The duration of entrustment should not exceed the economic life of the investment and the rate of return should reflect the commercial risk of that investment.
Introduction
Is a 55-year entrustment through a direct award compatible with EU rules? And is a profit rate that exceeds the risk-free rate of return reasonable? These are some of the questions that the General Court had to address in its judgment of 12 September 2019, in case T‑417/16, Achemos & Achema v European Commission.[1]
The case concerned the construction of a liquefied natural gas [LNG] terminal in Lithuania to reduce that country’s dependency on Russian energy supplies. In July 2010, Klaipėdos Nafta was appointed to construct the terminal. The Lithuanian government has a 72.3% shareholding in Klaipėdos Nafta.
The financing for the construction of the terminal infrastructure was underpinned by a state guarantee covering 100% of loans amounting to EUR 116 million obtained from the European Investment Bank and other financial institutions. A one-off fee was paid which was equal to 0.1% of EUR 116 million.
In June 2012 a special levy was imposed on all users of the natural gas transmission system. The levy was the so-called “LNG supplement”. That supplement was collected by the transmission system operator and transferred to Klaipėdos Nafta, after approval by the national regulatory authority, for the purpose of covering part of the costs of the construction of the LNG terminal and the related infrastructure.
At the same time, certain companies supplying heat and electricity were required to purchase a minimum amount of gas from the LNG terminal. This was the so-called “purchase obligation”.
In November 2013, the Commission, in decision SA.36740, found that the above three measures [i.e. the guarantee, the supplement and the purchase obligation] constituted State aid which was, however, compatible with the internal market. The guarantee on the financing of the investment [i.e. construction of the terminal] was compatible with Article 107(3)(c) TFEU, while the LNG supplement and the purchase obligation, to the extent that they covered operating costs, were compatible with Article 106(2) TFEU as the terminal was considered to provide a service of general economic interest. The Commission decision was reviewed on the StateAidHub in two parts on 21 June 2016
[View http://stateaidhub.eu/blogs/stateaiduncovered/post/6517]and on 5 July 2016
[View http://stateaidhub.eu/blogs/stateaiduncovered/post/6567].Achemos and Achema appealed against the Commission decision.
The applicants raised three pleas: 1) the Commission failed to initiate the formal investigation procedure; 2) the Commission did not assess correctly the compatibility of the aid with the internal market; and 3) the Commission did not apply the SGEI Framework correctly in relation to the operating aid.
Non-initiation of formal investigation
The General Court began its examination of the first plea by recalling established case law according to which when the Commission encounters serious difficulties and is unable to verify that a public measure either is not State aid or is aid compatible with the internal market, it has no other option but to open the formal investigation procedure laid down in Article 108(2) TFEU. The concept of serious difficulties is an objective one and depends on whether the information that is available to the Commission is sufficient for it to determine the absence of aid or its compatibility. The burden of proof is on those who claim the existence of serious difficulties.
In the present case, the applicants contended that the Commission failed to take into account the existence of an alternative investment project for an LNG terminal undertaken by Achemos. Their view was that there was no market failure and, therefore, no need for state intervention.
The General Court found that “(58) the applicants cannot validly criticise the Commission for not taking into consideration their project to construct an LNG terminal, given that at no time did they make this known to the Commission before the adoption of the contested decision. Although it is true that during the preliminary examination procedure the applicants do not have the right to submit observations, the fact remains that they did not at any point, spontaneously or by lodging a complaint, draw to the Commission’s attention the existence of their project to construct an LNG terminal, all the more so since, in the present case, they had taken the initiative of contacting the Commission concerning the aid measures at issue by letter of 27 September 2012, without, however, mentioning the existence of their project.”
“(60) In those circumstances, the Commission cannot be criticised for failing to take into account matters of fact or of law which could have been submitted to it during the administrative procedure but which were not, since it is under no obligation to consider, of its own motion and on the basis of prediction, what information might have been submitted to it”.
“(64) Contrary to what the applicants claim, the fact that the Commission was in receipt of a complaint […] and received the applicants’ observations in their letter of 27 September 2012 does not establish, in the present case, the existence of serious difficulties with regard to the compatibility of the aid measures at issue with the internal market. Even though complaints submitted by third parties may constitute indications of the existence of serious difficulties, their relevance very much depends on the evidence contained in those complaints and not on the mere fact that observations have been submitted”.
Infringement of Article 107(3)(c) TFEU: Necessity and proportionality of aid
Next, the applicants argued that the aid measures were not needed because the market [i.e. they] could provide gas.
The General Court thought it was appropriate to recall, first, the relevant principles in the case law. These principles were outlined in paragraphs 68 & 69:
First, aid is compatible with the internal market only if it can contribute to the attainment of one of the objectives in Article 107(3).
Second, the aid beneficiary should not be able to achieve those objectives under normal market conditions, using its own resources. That is, the aid must be necessary for the attainment of those objectives and at the same time induce the recipient to act so as to achieve them. It follows that aid which improves the financial situation of the recipient without being necessary for the attainment of the objectives in Article 107(3) is not compatible with the internal market.
Third, market failure is a relevant factor for declaring State aid compatible with the internal market, without being an absolute requirement.
It is not the first time during the past couple of years that the General Court has stated that market failure is not a necessary condition. By contrast, the Commission requires Member States to demonstrate market failure as a justification of their intervention. Whether there is a contradiction between the case law and the principles of compatibility defined by the Commission depends on how market failure is understood. There is no contradiction if market failure is understood to mean inability of the market to provide what the state wants it to provide. In other words, there may not be market failure as defined by economics, but still citizens may not receive what the state believes they need.
Market failure?
Therefore, given the different meanings of the term “market failure”, the General Court went on to make an important clarification. “(70) The absence of market failure does not necessarily mean that the conditions laid down in Article 107(3)(c) TFEU are not satisfied […] For example, State intervention may be considered to be necessary for the purposes of that provision where market forces are not capable by themselves of ensuring that the public interest objective of the Member State is achieved in sufficient time, even if, as such, that market cannot be considered to be failing”.
In other words, an aid granting authority should first establish that the market does not deliver the outcomes which are desirable from a public policy perspective. This means that in practice State aid rules do not prevent Member States from seeking to go beyond market outcomes, as long as such outcomes fall within one of the categories in Articles 93, 106(2) and 107(2) & (3) and they generate social benefits.
Then the General Court proceeded to find that “(71) the aid measures at issue pursued a legitimate objective of common interest, that is to say, ensuring security of gas supply in Lithuania […]. In that regard, the Commission considered that the aid measures were necessary and appropriate for attaining that objective given that the market in Lithuania would not have made it possible, without State support, to develop and finance a project capable of ensuring security of supply, irrespective of the fact that gas prices in Lithuania are among the highest in the European Union […]. According to the Commission, that lack of interest in investing in alternative energy infrastructure is mainly due to the particular situation in the Lithuanian gas market, dominated as it is by a single supplier, Gazprom, which enjoys a commercial margin sufficient to enable it to undercut LNG prices. For those reasons, […], the Commission took the view that, without the LNG supplement and the purchase obligation, the LNG terminal would be neither competitive, nor viable.”
“(73) It should be added that, according to the applicants’ own submissions, their LNG terminal was intended to cover only the gas needs of Achema. Such a project clearly does not meet the objective of security of supply for all Lithuanian consumers pursued by the aid measures at issue”.
“(77) The Commission examined a number of alternative measures, such as increasing the capacity of the interconnector between Lithuania and Latvia, increasing storage capacity in Latvia, building a storage facility in Lithuania and putting in place an interconnector between the gas networks of Poland and those of Lithuania, planned for 2018-2020, but concluded that none of them provided Lithuania with the same degree of security of supply. The applicants are therefore wrong to maintain that the Commission did not examine whether there were other types of investments capable of ensuring the same level of security of supply.”
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.
Incentive effect
Another argument advanced by the applicants was that the aid lacked incentive effect.
As explained by the General Court, in order for aid to be compatible with the internal market, it must have an incentive effect in the sense that in the absence of the aid, the target investment is not undertaken.
In the present case, the applicants contended that the State aid had no incentive effect because Klaipėdos Nafta was required by law to construct the LNG terminal and that the aid measures were adopted after the legal obligation had been imposed on Klaipėdos Nafta.
The General Court explained that the applicants’ arguments were based on the premise that when an undertaking is legally required to implement a project, aid for that project has no incentive effect. “(88) That premise is, however, incorrect. There is nothing in the file, nor do the applicants even contend, that Klaipėdos Nafta had already started to implement the project at issue or, at the very least, that it had the intention of doing so, irrespective of the aid measures at issue.” “(89) On the contrary, […] the construction of the terminal did not commence until after the Lithuanian State undertook to support the project financially.” “(90) A reading of the entire national legal framework, […], shows that the aid measures at issue amounted to a condition sine qua non for implementation of the project.” “(92) Thus, the fact that the obligation to implement the project was legally imposed on the recipient of the State aid concerned does not in any way signify that the aid measures at issue lack incentive effect.”
This is perhaps a surprising conclusion. In principle, any aid that reduces the cost of legally mandated investment of expenditure is incompatible with the internal market because it lacks incentive effect. The aid recipient is obliged to undertake the investment or incur the expenditure anyway. But I suppose that what the General Court really says here is that the obligation and the aid were granted at the same time because Klaipėdos Nafta was unable to construct the terminal without aid. This is the typical situation with the imposition of public service obligations and the granting of public service compensation.
Proportionality
Another argument advanced by the applicants was that the aid was disproportional.
The General Court, first, noted that “(95) in order to be capable of being declared compatible with the internal market pursuant to Article 107(3)(c) TFEU, aid must be aimed at the development of an activity that constitutes a public interest objective and must be appropriate, necessary and not disproportionate”.
In the present case, the applicants disputed that the State aid was limited to the minimum necessary because the capacity of the terminal [4 billion cubic metres per year] was much higher than the total amount of Lithuania’s gas needs.
In this connection, the Commission counter-argued that the size of the LNG terminal was not part of the proportionality assessment of the State aid in question.
Interestingly enough, the General Court disagreed with the Commission. “(98) The Commission’s argument that the capacity of the LNG terminal ‘is not […] part’ of the proportionality assessment of the State aid in question must be rejected. The capacity of that terminal is directly linked to the amount of the aid. The amount of both the investment and operational aid depends, in part, on the capacity of the terminal, given that its size, first, determines the amount and the parameters of the investment, and therefore its cost, and secondly, affects its operational costs, since any unused overcapacity is likely to generate additional costs.”
“(101) That being so, the Commission did not err in considering that the capacity of the LNG terminal did not go beyond what was necessary to attain the general interest objective of ensuring security of supply in Lithuania.” “(102) The Commission rightly considered … that the capacity of the LNG terminal had to ensure basic demand in Lithuania during the peaks associated with the winter months in the event of a disruption of supply. The applicants do not dispute the fact that, during those months, a daily capacity of 11 million cubic metres is necessary, which corresponds to the LNG terminal’s capacity of 4 billion cubic metres per year.”
Duration of entrustment
The applicants acknowledged that the LNG terminal provided a service of general economic interest, but argued that the entrustment period was too long.
The General Court first observed that that “(112) the Commission is not entitled to rule on the scope of the public service tasks assigned to the public operator, in particular the level of costs linked to that service, or the expediency of the political choices made in that regard by the national authorities, or on the economic efficiency of the public operator”.
“(114) According to paragraph 17 of the SGEI Framework, ‘the duration of the period of entrustment [for the provision of the SGEI] should be justified by reference to objective criteria such as the need to amortise non-transferable fixed assets’ and that ‘in principle, the duration of the period of entrustment should not exceed the period required for the depreciation of the most significant assets required to provide the SGEI’.”
In this case the Commission had considered that the period of entrustment of 55 years was equal to the period of depreciation of the pipeline connecting the terminal to the gas transmission system. But the applicants took the view that the Commission failed to explain why the period of entrustment had to correspond to the lifetime of the pipeline connecting the LNG terminal to the gas network, given that the aid was granted to a larger and more complex project, or why that pipeline should be regarded as one of the most significant assets required to provide the SGEI.
“(119) First of all, with regard to the issue whether the pipeline is a significant asset required to provide the SGEI, within the meaning of paragraph 17 of the SGEI Framework, the Commission found, […], that that was the case, given that the pipeline was one of the most important assets needed to implement the SGEI, because, without the pipeline, it would not be possible to supply gas through the LNG terminal. The applicants do not challenge that finding.”
“(120) Nevertheless, it must be stated that paragraph 17 of the SGEI Framework refers to the ‘most significant assets required’, thus in the plural, without laying down more precise rules where the project at issue, serving as the basis for the SGEI, is composed of a number of ‘significant assets required to provide the SGEI’ with different depreciation periods, as in the present case. In paragraph 60 of the contested decision, the Commission identified the different depreciation periods of the other components of the infrastructure of the LNG terminal, that is to say, machinery and equipment (13 years), metering station (18 years), jetty (2 years) and the floating storage and regasification unit (30 years).”
“(121) Insofar as all of those elements, or at least some of them, may be regarded as ‘significant assets required’ to provide the SGEI, within the meaning of paragraph 17 of the SGEI Framework, as with the pipeline, the question arises as to which period of entrustment should be used.”
“(122) In the absence of further details in that regard in the wording of paragraph 17 of the SGEI Framework, it is necessary to interpret it in the light of its objective which is to avoid any overcompensation. According to paragraph 16(e) of that framework, the entrustment act which a Member State must adopt in order to ensure that the SGEI at issue complies with Article 106(2) TFEU, must make provision, inter alia, for ways of avoiding any overcompensation.”
“(123) It must therefore be concluded that the duration of the period of entrustment must correspond to the depreciation period of one of the most significant assets required to provide the SGEI, provided that the resulting remuneration does not lead to any overcompensation.”
“(124) In the present case, the applicants do not advance any argument to demonstrate that the period of entrustment of 55 years would lead to any overcompensation in favour of Klaipėdos Nafta.”
“(125) Furthermore, […] As the depreciation period of the other assets gradually expires during the period of entrustment, the relevant depreciation costs will progressively decrease over time. That decrease will be carried forward to the calculation of the amount of the LNG supplement”.
The question of depreciation appears to be just another technical issue, but in fact it is rather important in the context of SGEI. The Court did not seem to appreciate its importance, perhaps because the Commission’s SGEI Framework does not underline its significance.
There are two facets here which are mixed up: the economic life of an asset and the period of recoupment of the investment in that asset. If an investor could recoup its investment before the economic life of the asset is over, then the market would provide the service in question without any need for state intervention. By contrast, the state needs to intervene, impose a public service obligation and offer public service compensation when the investment cannot be recouped within the economic life of the asset or when the price of the service which is necessary for the recoupment of the investment is too high and consumers cannot afford to pay it. For example, assume an asset has a life of 5 years and the annual net operating revenue [i.e. gross revenue minus operating costs] is 18 [we ignore inflation, time discounting and cost of capital]. The total net revenue over 5 years is only 90. The investor needs State aid of 10 to provide the service. But it can also be the case the net operating revenue is 25 per year so that the investment is recouped in 4 years. In this case the market can provide the service and no state intervention is necessary. However, even in this case the state may decide that the price charged to consumers is too high and impose at the same time price regulation to make the service affordable and offer compensation to cover the revenue shortfall. For example, the price of the service may be reduced to such a level that the net annual revenue drops to, say, 15 which would necessitate compensation of 25 [100 = (15×5) + 25].
This very simple example shows that the period of entrustment depends both on the economic life of the asset and the time which is necessary for the investment to be recouped which in turn depends on the revenue that can be generated which is also affected by possible price regulation.
It is for this reason that the General Court should have been more thorough in its analysis and should have asked whether the cost of the investment could have been recouped earlier than 55 years. The discussion of whether the pipeline was a significant asset was a diversion. The decisive element is the recoupment of the investment. The Commission, in its decision, found that the IRR was lower than the WACC over that 55-year period. But, as the Court acknowledged, the important point here is that the compensation through the LNG supplement was to be gradually reduced to prevent overcompensation. Hence, this implicitly proves that the length of entrustment was conditional on the amount of annual compensation. Lithuania could raise the supplement and reduce the period of entrustment. [In fact, as will be evident below in the last part of the judgment, the planned compensation of Klaipėdos Nafta was not expected to fully cover its operating costs. This necessarily affected the length of the period it could achieve full recoupment]. Therefore, the question that the applicants should have asked is whether Member States are free to lengthen the entrustment period by providing lower annual compensation and whether the calculations looking forward over a 55-year period can be credible. After all, is a 55-year period compatible with Article 106(2) which requires that trade must not be affected to such an extent as would be contrary to the interests of the Union? Is the award of a 55-year contract in the interests of competition and the Union?
Perhaps more importantly and disturbingly, there appears to be a logical gap in the reasoning of the General Court. The length of the period of entrustment has to be fixed on the basis of objective factors and should not result in overcompensation. But in this case the amount of annual compensation is determined by the entrusting public authority which sets the LNG supplement. Therefore, the period of entrustment is indirectly determined by the entrusting authority. The General Court did not seem to be aware of this implication.
Direct award
Another argument of the applicants was the direct entrustment to Klaipėdos Nafta violated public procurement rules and that there were less restrictive alternatives.
It should be recalled that paragraph 9 of the SGEI Framework stipulates that “aid will be considered compatible with the internal market on the basis of Article 106(2) TFEU only where the responsible authority, when entrusting the provision of the service to the undertaking in question, has complied or commits to comply with the applicable Union rules in the area of public procurement. This includes any requirements of transparency, equal treatment and non-discrimination resulting directly from the Treaty and, where applicable, secondary Union law. Aid that does not comply with such rules and requirements is considered to affect the development of trade to an extent that would be contrary to the interests of the Union within the meaning of Article 106(2) TFEU.”
The General Court observed at the outset that the Commission considered that the award to Klaipėdos Nafta could be classified either as a public service contract, falling within the scope of Directive 2004/18, or as a concession. Then it noted that regardless of its classification as a public contract or a concession, the award had to be made through a transparent and competitive procedure, either according to Directive 2004/18 or on the basis of the principles in the Treaty.
The Court went on to point out that “(133) given that the applicants dispute the direct award at issue on the basis of Article 14 of Directive 2004/18, it should be observed that, under that provision, the directive ‘shall not apply to public contracts when they are declared to be secret, when their performance must be accompanied by special security measures in accordance with the laws, regulations or administrative provisions in force in the Member State concerned, or when the protection of the essential interests of that Member State so requires.’”
But such derogations must be interpreted strictly. “(137) Accordingly, a Member State which wishes to avail itself of those derogations must establish that the protection of such interests could not have been attained within a competitive tendering procedure as provided for by Directive 2004/18”.
“(138) In the present case, the Commission found that the Republic of Lithuania had identified its essential security interests which it considered must be protected and the guarantees inherent in the protection of those interests. […] [T]he Commission considered that the Lithuanian State could validly conclude that the SGEI operator had to be controlled by the State and made subject to certain security conditions, and that Klaipėdos Nafta satisfied those conditions. The Commission therefore concluded that the direct award to Klaipėdos Nafta of the LNG terminal project as an SGEI complied with Article 14 of Directive 2004/18.”
“(140) In that regard, the Court considers that none of the alternative measures proposed by the applicants, […], is such as to ensure the protection of the essential interests of the Member State concerned.”
Appropriate rate of return
The last major argument of the applicants was that the rate of return of Klaipėdos Nafta was too high and non-compliant with paragraphs 36 and 38 of the SGEI Framework because Klaipėdos Nafta bore little commercial risk.
According to paragraph 21 of the SGEI Framework, “the amount of compensation must not exceed what is necessary to cover the net cost of discharging the public service obligations, including a reasonable profit”.
According to paragraph 33 of the SGEI Framework, “reasonable profit” means the rate of return on capital, or the internal rate of return which the undertaking obtains on its capital invested over the lifetime of the project.”
Paragraph 36 of the SGEI Framework specifies that “a rate of return on capital that does not exceed the relevant swap rate plus a premium of 100 basis points is regarded as reasonable”.
Paragraph 37 of the SGEI Framework provides that “where the provision of the SGEI is connected with a substantial commercial or contractual risk, for instance because the compensation takes the form of a fixed lump sum payment covering expected net costs and a reasonable profit and the undertaking operates in a competitive environment, the reasonable profit may not exceed the level that corresponds to a rate of return on capital that is commensurate with the level of risk”.
Lastly, paragraph 38 of the SGEI Framework requires that, if the SGEI is not connected with a substantial commercial or contractual risk [e.g. in case of full compensation], the reasonable profit should not exceed the swap rate plus 1%.
The General Court applied the provisions of the SGEI Framework to the present case as follows. “(150) With regard to the internal rate of return of the LNG terminal for the duration of the SGEI, that is to say, 55 years, […], that rate, […], is ‘somewhat’ above the risk-free rate of return (4.84%), but below the regulated weighted average cost of capital (WACC) (7.09%) and ‘clearly below’ the WACC for the sector […] It follows that the internal rate of return of the terminal accepted by the Commission for the duration of the SGEI is between 4.84% and 7.09%.” “(151) According to the applicants, the internal rate of return should, at most, have been set at the same level as the risk-free rate of return, that is to say, 4.84%, in accordance with paragraph 38 of the SGEI Framework, given that the Klaipėdos Nafta does not bear any commercial risk.”
“(153) The Commission considered that an internal rate of return above the ‘risk-free’ rate of return was justified for the following reasons. First, it stated that the SWAP rate of 4.84% was calculated on the basis of a 10-year maturity, while in the present case the duration of the SGEI was 55 years, and that the Commission did not have any data enabling it to calculate a SWAP rate for a maturity of 55 years. In those circumstances, the Commission found that the 10-year SWAP rate referred to above considerably underestimated the SWAP rate which should normally be used as a benchmark in the present case”.
“(154) Secondly, the Commission stated that the SWAP rate plus 100 basis points provided for in the SGEI Framework was not ‘fully appropriate’ in the circumstances of the present case, since the operation of the LNG terminal was subject to commercial risk, despite the compensation provided for. In that regard, the Commission stated that the operation of the LNG terminal takes place on a competitive market and that the purchase obligation applies only for the first 10 years at the most, while the provision of the SGEI at issue must continue for the remaining 45 years in the absence of such an obligation. In addition, according to the Commission, the SGEI provider is not compensated for the entire costs, since the compensation paid ex post covers fixed operating costs in full, but only ‘a small proportion of variable operating costs’, so that ‘at least a part of the variable operating costs’ is not covered by any compensation. In addition, according to the Commission, the project does not benefit from a ‘guaranteed rate of return’, so that if the activities of Klaipėdos Nafta yield less than the regulated WACC, Klaipėdos Nafta will not receive extra compensation”.
“(162) The SGEI compensation comes from two of the aid measures at issue. First, that compensation is financed by the LNG supplement which is determined so as to cover the fixed costs of operating the LNG terminal. Secondly, the SGEI compensation is financed by regasification revenues resulting from regasified volumes corresponding to the purchase obligation at regulated prices. Those revenues cover part of the variable operating costs. […] Thus, the SGEI compensation covers neither the variable operating costs for the regasification of quantities of LNG exceeding those covered by the purchase obligation for the first 10 years of operating the LNG terminal (that is to say, above 0.54 billion cubic metres per year), nor the variable operating costs incurred during the 45 remaining years of the life of the LNG terminal.”
“(164) All of those considerations justify the Commission accepting SGEI compensation of which the reasonable profit somewhat exceeds the applicable SWAP rate plus 100 basis points.”
The reasoning is correct but it does not really explain how the additional risk was translated into the extra return that was added to the risk-free return of the swap rate plus 1%.
———————————————-
[1] The full text of the judgment can be accessed at:
