Member States are free to determine the services they consider to be in the general economic interest. The Commission’s power is limited to ensuring absence of any “manifest error”. In its decision, the Commission conducts an extensive review of both the concept of SGEI and what it considers to be a manifest error.
It explains that “(198) the concept of service of general economic interest is an evolving notion that depends, among other things, on the needs of citizens, technological and market developments and social and political preferences in the Member State concerned. The Court of Justice has established that SGEIs are services that exhibit special characteristics as compared with those of other economic activities” .
Since the EU Treaty does not define SGEI, but on the contrary acknowledges that Member States enjoy autonomy in designating their SGEI, while the case law has only referred to the special characteristics of SGEI, the discretion of Member States is quite wide. However, this discretion is constrained by the Commission’s check of manifest error and by any other obligations laid down in secondary legislation such as the directives in the fields of telecoms, transport and energy.
The Commission further clarified that “(200) the entrustment of a particular public service task implies the supply of services which, if it were to consider its own commercial interest, an undertaking would not assume or would not assume to the same extent or under the same conditions.”
“(201) The Commission thus considers that it would not be appropriate to attach specific public service obligations to an activity which is already provided or can be provided satisfactorily and under conditions, such as price, objective quality characteristics, continuity and access to the service, consistent with the public interest as defined by the State, by undertakings operating under normal market conditions.”
The statement in paragraph 201 is often repeated in many Commission documents and decisions. In essence, before Member States designate an SGEI, they need to carry out a market survey and establish that indeed the service they consider essential for the needs of their citizens or a group of citizens [e.g. the old or the young] is not provided by the market.
What is not clear and what is more important for most Member States is whether they are free to combine in a single act of entrustment a bundle of services that contain both profitable and unprofitable market segments. On this point we only have a judgment of the General Court concerning bundling broadband services in profitable and unprofitable geographic areas around Paris [see case T-79/10, Colt Télécommunications France v Commission]. The judgment accepted this geographic bundling, but in the case of broadband which is a network it was rather natural that a network had to be seamless. Since that judgment dealt with very specific issues it is not clear whether it can be generalised.
With respect to the LNG terminal in Klaipeda, the Commission concurred with the views of the Lithuanian government that it was indispensable in ensuring security of supply through diversification of gas sources and that it was justified to impose PSO on the terminal operator.
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 Commission also addressed the arguments of a competitor that the operation of the terminal could not be designated as an SGEI because the LNG market was open and internationally competitive and that there was no evidence of market failure. “(206) The Commission would agree that in a competitive market where security of supply is already ensured by market forces, the designation of an LNG terminal as SGEI would not be warranted. However, the situation of Lithuania in this regard is very specific. While the Commission does not deny that the LNG market in general is an open and internationally competitive market, it also notes that Lithuania has so far no access to that market. Rather, it is fully dependent on one single supply source, which raises security of supply issues.” “(207)
Indeed, […] Lithuania is faced with the situation of a market failure. Despite very high gas prices, the market has so far failed to deliver the infrastructure needed to give Lithuania access to other competing supply sources.” “(208) Moreover, the mere construction of an LNG terminal would not ensure security of supply. Rather, this will require that it is maintained operational even under difficult market conditions and provides third party access to any operator wishing to import LNG with supplies at regulated tariffs, irrespective of the load mode.”
What Lithuania is defining as SGEI is not the provision of gas as such but rather the provision of gas from alternative sources. Its policy objective is diversification of suppliers. This too is a genuine SGEI in the eyes of EU law.
Once the Commission found that the LNG terminal was providing an SGEI, it proceeded to assess whether the aid measure conformed with the remaining conditions in the 2012 Framework.
It agreed that Lithuania had conducted a “public consultation” because the issue was debated repeatedly in parliament.
It also agreed that the PSO was properly entrusted and the obligations of the terminal operator as well as the method of compensation were accurately defined.
With respect to the duration of entrustment which is unusually long [55 years], the Commission concurred that it corresponded objectively to the period of depreciation or amortisation of the pipeline connecting the LNG terminal to the gas network. The Commission classifies the pipeline as “one of the most significant assets required to provide the SGEI (as no gas could be supplied […] without the pipeline)” [paragraph 221]. Although this is undoubtedly correct, is the physical necessity of a rather simple component of a much larger and more complex project the determinant factor for the depreciation of the whole project and therefore the duration of the whole PSO? Probably, the natural life of the pipeline was indeed 55 years. But financially, the minimum period of entrustment is the time that is needed to recoup the cost of investment in tangible and intangible assets. If you invest a 100 in an asset and you earn 10 per year, you need at least 10 years to recoup the investment, regardless of whether the physical life of the asset may be 12 years. Any residual value would have to be accounted at the end of the entrustment period to prevent overcompensation. The 2012 Framework stipulates that the minimum period of entrustment is “the period required for depreciation”. The minimum period required is the period in which an asset is fully re-paid, even if it has a longer useful life. These issues appear not to have received any attention by the Commission. It should have explained its views on the difference between physical and financial depreciation.
Then the Commission considered the problem of the direct assignment of the tasks to the terminal operator, which was majority state-owned, without a public procurement procedure or competitive selection. Lithuania argued that the relevant public procurement directive and internal market rules did not apply. The public procurement directive 2004/18 excluded contracts in case of “protection of the essential interests” of Member States while Treaty articles 52 and 62 allowed for restrictions on the freedom to provide services on grounds of public security.
There is no doubt that the project was of “major importance to Lithuania”. The question is whether the direct entrustment of the SGEI was necessary to protect the interests of Lithuania. In other words, would the operation of a terminal by another company, which would anyway be under the supervision of the energy regulator, have jeopardised national interests? The Commission stated in paragraph 232 that a different company could have or could have developed ties with Gazprom which could influence the market behaviour of that other company! I find this explanation astonishing for two reasons.
First, why would a competitor of Gazprom wilfully enter into a relationship that would enable the dominant company to influence its commercial policy or weaken its ability to fulfil its contractual obligations. Economically it appears to be irrational.
Second, the Lithuanian authorities could have excluded bidders with links to Gazprom or could have provided for penalties in case the operator would have developed ties with Gazprom afterwards. Legally it would have been possible for Lithuania to ensure the non-involvement of Gazprom’s business partners.
Then the Commission tuned its attention to the amount and method of compensation. According to the 2012 Framework, the compensation can be granted ex ante on the basis of expected costs and revenues or ex post on the basis of actual costs and revenue, or a combination of the two. The net costs must be calculated using the net avoided cost methodology. In this case, the whole terminal could not be constructed without aid. Therefore, all costs could be avoided.
The parameters of compensation were indeed determined in advance by the methodology for setting the gas levy and for calculating the subsidy that was necessary to enable the terminal operator to cover its large fixed costs, operating expenses and earn a reasonable return. [Please see above the sections on the guarantee and the levy.]
With regard to the reasonable profit that the SGEI is entitled to, the 2012 Framework defines as a safe harbour the relevant swap rate plus 1%. This can also be considered to be the risk-free rate of profit. If the provision of the SGEI is subject to commercial risk or contractual risk [e.g. because the amount of compensation is fixed or gradually reduced], the rate of profit may be raised but may not exceed a rate of return on capital that corresponds to the level of assumed risk. The corresponding rate may be set by comparison to rates in typical contracts or average rates in the sector or the SGEI provider’s WACC.
For Lithuania the swap rate plus 1% was 4.84%. The Commission, however, rejected this rate for underestimating the rate was suitable for the terminal operator because the swap rate was based on 10-year maturity while the entrustment period was 55 years. The Commission also thought that the terminal operator was subject to substantial risk, the project did not benefit from a guaranteed rate of return, while the subsidy granted to the operator covered a large part of the costs [fixed and some operating costs] but not all of the costs. [The problem, again, is that the relative size of these amounts are not revealed, so it is not possible to know whether indeed the remaining costs of the operator which were not covered by the subsidy did indeed pose a significant commercial risk.] In the end, the Commission accepted that the IRR of the project was a reasonable rate of profit.
The Commission also examined whether the terminal operator was obliged to improve its efficiency, as required by the 2012 Framework. It noted that the national regulator imposed upper limits on the compensation to be provided to certain expenses. Higher costs could not be compensated and lower costs could be compensated only up to their actual level. What the Commission did not explain in the decision was whether costs that fell below the ceiling set by the regulator in one year could be raised again in the next year up to the permissible ceiling or whether the change was only downwards which is in fact the meaning of efficiency gains. At any rate, the Commission also noted that compensation could be raised only up to half of the rate of inflation. This corresponds more closely to the “RPI – X” method of regulating prices of utilities, which force them to increase their efficiency.
Lastly the Commission accepted that the annual check by the national regulator complied with the requirement in the 2012 Framework for establishment of mechanisms to prevent overcompensation. The Framework stipulates that any overcompensation is corrected at the end of the period of entrustment. In this case, the period is too long [55 years]. Hence annual correction goes indeed beyond the legal requirement. In addition, the national regulator by setting the rate of return of the terminal operator would not allow it to gain from excess profits.
Conclusions
The funding of the LNG terminal in Klaipeda is an unusual measure because the same undertaking – the operator of the terminal – received both investment aid that was assessed on the basis of Article 107(3)(c) and compensation for public service obligations that was compatible with Article 106(2).
The Commission approved the two types of aid because they targeted different objectives and because the underlying costs were distinct. Certainly, operating aid could not be approved under Article 107(3)(c). But investment aid could be approved under Article 106(2). Hence it is puzzling why the Lithuanian authorities did not notify the whole amount of the aid on the basis of the 2012 SEGI Framework. It is not clear why the distinction was made between investment and operating aid given that the terminal operator was eligible for both types of aid in order to carry out the public service tasks assigned to it.
There are also two other puzzling aspects of this case. First, the period of entrustment was very long. The justification was that the pipeline connecting the terminal to the gas network had a long life. However, the long life of a component of a larger project does not necessarily make the life of the whole project longer too. And, there seemed to be confusion between physical depreciation [e.g. wear and tear] and financial depreciation.
Second, the contract was directly awarded to the terminal operator which was controlled by the state. Undoubtedly, diversification of gas supplies was of high national importance to Lithuania. This national importance justified deviation from public procurement rules. The question is whether supply diversification could be achieved only by company that was under the direct control of the state. The argument that the operator could have developed links with Gazprom is not convincing either economically or legally.
—————————————————–
[1] The full text of the decision can be accessed at: http://ec.europa.eu/competition/state_aid/cases/250416/250416_1542635_190_2.pdf.
