Tempus submits, in essence, that, having regard to recital 102 of the contested decision, the Commission failed to fulfil its obligation to assess the incentive effect of the aid scheme and its obligation to state reasons, in particular in relation to the penalty regime which is crucial for the effectiveness of CMUs’ obligations under the Polish capacity market. That was all the more necessary in view of the stress periods linked to the summer heat waves occurring every five years which present a major risk to security of supply. The penalty regime must be defined in an objective manner, have an incentive effect and induce the beneficiary of the aid to change its behaviour in order to improve the functioning of a secure, affordable and sustainable European energy market in accordance with paragraph (49) of the Guidelines. The DSR model reduces capacity gaps at peak load with higher efficiency than generators because losses due to transmission and distribution are reduced together with demand. The Polish capacity market ignores that aspect, but results in generating CMUs competing with DSR CMUs on the basis of capacity obligations. The Commission should also have had doubts as to the appropriateness of penalties as they are ineffective incentives which disadvantage DSR. In times of peak demand or scarcity, due to the negative CO2 balance of DSR, substituting a generating CMU emitting 400 kg CO2/MWh with a DSR CMU will lead to a reduction in CO2 emissions by 400 kg CO2/MWh. However, it is the generating CMU that will receive the Green Bonus, not the DSR CMU. Even if DSR CMUs shifted the demand towards non-peak demand hours, their net effect on the emission balance would most probably be negative because of how the merit order system influences the dispatch.

275    Tempus takes the view that the assessment set out in recital 45 of the contested decision is inconsistent with that contained in recitals 42 and 43 of that decision, according to which the Polish capacity market regards the aid sometimes as remuneration for the obligation to deliver the necessary capacity, such that the penalties should be aligned with the volume of capacity actually provided, regardless of the milestones of the CAPEX, and sometimes as investment aid for the construction, expansion or refurbishment of (new) capacity, such that the penalties may be aligned with the failure to comply with the investment milestones. By contrast, according to Tempus, if the Polish capacity market is a form of investment aid, no aid should be granted based on the amount of the investments incurred before that market is operational (see recital 44 of the contested decision). It adds that reduction of a contract length to one year is perceived by the Polish authorities as a penalty, when this is the normal duration of a contract for foreign capacity and many Polish DSR units. The Commission’s reasoning is therefore vitiated by errors of assessment with respect to the incentive effect of the penalties.

276    The Commission, supported by PGE, contends, in essence, that Tempus does not contest that, by the penalty regime, the Polish capacity market provides incentives for capacity availabilities in the event of network stress and thus contributes to the achievement of the objective of common interest. That regime is defined in an objective manner and it cannot be designed based on hypothetical scenarios of most probable peak load situations in the coming years. The penalties are aimed merely at ensuring that generation capacity is available at the start of the delivery year and, should a generating CMU be unavailable, the Polish authorities have the option of replacing the missing capacity in the ‘N‑1’ auctions. The aid therefore cannot be qualified as investment aid.

277    PGE adds that Tempus fails to appreciate that the penalties provide sufficient incentives to DSR operators and generators, and are in fact more beneficial for those operators. It notes that, if DSR operators reach 50-80% performance in the test procedure, while losing a corresponding portion of their collateral, they nevertheless receive a positive test result certificate – with the capacity reduced accordingly – and the capacity agreement remains valid. Those operators are therefore not obliged to pay a penalty and are considered to have passed the test procedure. Moreover, if they reach at least 80% performance during the test, they are remunerated for their total certified capacity. On the other hand, generators not meeting the 95% test lose all collateral and must pay financial penalties for each undelivered month for a maximum period of three years until the termination of the capacity agreement takes effect. The preferential treatment of DSR is also clear from a comparison with modernised units eligible for five-year capacity agreements. According to PGE, if such units do not satisfy a 95% test ratio, they will have their contract shortened and cannot be certified for up to two years.

278    The penalty regime and the testing regime are described in recitals 102 to 105 of the contested decision.

279    According to recital 102 of the contested decision, in particular, the penalty regime aims to provide capacity providers with incentives to deliver energy when needed. CMUs which perform below the expected level of performance will be penalised, while those that exceed the expected level will receive over-delivery payments so that, at the end of the year, the capacity payments will broadly reflect their performance. According to recital 103 of that decision, the penalties are the same for Polish and foreign CMUs.

280    Recital 104 of the contested decision states that the penalty regime is complemented by a system of performance demonstrations to ensure that capacity providers are able to deliver energy where needed and receive capacity payments only if they are reliable. That is especially important for those delivery years with no stress events in which testing providers’ performance ensures that they are physically capable of delivering as per their capacity obligations.

281    Similarly, according to recital 45 of the contested decision, a CMU which fails to deliver the capacity required will be exposed to penalties which, depending on the gravity of the failure, can be a financial penalty, a reduction to a one-year capacity agreement or a complete termination of the agreement. The latter two forms of penalty refer to those provided for in Article 46(3) and Article 47(1) and (2) of the Act, which are not disputed as such by Tempus.

282    On the other hand, the financial penalties for failure to perform or improper performance of capacity obligations, disputed in the present case, are laid down in Article 59(1) and (2) of the Act. The penalty regime and performance test regime, as described above, does not however derive directly from the Act, but is stipulated, in accordance with Article 68 of the Act, by ordinance of the Polish Minister of Energy.

283    Without specific provisions in the Guidelines on sanctions or penalties for non-compliance with capacity obligations, the appropriateness of the penalty regime must be assessed in the light of the general criteria governing the incentive effect of the aid scheme. Having regard to the reasons set out in recital 102 of the contested decision, Tempus’s vague assertion that the Commission failed to assess and state sufficient reasons as to whether the penalty regime designed to reward compliance with capacity obligations and to penalise non-compliance has the required incentive effect cannot succeed. That regime presents a coherent complement to the incentive effect linked to the duration of the capacity agreements and to the capacity payments (see, also, paragraph 169 above) by increasing the incentive for CMUs to comply with the obligations of capacity the availability of which is required in order to achieve the reliability standard at issue, as Tempus itself acknowledges. It follows that the applicant has not established that the penalty regime is not based on an objective definition or that it does not have an incentive effect within the meaning of paragraph (49) of the Guidelines, or that the Commission should have had doubts in that regard. Moreover, as the Commission and PGE point out, Tempus’s unintelligible argument based on the performance test regime relating to stress periods linked to the summer heat waves occurring every five years is irrelevant, since that test, as described in recital 104 of the contested decision, is intended precisely to ensure that capacity providers are able to withstand such stress. In addition, Tempus merely reiterates, in a qualified manner, as regards incentive effect, its arguments relating to the alleged discrimination of DSR CMUs as compared to generating CMUs, which have already been rejected, including under the criterion of incentive effect (see paragraphs 165 et seq. and 206 above). What is more, in that context Tempus has failed to contradict PGE’s arguments plausibly explaining the reasons for which the penalty regime does not place DSR operators at a disadvantage (see paragraph 277 above).

284    Furthermore, it must be held that, contrary to what Tempus claims, there is no inconsistency between recitals 42 and 43 of the contested decision, on the one hand, and recital 45 thereof, on the other, given that the capacity payments are operating aid remunerating the making available of capacity and not aid for investment in the construction of new capacity (see paragraph 258 above). Its argument based on the nature of investment aid must therefore, if only for that reason, be rejected as unfounded.

285    In those circumstances, Tempus has failed to demonstrate that the Commission should have entertained doubts as to the incentive effect of the penalty regime, or committed errors of assessment in that regard.

286    Consequently, the fourth subsection must be rejected as unfounded, as must the third section in its entirety.

(5)    Fourth section: alleged incompleteness of the assessment of the proportionality test

287    By the fourth section, in the first place, Tempus submits that the Commission should have had doubts about the proportionality of the Polish capacity market, within the meaning of paragraph (228) of the Guidelines. That market must allow for a reasonable rate of return without triggering windfall profits, and the Commission is obliged to verify that the bidding processes satisfy clear, transparent and non-discriminatory criteria. In addition, that market must be proportionate and limited to the minimum amount of aid necessary to resolve the resource adequacy problem and avoid overcompensation. In that regard, it is necessary to factor in the potential of DSR and interconnection in order to resolve that issue through the energy-only market. However, the design of the Polish capacity market deprives DSR CMUs and other CMUs from neighbouring markets of participation opportunities by dampening price signals in favour of future flexibility. Moreover, the Commission should have assessed the risk of incentives to inflate CAPEX requirements in order to obtain longer agreements, meet milestones and avoid penalties. The retroactive and arbitrary starting dates for CAPEX contradict the principle of proportionality.

288    In the second place, Tempus states that the Polish capacity market enables CMUs to provide extra capacity and to receive payments for such deliveries (recital 102 of the contested decision), with the result that the Commission should have examined whether that possibility could lead to overcapacity and overcompensation and, at the very least, have an impact on the ‘load following’ principle. In recital 147 of the contested decision, the Commission itself states that over-procurement leads to overcompensation and that such a situation is incompatible with the internal market. There is therefore no cap to prevent CMUs from providing extra capacity in excess of what is required to respond to the stress event and, in general, no limit on over-delivery, which is, moreover, contrary to the principles of incentive effect and proportionality. Last, Tempus observes that the contested decision does not clearly address the questions concerning the cumulation of aid and the risk of discrimination towards outside capacity. Unlike Polish co-firing installations using both RES and non-RES, foreign co-firing installations are not authorised to participate either in the RES support scheme through auctions for those renewable sources or in the capacity market. That discriminatory treatment of co-firing CMUs, which is not explicitly mentioned in the contested decision, was not subject to an examination on the part of the Commission.

289    The Commission, supported by PGE, counters that it conducted a full assessment of proportionality in accordance with paragraphs (228) to (230) of the Guidelines (recitals 170 to 177 of the contested decision). As to the alleged risk of overcapacity and over-compensation, it states that the extra capacity is to be rewarded in the form of a bonus resulting from the redistribution of funds from penalties for undelivered capacity paid by CMUs throughout the year. That bonus is calculated and paid once a year proportional to extra capacity delivered by the CMU and all extra capacity delivered by all CMUs. That bonus, paid in return for the delivery of such capacity, thus corresponds, at most, to the amount of the penalties imposed for undelivered electricity during stress events, such that there is no possibility of overcompensation. PGE adds that over-delivery is put in place in order to balance underperforming CMUs in the secondary capacity market. Those CMUs will be granted compensation only if there is a shortage in required and contracted capacity at the level of primarily contracted CMUs (recitals 89 to 92 of the contested decision). In such a case, the latter do not have to pay penalties, but their capacity payments are deducted according to the performance provided by the CMUs contracting or operating on the secondary market. There is therefore no overcompensation, but rather adequate remuneration on the basis of additional capacity availability. In addition, the Commission considers that it conducted a complete assessment, excluding the cumulation of capacity payments with other operating and investment aid (recital 176 of the contested decision). With regard to co-firing installations generating power from RES and fossil fuels, the fact that the capacities linked to non-RES are eligible does not in itself lead to a cumulation of capacity payments. RES support is granted solely for that same energy and, in the case of co-firing installations, only the capacity provided by non-RES generation may be eligible for the capacity mechanism. Thus, such capacity providers have to submit a declaration that they will not combine RES support with the capacity payments for the same volume of capacity (recital 18 of the contested decision). Furthermore, the aid amounts paid to a capacity provider under other aid schemes will be deducted from the capacity payments, ruling out a duplication of aid (recital 19 of the contested decision).

290    It should be recalled that, according to paragraphs (228) to (230) of the Guidelines, under the heading ‘Proportionality’, in essence, the calculation of the overall amount of aid should result in beneficiaries earning a rate of return, which can be considered reasonable under normal circumstances, where it is the result of a competitive bidding process based on clear, transparent and non-discriminatory criteria, effectively targeting the defined objective. Moreover, the aid measures should have built-in mechanisms to ensure that windfall profits cannot arise. Similarly, it is apparent, in essence, from paragraphs (69) and (70) of the Guidelines that energy aid is considered to be proportionate if the aid amount per beneficiary is limited to the minimum needed to achieve the energy objective aimed for, which is normally the case if the aid corresponds to the net extra cost necessary to meet that objective, compared to the counterfactual scenario, that is to say, in the absence of aid.

291    It is apparent from recital 18 of the contested decision that the Polish capacity market excludes capacity providers receiving operating aid, for example on the basis of the RES support scheme, and that those operators have to submit a declaration that they will not combine such aid with the capacity payments if they are successful in a capacity auction (see, also, recital 25(h) of that decision). By contrast, according to recital 19 of the contested decision, in essence, capacity providers in receipt of investment aid, inter alia, in the context of the European Union’s Emissions Trading System, may participate in the capacity market, but they must have the amount of that aid deducted from those capacity payments in order to avoid any overcompensation. That statement of reasons corresponds to the situation provided for in Article 62(1) of the Act.

292    So far as concerns proportionality, recital 176 of the contested decision states that specific rules are intended to regulate the situation in which capacity providers were to receive both operating or investment aid and the capacity payments (see, also, recitals 18 and 19 of that decision). The Polish capacity market excludes the participation of capacity providers in receipt of operating aid, and those providers must opt out of the operating aid scheme if they are successful in a capacity auction. By contrast, capacity providers in receipt of investment aid may participate in the Polish capacity market, but the investment aid is deducted from the capacity payments, including that granted under Article 10c(1) of Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community (OJ 2009 L 140, p. 63), such that that mechanism prevents the possibility of such overcompensation in case of cumulation of aid.

293    In the first part of this section, Tempus merely reiterates, in essence, its complaints relating to the alleged discrimination of DSR CMUs, as well as to the inappropriateness of the conditions for their participation in the Polish capacity market, including in the auctions, and including the CAPEX criteria and the allegedly retroactive and arbitrary manner in which they were set, which have already been rejected as unfounded in paragraphs 165 et seq. and 248 et seq. above.

294    Therefore, this first part cannot succeed from the perspective of proportionality, either, and must be rejected.

295    With regard to the second part concerning the alleged overcompensation due to the over-delivery payments, it is true that recital 102 of the contested decision states that CMUs which provide extra capacity will receive over-delivery payments. However, even if the parties have not specified the relevant executive provisions governing those payments, which are not provided for in Articles 60 to 68 of the Act, it is sufficient to note that Tempus has failed to challenge in its submissions the detailed explanations put forward by the Commission and PGE in this respect during the proceedings, namely that, on the one hand, the bonus for the delivery of extra capacity is both limited and financed proportionately from the redistribution of funds from the penalties paid by failing CMUs and, on the other hand, that bonus is intended to reward deliveries of capacity aimed at balancing the underperformance of CMUs on the secondary capacity market (see recitals 89 to 92 of the contested decision). In those circumstances, Tempus has not established that CMUs active on the Polish capacity market, including on its secondary part, are liable to obtain overcompensation or windfall profits within the meaning of paragraph (230) of the Guidelines. On the contrary, in the absence of a detailed and substantiated challenge on its part, the provisions of the Act governing the calculation of capacity payments (Articles 60 to 68), read in conjunction with those relating to auctions (Articles 29 to 40), indicate that those payments are intended to generate a reasonable rate of return in accordance with paragraphs (228) and (229) of the Guidelines.

296    Nor has Tempus succeeded in calling into question the considerations set out in recitals 18, 19 and 176 of the contested decision concerning the lack of cumulation of capacity payments with other operating and investment aid, even with regard to co-firing power plants. In this respect, the Commission explained in the course of the proceedings precisely why those power plants cannot obtain capacity payments for the part of the energy produced from renewable sources, subject to the RES support scheme, such that overcompensation is ruled out, without those explanations being challenged by Tempus in the reply.

297    Consequently, Tempus has not shown that the Commission should have entertained doubts as to the proportionality of the capacity payments within the meaning of paragraphs (228) to (230) of the Guidelines and the fourth part must be rejected as unfounded.

(6)    Fifth part: alleged incompleteness of the assessment of the limited nature of the negative effects on competition and trade

298    According to Tempus, following the Guidelines, the policy design of the Polish capacity market must allow for the participation of a sufficient number of CMUs in the auctions in order to establish a competitive price, but also the genuine inclusion of different technologies, including DSR and storage as well as the use of interconnectors. The restrictions on access for DSR operators, in particular foreign ones, and the discriminatory aspects referred to alone demonstrate that this is not the case. At the very least, the Commission should have examined whether the Polish capacity market actually ensures that unnecessary negative effects on competition and on trade between Member States are avoided and should have carried out an extensive review on that question. Tempus adds that the results of the first auction, for delivery in 2021, show that market concentration has been strengthened. It is therefore legitimate to doubt that the Commission undertook a complete assessment in the light of paragraph (233)(d) of the Guidelines. Tempus concludes from this that a thorough analysis of Section 3.9.6 of the Guidelines, and in particular of paragraph (232)(a) thereof, should have led the Commission to raise doubts as to the sufficient mitigation of the negative effects of the Polish capacity mechanism.

299    According to the Commission, Tempus does not make a particular claim about the completeness of the assessment set out in recitals 178 to 185 of the contested decision with regard to the avoidance of undue negative effects on competition and trade between Member States. The mere succinct repetition of the claims made with regard to the previous compatibility criteria does not suffice to give rise to doubts in that respect. Tempus may not rely on the results of the first auction conducted after the adoption of the contested decision to argue that the Commission should have had doubts as to the assessment of the negative effects of the aid scheme on competition and trade between Member States.

300    As the Commission and PGE submit, it is sufficient to point out that, under the fifth section, Tempus merely repeats, in essence, in the light of paragraph (232)(a) and paragraph (233)(d) of the Guidelines, under the guise of the criterion of negative effects on competition and trade, the same complaints which it has put forward under the second, third and fourth sections and which have already been rejected.

301    Therefore, the fifth section must also be rejected, as must the first plea in its entirety.

3.      Second plea: inadequate statement of reasons in breach of Article 296 TFEU

302    Tempus submits, in essence, that the Commission failed in its obligation to provide adequate reasons under Article 296 TFEU in several instances. It failed to give sufficient explanations of such kind as to provide the General Court with the exact reasons which led it to decide not to raise objections to a novel, complex capacity mechanism applicable in the entire Polish electricity market in the long term. A decision to approve such a mechanism, without initiating a formal investigation procedure, however, must contain an adequate statement of reasons. Far from mentioning the principal issues of law and of fact in relation to the applicable law in such a way that the essential part of its reasoning may be understood, the contested decision does not deal with any of the ambiguous points put forward in the first plea.

303    The Commission, supported by Enel X and PGE, contends that the present plea should be rejected.

304    In that regard, it is sufficient to point out that the second plea merely refers, in a general and abstract manner, to the substantive complaints, raised under the second part of the first plea, in respect of which the General Court has been able to carry out its review of legality in the light of the concept of doubts or serious difficulties and which have been rejected in their entirety, with the result that the present plea must also be rejected as unfounded.

305    Consequently, the action must be dismissed in its entirety.

IV.    Costs

306    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 applicants have been unsuccessful, they must be ordered to bear their own costs and to pay those incurred by the Commission, PGE, Enel X and Enspirion, in accordance with the form of order sought by them.

307    Under Article 138(1) of the Rules of Procedure, the Member States which have intervened in the proceedings are to bear their own costs. The Republic of Poland must therefore be ordered to bear its own costs.

On those grounds,

THE GENERAL COURT (Third Chamber)

hereby:

1.      Dismisses the action;

2.      Orders Tempus Energy Germany GmbH and T Energy Sweden AB to bear their own costs and to pay those incurred by the European Commission, PGE Polska Grupa Energetyczna S.A., Enel X Polska z o.o. and Enspirion sp. z o.o.;

3.      Orders the Republic of Poland to bear its own costs.

Collins

Kreuschitz

Steinfatt

Delivered in open court in Luxembourg on 6 October 2021.

E. Coulon

 

A.M. Collins

Registrar

 

President


Table of contents


I. Background to the dispute

A. The applicants

B. The administrative procedure and the contested decision

C. The Polish electricity market

D. The aid scheme

II. Procedure and forms of order sought

III. Law

A. Admissibility

B. Substance

1. Subject matter of the dispute and review of substantive legality

2. First plea: failure by the Commission to fulfil its obligation to initiate the formal investigation procedure in accordance with Article 108(2) TFEU

(a) First part of the first plea: existence of doubts as to the conduct and length of the procedure

(1) Summary of the main arguments of the parties

(2) Findings of the General Court

(b) Second part of the first plea: existence of doubts concerning the content of the contested decision

(1) Preliminary observations

(i) The alleged doubts or serious difficulties in the light of the provisions of the Guidelines

(ii) The legal nature of the Guidelines and the scope of the review of legality by the EU Courts in that regard

(2) First section: alleged incompleteness of the assessment of the objective of common interest and the need for State intervention

(i) First subsection: the objective of common interest

– The first complaint

– The second complaint

– The third complaint

(ii) Second subsection: the need for intervention by the Polish State

– The first complaint

– The second complaint

– The third complaint

– The fourth complaint

(3) Second section: alleged incompleteness of the assessment of the appropriateness of the aid scheme

(i) First subsection: error of assessment of the relevance of the Polish capacity mechanism in a highly concentrated and State-dominated electricity market

(ii) Second subsection: discrimination of DSR

– Preliminary observations

– The first complaint

– The second complaint

– The third complaint

– The fourth complaint

– The fifth complaint

(iii) Third subsection: incorrect assessment of the sufficiency of the incentives for new market participants

(iv) Fourth subsection: insufficient participation of foreign capacity

(v) Fifth subsection: inappropriateness of the aid scheme for guaranteeing security of supply

(4) Third section: alleged incompleteness of the assessment and alleged violation of the incentive effect of the aid

(i) First subsection: alleged retroactivity of the aid scheme

(ii) Second subsection: incomplete assessment of the incentives for new capacity

(iii) Third subsection: insufficient assessment of the cost recovery methodology

(iv) Fourth subsection: penalty regime

(5) Fourth section: alleged incompleteness of the assessment of the proportionality test

(6) Fifth part: alleged incompleteness of the assessment of the limited nature of the negative effects on competition and trade

3. Second plea: inadequate statement of reasons in breach of Article 296 TFEU

IV. Costs


*      Language of the case: English.


1      This judgment is published by extracts.