When the state acts in its capacity as legislator it is not presumed to direct the resources of undertakings it owns and over which it can exercise dominant influence as a shareholder.
On 13 September 2017, the Court of Justice replied to a request from a national court to advise it on the concept of state resources [case C‑329/15, ENEA v Prezes Urzędu Regulacji Energetyki]. The national court was adjudicating a dispute between ENEA, a Polish state-owned energy company, and the Polish energy regulator Urzędu Regulacji Energetyki [URE]. URE had imposed a fine on ENEA for failing to purchase electricity produced by cogeneration [i.e. electricity that was produced together with heat].
Under Polish law, electricity distributors or traders are obliged to buy electricity produced by cogeneration. If they fail to obtain a certificate that proves that indeed they purchased cogenerated electricity they can be subject to financial penalties. The proceeds of these penalties are paid into the national fund for environmental protection and water management.
In the domestic legal proceedings initiated by ENEA to contest the fine, ENEA claimed that the obligation to buy electricity produced by cogeneration constituted State aid, which was illegal for the simple reason that it had not been notified to the European Commission. ENEA claimed that the illegality of the aid implied that the imposition of the financial penalty was also unlawful.
The referring national court considered that the Polish measure in question conferred a selective advantage to producers of electricity from cogeneration, it possibly distorted competition and affected trade between Member States and that the purchase obligation was imputable to the state as it is imposed by law. However, it had doubts as to whether the intervention used state resources.
ENEA was required to sell to end users a minimum amount of electricity produced by cogeneration, either by producing such electricity itself or by purchasing it from third parties. If it bought from others, the purchase price of electricity was to be determined by mutual agreement between the undertaking subject to the purchase obligation [i.e. ENEA] and the producer of that electricity.
However, URE had the power, when approving tariffs charged by electricity companies to consumers, to fix the price of electricity produced by cogeneration at a level that it considered to be reasonable. In other words, URE had the power to prevent distributors from passing on the cost of more expensive electricity to end users.
Given this framework of the Polish regulatory system, the referring court asked whether Article 107(1) TFEU meant that a national measure placing an obligation on both private and public undertakings to purchase electricity produced by cogeneration constituted intervention by the state or through state resources.
Do you know we also publish a journal on State aid?
The European State Aid Law Quarterly is available online and in print, and our subscribers benefit from a reduced price for our events.
The main elements of the judgment
The Court of Justice noted, first, that “(21) in order to assess whether a measure is attributable to the State, it is necessary to examine whether the public authorities were involved in the adoption of that measure”.
The Court observed that “(22) the obligation […] to supply electricity produced by cogeneration, was imposed by the Law on energy, and that measure must therefore be regarded as attributable to the State”.
Then it explained that “(23) the condition that there must be intervention by the State or through State resources is satisfied not only where aid is granted directly by the State but also where it is granted by public or private bodies established or designated by the State with a view to administering the aid”. “(24) A measure consisting, inter alia, in an obligation to purchase energy may thus fall within the definition of ‘aid’ even though it does not involve a transfer of State resources”.
Paragraph 24 is confusing. If a measure does not involve state resources it cannot be State aid. What perhaps the Court of Justice meant was that a measure may be classified as aid even though the money may not come from a state budget. As is well established in the case law, state resources are also considered to be those resources which, even though they are not permanent assets of the public sector, remain constantly under public control and, consequently, are available to public authorities.
At any rate, the Court went on to distinguish this kind of resources from those that are used “mostly by private undertakings” which “(26) are not appointed by the State to manage a State resource, but are merely bound by an obligation to purchase using their own financial resources”.
The Court also noted “(27) that regard that the mechanism at issue […] required electricity suppliers to sell a quota of the electricity produced by cogeneration accounting for at least 15% of their annual electricity sales to end users.” “(28) The President of the URE set the maximum tariffs for sale of electricity to end users, so that the financial burden resulting from that purchase obligation could not be systematically passed on to end users by undertakings.” “(29) It is thus apparent from the information before the Court that, in certain circumstances, electricity suppliers purchased electricity produced by cogeneration at a higher price than that charged to end users, which resulted in extra costs for the suppliers.”
According to the Court, “(30) given that those extra costs cannot be passed on entirely to end users and are not financed by a compulsory contribution imposed by the State or by a full offset mechanism […] it must be concluded, […], that the supply undertakings were not appointed by the State to manage a State resource, but were funding a purchase obligation imposed on them by having recourse to their own financial resources.”
This is an important conclusion. But it is not clear why the Court felt that it had to refer to the fact that the extra costs could not be “passed on entirely” to end users. What appears to be sufficient for the conclusion of the Court was that there was no compulsory contribution [perhaps from users of electricity], nor any offset mechanism, full or partial. The distributors were not appointed by the state to levy a charge on consumers so as to generate revenue to offset the extra cost of cogenerated electricity, as happens in many other national schemes.
The Court also considered the argument put forth by both ENEA and the Commission that most of the undertakings under the purchase obligation were public undertakings and therefore that obligation could be regarded as being financed through state resources.
The Court responded by noting that “(31) the resources of public undertakings may be regarded as State resources where the State is capable, by exercising its dominant influence over such undertakings, of directing the use of their resources in order to finance advantages to the benefit of other undertakings”. “(32) The mere fact that the State held the majority of the capital in some of the undertakings subject to the purchase obligation does not lead to the conclusion that, […], the State exercised a dominant influence that enabled it to direct the use of the resources of those undertakings within the meaning of the case-law referred to in the preceding paragraph.”
Paragraphs 31 and 32 are also confusing because they conflate the concepts of control with imputability. A majority shareholder can always exercise dominant influence if it wishes to do so. Indeed it is not obvious how the state can have a majority shareholding without being able to exercise a dominant influence. The issue here is whether the state intervened to determine the decision of the state-owned companies. Nonetheless, the Court continued that “(33) it appears that the purchase obligation applied equally to all electricity suppliers, regardless of whether their capital was predominantly held by the State or by private operators.”
“(34) In addition, it is not clear from the information submitted to the Court, in particular from the information provided at the hearing, that ENEA’s conduct was dictated by instructions from public authorities. On the contrary, it was indicated that the decision to decline offers for the sale of electricity produced by cogeneration during the year 2006 was the result of wholly autonomous business decisions.” Indeed, this is a more powerful argument for the purpose of demonstrating that the state did not use its shareholder prerogatives to force electricity distributors to buy more expensive cogenerated electricity.
The Court also observed that “(35) contrary to the Commission’s submissions, the fact that the measure is attributable to the Member State concerned […] does not mean that it may be inferred that that Member State exercises a dominant influence over an undertaking in which it is the majority shareholder, within the meaning of the judgment of 16 May 2002, France v Commission (C‑482/99) […] . There is nothing in the State’s conduct as legislator to suggest that it exercised such influence in its capacity as majority shareholder in an undertaking.”
Case C-482/99 was the landmark “Stardust marine” judgment in which the Court of Justice defined a list of indicators which as a group were sufficient to prove attribution to the state, even they were insufficient if examined individually.
This is an important judgment because the Court of Justice made a distinction between the role of the state as a legislator and the role of the state as a shareholder. I wonder, however, how difficult it will be for a state to design laws that influence the decisions of undertakings for the benefit of third parties without having to give specific instructions to those undertakings. The Stardust marine judgment did not only refer to dominant influence exercised through shareholdings but also through other channels that enabled the state to direct decisions of undertakings.
In the landmark PreussenElektra judgment [C-379/98], the Court of Justice examined a similar arrangement as that in Poland and concluded that there was no State aid not because there was no attribution to the state but because there was no transfer of resources over which the state could exercise control.
If one could summarise 15 years of case law in a few sentences, it would be as follows:
- When the aid is paid from the budget of a public authority, there is both state control and attribution to the state.
- When the aid is paid by a company owned or controlled by the state or an agency of the state, there is state control, but there may not be attribution to the state if the company or the agency is acting independently of the state.
- When the aid is paid by third parties compelled to do so by the state and it transits through a channel set up by the state, there is both state control and attribution to the state, regardless of whether that channel is managed or operated by a private entity.
- When the aid is paid by third parties compelled to do so by the state and it transits through a channel which has not been set up by the state and which is not otherwise controlled by the state, there is no state control even though the measure is attributed to the state.
- Lastly, in the present case, when the aid is paid by third parties compelled to do so by the state and it transits through a channel which has not been set up by the state and which is not otherwise controlled by the state, there is neither state control, nor attribution to the state of the specific decisions of the manager of the aid, even though the measure has been established by law.
 The full text of the judgment can be accessed at: