According to Section 7(3) and (5) of the Planning Act, Femern A/S benefits from a State guarantee covering the obligations of Femern A/S in relation to loans and other financial instruments used to finance the preparation, investigation, planning and other necessary measures concerning the Fixed Link.

According to Section 4(2) of the Construction Act, Femern A/S benefits from a State guarantee covering the obligations of Femern A/S in relation to loans and other financial instruments used to finance and refinance the planning, construction, operation and other necessary arrangements for the purpose of construction and operation of the Fixed Link.

Femern A/S, similar to other public undertakings that obtain loans covered by a State guarantee, will be required to pay a guarantee premium to the Danish State, as notified in 2014 amounting to 0,15 % per annum on the outstanding debt covered by the guarantee.

The State guarantee covers the obligations of Femern A/S in relation to loans and other financial instruments used to finance and refinance the planning, construction, operation and other necessary arrangements for the purpose of planning, construction and operation of the Fixed Link.

The State guarantee can only be used to cover loans that Femern A/S obtains in order to finance the Fixed Link. As a special purpose company, Femern A/S cannot obtain loans for or engage in any other activity than the financing, planning, construction and operation of the Fixed Link.

The complainants also argued that every time Femern A/S enters into a new financial agreement, a new guarantee is granted, involving a new ad hoc aid measure.

The complainants further argued that the State loans as intended have no fixed repayment period and are, therefore, unlimited in time. In the absence of such a period, Femern A/S can continuously delay the repayment of the initial loans by successively obtaining new loans to refinance the initial ones.

In November and December 2018, when the Construction decision was still valid, Femern A/S obtained State loans for a nominal value of DKK 7,4 billion (EUR 1,0 billion) in order to ensure the necessary funding for initiating the construction phase. Those loans have been disbursed in the period November 2018 – April 2019.

The complainants also argued that every time Femern A/S raises a new State loan, a new measure is granted, involving a new ad hoc aid measure.

The Opening decision further noted Scandlines’ concern that Femern A/S is subject to a special tax regime under the Danish tax law, which was originally introduced by the Planning Act.

According to Section 8 of the Fehmarn Belt Treaty, Germany and Denmark shall make available the necessary areas of land for the construction and operation of the Fehmarn Belt Fixed Link project (Section 8, subsection 1), including the necessary water areas and seabed, free of charge (Section 8, sub-section 2).

As mentioned in the Opening decision, Scandlines argued that through those provisions, Femern A/S benefits from a financial advantage as, in their absence, it should normally pay a market fee for the use of the water areas and the seabed.

However, the Danish authorities submitted that there is no general rule or principle in Danish law requiring companies in a similar factual and legal situation as Femern A/S to pay fees to the State for making use of seabed and water areas. Hence, the same free access principle applies to other toll funded transport infrastructures in Denmark (such as the Øresund Fixed Link and the Great Belt Fixed Link). Similarly, though not in a directly comparable situation, no fees are paid to the Danish State by ferry operators or other ship freight companies for crossing any sea areas under Danish jurisdiction. In addition, the harbours pay no fees for using the seabed.

Further, the Danish authorities expect that revenues from the railway use will account for about 15 % of the total operating income, while the construction costs for the railway part of the Fixed Link account for 51 %, which means that the users of the road connection will ultimately pay for a part of the railway connection.

Since all railway operators using the Fixed Link are supposed to pay the railway fees, the Danish authorities indicated that the charging of railway fees would not constitute use of State resources granted to Femern A/S, as the transfer of funds from railway operators to Femern A/S would have no impact on the State budget.

Moreover, Scandlines et al. refer to the case-law of the European Courts and the Commission decision-making practice, arguing that it is irrelevant whether Femern A/S can determine its own prices or is subject to state-regulated prices. Furthermore, Femern A/S retains a wide discretion to grant steep discounts with the effect that the Minister for Transport does not determine the exact consideration but only sets maximum prices.

As the activities of Femern A/S are considered economic in nature, Scandlines et al. consider that in addition to the funding for planning, constructing and operating, also the funding for promotional/marketing/information activities constitutes State aid. Scandlines et al. further emphasise that Femern A/S has performed clear marketing activities, which go well beyond informing the general public about an infrastructure project.

Scandlines et al. note that according to the Commission’s decision-making practice, the direct appointment of Femern A/S as the sole constructor and operator of the Fixed Link, without a tender, in itself involves an economic advantage. Scandlines et al. argue that each aid measure listed in Section 2.4 of the Opening decision provides a selective advantage to Femern A/S.

As regards the capital injections, State guarantees and State loans, Scandlines et al. argue that the aid element consists of the entire amount guaranteed or lent by the State, rather than the difference between the premium or interest rate paid by Femern A/S and the market conform premium or interest rate. This argument is based on the reasoning that Femern A/S would be in such a weak financial position, that no private lender of guarantor would offer any loans or guarantees. Scandlines et al. deduce this weak financial position from their interpretation of the Enquiry of commercial interest in that no market operator would grant guarantees or loans to Femern A/S. On this basis, Scandlines et al. conclude that no market rate exists and the aid element is therefore the entire guarantee and the entire loan amount.

As regards the railway fees, Scandlines et al. argue that the only relevant parameter for price setting is costs, as no competitor is offering a similar service after the removal of railway tracks to the Scandlines’ ferries. Since the Danish State will pay Femern A/S an annual railway fee of DKK 350 million (EUR 46,9 million) through the State-owned railway operator DSB, without any downward adjustment mechanism regarding the number of trains actually using the Fixed Link, the price is not cost-based. FSS adds that the amount of DKK 350 million constitutes a specific share of the overall annual revenue generated from rail fees and that the application of this fee is not limited in time.

As regards the tax measures, Scandlines et al. maintain that all measures involve an advantage. They further note that the tax measures have been applied both to planning activities and to construction works.

As regards the free use of State property, Scandlines et al. argue that Femern A/S benefits from free use of water areas and part of the seabed, as explicitly provided for in the Planning and Construction Acts. They consider the claim of Denmark that in general no fees are to be paid to the Danish State by ferry operators or other ship freight companies for crossing any sea areas, as contradictory to the inclusion of the said provision in the Acts. FSS considers the fact that companies are actually not paying charges for the transit through national sea areas as irrelevant.

Scandlines et al. further argue that the actual European net benefit of the Fehmarn Belt Fixed Link is negative and that the rate of return of the Fehmarn Belt Fixed Link is clearly well below 4 %, which they consider as the established minimum return for an infrastructure project to be undertaken in Denmark.

FSS did not comment on this element.

NABU considers that the Fehmarn Belt Fixed Link project does not deliver the benefits claimed by the Danish authorities as it does not improve traffic conditions or lead to the claimed timesavings and thus does not contribute positively to the transportation between Germany and Scandinavia.

In the view of Scandlines et al., operating aid is prohibited and cannot be authorised. Scandlines et al. consider that operating aid is not related to a particular project phase (such as the operating phase) but to the nature of the costs. Any aid covering operating costs constitutes operating aid. Therefore, all relevant operating costs should be excluded from the funding gap calculation. FSS adds that, in their interpretation of the Øresund judgment, any loan taken out to refinance an earlier loan automatically becomes operating costs and that therefore the aid for such refinancing is prohibited. The authorization of any aid to Femern A/S can therefore not cover costs for refinancing loans. In FSS’ view, the toll prices should be cost-based and price dumping should be excluded. NABU did not comment on the operating aid element.

Scandlines et al. claim that the funding gap as described in the Opening decision is flawed for three reasons: (1) it includes inflated and ineligible costs relating to (i) dividends paid to A/S Femern Landanlæg for the hinterland connections; and (ii) operating costs; (2) it includes artificially low revenues by (i) excluding the TEN-T subsidies; and (ii) by accepting that Femern A/S sets its tolls below costs; and (3) the lifetime of the Fixed Link project is too short. Scandlines et al. consider that all those elements lead to an artificially high funding gap and thus to a maximum permissible aid amount that is also too high.

As regards the eligible costs, Scandlines et al. recall that the Commission previously held that, where an aided project involves both economic and non-economic activities, only costs linked to the economic activity may be included, when calculating the funding gap. On this basis, they argue that the hinterland costs and the operating costs cannot be included in the funding gap, as those costs do not form part of the Fixed Link project. Furthermore, as dividend payments are not a cost item but distribution of profits, they cannot be included in the funding gap analysis. FSS adds that the preparatory works that started in 2013 are construction costs and therefore they cannot be authorised as eligible costs for planning purposes. Furthermore, in the view of FSS, planning costs are not part of the construction costs as laid down in Table 1 of the Opening decision.

As regards the revenues, Scandlines et al. claim that TEN-T subsidies are not included in the 2014 analysis of the funding gap. Including those subsidies will result in a significantly lower funding gap. In addition, revenues are artificially low due to the fact that Femern A/S’ toll prices are not cost-based but based on Scandlines’ ferry prices. The Commission must therefore require that during the period that Femern A/S will benefit from State aid, the prices of the Fixed Link are set at a level that reflects all its costs including serving its debt. This last element was not brought forward by FSS. As to the lifetime of the project, Scandlines et al. recall that the funding gap and IRR analysis should be based upon the lifetime of the investment and not on the repayment period of 55 years, as also confirmed by the General Court. It would appear from the website of Femern A/S that the lifetime of the tunnel is 120 years.

According to Scandlines et al., the amount of aid is underestimated as it only refers to the difference between the interest/premium rates paid by Femern A/S for the State loans and State guarantees and the corresponding market rates.

Scandlines et al. further argue that the duration of the aid must be much shorter than 55 years. In addition, each State guarantee and State loan must be time limited. The aid for Femern A/S must expire when the revenues can cover the costs of operations and the costs related to the debt on marked-based terms. Scandlines et al. refer to the annual reports of the Great Belt and the Øresund fixed links to suggest that each individual State guarantee and State loan should be limited in duration to 4 to 5 years after the opening of the Fixed Link and that the possibility to obtain the State guarantees and State loans should also be limited to 4 to 5 years after the opening of the Fixed Link.

Scandlines et al. claim that Femern A/S initiated the construction works on the Fixed Link in 2013 although the Danish authorities committed to grant the aid at the earliest in 2015 with the adoption of the Construction Act. The aid to Femern A/S therefore would not have the required incentive effect. Scandlines et al. further criticise the Commission for finding that publicly owned companies should just be exempt from the requirement that the aid has an incentive effect and allege that this approach is an example of discrimination in favour of publicly owned companies, contrary to Article 345 of the Treaty.

Scandlines et al. consider the IRR and WACC calculations to be wrong. The IRR should be calculated over the lifetime of the project, not over 55 years, and non-eligible costs should be excluded. The WACC of 11 % is too high. According to Scandlines et al., the WACC should range between 5,68 % and 6,71 %. Scandlines et al. provided two studies further documenting their view on the WACC calculation, on the one hand a study dating from February 2018 and on the other hand references from Damodaran from 2014.

NABU added that the impact on the environment should be part of the balancing test. The negative effects of the aid should not be limited to the distortion of competition.

By reference to the Guarantee Notice, Scandlines et al. reminded the Commission that it is not entitled to authorise aid in the form of State guarantees unless the Commission knows beforehand what the conditions are for triggering those guarantees. FSS requires that the Danish authorities make a commitment that the conditions for the mobilisation of State guarantees will comply with the requirements of the Guarantee Notice.

In the view of the Danish authorities, the planning, construction and operation of the Fixed Link are classic examples of an exercise of public planning power, which are not, and ought not to be, covered by Article 107(1) of the Treaty. Consequently, the State measures, including notably the State guarantee and State loans granted in favour of Femern A/S fall outside the scope of EU competition policy.

Contrary to the methodology used by the Commission in the Opening decision, the Danish authorities submit that an overall assessment – in light of the nature, aim and applicable rules – of Femern A/S’ activities should lead to the conclusion that Femern A/S exercises public power – or at least that Femern A/S exercises public power until it starts operation.

The Danish authorities argue that the aim of connecting the Danish and German State-owned road and rail infrastructures is much wider than other public investments in, for example, the utility sector. Similarly, the nature of Femern A/S and its activities are also fundamentally different from, for example, ferry operators and ferry activities, as Femern A/S is created with the sole purpose to implement the Danish State’s decisions regarding the construction of a particular infrastructure (the Fixed Link). Femern A/S is thus a Special Purpose Vehicle (‘SPV’) which is not subject to a profit maximisation standard and not allowed to carry out unrelated activities according the applicable rules.

According to the Danish authorities, an appreciation of those wider policy considerations underlying the setting up of Femern A/S supports the notion that Femern A/S does not carry out economic activities but rather exercises public power.

According to the Danish authorities, it is clear that the collection of charges by Femern A/S, which according to law are determined by the Minister for Transport, cannot be separated from the exercise of activities connected to public authority, such as providing access to and maintaining State owned general road and rail infrastructure. Thus, all activities by Femern A/S should be seen together as an exercise of public authority.

The Danish authorities also clarify that the Minister for Transport has not yet decided how the user fees will be established or what the level of those fees should be. The fees referred to in the Opening decision and in the comments of the interested parties is simply a reflection of budgetary assumptions. It is further noted, that the main aim and function of the user fees is to balance public budgets.

The Danish authorities further request the Commission to carefully assess when the period of potential economic activities would start and they refer to case-law arguing that whenever it is possible to separate activities, each of those activities should be subjected to a separate State aid analysis.

In addition, they argue that it makes little sense to assume that when the Construction Act entered into force and authorised Femern A/S to construct and operate the Fixed Link, all financing support granted to Femern A/S prior to that date became State aid retroactively.

In the Danish authorities’ view, there is no doubt that Femern A/S’ information activities are a public task and not an economic activity. One of the tasks imposed on Femern A/S in the Planning Act is ‘information activities’ which should be conducted in cooperation with local municipalities. This is a task that no private commercial infrastructure operator has. It is an inseparable part of the public tasks connected with the planning and construction of public infrastructure projects that affect the local community. Furthermore, the Danish authorities note that if those information activities are not considered as a public task, they should be considered as inextricably linked to the realisation of the project and therefore eligible under the IPCEI Communication.

In the view of the Danish authorities, the Commission must carry out a specific assessment of whether competition is distorted or not. This assessment should include a thorough analysis of the organisation of the State road infrastructure in Denmark to enable a determination of whether State roads are in a real competitive relationship with other modes of transport such as ferries, airlines, train operators, etc.

On this basis, the Danish authorities submit that any possible economic effects, which Femern A/S’ activities may have on other commercial operators cannot be considered a distortion of competition in the sense of Article 107(1) of the Treaty. It is especially clear, that Femern A/S does not distort any competition on any market while it plans and constructs the Fixed Link.

The Danish authorities note that Femern A/S has not at any time used the previously applicable rules regarding depreciation of assets and carry forward of losses. Thus, regardless of whether the rules may potentially have granted Femern A/S a selective advantage, no such advantage actually occurred in the relevant period.

In its remarks to the comments of the interested parties, the Danish authorities explain that the Construction Act simply clarifies, for the avoidance of doubt, that Femern A/S should not pay a fee for its use of the seabed. They reiterate that these provisions do not treat Femern A/S differently from entities in a comparable legal and factual situation.

The Danish authorities consider that the new studies submitted by Scandlines and Stena Line are irrelevant and, in any event, incorrect. In relation to the Knud Erik Andersen study, they note that the author is a known opponent to the Fehmarn Belt Fixed Link project, and that the study is based upon wrong calculation assumptions in relation to traffic forecast and construction costs. The Danish authorities further reject the argument put forward by Scandlines and Stena Line that the 2015 studies have inflated Scandlines’ costs to the extent that it results in an assumption that Scandlines operates with a massive loss. The Danish authorities consider this an incorrect and distorted presentation of the calculations in the ‘Original study’ from 2015 because the study is based on the assumption that the revenues of the ferries are higher than operating costs (the difference between revenues and costs is estimated at around 15 % of the costs). In addition, the Danish authorities argue that the studies do not become flawed because sub-components or circumstances change, and they note that new calculations of socio-economic effects are not necessary each time a sub-component changes. As such, the Danish authorities also reject that the 2015 studies are flawed due to increased costs of the German hinterland.

In the Danish authorities view, operating aid is aid relating to the operating of the infrastructure. All expenses incurred in the construction phase, including expenses concerning the day-to-day administration of the construction project are considered as construction costs. Femern A/S will not need to obtain loans from the State or loans with a State guarantee to finance its day-to-day operations in the operations phase.

The construction costs in the 2014 notification amounted to DKK 54,9 billion (EUR 7,4 billion) in fixed 2014 prices. This amount corresponds to the construction costs mentioned in the 2015 Construction Act: DKK 55,1 billion (EUR 7,4 billion) in fixed 2015 prices. Femern A/S received bids in September 2015. Consequently, Femern A/S prepared an update of the construction budget and the forecast of planning and construction costs for the Fixed Link was reduced to DKK 52,6 billion (EUR 7,1 billion) in 2015 prices or DKK 62,2 billion (EUR 8,3 billion) in nominal prices. This includes a reserve requirement of DKK 7,3 billion (EUR 1,0 billion) in 2015 prices. The reserve has been calculated based on Femern A/S’ risk register and reflects costs the project most likely will incur. A ‘P-value’ of P80 was used. As the Danish authorities explained, this means that there is an 80 % probability that the reserve budget will be sufficient to cover the identified risks. Ernst & Young has performed an external quality assurance of reserves and risks associated with the Fixed Link.

The financial calculations assume that the Fixed Link will receive EU funding corresponding to 10 % of the construction costs throughout the construction period, which is the same level as was assumed in the 2016 financial analysis. The funding relating to the planning phase has also been taken into account in the financial model.

The traffic forecast had been prepared by the German consultancy firms Intraplan and BVU and crosschecked with other models such as the Trans Tools model of the European Commission. The traffic forecast was presented in November 2014. As stated in the 2016 financial analysis, COWI consultancy group performed an external quality assurance of the forecast and concluded that the assumptions are reasonable and in line with professional traffic forecasting practice. The basic assumption in the 2014 traffic forecast is that the existing parallel ferry service would not continue when the Fixed Link opens. At the same time, the 2014 traffic forecast includes a separate analysis of a scenario, with continued (parallel) 1-hour ferry service. Since 2014, Scandlines repeatedly confirmed its intention to continue operating after the opening of the Fixed Link. The Danish authorities therefore considered in the financial funding gap calculation model a 1-hour ferry service between Rødby and Puttgarden. The Danish authorities explained that this is in line with the PWC study prepared for the German authorities in the context of the German plan approval process for the Fixed Link, as provided by Scandlines et al. in an annex to their comments to the Opening decision. Furthermore, the study ‘Forretningsanalyse af færgefarten Rødby-Puttgarden’ of 24 January 2016 prepared by consultants from KPMG on behalf of the Danish Ministry of Transport confirmed that the ferry service can continue with a very small marginal return. The effect of the 1-hour continued ferry service on the traffic volumes, already estimated in the 2014 traffic forecast, is a reduction of 14 % in the volume of cars in the first year after opening of the Fixed Link, decreasing to a 12 % reduction 13 years after the opening. For lorry traffic, the share using the parallel ferry service was estimated at 15 %. A second update after the 2016 financial analysis concerns the effect of the toll reduction on the Great Belt Link by 25 % with full effect as from 1 January 2023. This slightly reduces the traffic volumes on the Fixed Link.

As to the rates for the road traffic, the Danish Minister for Transport will determine the final rates just before opening of the Fixed Link. The Danish authorities explained that the road traffic prices are an extrapolation of 2007 ferry prices, when the Rødby-Puttgarden ferry route was still a legal monopoly owned by the Danish and the German states. The Danish authorities explained that the ferry prices were considered as a reasonable and transparent benchmark – also taking into account what future users of the Fixed Link would be willing to pay without jeopardising the broader aims of the Fixed Link. The assumed and average rate for passenger cars is DKK 494 (EUR 66,2) and for lorries DKK 2 092 (EUR 280,4) in 2015 prices.

In the funding gap calculation model, railway revenues are estimated at around DKK 400 million (EUR 53,6 million) per year (2015 prices). The fees will be determined in accordance with Directive 2012/34/EU. The Danish authorities provided a report from Consultancy TetraPlan ‘Analysis of Rail Infrastructure Payment on the Fehmarn Belt Fixed Link’ from March 2003 and its update from 2008. The 2003 document contains a technical analysis on the railway’s ability to pay for the using of a Fixed Link. The purpose of the analysis was to substantiate whether the estimated payment for using the rail track could be expected to materialise. The TetraPlan analyses assume that the total railway fee for trains going from/to Hamburg via the Fehmarn Belt Fixed Link should equal the total railway fee for trains going via the Great Belt fixed link.

The costs for operation, maintenance and reinvestment of the Fixed Link correspond to DKK 468 million (EUR 62,7 million) per year (in 2015 prices). This estimation is in line with the 2016 financial analysis, as the Danish authorities did not consider it necessary to adjust these figures. The Danish authorities evaluated the effect of the reduced traffic volume on the costs for operation, maintenance and reinvestment. They concluded that the reduced traffic volume only has a minor impact on the operating and maintenance costs and does not have any noticeable effect on the State aid calculations and the guarantee period.

In reply to the concern expressed by the Commission that also an average risk-free rate of 3,5 % per year might be too high, the Danish authorities explained that, due to the urgent need for a State aid decision, the Danish authorities are willing to use a risk-free rate of 1,5 % for the purpose of determining the funding gap if the Commission were to consider this necessary for the aid to be compatible with the internal market.

Accordingly, the Danish authorities provided two funding gap calculation models. One was based on an average risk-free rate of 3,5 % (time varying starting from low levels and increasing over time) (‘first funding gap calculation model’) and another one assumed, as an alternative model, a risk-free rate of 1,5 % (‘alternative funding gap calculation model’).

The resulting WACC figure is therefore dependent upon the applicable model. The calculations based on the Ministry of Finance’s projection result in a WACC ranging between 4,2–8,3 % in the first funding gap calculation model, while the calculations based on a risk-free rate of 1,5 % (in the alternative funding gap calculation model) result in a fixed WACC of 5,59 %. The Danish authorities explained that the WACC level was estimated in terms of what a typical investor would require for a project with comparable risk characteristics. They provided full details on how this WACC was established.

The Danish authorities calculated the funding gap model on the basis of an operational period of 40 years. They did not include dividend payments relating to the hinterland.

In order to address the doubts raised in the Opening decision, the Danish authorities revised the notification and limited the use of State guarantees and State loans. The State guarantees and State loans will be limited to the financing needed for the costs incurred during the planning and construction phase. Femern A/S will terminate all loans with a State guarantee and repay all State loans at the latest 16 years after start of operations of the Fixed Link (‘maximum guaranteed period’). Start of operations of the Fixed Link is the point in time where the Fixed Link opens for road and rail traffic. Femern A/S will not be able to raise loans with a State guarantee and State loans, which, together, exceed a maximum amount of DKK 69,3 billion (EUR 9,3 billion) (‘maximum guaranteed amount’). The maximum amount is the expected total net construction debt for Femern A/S, including accrued interest costs during the planning and construction phase. To this amount, a liquidity reserve of DKK 2,0 billion (EUR 0,27 billion) and a ‘construction cost buffer’ of DKK 2,5 billion (EUR 0,34 billion) has been added (2015 prices). The liquidity reserve was added to cover the risk of short-term fluctuations in project payments. The Danish authorities explained that an additional construction cost buffer is needed because, if construction costs were higher than expected (expected construction costs were calculated with a risk probability of 80 %), Femern A/S would not be able – still in construction phase – to obtain loans on the market. The buffer corresponds to an increase in the risk probability to 99 %. The Danish authorities however committed to a recalculation mechanism which adjusts the maximum guaranteed amount and the maximum guaranteed period downward if the actual costs incurred in construction mean that the construction cost buffer is not needed thus ensuring that the total aid amount does not exceed the funding gap. The Danish authorities committed to performing this recalculation at the latest five years after start of operations.

Femern A/S is obliged to pay a premium in connection with the State guarantees and State loans. As stated in the preparatory notes to the Construction Act (Section 11.1), this fee amounts to 0,15 %. This corresponds to the premium to be paid by Femern A/S as to the first funding gap calculation model of the revised notification. In the alternative funding gap calculation model, this premium is higher and amounts to 2 %.

Furthermore, Femern A/S will not use State loans and loans with a State guarantee to finance its dividend payment obligation to its owner A/S Femern Landanlæg. Thus, it is only when Femern A/S’ debt is fully repaid and its possibility of adopting new State loans and use State guarantees to finance future debt is terminated, that Femern A/S may start paying dividends to A/S Femern Landanlæg.

In addition, the Danish authorities informed the Commission that Femern A/S’ swap/derivatives contracts will be carried out on market terms. Therefore, Femern A/S will pay a market premium on the notional of the individual transactions to the Danish State for the State guarantee and for the collateral Femern A/S may need to post when entering into swap/derivative transactions.

The Danish authorities further submitted that, as the maximum guaranteed amount is limited to the planning and construction costs (with related financial costs), they consider that operating aid is excluded. In order to further demonstrate this, the Danish authorities will ensure that Femern A/S annually reports an account of its cash flows.

In the view of the Danish authorities, it makes little sense to consider a counterfactual scenario for a general transport infrastructure such as the Fixed Link, as it is decided by two States for broader public purposes and will be implemented by a public SPV established by law with no other activities or interests. As such, any counterfactual scenario for this SPV would be fully speculative. The counterfactual scenario therefore exists in the absence of an alternative project.

In their comments to the arguments of Scandlines et al., the Danish authorities refer to the General Court’s judgments of 13 December 2018. They also argue that the Fixed Link is liable to create new competition rather than restricting it, that the potential abuse of market dominance is only hypothetical, that it is the Danish / German State that decides on removing rail access and on the capacity of the Fixed Link and that it is the Danish Minister for Transport that decides on the tariffs. On the refusal of the Danish Government to support an application for CEF support by Scandlines, the Danish authorities clarified that the Ministry of Transport could not support a project that aimed to explore the potential for significant new investments in ferry operations across the Fehmarn Belt since this was not in line with the transport policy based on Annex 1 of the CEF regulation. The Danish authorities further noted that the rejection of Scandlines’ application by the CEF Committee on 27 September 2019 was also based on reasons unrelated to the Danish authorities’ refusal to co-sign the application. Finally, the Danish authorities reject the allegation that State aid granted to Femern A/S can be used to finance, and hence downgrade, the access to Puttgarden, which they consider to be German hinterland connections. This is because Germany has the sole responsibility for upgrading and financing the hinterland connections to the Fixed Link.

The Danish authorities provided minimum conditions for mobilisation.

The Danish authorities committed to comply with the transparency obligation of paragraph 45 of the IPCEI Communication.

The Commission considers in this respect that an overall assessment is necessary and, to qualify as acting by exercising public power, Femern A/S’ activity should be connected with the essential functions of the State by its nature, its aim and the rules to which it is subject. The Commission does not agree with the reasoning of the Danish authorities that its preliminary analysis in recitals (85) to (88) of the Opening decision is erroneous. The arguments of the Danish authorities concerning the aim, nature and rules are not sufficient to invalidate that analysis and to conclude that Femern A/S acts by exercising public power.

According to settled case-law, the qualification of economic activity should be based upon factual elements, namely the provision of goods or services on a given market. Femern A/S, as the owner and operator of the Fixed Link, provides a transport service for remuneration to citizens and undertakings. Femern A/S will charge a fee (toll) from the users of the road section of the Fixed Link for crossing the Fehmarn Belt strait. In addition, the railway companies will pay fees for access to the railway infrastructure on the Fixed Link. Femern A/S’ revenues from road and rail are meant to finance the total cost of planning, construction and operation of the Fixed Link, but also, once the State guaranteed loans and State loans are reimbursed, the costs of the construction of the hinterland connections, through the distribution of dividends to the parent company.

There is a market for crossing the Fehmarn Belt strait, in particular because the service is already provided for remuneration by the existing ferry operator, which is a private undertaking clearly operating under market conditions. Hence the transport services provided by Femern A/S will be in competition with the transport services provided by ferry operators. Moreover, the Fixed Link will also compete with other transport links that constitute an alternative for crossing the Fehmarn Belt strait. Consequently, Femern A/S will be providing services on a market.

In any event, as evidenced by the 2014 Øresund decision, where the owner/operator was found to be engaged in an economic activity, it is clear that the Danish authorities have decided to introduce a market mechanism. As the Øresund bridge/tunnel, the Fixed Link will also be operated as a commercially exploited, toll-funded infrastructure. This goes against the argumentation that the activity would be an exercise of public power.

As the Commission could, in the Opening decision, not establish the nature of the main activity of Femern A/S, it was not in a position either to come to a final conclusion on the so-called promotional/marketing/information activities in which Femern A/S has been engaged.

As the Danish authorities consider the activities of Femern A/S related to the planning and construction of the Fixed Link as an exercise of public power, they also consider the promotional/marketing/information activities as an exercise of public power. Public authorities are often required to communicate and inform about their activities and projects and such information activities should in their view not be categorised as marketing as suggested by Scandlines et al. The Danish authorities also argue that a close dialogue with affected groups in Germany and Denmark and cooperation with NGOs and local authorities in Denmark and Germany are vital to the realisation of the Fehmarn Belt Fixed Link project.

The Commission considers that the promotional/marketing/information activities clearly contributed to the planning, construction and operation of the Fixed Link as both the activity descriptions by Scandlines et al. and by the Danish authorities seem to confirm. Marketing activities, as in the interpretation of Scandlines et al., would clearly contribute to the economic activity of Femern A/S. However, even if the promotional/marketing/information activities should be considered as information activities, as in the interpretation of the Danish authorities, those activities would also contribute to the economic activity of Femern A/S. As Scandlines et al. also submitted, it is not unusual that companies are required to inform the public about their activities. Those information obligations are not, in general, considered to constitute an exercise of public powers. The Commission considers that in this sense it is not relevant whether the activities are categorised as information, promotion or marketing activities. What matters for the State aid qualification is that those activities are contributing to the economic activity. The Commission therefore also considers the promotional/marketing/information activities as economic activities.

As the above measures (capital injection, tax measures, State loans and State guarantees and free use of seabed) were granted by Denmark, they are by definition imputable to that State.

The Commission already tentatively concluded in the Opening decision that, to the extent that Femern A/S has to be considered as an undertaking active on the market for transport services to cross the Fehmarn Belt strait, the grant of a selective advantage may strengthen its position as compared to other undertakings active on that market, such as, in particular, ferry operators and port operators.

However, in their comments to the Opening decision, the Danish authorities consider that the Fixed Link is not in direct competition with, for example, ferries, in the same way as railway infrastructure is not in direct competition with roads, busses or airports. In this respect, they refer to point 220 of the Notice on the notion of State aid. It is the opinion of the Danish authorities that the public financing of the Fixed Link project does not distort competition.

The Commission does not agree with the reasoning of the Danish authorities. It considers that the measures which confer a selective advantage on Femern A/S strengthen the position of that undertaking compared with other undertakings competing in trade between Member States. In particular, it cannot be denied that Femern A/S will compete in trade between Member States with undertakings providing alternative transport services.

Secondly, the Commission remarks that, when quoting paragraph 220 of the Notice on notion of State aid to support that competition is not distorted by public financing of roads, even toll-roads, (and bridges by association), such as the Fixed Link under certain listed conditions, the Danish authorities omitted to mention the linked footnote 327 which clearly states ‘An atypical situation in which State aid cannot be excluded would, for example, be a bridge or tunnel between two Member States, offering a largely substitutable service to the service offered by commercial ferry operators or the construction of a toll-road in direct competition with another toll-road (for example two toll-roads running in parallel to each other, thereby offering largely substitutable services).’

Therefore, for all the reasons mentioned above, the Commission considers that the measures entailing a selective advantage may be considered as affecting intra-Union trade and are liable to distort competition.

On the basis of this assessment, the Commission concludes that the capital injections, the State guarantees for loans and the State loans constitute State aid within the meaning of Article 107(1) of the Treaty. Concerning the alleged measures consisting of the depreciation of assets, the fiscal loss carry forward, the joint taxation regime, the railway fees and the use of State property (free of charge), the Commission concludes that they do not constitute State aid in favour of Femern A/S in the sense of Article 107(1) of the Treaty. In addition, the Commission considers that, in view of the revised notification, the guarantees for the derivatives do not constitute State aid.

• it must be a single project which is clearly defined in respect of its objectives as well as the terms of its implementation, including its participants and its funding128,
• it must be of common European interest129,
• it must be important quantitatively or qualitatively130.

The measures as notified in 2014 that were assessed in the Construction decision covered the planning, construction and operation of the Fixed Link, until the full repayment of the debt. In its judgments of 13 December 2018, the General Court stated that as the aid at issue covered the operating costs of the Fixed Link, it could not be ruled out that the aid to some extent might constitute operating aid, which, if present, should have been specifically assessed by the Commission.

The Commission therefore, in Section 5.3.2 of the Opening decision, raised the question whether the aid measures should be classified as investment aid or operating aid, or as a combination of both. In particular, it was unclear at that stage to what extent the measures that would be implemented during the operational phase of the project would cover financing needs related to (i) the repayment of the debt created during the planning and construction phase of the project; and/or (ii) the payment of operating costs during the operational phase of the project; and/or (iii) the payment of the dividends to the parent company; or (iv) all of the above.

According to Scandlines et al., operating aid is linked to the nature of the costs and is not linked to a particular phase (such as the operating phase) of the project. Any aid covering operating costs constitutes operating aid in their view, irrespective of the phase in which the costs occur. As operating costs are included in the funding gap analysis, Scandlines at al. consider that by definition operating aid is present. Furthermore, for Scandlines et al., such operating aid is prohibited.

It is necessary to determine first whether the measures involve operating aid. If they do, it needs to be analysed whether such potential operating aid can be declared compatible with the internal market, and on what basis.

When considering the cited definition of operating aid together with the measures as notified by the Danish authorities in 2014, it seems that Femern A/S could potentially benefit from operating aid. Considering the Planning and Construction Acts, it is clear from resp. Sections 7 and 4 that Femern A/S would have the possibility to raise State loans and benefit from State guarantees on its loans to finance and refinance planning, construction, operation and other necessary measures in relation to the construction and operation of the Fixed Link. The Construction Act did not exclude that Femern A/S would be released from costs it would normally have to bear in its day-to-day management of the Fixed Link. Since the above cited definition of operating aid clearly refers to the ‘day-to-day management’ and ‘ordinary activities’, and implicitly refers to a ‘continuous nature’ (as opposed to ad hoc nature for investment aid), the Commission considers that this potential operating aid can only be present once the Fixed Link is operational.

However, as a reply to the doubts raised in the Opening decision and as outlined in recital (177), the Danish authorities revised the 2014 notification in order to further limit the State aid and to limit the provision of State guarantees and State loans to the financing needed for the costs incurred during the planning and construction phase, and therefore to the planning and construction costs of the Fixed Link. The maximum amount of debt that can be secured by State guarantees or taken up by means of State loans has been limited to the planning and construction cost, including the financial costs related to those planning and construction costs.

On the basis of the revised notification, the Commission concludes that the aid under consideration concerns investment aid as the objective of the aid is clearly to make possible the planning and construction of the Fixed Link, and not to release Femern A/S from costs that it would normally have had to bear in its day-to-day management or ordinary activities. The fact that Femern A/S refinances frequently the loans taken up to cover the planning and construction costs, and not through for instance a single massive loan, is only a matter of the form and type of the financial structure of the project. It does not mean that State guarantees for refinanced loans relating to planning and construction costs involves operating aid. In addition, as the Commission demonstrates below in the compatibility analysis and assessment of the funding gap, the aid amount is substantially lower than the construction costs of the Fixed Link.

Although the Construction Act seemed to allow for operating aid, the Commission considers that this could only confer a potential advantage on Femern A/S once operational. On the basis of the revised notification, the potential for future operating aid has been removed.

Furthermore, the Danish authorities will ensure that Femern A/S annually reports an account of its cash flows with the purpose of demonstrating that the operation of the Fixed Link is not subsidised.

According to paragraph 28 of the IPCEI Communication, ‘the aid must not subsidise the costs of a project that an undertaking would anyhow incur and must not compensate for the normal business risk of an economic activity. Without the aid, the project’s realisation should be impossible, or it should be realised in a smaller size or scope or in a different manner that would significantly restrict its expected benefits.’

Footnote 2 of paragraph 28 of the IPCEI Communication also contains a formal incentive effect requirement providing that ‘The aid application must precede the start of the works, which is either the start of construction works on the investment or the first firm commitment to order equipment or other commitment that makes the investment irreversible, whichever is the first in time. Buying of land and preparatory works such as obtaining permits and conducting preliminary feasibility studies are not considered as start of works.’

Scandlines et al. further argued that what matters is whether Femern A/S acted (by constructing) before the aid was granted or in other words, whether Femern A/S actually started construction before receiving a commitment to have the aid. In the opinion of Scandlines et al., ‘Femern A/S started construction works in 2013, although the Danish authorities committed to grant the aid, at the earliest in 2015 with the Construction Act’. According to Scandlines et al. those works cannot be considered as planning activities. Moreover, and considering that aid for construction activities was not granted yet, the formal incentive effect would be violated. In addition, Scandlines argued in its submission of October 2019 that Femern A/S has also breached the incentive effect requirement by carrying out construction works after the annulment of the Construction decision. The Commission notes that the formal incentive effect requirement, in general, does not compare the date of start of works to the date of aid granting (as Scandlines et al. seem to argue), but compares the date of start of works to the date of aid application. In any event, as the Danish authorities explained, the so-called preparatory works, which are part of a programme under which the Danish authorities decided to initiate certain preparations at an earlier stage than originally envisaged, were based upon several approvals of the Danish Parliament of an adjustment of the planning budget. The Commission considers therefore that it cannot be argued that Femern A/S, as a specific purpose entity and given its articles of association, proceeded to those works without relying on the fact that those works would be State subsidised.

Scandlines et al. further argued that the Commission, in paragraph 163 of the Opening decision, was wrong in referring to the standstill obligation. The Commission agrees that the standstill obligation and the formal incentive effect requirement are two different concepts. Indeed, aid that has been granted in violation of the standstill obligation may still have an incentive effect.

The Commission’s position that the absence of an aid application is not decisive for the assessment of the incentive effect of the aid in this case is without prejudice to the requirement that the aid must not subsidise the costs that a project would anyhow incur (the substantive incentive effect, as provided for in paragraph 28 of the IPCEI Communication). The Commission considers that the reference by Scandlines et al. to Article 345 of the Treaty in that there would be a discrimination in favour of publicly owned companies is therefore not relevant. As explained, the Commission considers that the substantive incentive effect test of paragraph 28 of the IPCEI Communication is fully applicable but that given the specific circumstances as referred to in recital (299) of this decision, the aid can be considered as inherent in the establishment of Femern A/S within the context of the project. It should therefore be demonstrated that without the aid the Fixed Link would not be realised or at least not in the same scope.

Paragraph 29 of the IPCEI Communication requires the Member State to provide the Commission with adequate information concerning the aided project. A comprehensive description of the counterfactual scenario corresponding to the situation where no aid is awarded by any Member State is also required.

As already noted in the Opening decision, the Danish authorities submitted that no credible counterfactual or realistic description of an alternative project exists. The Danish Parliament has sole decision-making authority as regards the project’s scope and its means of financing. Hence, Femern A/S (and A/S Femern Landanlæg) have no power to decide to carry out an alternative project of a different scale.

The scope of the 2000 cost-benefit report was the economic and financial evaluation of a Fixed Link across the Fehmarn Belt comprising the assessment of different technical solutions for a Fixed Link in comparison with the continued existence of the current ferry system. The Commission acknowledges that the 2000 cost-benefit report concludes that the relative efficiency of the alternative of an improved ferry system was considerably higher. However, at the same time, the report concludes that the absolute magnitude of net benefits gained by the Fixed Link solution cannot be achieved by an improved ferry system (the highest contribution to those benefits coming from reduced travel times and savings in transport cost). Consequently, an improved ferry system was not an alternative with the same scope and achieving comparable expected benefits as the Fixed Link project that should be considered as a counterfactual scenario. As to the other scenarios, the 2000 cost-benefit report did not make a ranking or spell out a clear preference, and it can in any event not be concluded from the report that alternative scenarios would exist where, without aid, a Fixed Link solution could be realised with a comparable scope or achieving comparable expected benefits. The Commission therefore considers that it can rely on the Enquiry of Commercial Interest, as launched in 2001, to establish the scenario where the Fixed Link would get no public support.

On this basis, the Commission considers that first, no rational private investor would engage in the financing of such a project under normal market conditions and second, that the Fixed Link could only be completed with substantial public support. The fact that the final technical solution has changed since the 2001 Enquiry does not change this conclusion and nothing suggests that a counterfactual scenario without aid had become viable in the meantime. In addition, the provision of EU financial assistance under the CEF would be a complementary strong indication of the necessity of public funding for the realisation of the project. Hence, without the aid, the project would not be realised. The counterfactual scenario thus consists in the absence of an alternative project. In the absence of an alternative project, aid must be considered as necessary if, for example, it can be demonstrated that the project’s IRR is below the normal rate of return required by investors engaging in similar investment projects, or if the IRR is insufficient to cover the cost of capital required by the market.

To this end, in the context of the 2014 notification, the Danish authorities already explained that Femern A/S did not have investment projects of a similar kind or cost of capital as a whole that could be used to calculate whether the aid amount exceeds the level necessary for the project to be sufficiently profitable. Therefore, it was necessary to compare the project’s IRR with the cost of capital requirements generally seen in the industry concerned. The Danish authorities have provided an updated analysis reflecting current market conditions and reflecting the project specific risk.

As outlined in recital (175), the benchmark WACC is dependent upon the risk-free rate. In the alternative model, the WACC is lower and has been estimated by the Danish authorities to be 5,59 %.

The Danish authorities compared this cost of capital requirement of the industry with the internal rate of return that would be achieved with the Fixed Link project in the absence of aid. As the Danish authorities provided updated financial data as a reply to the doubts raised in the Opening Decision on the proportionality of the aid, the Commission refers to Section 0 of this decision for a detailed assessment of the assumptions of the models.

The results of those updated calculations show that the project’s IRR, without any State aid, would amount to 3,9 %, considering an economic lifetime of the investment of 40 years. This IRR is considerably below the cost of capital of 5,59 % required by the market that was calculated in the alternative funding gap model. By consequence, it is also lower that the cost of capital required by the market established under the first funding gap calculation model, which assumes a higher average WACC. This analysis confirms the conclusions of the 2001 Enquiry of Commercial Interest as to the necessity of State aid. In addition, at the request of the Commission, the Danish authorities conducted several sensitivity analyses on the basis of the updated financial model. It appears that the IRR remains below the cost of capital required by the market even if, as suggested by the complainant, a very long lifetime is considered (in the simulation, the model extended until year 2100).

On the basis of the above elements, the Commission concludes that the aid does not subsidise the costs of a project that would, in any case, have been incurred by Femern A/S on the assumption that the project would have been undertaken in the absence of the aid in question. The aid is therefore necessary for the construction of the said infrastructure.

According to paragraph 31 of the IPCEI Communication, ‘[t]he maximum aid level will be determined with regard to the identified funding gap in relation to the eligible costs. If justified by the funding gap analysis, the aid intensity could reach up to 100 % of the eligible costs. The funding gap refers to the difference between the positive and negative cash flows over the lifetime of the investment, discounted to their current value on the basis of an appropriate discount factor reflecting the rate of return necessary for the beneficiary to carry out the project notably in view of the risks involved. The eligible costs are those laid down in Annex […]’.

In order to address the doubts raised in the Opening decision, the Danish authorities provided an update of the funding gap analysis and limited the aid so that it does not exceed the funding gap.

It therefore needs to be assessed whether the underlying assumptions of the Fixed Link funding gap model submitted by the Danish authorities are appropriate in view of the nature, scope and risks of the project.

In this context, the Commission notes that the aid amount is directly linked to the underlying assumptions of the funding gap model, not only as a consequence of the limitation of the aid amount to the funding gap level but also due to the fact that the level of the debt, and thus the level of the aid amount, depends on factors such as the overall construction cost and the interest rate assumed. The Commission therefore finds it appropriate, given the revised notification including an update of the funding gap analysis, that the model is also based upon the most recent underlying data. The Commission accepts in this respect the methodology of the Danish authorities to use the latest fully revised financial analysis of the project, the 2016 financial analysis, as a starting point and to update that analysis with further relevant developments.

The Danish authorities restricted the funding gap model to the financing model of the Fixed Link and excluded all construction costs related to the hinterland connections from the analysis. As it appears from subsection 2 of Section 5 of the Construction Act that A/S Femern Landanlæg pays all expenses for planning, construction, operation and other necessary measures related to the construction and operation of the hinterland connections, the Commission is of the view that the Danish authorities took a correct approach in excluding those hinterland costs from the funding gap analysis for Femern A/S. The Danish authorities further proposed not to take into account the dividend payments Femern A/S might incur in favour of A/S Femern Landanlæg. Those dividend payments therefore do not appear as a cost item in the model. The Danish authorities also revised the 2014 notification as they committed to ensure that Femern A/S will only start paying dividends to A/S Femern Landanlæg after Femern A/S’ State subsidised debt is fully repaid.

The Danish authorities explained that the eligible costs are restricted to the construction costs budget. This construction costs budget includes the planning costs as well as the costs for the promotional/marketing/information activities. The Commission considers, since the timing of those information/promotional activities occurred about 10 years before the planned operational date, since there is clearly an information obligation imposed by the Danish State and since information activities were part of the Planning Act, that those costs are linked to the realisation of the project and therefore eligible for aid under the IPCEI Communication. The planning and construction cost budget amounts to DKK 62,2 billion (EUR 8,3 billion) (in nominal prices) which the Commission considers as an appropriate estimate. The budget includes a reserve requirement which is set so that there is a probability of 80 % that the budget including the reserve will be sufficient to complete the project despite any cost overruns. It reflects costs that the project will most likely incur. In addition, for the purpose of the calculation of the maximum funding gap, the Danish authorities added a construction buffer of DKK 3 billion (EUR 0,4 billion) (in nominal prices) to the construction costs. The buffer corresponds to an increase in the reserve requirement from 80 % to 99 %. This means that, not only in 80 % but in 99 % of future cost realisations, the budget including the reserve of DKK 7,3 billion (EUR 1,0 billion) as well as the buffer of DKK 3 billion (EUR 0,4 billion) will be sufficient to complete the project. However, this also means that there is a probability that the actual construction costs will not be that high and that the construction costs assumed for the purpose of determining the funding gap may therefore be overestimated by this construction cost buffer.

The Danish authorities updated the initial assumption on EU funding in its 2016 financial analysis to 10 %. This assumption is lower than the initial assumption of 18 %, which formed the basis of the Construction Act (equivalent to EU support of about DKK 7,5 billion or EUR 1,0 billion). The Danish authorities explained that in 2015, the EU granted EUR 589 million for the funding period 2016-2019 which is equivalent to approximately DKK 4,4 billion (in current prices). At the same time, there were uncertainties regarding the total funding that was to be allocated to Femern A/S, including uncertainties regarding the timing of commencement of the construction period. For those reasons, the EU support assumption was changed to 10 % in the 2016 financial analysis. However, the Commission notes that the assumption in the 2016 financial analysis was stated as cautious. The Danish authorities explained that, due to a delay in the German plan approval of more than one year and subsequent litigation before the German Federal Administrative Court, Femern A/S will not be able to use the full amount of EU grants even with a prolongation of the period of support by two years until the end of 2022 in the most optimistic scenario. Furthermore, the Danish authorities explained that they expect the general EU budget to be reduced in the financial perspectives 2021-2027 (due to Brexit) and at the same time the number of applications for EU support to infrastructure projects to be increased. Therefore, the assumption of 10 % EU funding should be considered as a realistic assumption. In the funding gap calculations, the total EU funding amounts to 12 % of planning and construction costs (in NPV terms) which, on the basis of the above explanations, the Commission considers as a reasonable assumption.

As to the revenues, nothing that has emerged in the formal investigation gives the Commission reason to put into question the road traffic forecast as it has been developed in 2014, crosschecked and subjected to external quality assurance. In addition, the Commission considers it as reasonable that the 2014 road traffic forecast was updated with the effect of the Great Belt toll reduction. For what concerns the assumption of the continued ferry service in the traffic figures, the Commission notes that this assumption is not in accordance with the baseline scenario of the cost-benefit analysis as developed in 2015. However, as the Danish authorities submitted, the risk that the ferry service would continue was already clearly identified at that point in time as the traffic analysis includes a specific sensitivity analysis on that element. Since then, Scandlines repeatedly confirmed, including in its reply to the Opening decision, its intention to maintain ferry operations with a frequency even exceeding a 1-hour service. The studies completed by consultants of PWC and KPMG referred to by the Danish authorities (see recital (168) of this decision) also seem to confirm the feasibility of a continued service. In view of all this, the Commission considers that a reasonable investor would assume a continued ferry service in its financial analysis and considers the assumption of a 1-hour ferry service to be appropriate.

As to the assumed prices, the Commission does not share the view of Scandlines et al. that price setting would be too low because ‘Femern A/S sets its tolls below costs’. The Commission observes that the expected operating revenues largely exceed operating costs over the entire operational period. It cannot be a requirement that prices are to compensate for the full costs (including construction costs) of the project. If it were possible to set prices at a level that compensates for all construction and operating costs, no funding gap would be present and no State aid would be needed. The Commission, however, considers that prices assumed in the funding gap model cannot be artificially low with the sole purpose of inflating the funding gap. The Construction Act refers back to the Planning Act, assuming that the price level for road traffic was expected to be at the level of the ferry prices for Rødby-Puttgarden in 2007, adjusted by the general increase in prices up to the time of opening. This same price level was the basis for the traffic projections and was considered as a realistic assumption by the respective studies. The traffic forecast uses a flat, technical average price for both passenger cars and lorries. It further appears from the 2016 financial analysis that, if a differentiated price structure were introduced, the effect on the overall revenues would be relatively limited. Based on the above, the Commission considers the assumed road traffic revenues to be plausible and appropriate.

The Commission further considers the basis for the calculation of the railway revenues as explained in recital (170) as reasonable.

The 2016 financial analysis incorporates a thorough review of the costs for operation, maintenance and reinvestment of the Fixed Link, leading to a serious reduction in those costs. Those new assumptions have also been included in the updated funding gap model. In the view of Scandlines et al., operating costs must be excluded from the funding gap calculation. This is based on their view that all aid covering operating costs is operating aid, which in their view cannot be allowed. The Commission already elaborated in part 6.4.2 of this decision on what constitutes operating aid. The question whether operating costs can be included in a funding gap calculation is however a different issue.

To compute the funding gap, an appropriate discount rate has to be used to discount the future cash flows. The Danish authorities used the WACC as discount rate. The WACC is the weighted average of cost of debt (‘Kd’) and cost of equity (‘Ke’) for a company. The weights used are the proportion of Equity (‘E’) and of Debt (‘D’) in the capital structure of the company (E+D). The WACC is computed as follows: WACC=Kd*D/(E+D)(1-t)+Ke*E/(E+D). The estimates for the cost of debt and cost of equity are obtained on the basis of standard approaches according to the following formulas: Kd = risk-free rate + risk premium on debt, and Ke = risk-free rate + levered beta * risk premium on equity + project specific risk premium. The tax rate is denoted by ‘t’.

The Danish authorities have estimated the WACC a typical investor would require for a project with comparable risk characteristics. The figures used for the risk premium on equity, risk premium on debt and unlevered beta were updated to the latest information available.

Furthermore, Denmark adds a project specific risk premium of 2 %. The project specific risk premium is justified with reference to the significant uncertainties in the project, for example uncertainties related to the final German plan approval, which could cause a significant delay and additional costs, but also other microeconomic and macroeconomic risks relating to the project and not taken into account in the budget assumptions. The value of 2 % is conservative when compared to what the 2018 study submitted by Scandlines et al. suggested. However the project specific risk premium should be evaluated together with the construction cost buffer used in the State aid model which already takes into account the risk of a cost overrun in the construction phase. On this basis the Commission considers the use of a project specific risk premium of 2 % as reasonable.

The Commission considers that the components of the WACC as outlined in the above recitals are reasonable in view of the size, the risks, the timing and the type of activity of the project.

In light of the current market conditions, the relatively long construction period and in particular the overall time span of the financial model (until 2068 and therefore more than 30 years), the Commission considers the assumption of 1,5 % in the alternative model as reasonable and justified. In addition, Scandlines et al. provided two different studies on the estimation of the WACC. The Danish risk-free rate assumption of 1,5 % is in line with the estimation of 1,6 % in the most recent study (February 2018 but based upon 2014 data) and substantially lower than the figure of 3,04 % presented in the previous study of early 2014).

It results from the alternative funding gap calculation model that the funding gap for the Fixed Link amounts to DKK 12 046 million (EUR 1 615 million). A comparison between the funding gap and the eligible costs gives a funding gap ratio of 27,3 %.

The gross grant equivalent of the aid includes the aid resulting from the capital injections, the State guarantees and the State loans. The methodology the Danish authorities used to calculate the aid element of the guarantees follows a similar approach as set out in Section 4.2 of the Guarantee Notice, which provides that, in case no market price is available, ‘the aid element should be calculated in the same way as the grant equivalent of a soft loan, namely as the difference between the specific market interest rate the company would have borne without the guarantee and the interest rate obtained by means of the State guarantee after any premiums have been taken into account’. In their calculation of the aid amount, the Danish authorities did not make a distinction between the value of State aid associated with the State loans and the value of State aid associated with the State guarantees. It also follows from the Construction Act that both are interchangeable and the Minister of Finance has discretion in deciding on the optimal mix of both instruments. The annual premium which Femern A/S has to pay to the State on the outstanding guaranteed debt is equal to the premium on the outstanding State debt. The aid elements resulting from the State guarantees and the State loans were therefore calculated in the same manner. The Danish authorities determined the yearly aid element by taking the difference between the WACC that a market investor would be expected to require (5,59 %) and the risk-free rate (1,5 %) adjusted for the premium that Femern is required to pay to the Danish State, multiplied by the sum of outstanding guaranteed debt and outstanding State debt.

The Commission considers it appropriate that, in this case, the aid element corresponds to the difference between the risk-free rate (adjusted for the premium) and the WACC. Thanks to the State guarantees and State loans, Femern is expected to pay the same rate as the Danish State, which equals the Danish risk-free rate. Therefore, Femern A/S’ expected actual financing cost is the risk-free rate, adjusted by the premium. Without any aid, Femern is expected to pay the WACC which corresponds to the weighted average between the cost of equity and the cost of debt. In turn, taking the difference between the two rates captures the overall advantage.

Typically, the cost of equity is higher than the cost of debt and a project/company is financed by a mix of debt and equity. It could be considered as an additional advantage for Femern to be able to finance its project almost entirely with debt without any substantial need for equity. Relying on the cost of debt only would underestimate the aid. The Commission thus considers that the benefit of (almost) only debt financing is taken into account in the aid element on the State loans and State guarantees by using the WACC as a reference instead of the cost of debt.

Furthermore the Commission considers that subsection 4 of Section 4 of the Construction act, stipulating that the Danish government guarantees the other financial commitments of Femern A/S related to the construction of the project, does not involve additional aid since, as the construction costs are already covered by State loans and loans with State guarantees, it does not provide an additional advantage to Femern A/S.

Scandlines et al. are of the view that the aid element equals the full amounts effectively covered by the State loans and loans with State guarantees. The Commission notes that this determination of the aid element in a guarantee is provided for in the Guarantee Notice only in exceptional circumstances and where a guarantee is provided for companies in difficulty where the likelihood that the borrower will not be able to repay the loan becomes particularly high. The Commission considers that such exceptional circumstances are not present in this case.

The net debt is the effective debt of Femern A/S. The effective debt reflects the accumulated amount of money spent by the company to cover planning, construction, interest payments, own costs, etc. reduced by the paid-in equity and the received EU support. The net debt builds up during the construction phase is expected to reach its maximum the first year of operation. In the operational phase, the net debt will gradually decrease with the free cash flow of the project. This net debt is covered by State loans and loans with a State guarantee for a certain number of years.

Since the IPCEI Communication requires that aid in the form of guarantees is limited in time, and aid in the form of loans is subject to repayment periods, the Danish authorities ensured, in the alternative model, that Femern A/S will not adopt State loans and State guarantees, which, together, exceed an amount of DKK 69,3 billion (EUR 9,3 billion) (nominal). This amount is referred to by the Danish authorities as the ‘maximum guaranteed amount’. Those State loans and State guarantees are strictly limited to the financing needed for the costs incurred during the planning and construction phase. It is the sum of the maximum net debt required to finance the net construction costs (i.e. net of the EU funding), as described above, and a liquidity reserve. The liquidity reserve is the short-term liquidity, held by the company normally as bank deposits or short-term investments. The Danish authorities explained that the purpose of this liquidity reserve is to have a buffer for unplanned liquidity needs, such as payment to contractors and to allow borrowing to be executed well before the actual need for the liquidity. An example is the refinancing of an existing loan which normally takes place some weeks before the maturity of that loan in order to reduce the risk of unforeseen adverse market conditions.

In the alternative funding gap calculation model, the Danish authorities will limit the period within which the Minister of Finance can issue State guarantees and State loans in favour of Femern A/S. They will also ensure that Femern A/S will have terminated all loans with a State guarantee and repaid all State loans at the latest 16 years after start of operations. This means that all State guarantees and State loans are time-limited. The Danish authorities in any event ensure that the guaranteed period will not exceed the actual debt repayment period. If the actual debt repayment period is shorter than 16 years, no further guarantees will be provided after the actual debt is repaid.

It could therefore be considered – in principle – that the aid is proportionate. However, and as already highlighted above, the Commission is concerned that this funding gap might be overestimated due to the inclusion of a P99 reserve budget in the eligible cost base (see recital (320)). The Danish authorities therefore, in the alternative model, combine the ex ante calculation of the funding gap (with the related maximum guaranteed amount and maximum guaranteed period) with a recalculation of this funding gap at the latest five years after start of operations if it appears that the ‘construction cost buffer’ of DKK 2,5 billion (EUR 0,3 billion) was not needed. The Danish authorities will recalculate the funding gap, the maximum guaranteed amount and the maximum guaranteed period. They will then reduce the maximum guaranteed amount and the maximum guaranteed period if the updated funding gap is smaller than anticipated thereby ensuring that the net present value of the total aid amount does not exceed the funding gap. The Danish authorities provided a simulation in which the construction cost buffer would not be needed at all. In that case the maximum debt would be reduced to DKK 66,1 billion (EUR 8,9 billion) and the maximum guarantee period would be limited to 11 years after start of operations. In any event, the Danish authorities will ensure that the State guarantees and the State loans are strictly limited to the financing needed for the actual costs incurred during planning and construction phase. The timing of the recalculation therefore does not allow for costs related to the operational period to be State subsidsed. It only grants a temporary time buffer to account for any financing, still related to the construction phase, but only occurring shortly after the end of the construction phase. The Danish authorities will submit this updated calculation to the Commission.

The maximum guaranteed period and guaranteed amount therefore directly result from the funding gap calculation. The lower the funding gap, the lower the guaranteed period and/or guaranteed amount and the sooner Femern A/S will have to borrow on the open market without the support of State guarantees or State loans. The Commission considers it more appropriate to link the limitation of guaranteed period and guaranteed amount to the limits of the funding gap than to an amount of annual cash flow, as Scandlines et al. suggested. The annual cash flow is in practice always fluctuating, due to for example replacement investments and cyclical maintenance, so in any event [a kind of] net present value calculation would be required. In addition, the aid measure consists of a multitude of loans, all with individual repayment periods, within the limits of the overall maximum guaranteed period. It is therefore possible that at a certain point in time State loans and State guarantees will coexist with market conform borrowing. In any event, the Commission does not consider aid in excess of the funding gap as proportionate.

Furthermore, the Commission considers that the choice of State guarantees and State loans as main instruments is a positive indicator as regards both the proportionality and the appropriateness of the aid and preferred over the granting of a lump sum amount As pointed out in paragraph 36 of the IPCEI Communication, ‘where lack of finance is the underlying problem, Member States should normally resort to aid in the form of liquidity support, such as loans or guarantees’. The Commission considers that the Danish financing model is in line with this principle. A guarantee measure is an effective instrument to ensure that Femern A/S is not overcompensated as it merely enables Femern A/S to keep its capital costs at an appropriate level in order to make the project feasible by closing the funding gap. The Danish authorities also revised the 2014 notification and will report annually on the developments in the repayment of Femern A/S’ debt.

Based on the above, and taking into account the alternative funding gap calculation model of the revised notification in addition to the Planning and Construction Acts, the Commission concludes that the State guarantees and State loans can be considered to be limited in time and amount. Furthermore, the Commission concludes, again based upon the alternative funding gap model in the revised notification, that the aid amount does not exceed the minimum necessary for the aided project to be sufficiently profitable.

According to paragraph 40 of the IPCEI Communication, ‘the Member State should provide evidence that the proposed aid measure constitutes the appropriate policy instrument to address the objective of the project. An aid measure will not be considered appropriate if other less distortive policy instruments or other less distortive types of aid instruments make it possible to achieve the same result’.

The Danish authorities submitted that since the underlying problem for the project was lack of access to finance, aid in the form of liquidity support, such as loans or guarantees constitutes the appropriate policy instrument. Following the revised notification, the State guarantees and the State loans are strictly limited to the financing needed for the costs incurred during planning and construction phase. There is no risk that the guarantee/loans can be used to subsidise other non-eligible costs and activities. The Commission therefore considers that the chosen financial support mechanism is the appropriate policy instrument.

According to paragraph 41 of the IPCEI Communication, ‘aid can be declared compatible if the negative effects of the aid measure in terms of distortions of competition and impact on trade between Member States are limited and outweighed by the positive effects in terms of contribution to the objective of the common European interest’. Paragraph 42 provides that, in assessing the negative effects of the aid measure, the Commission will focus its analysis on the foreseeable impact the aid may have on competition between undertakings in the product markets concerned, including up- or downstream markets, and on the risk of overcapacity. Paragraph 43 of the Communication sets out that ‘the Commission will assess the risk of market foreclosure and dominance […] projects involving the construction of an infrastructure must ensure open and non-discriminatory access to the infrastructure and non-discriminatory pricing’.

It follows that for the purposes of preventing undue distortion competition and the balancing test, the Commission should focus its assessment of the negative effects of the aid on the distortions of competition and impact on trade between Member States.

The Fixed Link is part of a wider plan to promote mobility, further integration and cultural exchange of people living on both sides of the Fixed Link, and to improve the connection between the Nordic countries and central Europe for passengers as well as road and railway freight. Those expected benefits have been recognised at European level by including the Fehmarn Belt Fixed Link project in the list of TEN-T priority projects. In this context, the Fehmarn Belt Fixed Link project will also generate positive effects on a number of economic sectors in the region, such as gas stations, retail, restaurants, hotels, amusement parks and rail and bus and transport. As recognised by the CEF programme, the Fixed Link will enhance the accessibility to the railway transport leading to a transfer of freight and passengers from road to rail.

However, the opening of the Fixed Link will have a negative impact on ferry operators serving the Rødby – Puttgarden route as well as other ferry routes in the region. Decreased ferry operations may also have a negative impact on the ports used by those ferries in terms of traffic volumes and revenues. As recognised and taken into account in the socio-economic studies, the presence of the Fixed Link entails the risk that the ferry operations on the Rødby – Puttgarden route will even disappear once the Fixed Link is operational. The Commission considers the power of Femern A/S to influence the operations of the ferry services as rather limited since the State aid is limited to the financing needed for the costs incurred during planning and construction phase with the funding gap as upper limit. In addition, it is the Danish Minister for Transport that will determine the tolls and railway charges to be collected from the users of the road and the rail connection of the Fixed Link. Even if Femern A/S has some influence on the price level by applying discount schemes, it is bound by the need to ensure its revenue level as the State aid has been limited to finance part of the planning & construction of the Fixed Link, up to the limit of the funding gap. In other words, Femern A/S will have to ensure that its revenues are large enough to pay back its loans for the full planning and construction costs and to pay for its operating costs. The Commission therefore considers that the main impact on the ferry operations is created by the mere decision to construct the Fixed Link, providing an alternative to existing modes of transports. The choice for this infrastructure and its technical solution is a choice made by the public authorities. It is therefore also not for the Commission to assess whether the Puttgarden access plans will, if at all part of the Fixed Link, be ‘downgraded’, as alleged by Scandlines. Also the decision of the Danish authorities not to support the Scandlines project proposal should be considered in the wider mobility plan and mobility choices and is therefore not relevant in the balancing assessment. The effects on the ferry operations and related markets are therefore inherent in this type of projects, through which the States seek to offer a quicker and more convenient alternative to ferry services. The Commission also refers to the remark of the Danish authorities that the planning of the rail lines on the Danish side and the prioritising of the road on the German side are public authority tasks – decided by the Danish State and the German State respectively and not Femern A/S.

Scandlines et al. further argued that the Fixed Link would develop much additional capacity to an already saturated market. The Commission observes that the creation of an alternative to the existing services that is different from and considered superior by the Danish authorities cannot be equated to adding capacity to a saturated market.

Therefore, as also confirmed by the General Court in its judgments of 13 December 2018, while it is reasonable to conclude that a project involving the construction of infrastructure that will provide an alternative to existing modes of transport entails the risk that the latter will have to significantly reduce their activities or even disappear, it is clear that the Fixed Link provides a solution which, on balance, has positive results. It is not for the Commission to call into question the choice made by the Danish authorities.

Taking into account the foregoing, the Commission considers that the aid, as further limited and reduced in the revised notification in response to the doubts raised in the Opening decision, only has limited negative effect on competition and trade that are outweighed by the positive effects in terms of contribution to the objective of common European interest.

In any event, the environmental impact of the project has been duly considered and mitigated by the national authorities and was found, in accordance with applicable Union and international law, not to preclude the project. In the decision making process on the Fixed Link, the Danish authorities duly considered the environmental impact of the project. The comprehensive and thorough environmental impact assessment that was carried out on the Danish side is described in detail in the preparatory notes to the Construction Act. It is clearly specified in those preparatory notes but also in the Fehmarn Belt Treaty, and therefore applicable both in Denmark and in Germany, that the requirements under EU and national law must be the basis for the preparation, construction and operation of the Fixed Link across the Fehmarn Belt. On the basis of the Planning Act, the Minister for Transport was authorised after negotiations with the Minister for Environment, to prepare the Environmental Impact Assessment (‘EIA’) reports for the construction project, including consultation and other necessary environmental assessment of the construction project. The Ministry of Environment, relevant authorities, companies and municipalities were involved and the EIA reports etc. were prepared in compliance with the EIA Directive.

It follows that the basis for the project has always been that it should be prepared, constructed and operated so that harmful effects on nature and the environment are prevented and considerable adverse impacts, especially regarding the European Natura 2000 Network, are countered adequately. The Commission therefore considers that, even if the environmental aspects had to be taken into account for the purposes of the balancing test, there is no indication that the alleged negative effects of the Fixed Link on the environment would be of such a magnitude that they are liable to change the outcome of the balancing test.

The Commission concludes that the negative effects of the aid measure in terms of distortion of competition and impact on trade between Member States are limited and outweighed by the positive effects in terms of contribution to the objective of the common European interest.

According to Section 5.3 of the Guarantee Notice, ‘The Commission will accept guarantees only if their mobilisation is contractually linked to specific conditions, which may go as far as the compulsory declaration of bankruptcy of the beneficiary undertaking, or any similar procedure. These conditions will have to be agreed between the parties when the guarantee is initially granted. In the event that a Member State wants to mobilise the guarantee under conditions other than those initially agreed to at the granting stage, then the Commission will regard the mobilisation of the guarantee as creating new aid which has to be notified under Article 88(3) of the Treaty.’

1. (a)

   Femern A/S has failed to duly pay on the ordinary due date a sum payable under the guaranteed agreement;

2. (b)

   the lender must give written notice to the guarantor as set out in the relevant guarantee;

3. (c)

   such notice is not to be given until all applicable remedy periods under the guaranteed agreement have expired;

4. (d)

   the lender is not entitled to grant Femern A/S an extension of time for fulfilling its obligation under the guaranteed agreement;

5. (e)

   the guarantor shall have at least four banking days from receipt of such notice to pay the amount due under the guaranteed agreement.

As already noted in paragraph 203 of the Opening decision, and with regard to the State guarantee covering non-financial obligations foreseen in Section 4(4) of the Construction Act, the Danish authorities submitted that the conditions for mobilising this type of guarantee are based on an interpretation of the guarantee in light of the general principles of Danish law. The Danish authorities provided further clarifications on this issue. The Danish law on guarantees distinguishes between ‘simpel kaution’ and ‘selvskyldnerkaution’. A ‘simpel kaution’ means that the guaranteed party must show to the guarantor that the debtor is unable to pay his obligations. Normally this requires either (i) that it has been established during an execution (i.e. an attempt to execute a claim against the principal’s assets) that the principal is unable to pay his obligations as they fall due; or (ii) that the principal is subject to bankruptcy or similar insolvency proceedings. A ‘selvskyldnerkaution’ means that the guaranteed party can ask the guarantor to pay if the principal has failed to make payment in due time. According to Danish jurisprudence on guarantees, a guarantee is normally interpreted as a ‘simpel kaution’ unless there is a clear basis for interpreting the guarantee as a ‘selvskyldnerkaution’. Thus, unless the guarantee is clearly described as a ‘selvskyldnerkaution’, the guaranteed party will have to show to the guarantor that the principal is unable to pay his obligations as they fall due. As regards the provision in Section 4(4), this entails that, since it is not provided in the Construction and Operation Act that the conditions for mobilisation of the guarantee in a specific case will be determined in, for example, ministerial orders or in specific contracts covering each case, the State guarantee in Section 4(4) will be considered – by default – to be a ‘simple kaution’. Therefore, the guaranteed party will have to show that Femern A/S is unable to pay its obligations as they fall due (as described above) before the guarantee can be mobilised. On this basis, the Commission considers that the mobilisation of this type of guarantee is contractually linked to specific conditions.

Although the Commission was not informed of the exact mobilisation conditions at the time of granting the aid, it considers that it can accept the guarantees in the light of Section 5.3 of the Guarantee Notice.

According to paragraph 45 of the IPCEI Communication Member States shall ensure the publication, on a comprehensive website, at national or regional level, of at least the following information: the text of the aid measure and its implementing provisions, or a link to it; the identity of the granting authority or authorities; the identity of the individual beneficiary, the form and amount of the aid, the date of granting, the type of undertaking (SME/large undertaking); the region in which the beneficiary is located (at NUTS level II); and the principal economic sector in which the beneficiary undertaking has its activities (at NACE group level). The Danish authorities have committed to comply with this requirement.

According to point 49 of the IPCEI Communication ‘the execution of the project must be subject to regular reporting’. In this respect the Danish authorities have committed to submitting annual reports including figures on the evolution of the debt. Therefore, the Commission concludes that this condition is complied with.