The Commission must open the formal investigation procedure when it has serious doubts about the compatibility of State aid. The aid that is necessary to induce a company to locate its investment in an assisted region is the amount that covers the difference in costs between the assisted region and the best alternative location. This amount is also proportional.
If a project costs 100 and is expected to generate revenue of 80, a company would need to receive aid of 20 in order to be willing to carry out the project. The aid intensity is 20% [20/100]. But the possibility of receiving aid may lead the company to expand the project, say, by 10 extra units of cost to make its product fancier and generate more revenue. It may then ask for another 20% of aid which would correspond to 2 over the extra costs of 10. Should the company be given this extra aid of 2? If the granting authority wants to keep aid to the minimum amount that can enable the project to go ahead, then the extra 2 units are not necessary. The counterfactual in this example is not the project costing 100, instead of 110, but no project at all. It follows that the amount of aid that is necessary is the minimum amount that makes the project viable; i.e. 20, not 22.
The necessity and proportionality of aid were two of the issues that were considered by the General Court in its ruling of 12 September 2017 in case T‑671/14, BMW v European Commission. BMW applied for partial annulment of Commission decision 2016/632 [SA.32009] concerning a large investment project in Leipzig.
Germany intended to grant an amount of EUR 45.3 million to BMW to invest in a new plant in Leipzig for the production of the fully electric and hybrid models i3 and i8, respectively. The Commission found that only EUR 17 million of that amount was compatible with the internal market. The remaining amount of EUR 28.3 million could not be granted.
BMW had considered several locations for the new plant. The best and least costly location was at its headquarters in Munich. The next best location was in Leipzig. But production in Leipzig would require extra EUR 17 million resulting from additional expenses in investment, planning and start-up costs, production costs, supply costs, logistics costs and shipment costs.
The project would contribute to regional development primarily through the creation of 800 jobs. The total cost of the project was EUR 392 million in nominal terms and EUR 368 million in real terms. At that time, the aid ceiling for large enterprises in Leipzig was 30%. The aid amount of EUR 45.3 million was equivalent to an intensity rate of 12.3% in relation to the discounted costs of EUR 368 million.
Although the aid intensity was far below the maximum ceiling of 30%, the Commission concluded that the cost difference of EUR 17 million proved the necessity of aid and that an equivalent aid amount would conform with the principle of proportionality.
Although the aid was to be granted under a GBER-based measure, it had to be individually notified because the gross amount of aid exceeded the threshold of EUR 22.5 million which corresponded to the adjusted amount of aid that could be granted to a project of EUR 100 million [adjusted amount = (50 million x 30%) + (50 million x 30% x 0.5) = 15 + 7.5 = 22.5].
Under the regional aid guidelines in force at the time, when the Commission assessed individually notified aid, it also took into account the market share of the aid recipient.
Do you know we also publish a journal on State aid?
The European State Aid Law Quarterly is available online and in print, and our subscribers benefit from a reduced price for our events.
The discretion of the Commission at the stage of the preliminary investigation
The General Court considered first the claims of BMW that the Commission had committed a number of procedural errors. The General Court began with a summary of the case law on the obligations of the Commission concerning the opening of the formal investigation procedure [paragraphs 36-40].
- The Commission can find aid to be compatible with the internal market at the preliminary examination stage only if it does not encounter any “serious difficulties”.
- In order to find aid to be compatible, the Commission must be able to have a firm view after the end of the initial investigation.
- The Commission may not avoid the formal investigation procedure by relying on other criteria such as third-party interests, or administrative or political convenience.
- When it encounters serious difficulties, the Commission must initiate the formal procedure. It has no discretion in this regard.
- The notion of serious difficulties is objective.
- If the assessment carried out by the Commission during the preliminary examination stage is insufficient or incomplete, this constitutes evidence of the existence of serious difficulties.
Then the General Court recalled the purpose of the Commission assessment. “(41) When the Commission assesses the compatibility of State aid with the internal market […], it must take into account the Union interest and may not refrain from assessing the impact of those measures on the relevant market or markets in the EEA as a whole. In such cases the Commission is bound not only to verify that the measures are such as to contribute effectively to the economic development of the regions concerned, but also to evaluate the impact of the aid on trade between Member States, and in particular to assess the sectorial repercussions they may have throughout the Union”.
Under the then regional aid guidelines [RAG], the Commission was committed to open the formal investigation procedure when the aid recipient had a market share that exceeded 25%, even when it thought that the aid was prima facie compatible with the internal market. BMW argued that since its market share did not exceed 25%, the Commission should not have opened the procedure. But the Court observed that “(44) it is not apparent from that rule that the initiation of the formal investigation procedure will be precluded where those thresholds have not been exceeded and that the Commission will be bound, in such a case, to make a direct finding that the aid is compatible with the internal market […] . In the latter scenario, the Commission will always have the option not to open the formal investigation procedure, but it cannot justify that decision by claiming that it is required by paragraph 68 of the Guidelines not to do so”. At any rate, the Commission had stated in its decision that it could not rule out that the threshold had not been exceeded.
Then the Court considered how the Commission measured the market share of BMW. In the end the decisive element was the finding that “(65) the mere fact that the thresholds laid down in the Guidelines have not been exceeded does not have the automatic consequence that the aid is compatible with the internal market. On the contrary, even where those thresholds have not been exceeded and the other conditions laid down in the Guidelines are met, the Commission may still ascertain whether the advantages in terms of regional development outweigh the disadvantages occasioned by the project in question in terms of distortion of competition”.
However, the General Court did not indicate at all what could prompt the Commission to have doubts about the compatibility of the aid if the “other conditions laid down in the Guidelines are met”. By definition, compliance with all of the requirements of the RAG entails compatibility. In this case the Commission had doubts on the necessity and proportionality of the aid.
Necessity of aid
BMW argued that the Commission was wrong to approve only EUR 17 million. Yet Germany initially notified only EUR 17 million. At some point later and after BMW had taken the decision to invest at the site in Leipzig, BMW decided to expand the planned production facilities and asked for more aid. As put by the Court, “(83) the principal question is whether the amount of the aid that could be held to provide an incentive and also proportionate was EUR 17 million, as argued by the Commission, or EUR 49 million, as argued by the applicant.”
The Court went on to find that “(85) regarding the incentive effect of the aid, […], the Commission examined, correctly, the question whether the aid actually was a factor in the recipient’s changing its plans to the point where it decided, as a result of the aid, to invest in the region targeted.” “(90) Without the aid in question, the investment costs for Munich would have been EUR 17 million lower than for Leipzig.”
BMW argued that the aid amount of EUR 49 million “(104) was divided into two parts. […], the first part comprised the amount compensating for the disadvantages of the site location, whilst the second was aimed at covering other disadvantages arising during the implementation of the project and the unquantified (and partly unquantifiable) disadvantages necessarily linked to the risks of an investment in a completely new car, budgeted as exactly as possible and manufactured not at the undertaking’s principal location but at another production site.”
The Court replied that it agreed “(105) with the Commission in respect of the second part and finds that it is not appropriate to use State aid to eliminate all investment risk associated with a particular project. On the contrary, such risks are the responsibility of the undertaking receiving the aid, especially since they were not quantified or in part not even quantifiable in advance. The only relevant criterion in the assessment of the appropriateness of an aid measure is whether the aid was necessary in order for the investment project to be carried out in the assisted region in question.”
It must be said, however, that this reasoning of the Court ignores that the purpose of state aid is often to reduce the risk of investment. The question is not whether aid reduces that risk, but by how much that risk should be reduced in order to induce the company to carry out the investment.
The Court could have also observed, but did not, that BMW’s argument was inconsistent. If there was risk linked to producing a car in a new location, that surely contributed to the locational disadvantage of the site in Leipzig. That monetary value of that risk should have been included in the first part of the aid.
BMW then contended that the Commission had conflated the principles of incentive effect and proportionality. The General Court agreed with the Commission that “(108) the net costs […] which are considered to be related to the regional handicaps result in a lower profitability of the investment’ and that, ‘[f]or that reason, calculations used for the analysis of the incentive effect, can also be used to evaluate whether the aid is proportionate’.”
The Court also noted that “(109) according to the case-law the Commission can declare aid compatible with Article 107(3) TFEU only if it can establish that the aid contributes to the attainment of one of the objectives specified, something which, under normal market conditions, the recipient undertakings would not achieve by their own actions. In other words, the Member States must not be permitted to make payments which, although they would improve the financial situation of the recipient undertaking, are not necessary for the attainment of the objectives specified in Article 107(3) TFEU”. “(110) According to that same case-law, it is not acceptable for aid to include arrangements, in particular as regards its amount, whose restrictive effects exceed what is necessary to enable the aid to attain the objectives permitted by the Treaty”. “(111) Similarly, the finding that an aid measure is not necessary can arise in particular from the fact that the aid project has already been started, or even completed, by the undertaking concerned prior to the application for aid being submitted to the competent authorities, which precludes the aid concerned from operating as an incentive”.
The General Court confirmed that “(116) the additional costs associated with the investment in Leipzig could be covered by a lower amount of State aid than what was sought in BMW’s application.” And that “(120) the Commission was correct in finding, […], that the principle of proportionality implies that aid in excess of the minimum necessary to trigger the decision to locate the investment in the assisted area must be considered superfluous, because it constituted an unconditional financial subsidy to the aid recipient and served no purpose that would be compatible with the State aid rules.”
BMW then argued that the incentive effect of aid had to be determined ex ante, while the proportionality of aid ought to be determined ex post. That is, the final amount of aid that a company needs should be set after it finalises its project and establishes the actual rather than the planned expenditure. This, of course, is like asking for carte blanche from the state and transferring responsibility to the state for overruns and mismanagement of investment projects to taxpayers. The Court dismissed this argument in its entirety. “(128) Not only must the ‘incentive’ effect be assessed even before any investment decision […]. Contrary to what the applicant suggests, it is impossible to separate those criteria [of necessity and proportionality] and assess one a priori and the other a posteriori.”
Distortion of competition
BMW contended that the Commission failed to establish that the additional aid that Germany was willing to grant would distort competition.
The General Court replied that “(146) since the ‘proportionality’ criterion had not been shown as being met for the tranche of aid exceeding EUR 17 million, the Commission did not err in law or make a manifest error in finding that this additional aid amount of EUR 28 257 273 would have negative, i.e. highly distortive effects on competition, as it might, in particular, discourage competitors to invest in similar products, thus contributing to the crowding out of private investment in the relevant market. The possibility cannot be ruled out that that part of the amount of the aid allowed only to rule out the financing of a ‘risk’ associated with the investment, thereby simply providing free money to the recipient undertaking. In those circumstances, the Commission could assume that there would be a negative effect consisting in a possible distortion of competition and a dissuasive effect for competing private investments, since the aid provides an undue reinforcement of the undertaking’s position on the market by allowing it to finance its needs above what is necessary to pursue the stated purposes, being to provide an incentive for investment in the assisted region. Therefore, the Commission could rightly find that the aid was incompatible with the single market without analysing potential additional positive effects.”
There is no doubt that the extra amount of aid would distort competition. But this case also raises another question in relation to regional aid that compensates for locational disadvantages of investment that would have been carried out at an alternative location. Since the investment was profitable at the alternative location in Munich, would the siting of the plant in Leipzig have any extra impact on competition at all? By its very design, the aid merely compensated [or even undercompensated] for the cost of moving the plant from one site to another. Given that the aid had no effect on the decision to carry out the investment, the aid itself had no adverse effect on competition in the car market. Competition would have been affected only if the siting of the plant in Leipzig had an extra impact that would not have materialised had the plant been located in Munich.
Amount of allowable aid
As noted in the introduction, Member States were obliged to notify individually amounts of aid when they exceeded the threshold of EUR 22.5 million. This was the adjusted amount of aid that corresponded to projects of EUR 100 million in areas eligible for aid at a rate of 30%. Had Germany kept the amount of aid below EUR 22.5 million, it would not have been necessary to notify it and perhaps no one would have known anything about the assistance to the plant in Leipzig. Perhaps Germany made a tactical error.
At any rate, BMW put forth an interesting argument. The Commission should have allowed aid of EUR 22.5 because this was the maximum amount that the GBER permitted without Commission assessment.
The General Court rejected this argument. Since the aid “(176) exceeded the threshold for triggering mandatory notification, the Commission was correct in finding that it had to assess it as individual aid and not as aid coming under the regulation on block exemptions or as authorised existing aid”. “(177) The Court holds that, in the present case, none of the provisions referred to by the applicant or the intervener substantiates their assertion that the Commission exceeded its powers in assessing, as a whole, the notified aid in terms of its compliance with the internal market and not solely in terms of the amount exceeding the notification threshold. It should be noted in particular in that regard that the fact that the Commission takes the decision to authorise aid notified to it entails significant, mandatory legal consequences, subject only to potential challenge before the EU Courts. It is a very different procedural scheme from the one applying to block exemptions, which involves an approach based on a presumption of compliance with the internal market. In circumstances such as those of the present case, a presumption of compliance of the aid up to a certain threshold cannot prevail over an individual assessment, as it is common ground that the latter came under the notification obligation, that is to say, it was judged prima facie to be of particular importance in terms of its potential effects on competition.”
On the basis of the above findings, the General Court dismissed the appeal.
 The full text of the judgment can be accessed at: