This is yet another very interesting and well-decided case delivered by the EFTA Court, on alleged State aid granted to the recipient undertaking, which won a tender procedure and was awarded a lease contract. The Court referred to settled case law of the CJEU on private vendor test (PVT) in its attempt to prove that the EFTA Surveillance Authority (EFTA) conducted its preliminary examination in a faulty manner and, therefore, unlawfully upheld in its contested decision that there was no State aid, without initiating the formal investigation procedure. The part of the judgment that has to do with the PVT test refers to the recent landmark case of Land Burgenland[2] which is the main authority on the applicability of the PVT, along with earlier delivered cases[3] of the CJEU.
Facts
The judgment of the EFTA Court under scrutiny relates to an alleged State aid measure granted by the Icelandic Defence Agency to Vodafone. The aid measure consisted in the conclusion of a 10-year lease contract between Icelandic authorities and Vodafone, whereby the latter private operator obtained the use of one optical fibre on the basis of a tender procedure. The tender procedure was originally held to lease two optical fibres, but the second winner of the bidding procedure decided to pull out of the project. The special aspect of that case was the way the selection criteria of the bidding procedure were selected in connection to their appropriateness for warding off any State aid concerns. The assessment of tenders was based on certain criteria, taking into consideration the objectives of the project. ‘Stimulation of competition’ was awarded 40 out of 100 points, ‘rental charge’ 15 points, ‘commencement of services’ 10 points, ‘supply of services’ 10 points, ‘number of network termination points’ 15 points and ‘one tariff throughout the country’ 10 points[4]. What is more, another equally problematic part of the tender procedure, always speaking from a State aid view, was the controversial setting of minimum consideration at ISK 19 million per year per fibre[5]. Although a complaint was lodged with ESA by Mila ehf., a competitor of Vodafone as the beneficiary of the aid, and despite the fact that a 2-year round of information exchange between all the participating parties to the preliminary examination took place, the ESA decided to close the case without to initiate the formal investigation procedure because the lease for the use and operation of an optical fibre did not involve State aid within the meaning of article 61 EEA. Then, the complainant lodged an action with the EFTA Court. In its annulment action relied upon two pleas. First, it argued that ESA erred by not initiating the formal investigation procedure. Second and in the alternative, Mila ehf. asserted that ESA infringed its obligation to provide sufficient reasoning for its findings. The Court went over the first plea of annulment and dispensed with the second. Obviously, it was needless to also examine the second plea on faulty reasoning while it had already found that ESA had actually erred in not initiating the formal investigation procedure. That being its final conclusion it annulled the relevant ESA decision.
Puzzles of the private vendor test (PVT) applicability underlying infringements of an interested party’s procedural rights
At first sight, the judgment seems to address only the alleged violation of the procedural rights of the applicant, by reason of a precocious closing of the preliminary investigation phase, without to initiate the formal investigation procedure. However, that procedural allegation leads to another State aid issue of more substantial nature, that is to say the applicability of the private vendor test[6]. The connection between the actual plea of the applicant (infringement of its procedural right due to the omission to open the formal investigation procedure which was of procedural nature) and of the applicability of the private vendor principle was basically made because the applicant decided to establish the requisite level of serious doubts that ESA should have had on the involvement of State aid, by focusing on the content of the competition authority’s decision[7]. The private vendor test came into play when Mila ehf claimed that ESA made a manifest error of assessment in concluding that the measures did not involve State aid. That manifest error of assessment in essence comprised (apart from an inadequate assessment of facts) a misapplication of case law[8]. Specifically, Mila ehf maintained that the ESA misapplied the market economy investor test (MEIT). Therefore, the applicant tried to show the existence of serious difficulties in determining whether State aid was involved throughout the preliminary examination, by demonstrating that the assessment carried out by ESA was insufficient or incomplete, inasmuch as it misapplied the private investor principle (PIP)[9]. But let’s see whether and most importantly why the private investor test was actually misapplied.
The PIP applies, in general, when public investment, having as a target either a public or a private undertaking, leads to the assumption of risk, for example through the purchase of shares, injection of capital, granting of a loan or provision of a guarantee. On the contrary, the PVT is implemented when a public authority reduces its exposure by, for example by selling capital assets or land[10]. The critical question in our case is which of the two abovementioned tests fits better to the factual scenario where the State opts for leasing its property for a certain time span. The correct answer seems to point to the PVT. The State establishes a leasing contract though which it pursues to eliminate its risk rather than making any sort of investment. As is obvious, all transactional arrangements of the State (either as an investor, vendor or lessor) must rule out any public interest objectives. The State must focus only on one objective: to obtain a market rate of return or profit[11]. In our case, ESA first claimed that it was an inalienable right of the Icelandic authorities to design their tender criteria in a way that would have the ultimate effect of stimulating competition on the relevant market. It further admitted that a number of criteria were used that were not price-related and also contended that the right for a State to determine what it sees as valuable is long established[12]. That statement is absolutely sensible, in terms of the essential role that the national public authorities reserve for them, that is to exercise their public powers as the incumbent regulator and the unique public policy maker. Nevertheless, this view is manifestly extraneous in the context of the State aid assessment. To set as an overarching criterion the stimulation of competition, by requiring applicants not to hold any remarkable market share in the relevant market of the tendered lease contract, unambiguously reflects public interest considerations which must not be taken into account in the context of the PIP to justify further subsidies.
The conditional efficiency of tenders to ensure the market price threshold of the PIP
It is a well-settled principle in both State aid case law and legislation that the results of a tender procedure are a valid indicator of the market price, only where the call for tenders is open, transparent and unconditional[13]. In connection to that, the Court pronounced in the Land Burgenland judgment that where a public authority proceeds to sell an undertaking belonging to it by way of an open, transparent and unconditional tender procedure, it can be presumed that the market price corresponds to the highest offer, provided that it is established, first, that that offer is binding and credible and, secondly, that the consideration of economic factors other than the price is not justified[14]. We also read in the new Draft Commission Notice on the Notion of State Aid Pursuant to Article 107 (1) TFEU that when applying the MEO[15] test, a transaction’s compliance with market conditions can be empirically established through specific market data when it concerns the sale and purchase of assets, goods and services (or other comparable transactions) carried out through an open, transparent, non-discriminatory and unconditional tender procedure[16].
A number of peculiar features of the tender procedure at issue reveal a secret flaw and raise serious doubts as to the validity of the general concept which prescribes that a presumption of market price always exists where the public intervention into the market is made through an open, transparent and unconditional tender. Such a flat or streamlined version of the above presumption is improper to ensure that an advantage granted by State authorities is made on market terms or on the basis of the prevailing market price. Tender clauses which impose minimum price or selection criteria which neatly serve public policy objectives comprise counter-indicators to a State-aid free transaction and deterrent factors on the PIT/PVT applicability respectively and they should make competition authorities more suspicious of possible distortive effects on normal competition conditions. It is actually true that no private operator seeking to lease an infrastructure would set a minimum price or it would regard public policy considerations in its quest for a leasee. It is also settled case law that if any public authority relies on the private investor test during the administrative procedure, it must, where there is doubt, establish unequivocally and on the basis of objective and verifiable evidence that the measure implemented falls to be ascribed to the State acting as shareholder[17]. In that regard, it may be necessary to produce evidence showing that the decision is based on economic evaluations comparable to those which, in the circumstances, a rational private investor in a situation as close as possible to that of the Member State would have had carried out, before making the investment, in order to determine its future profitability[18]. The unfortunate aspect of the case at hand is that none of the above case law prescription was followed by the Icelandic authorities at first place or implemented by the ESA during the preliminary examination of the alleged State aid measure. In the light of the facts at hand, ESA’s main responsibility would be to establish whether the action of the Icelandic public authorities to lease an optical fibre and, primarily, the conditions under which such a lease would be effected ,constituted an investment which could be made by a private investor, hence an action to be attributed to the State acting as an economic operator in the same way as a private operator, or whether it constituted action taken by the State acting as a public authority, thus precluding application of the private investor test[19]. It should have been determined whether the measure in question was of that kind which a market investor, who counts on making a profit in the short or long term, could have granted[20]. In order to assess whether the same measure would have been adopted in normal market conditions by a private vendor in a situation as close as possible to that of the State, only the benefits and obligations linked to the situation of the State as shareholder – to the exclusion of those linked to its situation as a public authority – are to be taken into account[21]. When carrying out that assessment, the manner in which the advantage is provided and the nature of the manner by which the State intervenes are irrelevant where the Member State concerned conferred that advantage in its capacity as shareholder of the undertaking concerned[22]. The ESA should have found that a private lessor would not have entered into such a transaction, framed by those specific terms (minimum price and determination of stimulation of competition and other regulatory goals as prevailing for the selection procedure), and, therefore, it should have pronounced the private investor test as inapplicable in its entirety because the Icelandic authorities pursued, by definition, objectives other than that of making a profit from the resources granted to an undertaking and that those resources were, in principle, granted by the State exercising its prerogatives as a public authority[23]. In that way, it should have concluded that the preliminary examination raised doubts as to whether the lease agreement conferred an economic advantage on Vodafone.
Therefore, the Icelandic authorities failed in the plainest way to design a public tender which could be compatible with the basic principles of the PVT. The price offered by Vodafone might be credible and binding, the consideration of other economic factors might be unnecessary but the tender was still incompetent to guarantee the achievement of a price near to normal market levels. That specific shortfall can be attributed to the choice of setting a minimum price as part of the tender procedure. As previous case law has stressed, the variable of price in tender procedures is a central one with rather profound implications on State aid assessment. First of all, as a corollary and a direct inference, to set a minimum price clause in a public tender procedure has the result of making the whole transfer mechanism unfit for State aid purposes. To spell out this position, by posing a minimum threshold on price which, most importantly, falls short to a great extent from comparable market prices (which was exactly the case here[24]), deprive the tender procedure of its putative market-like character. Thereby, the State fails to qualify as a shareholder or private operator, vendor or even lessor which acts on the basis of rational market conditions or profitability of its investment. No private lessors would have set a minimum price which was so remote from the prevailing market prices. In fact, the real difference between the Land Burgenland principle (point 94 of the judgment: where a public authority proceeds to sell an undertaking belonging to it by way of an open, transparent and unconditional tender procedure, it can be presumed that the market price corresponds to the highest offer, provided that it is established, first, that that offer is binding and credible and, secondly, that the consideration of economic factors other than the price is not justified) and the present case is that in the former, the State still had the discretion to choose, of course under the above highlighted conditions, the lowest bid between two voluntary proposed offers which were shaped freely by the bidders, while in the latter, the State itself ruled out in an ex ante way any sense of a “market power effect” on the proposed price by Vodafone, insofar it imposed as a selection criterion minimum prices which were far from being classified as market prices.
However, the defective nature of the tender procedure, as far as the minimum price is concerned, is further supported by another case law principle. In Land Burgenland case, the Court held that the prices achieved through tender offers do not reflect real market prices, when in the context of a tender procedure which is unlawful on account of the presence of unlawful conditions (here the unlawful condition is the minimum price setting), the latter conditions have the effect of pushing lower the amount of bid[25]. In the opposite situation, where the defective conditions have as a consequence to increase the amount of the bid, there exists no price distortion and the private vendor test normally applies. Under such circumstances, the private vendor would opt in principle for the highest offer, where that offer is binding and credible, regardless of the reasons which led the potential buyer to submit that offer or of the fact that the bid appears to be exorbitant[26]. The vendor must overlook the motives of the individual bidder[27]. Therefore, because the minimum price criterion of the tender procedure was liable for receiving bids that were fairly lower than what free market powers would have delivered, the EFTA Court and the Commission were right to show the red card to ESA and eventually conclude that its preliminary examination misconstrued the conditions of applicability of the private investor test.
Concluding remarks
The very essence of the whole commentary can be boiled down in the following statement: The role of the State as the seller or lessor and its obligations in its capacity as a public authority should not be mixed up[28]. The main lesson to be learnt from the case at hand is that the selection criteria of a public tender which is considered to resemble normal market price mechanisms, should have as an overarching priority, first, to have price as the primary selection criterion, and, second, to avoid any price-related restriction which would lower the amount of the bid compared to the expected normal figures. Finally, the present case also uncovers the weakness of the Land Burgenland principle. It does not suffice to have only a credible and binding offer and to exclude consideration of any other economic factors. That principle is well-designed and efficient to address a great number of cases but its application should be preceded by the ascertainment of whether the tender process is really unconditional, in terms of restrictions on the price selection criterion, as the present case introduced.
[1] Judgment of 27 January 2014, Case E-1/2013, Mila ehf. v EFTA Surveillance Authority (Mila ehf case).
[2] Judgment of 24 October 2013, Joined Cases C‑214/12 P, C‑215/12 P and C‑223/12 P, Land Burgenland v Commission […], ECR (Land Burgenland case)
[3] Judgment of 5 June 2012, Case C‑124/10 P, Commission v EDF and Others, [2012] ECR I‑0000 (EDF case), Case T‑282/08 Grazer Wechselseitige Versicherung v Commission [2012] ECR I-0000, Joined Cases C‑278/92 to C‑280/92, Spain v Commission, [1994] ECR I‑4103 and Case C‑334/99, Germany v Commission, [2003] ECR I‑1139
[4] Mila ehf. case, paragraph 15
[5] Mila claimed during the Court proceedings that under normal market conditions the average market price which a market player would empirically be willing to spare was around 5 times higher (70 to 100 million ISK per year).
[6] The private vendor test is one of the ‘sub-facets’ of the original private investor test. In the present case, it seems that the private investor test could be further compartmentalized into another distinct test, that of the private lessor test. However, this division could be said to be only of theoretical value, since the conditions under which the State is emulated to a private lessor share the same principles and rational with the situation of a private vendor test. The crucial determinant between, on one part, the private investor test and, on the other part, the private vendor and lessor test, which can be seen in unity, is whether the State as a private operator increase or eliminate its risk by entering into a transaction with a third private or public operator.
[7] The existence of serious doubts that necessitate the initiation of the formal investigation procedure can be demonstrated by reference to a body of consistent evidence concerning, first, the circumstances and the length of the preliminary examination procedure and, second, the content of the contested decision. See to that effect, Mila ehf. paragraph 80 & Judgment of 11 December 2012, Case E-1/12, Den norske Forleggerforening v EFTA Surveillance Authority, paragraph 108
[8] Mila ehf case, paragraph 62
[9] Mila ehf case, paragraphs 91 & 101
[10] State aid blog lexxion , Phedon Nicolaides – Private Vendor Principle: How to Organise a State-Aid-Free Sale of Public Assets, available at: http://www.lexxion.eu/training/stateaidblog/2013/11/19/81-private-vendor-principle-how-to-organise-a-state-aid-free-sale-of-public-assets
[11] Ibid
[12] Mila ehf case, paragraph 72
[13] Land Burgenland case, paragraph 89
[14] Land Burgenland case, paragraph 94
[15] MEO (Market Economy Operator) test is an alternative expression for the Private Investor Test (PIT). The concept of the PIT is expressed in practice with multiple terms which are used interchangeably. See Phedon Nicolaides – Essays on Law & Economics of State aid, Chapter 14, Re-introducing the Market in the “Market Economy Investor” Principle on the succession and substitution of the term MEO by the PIT.
[16] See paragraph 87 of the Draft Commission Notice on the notion of State aid pursuant to article 107 (1) TFEU
[17] EDF case, paragraph 82
[18] EDF case, paragraph 84
[19] EDF case, paragraph 30
[20] Land Burgenland case, paragraph 48
[21] Land Burgenland case, paragraph 52
[22] Land Burgenland case, paragraph 53
[23] Land Burgenland case, paragraph 56
[24] See footnote no.5 above
[25] Land Burgenland case, paragraph 96
[26] Land Burgenland case, paragraph 99
[27] Phedon Nicolaides, Private Vendor Principle: How to Organise a State-Aid-Free Sale of Public Assets, available at: http://www.lexxion.eu/training/stateaidblog/2013/11/19/81-private-vendor-principle-how-to-organise-a-state-aid-free-sale-of-public-assets
[28] Land Burgenland case, paragraph 16