COMMISSION DECISION (EU) 2018/556

of 25 August 2017

on the state aid SA.35356 (2013/C) (ex 2013/NN, ex 2012/N) implemented by Poland for Autostrada Wielkopolska S.A.

(notified under document C(2017) 5818)

(Only the Polish text is authentic)

(Text with EEA relevance)

THE EUROPEAN COMMISSION,

Having regard to the Treaty on the Functioning of the European Union, and in particular the first subparagraph of Article 108(2) thereof,

Having regard to the Agreement on the European Economic Area, and in particular Article 62(1)(a) thereof,

Having called on interested parties to submit their comments pursuant to the provisions cited above(1) and having regard to their comments,

Whereas:

1.   

PROCEDURE

(1) By electronic notification of 31 August 2012, registered on the same day, Poland notified to the Commission under Article 108(3) of the Treaty on the Functioning of the European Union (‘TFEU’) State aid in the form of a financial compensation to Autostrada Wielkopolska S.A (‘Autostrada Wielkopolska’). The measure was registered under the State aid case number SA.35356.

(2) By letter dated 26 October 2012, the Commission requested further information on the notified measure. Poland provided the requested information by letter of 28 November 2012.

(3) By letter dated 29 January 2013, the Commission informed Poland that it would transfer the case to the register of non-notified cases (NN case), due to the fact that the compensation notified to the Commission had already been granted. In addition, the Commission requested further information. Poland provided the requested information by letters dated 18 February and 16 April 2013.

(4) On 11 July 2013, the Commission and Poland had a meeting to discuss the measure, after which the Commission requested further information by letter dated 22 August 2013. Poland provided the requested information by letter dated 20 September 2013.

(5) By letter dated 25 June 2014, the Commission informed Poland that it had decided to initiate the procedure laid down in Article 108(2) TFEU in respect of the notified measure.

(6) By letter dated 5 September 2014, Poland submitted its comments on the decision to initiate the procedure laid down in Article 108(2) TFEU.

(7) The Commission's decision to initiate the formal investigation procedure was published in the

Official Journal of the European Union

on 20 September 2014(2). The Commission invited interested parties to submit their comments on the measure.

(8) On 7 October 2014, the Commission received comments from Autostrada Wielkopolska. By letter dated 26 November 2014, the Commission forwarded those comments to Poland which was given the opportunity to react.

(9) By letter dated 5 December 2014, Poland requested from the Commission a translation into Polish of the comments of Autostrada Wielkopolska. The Commission provided Poland with the requested translation by letter dated 23 January 2015.

(10) Poland submitted its comments by letter dated 23 February 2015.

(11) By letters dated 26 June 2015 and 20 April 2016, the Commission requested further information. Poland replied to these requests by letters dated 10 and 17 July 2015 and 18 May 2016.

(12) On 7 December 2016, the services of the Commission and the Polish authorities had a conference call. By letter dated 12 May 2017, the draft minutes of the conference call were sent to Poland by the Commission. At the same time, the Commission requested additional information from Poland.

(13) Poland confirmed the minutes from the conference call and provided additional information by letter dated 23 May 2017.

2.   

DETAILED DESCRIPTION OF THE MEASURE

2.1.   

Introduction

(14) The measure consists in the grant of financial compensation by Poland to Autostrada Wielkopolska(3), the concession holder of a part of the Polish A2 motorway, between Nowy Tomysl and Konin, for a legislative change excluding heavy goods vehicles (hereinafter: ‘HGVs’) from the obligation to pay a toll for using the motorway in the period from 1 September 2005 to 30 June 2011.

(15) The concession for the construction and operation of this part of the A2 motorway was granted to Autostrada Wielkopolska on 10 March 1997 following a public tender. The concession agreement (hereinafter: ‘the Concession Agreement’ was signed by the relevant minister and Autostrada Wielkopolska on 12 September 1997. The concession was granted for a period of 40 years, i.e., until 10 March 2037.

(16) Under the Concession Agreement, Autostrada Wielkopolska committed to obtain (at its own cost and risk) external funding for the construction, operation and maintenance of the A2 motorway section between Nowy Tomysl and Konin. In exchange, Autostrada Wielkopolska received a temporary right to operate the constructed motorway section, including the right to collect and retain all tolls from all its users.(4)

(17) Five rates were applied to users of the motorway, depending on the vehicle category defined by Polish law. Under the Concession Agreement, the rates could be increased by the concession holder to maximise revenue but they could not exceed the maximum rates specified in Table 1 below.

Table 1

Maximum toll rates according to Annex P to the Concession Agreement (net of VAT

(5) )

2.2.   

The introduction of the compensation mechanism

(18) Upon joining the European Union in 2004, Poland had the obligation to transpose Directive 1999/62/EC of the European Parliament and of the Council(6) into Polish law. Article 7, paragraph 3 of this Directive provides that: ‘[t]olls and user charges may not both be imposed at the same time for the use of a single road section.’

(19) Therefore, in 2005, the Polish Parliament adopted an amendment to the Act on Toll Motorways and the National Road Fund (hereafter: ‘the Amendment Act’)(7) that eliminated the double charging for HGVs(8) for the use of a single road section. Accordingly, HGVs paying the vignette (a toll card) for using national roads in Poland were exempted, as from 1 September 2005, from tolls on motorways covered by concession agreements.

(20) The Amendment Act provided that the various concession holders should be compensated for the lost revenue by the National Road Fund (hereafter: ‘NRF’).

(21) The compensation method (the principles of which were set up by the Parliament) was based on the introduction of a shadow toll, namely a contractual payment made by the State to the concession holders per HGV using a toll motorway.

(22) Under the Amendment Act, the concession holders were entitled to a reimbursement equivalent to 70 % of the amount obtained by multiplying the actual number of journeys of HGVs on the toll motorway by the shadow tolls negotiated with the concession holders for HGVs depending on their categories. The reduction to 70 % in the shadow tolls was introduced to compensate for the expected but unknown increase in traffic of HGVs once they are exempted from paying tolls.

(23) According to the Amendment Act, the compensation method and the deadlines for reimbursement of funds were to be specified in each concession agreement.

(24) In the case of Autostrada Wielkopolska, following negotiations the compensation method and the shadow tolls were set out in Annex 6 to the Concession Agreement (hereafter: ‘Annex 6’), concluded on 14 October 2005.

2.3.   

Basic rules for the calculation of the compensation

(25) Poland explained that the compensation method described in Annex 6, in line with the Amendment Act, was based on the principle that the expected financial situation of the concession holder should not change. This means that the expected financial situation of the concession holder with the compensation should be neither better nor worse compared to the financial situation it could expect just before the legislative change (‘counterfactual’). According to Poland, it was agreed that this objective would be met if the expected Internal Rate of Return (hereafter: IRR(9)) for the investment of Autostrada Wielkopolska in the relevant section of the A2 motorway stayed at the same level as it would have been just prior to the legislative change (i.e., without the revenues foregone due to the legislative change).

(26) The parties also agreed that the calculation of the compensation to the concession holder would be the outcome of a two-step procedure conducted on the basis of financial models prepared by an independent expert. The financial models would show the actual up-to-date cash flow as well as the foreseen cash flow (i.e. costs incurred and to be covered by the concession holder and actual and foreseen revenues from tolls) and would allow to calculate the IRR for the investment. In the first step of the procedure the (shadow) toll rates for HGVs which the state had to pay to the concession holder were to be negotiated and determined (

setting of the shadow toll rates

). After an initial period of validity, in a second step, these rates were to be verified and changed if necessary (

verification of the shadow toll rates

). These basic principles were applied to all motorway concessions.

(27) In accordance with the provisions of the Amendment Act, Annex 6 distinguished between two types of rates:

1.   

Applied Rates

: the rates applicable on the effective date of the Amendment Act (1 September 2005) to all vehicles using the motorway. These rates were set according to the original provisions of the Concession Agreement.

2.   

AGRi Rates

(or shadow toll rates)

: the rates negotiated for HGVs with a vignette in Annex 6 on the basis of which the compensation to be paid to the concession holder is calculated.

(28) According to Article 3 of the Amendment Act, the AGRi Rates for each vehicle category could not exceed the Applied Rates for this vehicle category. In line with the Amendment Act, Autostrada Wielkopolska was to receive 70 % of the AGRi rate per HGV with a vignette entering the operated motorway as already mentioned in recital 22 above. Annex 6 to the Concession Agreement specifies in paragraph 4 point c that the AGRi rates are net rates and will be increased with the equivalent of the applicable value added tax (VAT).

2.4.   

Step one – setting the AGRi Rates

(29) In the first step of the negotiations of the compensation, prior to the effective introduction of the shadow toll, Autostrada Wielkopolska presented three financial models on the basis of which it justified the AGRi rates it proposed. The models were verified for accounting correctness by the financial advisor of the Polish authorities — Ernst & Young Corporate Finance Sp. z o.o. The three models showed the financial parameters of the concession including the IRR in three different scenarios and points in time:

(a) the Base model

shows the financial situation of the concession holder at the moment of financial closure in 2000(10). It assumes real toll collection from the beginning until the end of the concession. This model is based on traffic, revenue and cost assumptions valid at the time of financial closure. The expected IRR in this model is […] %;

(b) the Real Toll model

describes the financial situation that would prevail as of December 2004 if

HGVs were not exempted

from toll. It assumes real toll collection for all vehicle categories including HGVs until the end of the concession. The model presents actual cash flows for the period from the start of the Concession until the end of 2004 (historical period) and forecasts the cash flows for the period from 2005 until 2037 (forecast period)(11). The expected IRR in this model is […] %. This was the most up-to-date financial model at the disposal of the negotiating parties, presenting the financial situation of Autostrada Wielkopolska prior to the Amendment Act which showed the expected return if HGVs were not exempted from toll;

(c) the Vignette model

describes the financial situation that would prevail as of June 2005 if

HGVs were exempted

from toll. It assumes that the revenues of the concessionaire include the shadow toll compensation for HGVs (70 % of the maximum allowed AGRi rates) from 1 September 2005 until the end of the Concession Agreement and real toll collection for all other vehicles until the end of that agreement. The model presents actual cash flows for the period until June 2005 (historical period) and forecasts the cash flows for the period from July 2005 until 2037 (forecast period)(12). The shadow toll rates used in this model are set at the maximum level(13) allowed in the Concession Agreement. The expected IRR in this model is […] %(14).

(30) With these models Autostrada Wielkopolska demonstrated that even with the maximum allowed shadow toll rates the IRR of […] % from the Real Toll model would not be reached. For this reason the concession holder set the AGRi Rates at the maximum allowed levels in the Concession Agreement (see footnote 12)(15). These levels were subsequently subject to indexation on the terms provided for in Annex 6.

(31) As from 1 September 2005, HGVs which had settled the vignette fee became exempted from tolls on the A2 motorway and the concession holder started to receive each month compensation calculated on the basis of the number of HGVs with a vignette entering the motorway and the agreed AGRi Rate.

2.5.   

Step two – verification of the AGRi Rates

(32) In Annex 6 to the Concession Agreement, the parties agreed to verification of the AGRi Rates by 30 November 2007. The aim was to check how HGVs traffic developed as a result of the disappearance of toll payments and adjust the AGRi rates accordingly in order to avoid over- or underpaying the concession holder. The review of the AGRi Rates was obligatory and could lead to a decrease or an increase of the AGRi Rates and hence to the decrease or increase of total compensation disbursed to the concession holder.

(33) By 1 July 2007 at the latest, the concession holder was to submit an up-to-date financial model (‘Verification model’) to the General Directorate for National Roads and Motorways (Generalna Dyrekcja Dróg Krajowych i Autostrad; hereinafter: ‘GDDKiA’), indicating the impact of the AGRi Rates on the basic financial indicators of the Concession Agreement including the IRR(16).

(34) If the IRR in the Verification model exceeded the IRR in the Real Toll Model, the AGRi Rates would have to be lowered in order to liquidate the excess rate of return. Similarly, if the IRR in the Verification model was lower than the one envisaged in the Real Toll model, the AGRi Rates would have to be increased. The concession holder had to make a proposal to GDDKiA for the adjustment of the AGRi Rates.

(35) By 15 October 2007 at the latest, GDDKiA was required to notify to the concession holder whether it accepted the proposed adjusted AGRi Rates. If the proposed adjusted AGRi Rates were not accepted by this date, the parties were required to take action to reconcile the different positions. In case of lack of agreement by 30 November 2007, the amount of compensation to be paid to the concession holder was to be calculated based on the hitherto applicable AGRi Rates. Each party was entitled to refer the issue of the adjustment of the AGRi Rates for settlement according to the procedure stipulated in Article 24.3 of the Concession Agreement (i.e., arbitration).

(36) Hence, verification of the AGRi Rates was supposed to be based on the comparison of the IRR expected in the Real Toll Model with the IRR expected in the Verification Model. In 2007 Autostrada Wielkopolska provided the

Verification model

. This model shows the financial situation of the concession holder that would prevail as of December 2006. It is built on actual data for both light vehicles and HGVs as well as costs for the period from the start of the Concession until December 2006 and traffic, revenue and costs projections for the period afterwards until the end of the Concession in 2037. The Verification model assumes that the shadow toll system is in place until the end of the Concession. It calculates the expected IRR in June 2006 at […] %.

(37) The related verification report(17) submitted by Autostrada Wielkopolska suggested the AGRi Rates should be increased (thereby increasing the amount of compensations payable to Autostrada Wielkopolska). GDDKiA's advisors expressed no material reservations concerning that verification report.

(38) Autostrada Wielkopolska and GDDKiA however failed to achieve agreement on the adjusted AGRi Rates by 30 November 2007. By letter of 28 November 2007, GDDKiA informed Autostrada Wielkopolska that in view of doubts regarding the correctness of assumptions made for the purposes of Annex 6 — which could not be corrected as part of the verification report — GDDKiA did not accept the proposed adjusted AGRi Rates.

(39) This started a disagreement between the parties regarding the correctness of the AGRi Rates and the level of the actual amount of compensations payable to Autostrada Wielkopolska. However, in line with the provisions of Annex 6 to the Concession Agreement, the concession holder still received every month compensation payments based on the shadow toll.

2.6.   

Dispute concerning the Real Toll Model

(40) According to Poland, Autostrada Wielkopolska had overvalued the expected rate of return in the Real Toll model by using outdated traffic and revenue forecasts. Autostrada Wielkopolska has used the traffic and revenue study for the A2 motorway made by the consulting firm Wilbur Smith Associates (WSA) in 1999(18) (hereinafter ‘1999 WSA study’) instead of the updated study by WSA from June 2004(19) (hereinafter ‘2004 WSA study’) which was available at the time of the negotiations of Annex 6. The 2004 WSA study predicted significantly lower traffic and revenues than the 1999 WSA study. Consequently, the IRR of the Real Toll model would have been lower if it had been calculated on the basis of the up-to-date data of the study from 2004 instead of the outdated study from 1999. In other words, Autostrada Wielkopolska had presented, according to Poland, its financial situation immediately before the entry into force of the Amendment Act in a dishonest manner, which led to compensation in amounts exceeding the revenue that Autostrada Wielkopolska would have been capable of generating had the 2004 study been used.

(41) According to a report(20) procured by GDDKiA and the Ministry of Infrastructure, prepared by PricewaterhouseCoopers (hereinafter: ‘PwC report’), using the traffic and revenues assumptions from the 2004 WSA study instead of the 1999 WSA study (coupled with small methodological and tax corrections) reduces the IRR of the Real Toll model from […] % to […] %.

(42) On 13 November 2008, the Minister of Infrastructure made a statement of avoidance(21) of Annex 6 in the form of a letter addressed to AWSA, claiming, inter alia, that the Minister of Infrastructure was in error when concluding that agreement, since he had not been aware that Autostrada Wielkopolska failed to take into account the forecasts of traffic and revenue, for the section of A2 motorway between Nowy Tomysl and Konin, prepared in 2004, and hence more up-to-date than the forecasts prepared within this scope in 1999, used during negotiations on Annex 6. In the Minister's opinion, the fact that Autostrada Wielkopolska did not use the most up-to-date data on traffic and revenue levels on the date of signature of Annex 6 caused this annex to erroneously reflect Autostrada Wielkopolska's financial situation before the Amendment Act introduced the shadow toll system and the associated obligation to compensate the concession holder.

(43) In the opinion of the Minister of Infrastructure, supported later by the PwC report, Autostrada Wielkopolska received excessive compensation. The overpayment for the period from 1 September 2005 to 30 June 2011 amounted to PLN 894 956 889 gross(22) (around EUR 224 million) (hereinafter: the ‘Overcompensation’).

(44) Since Autostrada Wielkopolska refused to repay the Overcompensation, the Minister and GDDKiA, in a letter of 8 November 2010, applied to the General Counsel to the Treasury (Prokuratoria Generalna Skarbu Państwa) for commencement, on behalf of the State Treasury, of legal proceedings and for legal representation within the scope of recovery of amounts unduly paid to Autostrada Wielkopolska as compensation.

(45) At the same time, Autostrada Wielkopolska contested the repudiation of Annex 6 by bringing the case before an arbitral tribunal.

(46) In March 2013, the Arbitral Tribunal decided in favour of the concession holder, stating that Annex 6 is valid and should be respected by the State. In June 2013, the Polish authorities decided to challenge the arbitral award by lodging an annulment action in the competent national court. The case is pending.

2.7.   

End of compensation mechanism

(47) The need for a shadow toll compensation expired on 30 June 2011, which is when Poland introduced an electronic toll collection system called ‘via TOLL’ that replaced the vignettes. The ‘via TOLL’ system is obligatory for all HGVs and allows its users to perform electronic payment settlement for the use of toll road networks based on kilometres and on a road category.

(48) The ‘via TOLL’ system covers only selected roads and not the whole road network in Poland (as the vignettes did), thereby eliminating the possibility of double charging. Therefore, as of 1 July 2011, the concession holders could return to charging all HGVs entering toll motorways according to the rules set up in the concession agreements.

2.8.   

Grounds for initiating the procedure

(49) In its opening decision, the Commission raised doubts about the compatibility of the measure with the internal market. In particular, it highlighted the following questions:

— First, whether the compensation payments, the methodology for their calculation and/or the way the methodology was applied, granted an economic advantage to Autostrada Wielkopolska that went beyond what was necessary to preserve the economic balance of the Concession Agreement. In particular, the Commission asked whether:

(a) the Concession Agreement gave Autostrada Wielkopolska a right to be compensated in the event of legislative changes;

(b) the application of the compensation mechanism resulted in the State taking over, at least in the period from 1 September 2005 to 31 December 2006, of traffic and revenue risks related to light vehicles, and what was the financial benefit to the concession holder of this transfer;

(c) the use of the 1999 study of traffic and revenue for the calculation of compensation resulted in an additional economic advantage to AW S.A that was not necessary to remove the negative effects of the legislative change for the concession holder and what was the scale of this advantage.

— Second, whether the compensation paid to Autostrada Wielkopolska for the legislative change exempting HGVs from the obligation to pay a toll for using the motorway in the period from 1 September 2005 to 30 June 2011 is in whole or part an incompatible operating aid.

3.   

COMMENTS FROM POLAND

3.1.   

Economic advantage

3.1.1.   

Autostrada Wielkopolska's right to compensation

(50) Poland is of the opinion that Autostrada Wielkopolska had a right to compensation. Firstly, Poland points out that the Amendment Act was enforced because the Commission requested Poland to eliminate double payments by HGVs using toll motorways. In the interest of the correct implementation of the applicable Directive 1999/62/EC, the Amendment Act had to include a compensation mechanism to reimburse motorway operators for revenue lost due to the change in law.

(51) Alongside the Amendment Act, also the Concession Agreement, in Article 12.6, granted to Autostrada Wielkopolska the right to be compensated. This article stipulates that: ‘The Concession Holder shall be entitled to compensation from the State Treasury for any loss of Revenues from Operation incurred by the Concession Holder if due to any Changes in Requirements or otherwise the Concession Holder is prevented by the Public Authority's action from changing and adjusting the Toll Rates as provided for in Article 12.6.’ In terms of changing and adjusting the Toll Rates, Article 12.6. gives the concession holder the right to increase and decrease the toll rates at its own discretion within the limits of the regulations referred to therein(23). Poland argues that the Amendment Act, which was not foreseeable at the time of signing the Concession Agreement, affects the concession holder and, as such, constitutes a Change in Requirements.

(52) Poland argues that Autostrada Wielkopolska's right to compensation is also supported by the risk provisions of the Concession Agreement which assign risk of traffic and revenue from tolls to the concession holder. In return, the concession holder is granted the right to collect tolls. As the Amendment Act deprived Autostrada Wielkopolska of its right to collect tolls relating to HGVs, a compensation mechanism was necessary to re-establish the economic balance of the Concession Agreement.

(53) Poland concludes by pointing out that in the decision of December 2013 concerning the case SA. 29584, the Commission assumed that the concession holder of the A4 motorway had a right to a similar compensation which was caused by and calculated according to the same Amendment Act.

3.1.2.   

Light vehicles traffic

(54) On the transfer of the risk of light vehicles traffic between 2005 and the verification in 2007, presented in recitals 65-75 of the decision to open formal investigation procedure, Poland states that the mechanism used to calculate and to verify the shadow toll for the concession holder of the A2 motorway was different that the one accepted by the Commission in case SA.29584 for the concession holder of the A4 motorway.

(55) While for the A4 motorway, after the period during which the shadow toll level was verified, only data concerning HGVs were updated in the applicable financial model and all other data (including those pertaining to light vehicles) remained unchanged, for the A2 motorway after the shadow toll test period all data (including those related to light vehicles) were updated.

(56) Poland considers that in the case of A2 motorway the application of the compensation mechanism could potentially be seen as taking over the risk of traffic and revenue from light vehicles in the period from September 2005 to 31 December 2006 but in Poland's view it was necessary because the adopted methodology, that was supposed to preserve the financial situation of Autostrada Wielkopolska, required that the financial model covers the project as a whole (i.e., also light vehicles). In Poland's view, it was impossible to calculate the IRR for the project looking at HGV data only.

3.1.3.   

Traffic Studies — Internal Rate of Return

(57) As regards recitals 76 to 81 of the decision to open formal investigation procedure, Poland states that the calculation mechanism of the compensation, provided for in the Amendment Act and put in place by Poland, is correct. However, the concession holder was overcompensated due to the use of incorrect input data.

(58) Poland explains that to calculate the IRR, the traffic and revenue study from 1999 was used although a more up-to-date version of that study from 31 December 2004 existed. When using the 2004 study, the calculated IRR is significantly lower - […] % instead of […] % if calculated with the 1999 study. Poland argues that through this, ‘the concession holder was restored to its financial situation that existed in 2000 instead of that which had existed prior to the effective date of the Amendment Act and the date of Annex 6 (2005)’ and thereby the business risk of operations between 2000 and 2005 was transferred to the State. Poland had not guaranteed, however, that any specific IRR would be achieved at the conclusion of the Concession Agreement.

(59) Poland argues that the above issue was analysed in detail by Scott Wilson Sp. z o.o. (hereinafter: ‘Scott Wilson’), which came to the conclusion, when analysing all the traffic and revenue forecasts presented by Autostrada Wielkopolska in the period from 1998 to 2004, that the use of the study from 1999 was not appropriate given that a more up-to-date study from 2004 existed, as it lead to a significantly higher IRR as compared to the use of the study from 2004.

(60) Because Autostrada Wielkopolska was compensated to attain the IRR expected in 2000 and not the IRR Autostrada Wielkopolska could realistically have expected to achieve right before the Amendment Act entered in force, Autostrada Wielkopolska was put in a better situation than it would have been without the change of law and this constitutes an economic advantage.

(61) Moreover, Poland points out that the Concession Agreement in Article 25.9 limits the compensation to Autostrada Wielkopolska to what it would have obtained without the legislative change.

3.2.   

Compatibility Assessment

(62) In the view of Poland, no grounds exist to consider the aid in question compatible with the internal market.

4.   

COMMENTS FROM INTERESTED PARTIES

(63) The Commission received comments from Autostrada Wielkopolska as an interested party.

(64) In the introduction, Autostrada Wielkopolska states that Poland's notification of the compensation as incompatible State aid is simply a move in a commercial dispute after Poland lost the proceedings in front of the Arbitral Tribunal and is also now trying to set aside the award. […]. Autostrada Wielkopolska perceives the notification to the Commission as forum shopping on the side of Poland which should not be supported as this is contrary to the spirit and purpose of the EU's State aid policy. Autostrada Wielkopolska considers it would be contrary to public interest if Member States were allowed to abuse State aid rules to void agreements that they entered into at arm's length in commercial settings.

4.1.   

Economic advantage

4.1.1.   

Autostrada Wielkopolska's right to compensation

(65) Autostrada Wielkopolska considers that under the Concession Agreement, it was entitled to seek compensation after being deprived of the right to collect tolls from HGVs and therefore deprived of that source of income.

(66) Additionally, Autostrada Wielkopolska points out that it was never disputed that Autostrada Wielkopolska had a right to compensation nor was it disputed that the compensation would be in the form of a shadow toll.

4.1.2.   

Light vehicles traffic

(67) According to Autostrada Wielkopolska, the inclusion of light weight traffic in the shadow toll verification method was necessary to correctly assess the compensation amount and therefore it does not create an advantage to Autostrada Wielkopolska

(68) Firstly, the company explains that its situation has to be contrasted with that of the A4 motorway's concession holder. The A4 and A2 motorways differ in the way that the A4 is mainly a commuter road where only 20 % of the traffic is made up of HGVs (which Autostrada Wielkopolska estimates represent about 30 % of revenues). For the A2 motorway, being an export corridor, a bigger share, namely […] % of vehicles were HGVs (representing […] % of all revenues). It was clear to all parties involved that the abolition of a real toll payment would lead to a significant increase in HGVs traffic. As a consequence, the attractiveness of the motorway would decrease in a shadow toll scenario (increased congestion, higher maintenance needs resulting in more works with reduced traffic speed), which would result in different route choices and a considerable decrease in the light vehicle flows on the motorway. These factors have a significantly higher impact on a motorway in an export corridor than on a commuter road. Therefore, the number of light vehicles had to be updated in the case of the A2 motorway.

(69) Secondly, Autostrada Wielkopolska brings forth that Autostrada Wielkopolska was under an obligation to undertake construction works as a function of certain traffic thresholds being exceeded (e.g. a third lane on the Poznan bypass) leading to substantial capital expenditures. A verification of only HGVs would have ignored the implications that the introduction of shadow tolls had on other traffic and the milestones for additional construction work under the Concession Agreement.

4.1.3.   

Traffic Studies — Internal Rate of Return

(70) First of all, Autostrada Wielkopolska underlines that when entering, in 2005, the renegotiation of the Concession Agreement it targeted a level of compensation that would allow for an IRR estimate similar to that at the financial closure in 2000, which according to AW S.A was […] %.

(71) While it was agreed by the parties that the AGRi Rate, after the 2007 verification, would be set to target an IRR of […] %, in Autostrada Wielkopolska's opinion it is of critical importance that the AGRi Rate was also subject to an absolute cap on toll rates.

(72) Autostrada Wielkopolska estimated

ex ante

in 2005 that under the shadow toll mechanism the absolute cap on toll rates contained in Annex 6 limited Autostrada Wielkopolska's expected IRR to […] % (i.e., well below […] %). In 2007, the Verification Report increased the AGRi Rate charged in the interim period up to the cap. As the result, after the 2007 verification, Autostrada Wielkopolska's expected IRR was […] %.

4.1.3.1.   

London Underground

(73) Autostrada Wielkopolska states that the Concession Agreement is a long-term and a particularly complex contract as it involves the construction and operation of a motorway. To determine if an alteration of such a complex agreement between a State and a private party constitutes State aid, the principles established by the Commission in the London Underground case(24) should be applied. According to those principles, an alteration does not constitute State aid if:

(a) The original Concession Agreement does not entail State aid;

(b) The alteration either does not lead to a material improvement for the concession holder or the improvement is at market rate.

(74) Autostrada Wielkopolska explains that since the Concession was awarded following an open, transparent and non-discriminatory tender, the Concession Agreement benefits from the presumption of being concluded on market terms and therefore does not constitute State aid. According to AW S.A, in the case in question, the expected real, post-tax IRR on shareholders' capital (equity and subordinated debt) resulting from the initial Concession Agreement was […] % at the time of financial closure. As this was the expected rate of return implicit in the original agreement, such return should be considered market conform.

(75) Autostrada Wielkopolska compares the IRR of […] % expected after the conclusion of Annex 6 with the IRR resulting from the initial tender of […] %. Autostrada Wielkopolska explains the increase of […] % results from external financial considerations, namely replacing budget forecasts with actual financial results and updating inflation and foreign exchange forecasts. According to AW S.A, the increase was therefore entirely justified. Moreover, there was no material improvement in Autostrada Wielkopolska's position purely as a consequence of the switch from real tolls to shadow tolls (irrespective of any change in the IRR). In particular, the switch did not reduce Autostrada Wielkopolska's exposure to traffic risks.

(76) Autostrada Wielkopolska argues further that even if the alteration that Autostrada Wielkopolska sought could be regarded as ‘material’ within the meaning of London Underground (which Autostrada Wielkopolska does not concede), it would still have resulted in Autostrada Wielkopolska having an expected rate of return which is by no means above market rates. AW S.A notes that in the case N 462/2009, Poland indicated and the Commission accepted that the market return for motorway projects in Poland was around […] %(25).

(77) Moreover, Autostrada Wielkopolska argues that the IRR of […] %, which Poland states should have been the benchmark for calculating the shadow toll, is not a market rate as no rational private investor would have agreed to the Concession Agreement in 2005 if the expected rate of return had been as low as […] %.

(78) In Autostrada Wielkopolska's view Poland appears to take the position that to be compliant with State aid rules, any compensation, as a matter of law, necessarily has to reflect the economic situation at the moment immediately prior to the event, triggering the compensation. In Autostrada Wielkopolska's view that approach would deprive the private party of achieving long-term goals and could encourage Member States to renegotiate PPPs when business forecasts display negative trends.

(79) Furthermore, Autostrada Wielkopolska points out that in the past, the Commission assessed whether an agreement between a State and a private party was necessary and proportionate. Autostrada Wielkopolska states that the measures agreed between Poland and Autostrada Wielkopolska were proportionate, especially since the Concession Agreement contained safeguard mechanisms to prevent that Autostrada Wielkopolska would be overcompensated. These measures were laid down in Annex 6 as well as in the original Concession Agreement. In Annex 6, the parties agreed that if the Verification Model shows that the projected IRR is higher than the one calculated in the Real Toll Model, the AGRi rate will be adjusted downwards. Annex 6 also provided for an absolute cap on the shadow toll rates.

(80) Other provisions of the Concession Agreement, unchanged by Annex 6, specified that […]

4.1.3.2.   

MEOP

(81) Autostrada Wielkopolska, furthermore, argues that even if Annex 6 entails a material improvement it would still not constitute State aid because it was negotiated between the Polish State and Autostrada Wielkopolska on a commercial basis and at arm's length. Moreover, Poland agreed to terms that would have been acceptable to a private investor, similarly placed.

(82) Autostrada Wielkopolska argues that the State's submissions to the Commission consistently confuse Autostrada Wielkopolska's opening position in the 2005 negotiations (an expected return of […] %) with their actual outcome (an expected return of […] %).

(83) In Autostrada Wielkopolska's opinion, the State, as it was entitled to do in the context of a commercial negotiation, did not accept Autostrada Wielkopolska's proposed terms but insisted on two very important elements of the funding mechanism: (a) a verification of the shadow tolls after two years and (b) an absolute cap on the shadow tolls without any adjustment for inflation or exchange rate fluctuations until the end of the concession period. In other words, the parties agreed on a calculation methodology which could have resulted in a maximum expected IRR of […] % at the time of the verification exercise in 2007. However, from an

ex ante

perspective, the absolute cap on the shadow toll rendered that target merely theoretical. Actually, the IRR in 2005 was at […] % and in 2007 at […] %. Therefore, the actual result of the negotiations was closer to Poland's than Autostrada Wielkopolska's demand.

(84) Furthermore, Autostrada Wielkopolska stresses that the ‘outcome of a negotiation between two parties with different starting points is typically somewhere in between the two respective starting points.’ Therefore, it should be assessed whether the outcome of the negotiations was within a ‘range’ of reasonable behaviour compatible with the conduct of a hypothetical private investor. Autostrada Wielkopolska states that if a State's behaviour is not manifestly outside that range, then according to the Court Judgment in Case C-256/97(26), it can be expected that a hypothetical private investor would have behaved in the same way.

(85) Autostrada Wielkopolska states that a letter from Ernst & Young to GDDKiA of 14 October 2005, which was presented to the Arbitral Tribunal, shows that the State during the negotiations had to take into account the risk of Autostrada Wielkopolska not agreeing to Annex 6, which would have resulted in terminating the agreement and suing for damages. This was especially so, since an IRR equal to […] % was not sufficient to repay the EIB loan due in 2018. Autostrada Wielkopolska contends that a reasonable private operator would weigh the danger of having to find a new investor at a much higher market price. Taking all this into account, agreeing to the negotiated Annex 6 and IRR is in the range of reasonable behaviour of a hypothetical private investor.

4.1.3.3.   

The 2004 WSA study

(86) Autostrada Wielkopolska criticizes Poland's approach to use the 2004 study as basis for the calculations of Annex 6. Autostrada Wielkopolska states that there can be no doubt that the State's financial advisers had access to the 2004 study (and all other relevant data) and were well aware that the financial projection leading to an IRR of […] % was based on the 1999 study. This was confirmed by the Arbitral Tribunal.

(87) Furthermore, Autostrada Wielkopolska argues that the State had at its disposal a number of traffic studies, i.e. Autostrada Wielkopolska's traffic adviser WSA's studies from 1999 and 2004, the banks' traffic adviser SDG's study from 2005 and the State's own traffic adviser Faber Maunsell's study from 2005. In addition, the State had at its disposal the actual 2005 traffic counts. According to Autostrada Wielkopolska the fact that there were more recent studies and traffic counts must be of relevance when the State's argument is chiefly based on the claim that the WSA 2004 study is the most recent study available.

(88) Autostrada Wielkopolska also argues that the 2004 WSA study was principally prepared as part of the preparatory works of the second segment of the A2 motorway (Swiecko — Nowy Tomysl) and with a view to reach the financial closure for that section. When drafting the study between December 2003 and the preliminary version in April 2004, WSA was able to calculate the actual traffic flow figures only on the two sections of the motorway that had been built and were under operation at that time, i.e. Krzesiny — Wrzesnia, Wrzesnia — Konin as well as the Poznan Bypass. The section Nowy Tomysl — Komorniki was only opened to traffic in October 2004. This had an obvious impact upon the perception of the usefulness of the study in the view of the parties and lenders in late 2005. Also, Autostrada Wielkopolska points out that traffic forecasts are usually construed for a five-year period and not for 30 years as in the 2004 WSA study.

4.2.   

Compatibility Assessment

(89) Autostrada Wielkopolska states that if the Commission, contrary to Autostrada Wielkopolska's arguments, finds the compensation to be State aid, it is compatible under Article 107(3)(c) TFEU.

(90) Autostrada Wielkopolska points out that in the case N 462/2009(27), the Commission found State aid granted to Autostrada Wielkopolska for the construction and operation of the section Swiecko — Nowy Tomysl of the A2 Motorway (i.e. the adjacent part of the A2 motorway) to be compatible with the internal market. According to Autostrada Wielkopolska, State aid granted under Annex 6 would therefore have to be compatible as well, for similar reasons.

(91) In addition, Autostrada Wielkopolska states that since the A2 motorway is located in a region covered by the derogation set out in Article 107(3)(a) TFEU, the aid could also be found compatible with the internal market under this article. The Commission's Regional Aid Guidelines permit operating aid to be granted in region covered by the derogation set out in Article 107(3)(a) TFEU provided that two criteria are satisfied: (i) it is justified in terms of its contribution to regional development and its nature and (ii) it is proportionate. Autostrada Wielkopolska submits that the contributions of the motorway to regional development are multiple (e.g. the development of the logistical companies around Poznan and Konin, higher employment, etc.).

(92) For the criterion of proportionality, AW S.A explains that […].

(93) Autostrada Wielkopolska describes that in case of a negative decision, the EIB loan could not be refinanced and the EIB would invoke the State guarantee. As the State would then claim the amount from Autostrada Wielkopolska, which it would not be able to pay, Autostrada Wielkopolska would be forced into bankruptcy and its investors would not have any profit whatsoever. Therefore, the Commission was not correct, in Autostrada Wielkopolska's view, to assume that the investors were benefitting from the compensation payments – rather the money was all spent on the motorway.

5.   

COMMENTS FROM POLAND ON THE INTERESTED PARTY COMMENTS

(94) Poland informs that it does not agree with comments made by Autostrada Wielkopolska and that this disagreement relates both to the facts and to the legal assessment.

5.1.   

On the Relevance of the Arbitral Award

(95) Poland remarks that the Arbitral Award is irrelevant to the State aid proceedings by the European Commission because: (i) it concerned only the validity of Annex 6, (ii) only the European Commission is authorised to evaluate whether a measure constitute State aid or not.

5.2.   

On the economic advantage

5.2.1.   

On Autostrada Wielkopolska's right to compensation

(96) Poland repeats that Autostrada Wielkopolska had a right to seek compensation for the losses incurred due to the legislative amendment releasing HGVs from paying tolls. However, the compensation could not put the Concession Holder in a better position than it would have been without the Amendment Act.

5.2.2.   

On the light vehicles traffic

(97) In reply to Autostrada Wielkopolska's argument that the light vehicles traffic had to be updated because it depends on HGVs traffic which increase was difficult to predict when the shadow toil was introduced in 2005, Poland argues that the real data did not confirm that this dependence indeed existed (i.e. while there was a sharp increase of HGVs traffic observed, the light vehicles traffic was stable in the same period).

(98) Nevertheless, Poland maintains its position that including light vehicles traffic was necessary because ‘the adopted methodology required that the financial model covers the project as a whole’. In Poland's view, it was not possible to calculate the IRR for the project taking into account the HGVs traffic only.

(99) In reply to Autostrada Wielkopolska's argument that an update of traffic for other than HGVs was necessary because the concession holder was obliged to carry out works in function of exceeding traffic thresholds (e.g. third lane on the bypass of Poznan), Poland argues that this is irrelevant since the expected increased costs were already considered in the assumptions to the vignette financial model at the signing of Annex No 6 and they do not have any impact on the light vehicle traffic numbers.

5.2.3.   

On the traffic studies – Internal Rate Return

(100) Poland does not agree with Autostrada Wielkopolska's opinion that the amount of compensation negotiated in September 2005 should have been determined so as to allow Autostrada Wielkopolska to achieve the IRR assumed in 2000. Poland points out that firstly, the IRR expected at the time of the financial closure, as specified by Autostrada Wielkopolska, was based on the concession holder's estimates and, secondly, under the Concession Agreement, Poland did not guarantee that any rate of return would be achieved by the non-public party. Secondly, during the negotiations preceding the conclusion of Annex 6, Autostrada Wielkopolska did not make known any expectation as to the achievement of the rate of return which was expected on the date of the financial closure in 2000.

(101) Poland is of the opinion that the economic situation of Autostrada Wielkopolska that existed immediately before the entry into force of the Amendment Act should be taken as a benchmark for determining the amount of damages payable to Autostrada Wielkopolska

(102) In Poland's view Autostrada Wielkopolska is inconsistent because on the one hand it argues that it was entitled to seek compensation securing the IRR at the time of the financial closure ([…] %), while on the other hand it voluntarily consented to the compensation securing the IRR (in 2005) at the level of ‘merely’ […] % and finally (following the 2007 Verification Report) at the level of […] %. For Poland, it is interesting that: Autostrada Wielkopolska has never raised any claims for compensation for loss of Revenue from Operation relating to the difference between the compensated IRR level ([…] %) and the allegedly guaranteed IRR level of […] %. In Poland's view: ‘This should be explained by the fact that in 2005 the expected rate of return calculated on the basis of the data available at that time stood at the level of […] %, which was significantly lower than the IRR level of […] % resulting from the compensation disbursements’.

5.2.3.1.   

London Underground

(103) In response to the London Underground argumentation, Poland agrees that the original Concession Agreement is free of State aid. However Poland objects to Autostrada Wielkopolska's argument that an improvement of Autostrada Wielkopolska's business situation, due to compensation, can only be seen in the […] % (i.e. difference between […] % - IRR expected at financial closure in 2000 and […] % - IRR included in Annex 6), as, according to Poland, improvement lies in the difference between the actual situation of Autostrada Wielkopolska (IRR of […] %) in 2004 and the agreed IRR in Annex 6 ([…] %). According to Poland this increase is not marginal. Poland also does not agree that IRR of […] % is the market rate for motorway projects in Poland as the Commission once approved […] % to be the market rate. Poland argues that the market rate is ‘what actually happens’ and should be determined for ‘each case individually’. The market rate for Autostrada Wielkopolska therefore should be […] %.

(104) Poland supports its argumentation that Autostrada Wielkopolska enjoyed a material improvement compared to its financial situation both before and after the period of application of the shadow toll by providing three graphs. These graphs show that application of the shadow toll led to revenues for Autostrada Wielkopolska considerably higher than the revenues originally projected and, furthermore, that Autostrada Wielkopolska's revenues returned to the originally projected level after the end of the shadow toll.

5.2.3.2.   

MEOP

(105) Poland states that a private investor would only pay damages he is legally obliged to pay. The damages in this scenario come to the amount Autostrada Wielkopolska was missing to achieve the IRR it would have achieved without the Amendment Act. According to Poland, a private investor would know that Autostrada Wielkopolska could only claim damages limited to the amount necessary to repair the loss (which in this case would be compensation to achieve an IRR of […] %) and would not have entered into a settlement granting higher damages to the counter party.

(106) Poland strongly rejects Autostrada Wielkopolska's argument that Poland was under pressure during the negotiations arising from the risk that the State would have to replace Autostrada Wielkopolska. In Poland's view, if the contract with Autostrada Wielkopolska would have been terminated, the State Treasury could have managed the motorway itself or would have sought a replacement but against much lower costs.

5.2.3.3.   

The 2004 WSA Traffic and Revenue study

(107) Poland stresses that it did not, contrary to Autostrada Wielkopolska's claim, arbitrarily select one of the available traffic and revenue studies, nor did it intentionally fail to make known to the Commission other, allegedly alternative studies. Poland reminds that it had submitted the opinion of traffic advisor Scott Wilson, which took into account all available traffic and revenue studies and indicated that the 2004 WSA study should be used.

(108) According to Poland, the 2004 WSA study should have been used in 2005 as it was the most up-to-date study. The study had been prepared at the request of Autostrada Wielkopolska and was an updated version of the study from 1999, used as the basis for estimating the IRR applicable to the project on the date of the financial closure.

(109) Poland does not accept Autostrada Wielkopolska's argument that the 2004 WSA study should not be used because the forecasts are usually formulated in the perspective of five rather than 30 years. Poland notes that while the 2004 WSA study covers 30-year period, the study from 1999, which Autostrada Wielkopolska took as basis for calculating the IRR, covers a 35-year time horizon.

(110) Poland also points out that although at the time the 2004 study was prepared there was only data on actual traffic flows for a part of the motorway available, at the time the 1999 study was prepared, there were no sections of the motorway in use at all. Therefore, according to Poland, it was more appropriate to use the study from 2004 than the one from 1999.

5.3.   

Compatibility assessment

(111) In reply to AW S.A's arguments that in the case in question the compatibility assessment would be in general the same as for the case – N 462/09 (positively assessed by the Commission in the past), Poland notes that the aid in that case was granted to facilitate the construction of the road infrastructure while in the present case the motorway had been already constructed and opened in 2004, i.e. before the adoption of the Amendment Act in 2005. Hence, Poland argues, the incentive effect is missing.

(112) Moreover, State aid obtained by Autostrada Wielkopolska affected only the profitability of the project and it had no effect on the promotion of any economic activities or economic areas. Therefore, Article 107(3)(c) TFEU does not apply to the present case. In Poland's view since the aid did not contributed to regional development, it cannot also be declared compatible State aid under the Regional Aid Guidelines.

6.   

ASSESSMENT OF THE MEASURE

(113) The Commission's assessment in the present case is limited to the compensation granted by Poland to Autostrada Wielkopolska under Annex 6 of the Concession Agreement for excluding heavy goods vehicles from the obligation to pay toll for using the A2 motorway between Nowy Tomysl and Konin from 1 September 2005 to 30 June 2011.

6.1.   

Existence of State aid

(114) Under Article 107(1) TFEU ‘any aid granted by a Member State or through State resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods shall, in so far as it affects trade between Member States, be incompatible with the internal market.’

(115) The criteria provided for in Article 107(1) TFEU are cumulative. Therefore, in order to determine whether the measure in question constitutes aid within the meaning of Article 107(1) TFEU, all of the following conditions need to be fulfilled. Namely, the financial support must:

(a) be granted to an undertaking (economic activity);

(b) be financed through State resources and be imputable to the State;

(c) grant an economic advantage;

(d) be selective;

(e) distort or threaten to distort competition and affect trade between Member States.

6.1.1.   

Economic activity and notion of undertaking

(116) According to settled case law, the Commission must first establish whether Autostrada Wielkopolska is an undertaking within the meaning of Article 107(1) TFEU. The concept of undertaking covers any entity engaged in an economic activity, regardless of its legal status and the way in which it is financed(28). Offering goods or services on a given market is an economic activity(29).

(117) In this regard, the Commission notes that the A2 motorway between Nowy Tomysl and Konin is operated on a commercial basis as any vehicle that can transit on motorways can use the A2 motorway against payment in the form of a toll that is income of the concession holder. It follows that Autostrada Wielkopolska is offering services on the market and constitutes an undertaking for the purposes of Article 107(1) TFEU(30).

6.1.2.   

State resources and imputability to the State

(118) The concept of State aid applies to any advantage granted through State resources by the State itself or by any intermediary body acting by virtue of powers conferred on it(31).

(119) In the present case, the compensation was paid to Autostrada Wielkopolska by the National Road Fund (NRF). The NRF is a state fund established within the Bank Gospodarstwa Krajowego (hereinafter: ‘BGK’) - the State Development Bank of Poland. The main source of NRF's funding are the incomes from the fuel fee (a parafiscal tax(32)) imposed on motor fuels and gas introduced into the Polish market (paid by the producers and importers of motor fuels) and the motorway toll collected directly by GDDKiA — the central administration authority in charge of the national road system. Therefore, the Commission considers that the compensation payments to Autostrada Wielkopolska were financed from State resources.

(120) The NRF was designated as the source of compensation payment by the Polish Parliament in the Amendment Act. Moreover, the Amendment Act specified that the compensation payments from the NRF to Autostrada Wielkopolska were to be executed at the request of the General Director for National Roads and Motorways(33). Thus, the Commission considers that the compensation payments to Autostrada Wielkopolska are imputable to the State.

6.1.3.   

Economic advantage

(121) In this section, the Commission first examines and establishes the right for the concession holder to compensation due to the legislative change. The Commission acknowledges that only the excess to that compensation gives the concession holder an undue advantage. Secondly, the Commission discusses the models provided by the parties in the process of the negotiations. Thirdly, the Commission establishes that the use of the outdated traffic study led to overcompensation. Fourthly, the Commission examines and approves the process of the verification of the AGRi rates performed in 2007. Fifthly, the overcompensation amount is established based on a report of PwC. Additionally some comments raised by the concession holder are addressed. Finally, it is concluded on the existence of an economic advantage.

6.1.3.1.   

The right to compensation

(122) The Commission notes at the outset that tolls were the main source of income for the concession holder. The concession holder was free to set the toll rates at its own discretion within the limits set by the Concession Agreement.

(123) Both Poland and Autostrada Wielkopolska indicate that the Concession Agreement gave the concession holder the right to be compensated for damages due to the legislative change at stake. The Concession Agreement clearly states that the State has to compensate the concession holder in case it is prevented by actions of the State from changing or adjusting the toll rates (see Article 12.6 of the Concession Agreement cited in recital 51 above). The Amendment Act deprived the concession holder not only of the possibility to adjust the toll rates for HVGs (within certain limits), but of the very right to charge tolls for the use of the motorway by HGVs.

(124) In view of the above, the Commission considers that, in the specific circumstances of the present case (among others, the fact that the concession was the object of a tender), the concession holder had a right to be compensated for the change of the law which deprived it of its right to charge tolls in a drastic manner.

(125) The Commission also recognises that a compensation which merely restores the effects of the legislative change does not constitute as such State aid. However the Commission considers that a compensation that exceeds what is normally due to restore the effects of the legislative change confers a selective advantage to the undertaking.

(126) In the current case the right for compensation of the effects of the legislative change means the right of the concession holder to receive from the State a compensation which restores its expected financial situation just before the legislative change. However, if the envisaged compensation improved the expected financial situation of the concession holder by going beyond a compensation linked to the direct effects of the legislative change, it received an undue advantage constitutive of State aid within the meaning of Article 107(1) TFEU(34).

6.1.3.2.   

The models used for establishing the compensation

(127) To assess the question whether the level of compensation paid to Autostrada Wielkopolska entailed an undue advantage, by improving its financial situation as compared to the situation prevailing without the relevant legislative change, the Commission examined in detail the process through which the compensation payments were determined as well as the financial models used to that end. This analysis is presented below.

(128) A possible indicator of the financial situation of the concession holder is the internal rate of return (IRR) of the cash flows which the concession holder expected just before the legislative change. Among the feasible IRR indicators, the Polish State and Autostrada Wielkopolska agreed in Annex 6 to the Concession Agreement to use the real annual total shareholder return on equity and subordinated debt expressed in EUR(35). A widely accepted standard methodology to determine the return on investments is to calculate the IRR.(36) Therefore the Commission considers the use of this indicator to be reasonable and, therefore, does not object to it.

(129) During the negotiations of the shadow toll rates between the concession holder and the Polish authorities, Autostrada Wielkopolska presented three financial models(37). These are the Base model, the Real Toll model and the Vignette model presented in recital 29 above.

(130) The

Base model

presents the financial situation of the concession holder with an IRR of […] %

at Financial Closure in 2000 as described in recital 29(a) above

. The Commission considers however that the IRR that could be expected by the concession holder at the beginning of the concession is not relevant in the present case, as the Concession Agreement does not guarantee any level of IRR during the term of the concession, but rather transfers the market and financial risks as well as the opportunities to the concession holder. Therefore the expected IRR can be different in each moment in time of the duration of the concession depending on the realisation of the risks and opportunities. For that reason the Commission considers that the relevant IRR in this case can only be the one that could be expected by the concession holder

just before the legislative change and the related introduction of the shadow toll mechanism in 2005

. This return could be higher or lower than the one that could be expected at the beginning of the concession.

(131) The

Real Toll model

(38) as described in recital 29(b) above

(see for more details model 1 in Table 2 below) uses the traffic and revenues forecasts from the 1999 WSA study(39) for the forecast period (from 2005 until the end of the concession in 2037). It represents the financial situation of the concession holder as of 31 December 2004 in the counterfactual situation without the legislative change, namely the scenario in which the concession holder would collect tolls from all vehicles using the A2 motorway until the end of the concession. According to this model the expected IRR was […] %.

(132) The

Vignette model

(40) described in recital 29(c) above

(see for more details model 2 in Table 2 below) represents the financial situation of the concession holder with the legislative change, namely a scenario in which HGVs would use the A2 motorway free of charge and shadow toll compensation would be paid by the State at 70 % of the maximum allowed toll rates. In this model Autostrada Wielkopolska used its own traffic assumptions reflecting the use of the A2 motorway free of charge by HGVs with a valid vignette. The Polish authorities claim that these traffic assumptions were not agreed between the parties at the time(41). According to this model the expected IRR is […] %. With this model Autostrada Wielkopolska demonstrated that even with the maximum allowed shadow toll rates it would not achieve the IRR of […] % it could have expected to achieve just before the legislative change.

(133) The

Verification model

(42) presented in recital 36 above (see also model 5 in Table 2) was provided by Autostrada Wielkopolska in 2007. It was used for the verification of the shadow tolls in 2007. In this model Autostrada Wielkopolska used the 2006 WSA study(43) for the projection period(44). The Verification model assumed that the shadow toll system was in place until the end of the concession. It calculated the expected IRR in 2007 at […] %. Therefore according to Autostrada Wielkopolska, the Verification model showed that the IRR of […] % prior to the legislative change (see Real Toll model) was not reached.

(134) Table 2 below shows all financial models used by the parties in the negotiations with their characteristics as well as the versions of the models adjusted by PwC and used in their report to estimate the overcompensation (see below point 6.1.3.5):

Table 2

Financial models used by the parties

6.1.3.3.   

The use of the wrong traffic study and establishing the existence of overcompensation

(135) The Commission notes that, as was mentioned above, the Real Toll model uses the traffic and revenues forecast of the 1999 WSA study in relation to the period 2005 to 2037(45). However, as this model was assumed to represent the financial situation of the concession holder at the time just prior to the legislative change in 2005 and, as such, the benchmark for setting the shadow toll, it should have used a contemporaneous traffic and revenues forecast. The Commission, therefore, takes the view that Autostrada Wielkopolska should have used the available up-to-date traffic and revenues forecast of the 2004 WSA study(46) instead.

(136) The 2004 WSA study is an updated version of the 1999 WSA study. The 1999 WSA study formed the basis for the preparation of the financial model estimating the IRR applicable to the project at the time of the financial closure back in 2000. The Commission agrees with Poland that in order to achieve an updated level of the expected IRR at the time of the legislative change, it is appropriate to use an update of the traffic and revenue study. The 2004 WSA study provides such an update. It is based on the experience of the actual traffic and revenue development on the initial portion of the motorway as well as the economic development of Poland up to that moment. Therefore, it reflects more accurately the market reality at the time of the legislative change. For that reason, according to the Commission, the 2004 WSA study represents a more accurate traffic and revenues forecast for the moment of the legislative change than the one reflected in the 1999 WSA study.

(137) Compared to the 1999 WSA study, the 2004 WSA study shows slightly higher traffic numbers for vehicles of category 1 and significantly lower traffic numbers for vehicle categories 2 and 3. The projected optimal real toll rates for categories 2, 3 and 4 in the 2004 WSA study are much lower than the projected optimal real toll rates for these categories in the 1999 WSA study. In effect the optimal revenues from HGVs forecasted in the 2004 WSA study are significantly lower than the optimal revenues from HGVs forecasted in 1999 WSA study. Therefore the use of the 1999 WSA study as a base for the Real Toll model leads to a higher IRR than the IRR which would have been achieved had the 2004 WSA study been used instead.

(138) The Commission also notes that the two 2005 studies referred to by Autostrada Wielkopolska could not be used instead of the 2004 WSA study. First, the 2005 SDG (see recital 87 above) study was prepared for the shadow toll (vignette) scenario (i.e., the scenario in which the concession holder receives shadow toll payments for HGVs) while the assessment of the concession holder's financial situation immediately before the entry into force of the Amendment Act required using a study for the real toll scenario (i.e., the scenario in which the concession holder collects real fees from all the users of the A2 motorway until the end of the concession period). Moreover, Poland informed the Commission that the 2005 Faber Maunsell study, referred to, in its comments, by Autostrada Wielkopolska (see recital 87 above), focused only on traffic projections and not on revenue projections (i.e., it did not take account of the rate optimisation to maximise revenue). Therefore it was unusable for the purpose of calculating the IRR of the project.

(139) For these reasons, the Commission takes the view that relying on the Real Toll model based on the old 1999 WSA study has led to a higher IRR than could be reasonably expected at the time of the legislative change. This has resulted in overcompensation in the form of higher shadow toll payments.

6.1.3.4.   

Verification of shadow tolls in 2007 — acceptance of the verification

(140) During the negotiations of the shadow toll rates in 2005, the Polish authorities agreed with Autostrada Wielkopolska to verify the IRR in 2007 and amend the shadow toll rates if the targeted IRR was not achieved as explained in point 2.5 above. The Commission notes that the verification was agreed upon because of the uncertainty about the traffic development after the opening of the A2 motorway free of charge to HGVs with a valid vignette.

(141) The Commission expressed doubts in the opening decision as to whether the application of the compensation mechanism resulted in taking over by Poland, at least in the period from 1 September 2005 to 31 December 2006, of traffic and revenue risks, but only as regards light vehicles. The Commission notes that, through the verification, the Polish State effectively assumed the risk of traffic (and revenue) development for the period from the introduction of the shadow toll system until the verification (a period of around 16 months). The Commission notes that, by doing so, the Polish State could be seen to have deviated from the nature of the initial concession contract which did not guarantee to the concession holder any level of return.

(142) However, the Commission accepts the verification of the shadow toll rates as it allowed the State to re-set the shadow toll rates at such a level which avoids overpaying the concession holder for HGVs. Moreover the verification was done after a very limited period of time (around 16 months – from September 2005 until December 2006) which allowed the parties to gather real traffic data for heavy as well as for light vehicles, make traffic forecasts on that basis and set the shadow toll rates accordingly. Compared with the whole duration of the Concession Agreement of 40 years, this is a limited period of time which in fact was reasonable in the sense that it just allowed the parties to gather the necessary traffic data and on that basis build reliable traffic forecasts for the future.

(143) In this context, it is important to recall that the initial level of the shadow toll was set based on the assumptions of the concession holder for the future traffic developments (used in the Vignette model). The Commission notes that at that very first moment there was no actual traffic data for HGVs using the motorway for free. For that reason it was not possible to precisely assess the impact of the legislative change on traffic, and hence to assess the accuracy of the concession holder's assumptions made on future traffic developments.

(144) The Commission notes that there was no further verification of the shadow tolls foreseen until the end of the concession or the end of the shadow toll regime. For that reason once the shadow toll rates were to be set at their final level after the verification in 2007 the return of the concession holder was again exposed to the risks of traffic developments similarly to the initial concession contract.

(145) The Commission notes that in the case of A2 motorway the share of HGV traffic is larger than the share of HGV traffic on the A4 motorway. The share of HGV traffic on A2 was estimated at […] %(47) whereas the share of HGV traffic on A2 was only […] %. The Commission also notes that because of the higher proportion of HGVs on A2 the interdependency between the two types of traffic may be larger than in the case of the A4 motorway. While the legislative change only applied to HGVs, the Commission nevertheless accepts in the specific circumstances of this case that the verification covers not only HGVs but also light vehicles traffic.

6.1.3.5.   

Calculation of the overcompensation — PwC report

(146) As part of their notification, the Polish authorities presented the PwC report (see recital 41). In this report PwC updates the Real Toll model of Autostrada Wielkopolska with the traffic and revenues forecasted of the 2004 WSA study. It also introduces some small corrections (arithmetical, logical and fiscal) to the model. As a result of these changes the IRR that could have been expected at the time of the legislative change falls from […] % to […] %(48) (see PwC Real Toll model in table 2).

(147) The Commission considers that the replacement of the traffic and revenue forecast introduced by PwC in the Real Toll model is reasonable, as it properly represents the updated forecasts at the time of the relevant legislative change. The Commission therefore agrees that the IRR level of […] % can be seen as the IRR that could have been expected by the concession holder just before the legislative change.

(148) Compared to […] %, the IRR of […] % claimed by Autostrada Wielkopolska and actually used in the negotiation of the compensation payments is excessive. The IRR of the Vignette model of […] % is also above […] %.

(149) For establishing the overcompensation amount for the period until the verification, PwC uses the Vignette model to recalculate the AGRi Rates which should have been applied initially as of September 2005 in order to achieve the lower expected IRR equal to […] % instead of the IRR of […] % in the Vignette model presented by Autostrada Wielkopolska in 2005 (see PwC Vignette model in Table 2). In its report PwC substitutes the IRR of […] % of the Vignette model with […] % and estimates the AGRi Rates which would be sufficient to reach this lower IRR. The amount of compensation due to Autostrada Wielkopolska for the period from September 2005 to October 2007 based on the recalculated AGRi Rates was determined at PLN […] gross(49). This amount was compared with the actual payments made to Autostrada Wielkopolska for this period, which amounted to PLN […] gross(50). The calculations show that in the period between September 2005 and October 2007 the estimated overpayments amounted to PLN […] gross(51), which represents approximately […] % of the total payments made to Autostrada Wielkopolska in this period.

(150) For the period after the verification (from November 2007 to June 2011) PwC uses the Verification model to estimate the overcompensation received by Autostrada Wielkopolska(52). To this end PwC recalculates the shadow toll rates so that an IRR of […] % instead of […] % is reached in the Verification model (see PwC Verification model(53) in Table 2). The amount of overcompensation corresponds to the difference between the shadow toll rates recalculated by PwC and the shadow toll rates actually paid to Autostrada Wielkopolska per vehicle during the relevant period. The amount of compensation due to Autostrada Wielkopolska for the period from November 2007 until June 2011 based on the recalculated shadow toll rates was determined at PLN […] gross(54). This amount was compared with the actual payments made to Autostrada Wielkopolska for this period, which amounted to PLN […] gross(55). The calculations show that in the period between November 2007 and June 2011, the estimated overpayments amounted to PLN […] gross(56), which represents approximately […] % of the payments.

6.1.3.6.   

Additional comments on the economic advantage

(151) Autostrada Wielkopolska argued that, even if the compensation in question entailed a material improvement, it was negotiated between Poland and Autostrada Wielkopolska on a commercial basis and at arm's length. Therefore, the Market Economy Operator Principle (MEOP) would be applicable and met in this case.

(152) The Commission considers that the Amendment Act and the Concession Agreement imposed on Poland an obligation to compensate the concession holder only up to the amount equal to the estimated loss of revenue due to the relevant legislative change. It cannot be argued that paying more than what was required by law and the Concession Agreement would be acceptable to a hypothetical, rational private operator. Moreover, it is highly questionable that a rational private operator would agree to calculate the compensation using the 1999 WSA study on traffic and revenues and not the most up-to-date 2004 WSA study. Therefore, the Commission does not agree that the MEOP is met in the given case.

(153) Furthermore, Autostrada Wielkopolska referred to the 2002 London Underground State aid case in which the Commission concluded that an alteration to a long-term contract does not constitute State aid if: (i) the original contract did not entail State aid and (ii) the alteration either does not lead to a material improvement of the concession holder or the improvement is at market rate(57).

(154) While the Commission does not possess any information that would suggest that the initial contract entailed State aid, the analysis of the financial models presented above shows that the shadow toll payments led to a material improvement for the concession holder as they were based on higher expected returns than those that could have been expected before the legislative change. The Commission considers that the market price for the concession held by Autostrada Wielkopolska was established in 1997 as the result of the open tender procedure. According to the provisions of the Concession Agreement the IRR of the project was not guaranteed to the concession holder. While the IRR is used in the compensation method, it is used only as a benchmark allowing the calculation of the correct compensation, i.e. the compensation restoring the financial situation that the concession holder would have been in without the Amendment Act. The IRR is only an indicator of the concession holder's financial situation just before the legislative change and not a price paid to Autostrada Wielkopolska by the State for accepting Annex 6 and therefore it cannot be used for price to price comparisons.

(155) The Commission also notes that for its assessment of the existence of an economic advantage to the concession holder, it is not relevant if this advantage was granted intentionally or unintentionally by the Member State.

(156) Moreover, the Commission notes that the proceedings and the award of the Arbitral Tribunal do not address the State aid aspect of the case(58). It is in any event for the Commission to assess the existence of State aid and its compatibility with the internal market.

6.1.3.7.   

Conclusion on the economic advantage

(157) In light of the above, the Commission considers that the concession holder has received an economic advantage in the form of overcompensation for HGVs for the period from September 2005 until June 2011, the date at which the shadow toll was brought to an end. The Commission agrees with the method of estimating the overcompensation used by PwC.

6.1.4.   

Selectivity

(158) The Commission notes that the compensation for each concession holder was granted following individual negotiations between the concession holder and the State. It resulted therefore in intrinsically specific measures tailor-made for each concession holder. Thus, to the extent that such measures confer an advantage on the concession holder, they are selective.

(159) As it has been shown above that the compensation granted by Poland confers an advantage on Autostrada Wielkopolska, that measure must be consider as selective within the meaning of Article 107(1) TFEU.

6.1.5.   

Distortion of competition and effect on trade

(160) When aid granted by a Member State strengthens the position of an undertaking compared with other undertakings competing in the internal market, the latter must be regarded as affected by that aid(59). It is sufficient that the recipient of the aid competes with other undertakings on markets open to competition(60).

(161) The economic advantage granted by the present measure to Autostrada Wielkopolska strengthens its economic position, as it increases the company's revenues. Moreover, the market for the construction and operation of motorways in Poland is open to any economic operator active in the EU. Therefore, the advantage to Autostrada Wielkopolska has the potential to distort competition and affect trade between Member States.

6.1.6.   

Conclusion

(162) In view of the above, the Commission considers that the compensation paid to Autostrada Wielkopolska for the legislative change excluding HGVs from the obligation to pay a toll for using the A2 motorway between Nowy Tomysl and Konin, in the period from 1 September 2005 to 30 June 2011 constitutes State aid within the meaning of Article 107(1) TFEU in so far as it overcompensated that company for the loss in revenue resulting from the Amendment Act.

6.2.   

Legality of the aid

(163) As the compensation was put at the disposal of Autostrada Wielkopolska prior to the notification, the Commission considers that Poland did not respect the prohibition of Article 108(3) TFEU(61). The State aid granted by Poland is therefore unlawful.

6.3.   

Compatibility of the aid

(164) According to the established case law of the Court(62), it is up to the Member State to invoke possible grounds for the compatibility of the aid, and to demonstrate that the conditions for such compatibility are met. The Commission notes that Poland did not put forward any grounds for the compatibility of the aid in question. On the contrary, Poland informed the Commission that in its view the aid is not compatible with the internal market.

(165) The Commission has nonetheless examined whether the aid in question can be deemed compatible with the internal market. Article 107(3) TFEU provides for certain exemptions to the general rule set out in Article 107(1) TFEU that State aid is not compatible with the internal market.

(166) First, the Commission has assessed the aid in question against the exemption laid down in Article 107(3)(c) TFEU. Article 107(3)(c) TFEU stipulates that ‘aid to facilitate the development of certain economic activities or of certain economic areas, where such aid does not adversely affect trading conditions to an extent contrary to the common interest’, may be considered to be compatible with the internal market. In this regard, the Commission also notes that none of the State aid guidelines directly apply to motorway infrastructure and its operation.

(167) The Commission notes that the assessment of the compatibility of aid can differ depending on its classification as investment or operating aid.

(168) The Commission further notes that at the time when the aid in question was granted, the whole part of the A2 motorway operated by Autostrada Wielkopolska had been opened for almost one year. Therefore, contrary to AW S.A's reference to the Commission's earlier decision on construction and operation of the adjacent section Swiecko — Nowy Tomysl of the A2 Motorway (case N 462/09), it cannot be concluded in the case at hand that it was an investment aid to facilitate the construction of the road infrastructure. The State aid in case N 462/09 had been paid to the beneficiary before the construction works began and had an incentive effect for the beneficiary to carry out the relevant investment project. By contrast, the Commission does not find an incentive effect in the case at hand, as the construction of the motorway was already finished. Neither Poland nor Autostrada Wielkopolska pointed to any other investment project to which the aid in question could contribute.

(169) Moreover, the Commission notes that, contrary to investment aid, the aid in the present case was not linked to the costs of any investment project but was dependant on the amount of vehicles using the motorway and had a recurring character. As such, it did not contribute to the financing of an investment project, but to the recurring operating costs of the beneficiary. Therefore, the Commission considers that the aid in question constitutes operating aid. According to the case law of the Court(63), such operating aid is in principle incompatible with the internal market.

(170) The Commission, however, notes that the A2 motorway between Nowy Tomysl and Konin is located in an underprivileged region covered by the derogation set out in Article 107(3)(a) TFEU. The Commission has, therefore, assessed whether the operating aid at stake can be considered compatible under the Regional Aid Guidelines (hereinafter ‘RAG’).

(171) The Commission has considered that it should assess the aid in question under the Guidelines valid at the time when the decision to grant the aid in question was taken by signing Annex 6 to the Concession Agreement, namely in 2005. For aid granted at that time, the Guidelines on National Regional aid adopted by the Commission in 1998 apply (hereinafter ‘RAG 1998’)(64).

(172) According to paragraph 4.15 of RAG 1998, operating aid that is normally prohibited can be exceptionally granted in regions eligible under the derogation in Article 92(3)(a) (now – Article 107(3)(a)) provided that (i) it is justified in terms of its contribution to regional development and its nature and (ii) its level is proportional to the handicaps it seeks to alleviate. It is for the Member State to demonstrate the existence of any handicaps and gouge their importance.

(173) In this context the Commission notes that Poland did not demonstrate the existence of any handicaps that the aid in questions could alleviate. Contrary, Poland is of the opinion that the aid did not contribute to regional development.

(174) The Commission considers that the operating aid in this case resulted exclusively in an increase of the IRR of the project for the investors and, as such, does not contribute to regional development of the regions where the toll motorway is located. Thus, the Commission considers that the criteria set out in the paragraph 4.15 of the RAG are not met.

(175) Further, it is irrelevant for the purpose of the Commission's assessment that the compensation to AW S.A has not been and will not be used as profit distribution (dividends) to shareholders, but instead for repayment of the EIB loan. In fact the repayment of the EIB loan is an obligation of the concession holder stemming from the financial arrangements the concession holder had to make in order to fulfil the Concession Agreement. If the compensation paid to Autostrada Wielkopolska was such that it enabled the company to repay its loans earlier than foreseen (i.e., in the case without the change in the law and the need for compensation), then the compensation provided an advantage to the company as it made resources available for other purposes like for example distribution of dividends which could not have been paid out otherwise. Such distribution of dividends however had no contribution to the development of the region in which the motorway is located.

(176) The Commission also notes that the compatibility criteria for operating aid included in paragraph 76 of the Regional Aid Guidelines for the period 2007 – 2011(65) (hereinafter ‘RAG 2007’), that applied as of 1 January 2007, are the same. Thus, the compatibility assessment for the aid in question is the same under RAG 2007.

6.3.1.   

Conclusion on compatibility

(177) The Commission has not identified any other provision regarding the compatibility of aid with the internal market that could provide a basis for considering the aid in question compatible with the Treaty. Nor has Poland invoked any provision concerning compatibility with the internal market or provided any arguments that would allow the Commission to consider the aid in question compatible. Similarly, the comments submitted by the third party do not allow the Commission to consider the aid in question compatible.

(178) Consequently, the Commission concludes that the State aid that Poland has granted to AW S.A is incompatible with the internal market. Poland has unlawfully implemented the aid in question in breach of Article 108(3) TFEU.

7.   

RECOVERY

(179) According to the Treaty and the Court's established case-law, the Commission is competent to decide that the Member State concerned must abolish or alter aid when it has found that it is incompatible with the internal market(66). The Court has also consistently held that the obligation on a Member State to abolish aid regarded by the Commission as being incompatible with the internal market is designed to re-establish the previously existing situation(67).

(180) In this context, the Court has established that this objective is attained once the recipient has repaid the amounts granted by way of unlawful aid, thus forfeiting the advantage which it had enjoyed over its competitors on the market, and the situation prior to the payment of the aid is restored(68).

(181) In line with the case-law, Article 16(1) of Council Regulation (EU) 2015/1589(69) states that ‘where negative decisions are taken in cases of unlawful aid, the Commission shall decide that the Member State concerned shall take all necessary measures to recover the aid from the beneficiary […]’.

(182) The Commission has concluded that Poland has overcompensated AW SA. for the period from 1 September 2005 to 30 June 2011. The Commission has also established that this overcompensation constitutes unlawful aid that is incompatible with the internal market. As a result, the aid must be recovered in order to re-establish the situation that existed on the market prior to its granting.

(183) As stated in recitals from 146 to 150 above (see also the Annex to this decision), the Commission concludes that for the period from 1 September 2005 to 30 June 2011 the total aid amounts to PLN 894 956 888,88(70) gross(71). Recovery should cover the time from when the advantage accrued to the beneficiary, that is to say the date at which the aid was put at the disposal of the beneficiary, until effective recovery. Consequently, the total amount of aid to be recovered shall bear interest as of the moment that each of the monthly disbursements referred to in recitals from 146 to 150 and the Annex to the decision took place and until the moment of its reimbursement.

(184) To the extent that Autostrada Wielkopolska has paid tax on the aid received, Poland may, in accordance with its national tax rules, take account of the earlier payment of tax by recovering only the net amount received by Autostrada Wielkopolska(72). However, in that case, the national authorities should ensure that Autostrada Wielkopolska will not be able to enjoy a further deduction by claiming that the reimbursement has reduced his taxable income, since this would mean that the net amount of the recovery was lower than the net amount initially received.

(185) Poland has a period of two months from the date of adoption of this decision within which it should provide the Commission with evidence of any tax paid on the aid received and the date at which the payments referred to in recitals from 146 to 150 and the Annex to the decision were made. The Commission considers that from those dates, the beneficiary has at its disposal the aid.

HAS ADOPTED THIS DECISION:

Article 1

The overcompensation for the period from 1 September 2005 to 30 June 2011 amounting to PLN 894 956 888,88, granted by Poland to Autostrada Wielkopolska S. A. on the basis of the Amendment Act, constitutes State aid within the meaning of Article 107(1) of the Treaty.

Article 2

The State aid referred to in Article 1 is unlawful as it was granted in breach of the notification and standstill obligations stemming from Article 108(3) of the Treaty.

Article 3

The State aid referred to in Article 1 is incompatible with the internal market.

Article 4

1.   Poland shall recover the aid referred to in Article 1 from the beneficiary.

2.   The sums to be recovered shall bear interest from the date on which they were put at the disposal of the beneficiary until their actual recovery.

3.   The interest shall be calculated on a compound basis in accordance with Chapter V of Commission Regulation (EC) No 794/2004(73) as last amended by Regulation (EC) No 271/2008(74).

Article 5

1.   Recovery of the aid referred to in Article 1 shall be immediate and effective.

2.   Poland shall ensure that this decision is implemented within four months following the date of notification of this Decision.

Article 6

1.   Within two months following notification of this Decision, Poland shall submit the following information to the Commission:

(a) the total amount (principal and recovery interests) to be recovered from the beneficiary;

(b) a detailed description of the measures already taken and planned to comply with this Decision;

(c) documents demonstrating that the beneficiary has been ordered to repay the aid.

2.   Poland shall keep the Commission informed of the progress of the national measures taken to implement this Decision until recovery of the aid referred to in Article 1 has been completed. It shall immediately submit, on simple request by the Commission, information on the measures already taken and planned to comply with this Decision. It shall also provide detailed information concerning the amounts of aid and recovery interest already recovered from the beneficiary

Article 7

This Decision is addressed to the Republic of Poland.

Done at Brussels, 25 August 2017.

For the Commission

Margrethe VESTAGER

Member of the Commission

(1)  

OJ C 328, 20.9.2014, p. 12

.

(2)  See footnote 1.

(3)  Autostrada Wielkopolska is a private company whose exclusive activity is the construction and operation of the A2 motorway section between Nowy Tomysl and Konin. The company's main shareholder is AWSA Holland I B.V., which holds 98,85 % of Autostrada Wielkopolska's share capital. The sole shareholder of AWSA Holland I B.V. is AWSA Holland II B.V., whose main shareholders are Kulczyk Holding S.A. (registered in Warsaw, 24,10 % of share capital), Meridiam Infrastructure S.a.r.l (Luxembourg, 20,12 %), PGE S.A. (Warsaw, 20,00 %), Strabag AG (Spittal/Drau, 10,12 %), KWM Investment (Vienna, 9,32 %), Bank Zachodni WBK S.A. (Wroclaw, 5,44 %), and TUiR ‘Warta’ S.A. (Warsaw, 4,81 %).

(4)  Under the Concession Agreement, the concession holder first modernised the existing 47,7 km-long sub-section of the motorway (sub-section I: between Konin and Wrzesnia) and then constructed two sub-sections: between Poznan and Wrzesnia (sub-section II – 37,5 km) and between Nowy Tomysl and Komorniki (sub-section III – 50,4 km).

(5)  These rates are based on values from July 1999. They are subject to an increase by the inflation rate between July 1999 and the respective dates of commencement of operation of individual toll collection points. Moreover, after commencing toll collection, the maximum toll rates are annually adjusted by the inflation index (annual consumer price index) and foreign exchange rate difference index (between EUR and PLN).

(

*1

)

  Covered by the obligation of professional secrecy.

(6)  Directive 1999/62/EC of the European Parliament and of the Council of 17 June 1999 on the charging of heavy goods vehicles for the use of certain infrastructures (

OJ L 187, 20.7.1999, p. 42

).

(7)  The Act of 28 July 2005 amending the Act on Toll Motorways and the National Road Fund, and the Act on Road Transport (Journal of Laws No 155, item 1297).

(8)  Namely vehicles in categories 1-5 with weight exceeding 3,5 tonnes; in practice, the share of HGVs in category 1 was negligible.

(9)  The IRR mentioned in the text of this decision refers to the expected total return on shareholders' capital (equity and debt) real, annual, post-tax, based on the cash flows in EUR, unless otherwise specified.

(10)  This is the moment when the financial agreements needed for the project fulfilment were already in place.

(11)  As described in PwC report ‘Estimation of the adjusted AGRi rate and overpaid compensation on the basis of the 2004 WSA Traffic and Revenue Study’, page 36.

(12)  As described in PwC report ‘Estimation of the adjusted AGRi rate and overpaid compensation on the basis of the 2004 WSA Traffic and Revenue Study’, page 61.

(13)  Toll rates for vehicles of cat. II – IV: PLN: […],[…],[…] with VAT of 22 % and from which the concession holder received 70 % (i.e. PLN […],[…] and […]).

(14)  In their submissions to the Commission Autostrada Wielkopolska refers to an IRR of […] % from the Vignette model. However the Vignette model which Poland submitted to the Commission shows an IRR of […] %. Poland confirms that […] % refers to calculations conducted by a consultancy for Autostrada Wielkopolska but does not know how it was calculated. Therefore the IRR of the Vignette model used in the following text is […] %. The conclusions made in the text remain valid independently of which one of these IRR figures is used.

(15)  In the anticipation of this prior to the effective date of the Amendment Act the concession holder increased the Applied Rates collected from drivers of vehicles of all vehicle categories to the maximum rates allowed at that time by the Concession Agreement. However, in practice, toll rates to be paid remained lower since Autostrada Wielkopolska, in parallel, introduced rebates applicable to all vehicle categories.

(16)  Apart from the IRR, debt service coverage ratio, loan life cover ratio, minimum rate of cover for the debt guaranteed by the state were analysed.

(17)  Verification Report from 8 June 2007, prepared by […] for Autostrada Wielkopolska.

(18)  Poland A2 Toll Motorway Project Traffic and Revenue Update Study, November 1999, prepared by Wilbur Smith Associates for […].

(19)  A2 Toll Motorway Project Phase VI Final Traffic and Revenue Study, June 2004, prepared by Wilbur Smith Associates for Autostrada Wielkopolska S.A.

(20)  Estimation of the adjusted AGRi rate and overpaid compensation on the basis of the 2004 WSA Traffic and Revenue Study, by PricewaterhouseCoopers, 24 September 2010.

(21)  See Articles 84 and 88 of the Polish Civil Code.

(22)  Gross amounts include the equivalent of the applicable VAT.

(23)  The entitlement to compensation as described herein is provided in Annex 2 to the Concession Agreement.

(24)  Commission decision of in the case N 264/2002: London Underground public private partnership.

(25)  Commission decision of 29 July 2009 in the case N 462/2009: Aid for transport infrastructure — the construction and operation of the A2 motorway section Swiecko — New Tomysl, footnote 6.

(26)  Case C 256/97 — DM Transport [1999] ECR I-3913.

(27)  The Commission decision of 2 December 2009 in the case N 462/2009 — Aid for transport infrastructure — the construction and operation of the A4 motorway section Swiecko — Nowy Tomysl. It should be noted that the Commission decision concluded in fact to the absence of State aid in that case.

(28)  Judgment of the Court of Justice of 18 June 1998,

Commission

v

Italy

, C-35/96, ECLI:ECLI:EU:C:1998:303; Judgment of the Court of Justice of 23 April 1991,

Höfner and Elser

v

Macrotron

, C-41/90, ECLI:EU:C:1991:161; Judgment of the Court of Justice of 16 November 1995,

FFSA and Others v Ministère de l'Agriculture and de la Pêche

, C-244/94, ECLI:EU:C:1995:392; Judgment of the Court of Justice of 11 December 1997,

Job Centre

, C-55/96, ECLI:EU:C:1997:603.

(29)  Judgment of the Court of Justice of 16 June 1987,

Commission v Italy

, C-118/85, ECLI:EU:C:1987:283; Judgment of the Court of Justice of 18 June 1998,

Commission v Italy

, C-35/96, ECLI:EU:C:1998:30.

(30)  As for motorway concession holders the Commission presented the same conclusion e.g. in the decision of 27 October 2014 in case SA.39224 – Greece – Reset of Greek Motorway concession projects – Moreas Motorway (

OJ C 460, 19.12.2014, p. 1

).

(31)  Judgment of the Court of Justice of 16 May 2002,

France v Commission

(hereafter: ‘Stardust Marine’), C-482/99, ECLI:EU:C:2002:294.

(32)  The fuel fee was introduced by the Polish Parliament with the Act of 27 August 2004 amending the Act on Toll Motorways and the National Road Fund.

(33)  The General Director for National Roads and Motorways is the central administration authority for issues related to the national road system. The General Director performs its tasks with the support of GDDKIA.

(34)  Similarly, in its decision of 4 December 2013 in case SA.29584 — Shadow toll compensation to Stalexport Autostrada Małopolska S.A. (SAM S.A.) — A4 motorway (Katowice - Krakow), the Commission concluded that the compensation to the concession holder did not constitute State aid as, according to the concession agreement for the given section of A4 motorway, the responsibility for the risk related to the change of law which directly or indirectly affected the concession holder lied with the State and the compensation was limited to the loss suffered by the concession holder (

OJ C 172, 6.6.2014, p. 2

).

(35)  See Annex 6, page 9.

(36)  See Commission Notice on the Notion of State Aid (

OJ C 262, 19.7.2016, p. 1

), paragraph 102.

(37)  These three financial models were presented and verified by Ernst & Young in a report from 19 September 2005.

(38)  File name: 050518 A2-CB-2004-12 V1.51.123

(39)  The 1999 WSA study is based on the assumption that all vehicles pay real tolls for using the motorway. See 1999 WSA study, page 17, points 3 and 4.

(40)  File name: 050915 A2-COMPENSATION V.3.71 (awsa)_EY.123

(41)  The Commission notes as well that these traffic assumptions were not based on real traffic data for the case of a toll free motorway and were not based on a proper traffic and revenues study.

(42)  File name: Ref 071023-Combined Phase I & II — Poland A2.xls

(43)  A2 Toll Motorway Traffic and Revenue Update Study, November 2006.

(44)  The WSA study of 2006 is based on initial real traffic data and assumes a shadow toll system for the whole period (free use of A2 for HGVs).

(45)  The WSA traffic and revenue studies include a forecast for the traffic on A2 based on an optimal schedule of real tolls valid at the time each forecast is made.

(46)  The 2004 WSA study is based on the assumption that all vehicles pay real tolls for using the motorway. See 2004 WSA study, page 5-10, point 4.

(47)  As estimated by Autostrada Wielkopolska based on averages for the period 2005-2011.

(48)  The largest portion of the decrease (324 basis points) is due to the update of the traffic forecast and only 11 basis points are due to the other smaller corrections.

(49)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(50)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(51)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(52)  This is done in the PwC report of 2010 for the period until August 2010 and in an additional PwC report of 2011 for the period from September 2010 to June 2011.

(53)  File name: […]

(54)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(55)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(56)  These amount corresponds to EUR […] gross (at an exchange rate of PLN/EUR = 4).

(57)  Case N 264/2002 London Underground Public Private Partnership, paragraph 79.

(58)  In point 4.25, the Arbitral Award says: […].

(59)  Judgment of the Court of Justice of 17 September 1980,

Philip Morris ν Commission

, C-730/79, ECLI:EU:C:1980:209, paragraph 11; Judgment of the Court of First Instance of 30 April 1998,

Het Vlaamse Gewest v Commission

, T-214/95, ECLI:EU:T:1998:77, paragraph 50.

(60)  Judgment of the Court of Justice of 21 March 1991,

Italy ν Commission

, C-303/88, ECLI:EU:C:1991:136; Judgment of the Court of First Instance of 30 April 1998,

Het Vlaamse Gewest v Commission

, T-214/95, ECLI:EU:T:1998:77, paragraph 49.

(61)  Judgment of the Court of First Instance of 14 January 2004 Fleuren Compost v Commission, T-109/01, ECLI:EU:T:2004:4.

(62)  Judgment of the Court of Justice of 28 April 1993,

Italy v Commission

, C-364/90, ECLI:EU:C:1993:157.

(63)  Judgment of the Court of First Instance of 8 June 1995,

Siemens SA v Commission

, T-459/93, ECLI:EU:T:1995:100, Judgment of the Court of Justice of 19 September 2000,

Germany v Commission

, Case C-156/98, ECLI:EU:C:2000:467, point 30.

(64)  

OJ C 74, 10.3.1998, p. 9

.

(65)  

OJ C 54, 4.3.2006, p. 13

.

(66)  Judgment of the Court of Justice of 12 July 1973,

Commission v Germany

, C-70/72, ECLI:EU:C:1973:87, paragraph 13.

(67)  Judgment of the Court of Justice of 14 September 1994,

Spain v Commission

, joined cases C-278/92, C-279/92 and C-280/92, ECLI:EU:C:1994:325, paragraph 75.

(68)  Judgment of the Court (Sixth Chamber) of 17 June 1999,

Belgium v Commission

, C-75/97, ECLI:EU:C:1999:311, paragraphs 64 and 65.

(69)  Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union (

OJ L 248, 24.9.2015, p. 9

).

(70)  This amount corresponds to the sum of the amount mentioned in recital 148 — PLN […] and the amount mentioned in recital (150) - PLN […] which are gross and correspond approximately to a total of EUR 223,74 million (at an exchange rate of PLN/EUR = 4).

(71)  See footnote 21.

(72)  Judgment of the Court of First Instance of 8 June 1995,

Siemens v Commission

, T-459/93, ECLI:EU:T:1995:100, paragraph 83. Please also see Judgment of the Court of Justice of 15 December 2005,

Unicredito Spa v Agenzia delle Entrate, Ufficio Genova

, C-148/04, ECLI:EU:C:2005:774, paragraph 117-120.

(73)  Commission Regulation (EC) No 794/2004 of 21 April 2004 implementing Council Regulation (EU) 2015/1589 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union (

OJ L 140, 30.4.2004, p. 1

).

(74)  Commission Regulation (EC) No 271/2008 of 30 January 2008 amending Regulation (EC) No 794/2004 implementing Council Regulation (EC) No 659/1999 laying down detailed rules for the application of Article 93 of the EC Treaty (

OJ L 82, 25.3.2008, p. 1

).

ANNEX

OVERCOMPENSATION PAID TO AUTOSTRADA WIELKOPOLSKA