2024/2057

1.8.2024

COMMISSION DECISION (EU) 2024/2057

of 28 February 2023

on the State aid SA.48556 (2019/C) (ex 2018/N) – Hungary – LIP – Regional investment aid to Samsung SDI

(notified under document C(2023) 1478)

(Only the EN version is authentic)

(Text with EEA relevance)

THE EUROPEAN COMMISSION,

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

Having regard to the Agreement on the European Economic Area,

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

Whereas:

1.   

PROCEDURE

(1) By electronic notification registered on 16 May 2018, Hungary notified the Commission of its intention to grant regional investment aid in favour of Samsung SDI Magyarország Zrt. (‘Samsung SDI’ or ‘the beneficiary’).

(2) By letter of 14 October 2019 (‘the opening decision’), the Commission informed Hungary that it had decided to initiate the procedure laid down in Article 108(2) of the Treaty on the Functioning of the European Union (‘TFEU’) in relation to this regional investment aid measure (‘the aid measure’) and invited Hungary to submit its comments within 1 month.

(3) The opening decision was published in the

Official Journal of the European Union

 (2). In the same publication, the Commission called on interested parties to submit their comments.

(4) Hungary, Samsung SDI and South Korea submitted comments on the opening decision.

(5) By letter of 29 June 2021, the Commission informed Hungary that it had decided to extend the scope of the opening decision (‘the extension decision’) in respect of the aid measure (3). It invited Hungary to submit its comments within 1 month. The extension decision was published in the

Official Journal of the European Union

 (4).

(6) Hungary, Samsung SDI and South Korea submitted comments on the extension decision. Hungary provided comments on that decision by letter of 9 August 2021. Samsung SDI sent its preliminary comments on that decision on 28 January 2022 and the final comments on 14 March 2022. On 18 March 2022, South Korea sent its comments.

(7) On 21 March 2022, Hungary sent its comments on Samsung SDI’s submission and confirmed that it had no comments on the submission of South Korea.

(8) The Commission sent further requests for information to Hungary, the most recent in December 2022, to which Hungary replied in due time.

(9) By letter of 17 February 2023, Hungary agreed to have this Decision adopted and notified in the English language.

2.   

DETAILED DESCRIPTION OF THE AID MEASURE

2.1.   

Objective of the aid measure

(10) Hungary intends to promote regional development by providing regional aid for the extension of the capacity of an existing establishment of Samsung SDI, which produces battery cells for electric vehicles (‘EVs’) in Göd, in Pest County, Central Hungary. Göd is located in an area eligible for regional aid under Article 107(3), point (c) TFEU, with a standard regional aid ceiling of 35 % under the Hungarian regional aid map for the period 2017 to 2020 (5).

2.2.   

The beneficiary

(11) The recipient of the aid is Samsung SDI, a fully owned subsidiary of Samsung SDI Co. Ltd. (‘Samsung SDI Group’) headquartered in South Korea. Samsung SDI Group is a large undertaking with almost 19 000 employees globally and in 2016 it recorded a turnover of about EUR 4 billion.

(12) Samsung SDI Group is involved in the manufacturing of rechargeable batteries used in the IT industry, EVs, and energy storage systems (‘ESS’).

(13) In 2016, Samsung SDI set up in Göd an EV battery plant (‘Investment 1’) with a maximum capacity of [100 000–400 000] (6) EV battery cells per month, which is equivalent to an annual capacity of [0,1–0,5] GWh. This manufacturing facility started trial production in the second quarter of 2017 and mass production in May 2018. No State aid was granted for Investment 1.

(14) Hungary confirmed that the beneficiary is not an undertaking in difficulty within the meaning of the Guidelines on State aid for rescuing and restructuring non-financial undertakings in difficulty (7). Hungary committed to suspend the award and/or payment of the notified aid to the beneficiary if it has received earlier unlawful aid declared incompatible by a Commission decision until the beneficiary has repaid or paid into a separate account the unlawful and incompatible aid and the corresponding recovery interest.

2.3.   

The investment project and eligible costs

2.3.1.   

The investment project

(15) The project supported by the aid measure (‘Investment 2’) concerns a capacity extension of the EV battery cell production facility within an existing establishment owned by the beneficiary in Göd (i.e. within the establishment set up under Investment 1). With the envisaged extension, Samsung SDI’s production capacity in Hungary was planned to reach 6 million battery cells per month in […], which is equivalent to an annual capacity of [17-20] GWh (8). Samsung SDI intended to use [50-90] % of this capacity to serve the market of the European Economic Area (‘EEA’) and [10-40] % to serve […] (9).

(16) The aid measure supports investments into equipment and machinery relating to the production of battery cells, as well as into the necessary facilities for producing those battery cells. Samsung SDI intended to complement the investment into cell production by additional investments into battery modules and battery packs (10) production for which no aid would be requested. Investment 2 was to be implemented in two phases (11) (see table 1, recital (16) of the opening decision).

(17) Works on Investment 2 started on 1 December 2017, after the beneficiary had submitted the relevant aid application to Hungary on 13 September 2017. Investment 2 was expected to create 1 200 new direct jobs.

(18) Hungary confirmed that Samsung SDI completed the purchasing of equipment needed for Investment 2 by the end of 2021 and that Investment 2 reached full production capacity in January 2022, as planned.

(19) According to Hungary, Samsung SDI intended to develop and to carry out mass production of EV battery cells using significant changes in techniques, equipment and software.

(20) The aim of Samsung SDI’s new process innovations was to decrease the time and costs for manufacturing battery cells. The produced battery cells and some of the input materials used in the production process would be modified as a result of the implementation of the process innovations during both phases of Investment 2.

(21) Hungary further explained that, on the EV battery market, there exist three basic distinct cell structures (i.e. prismatic – which corresponds to those produced by Samsung SDI in Göd following Investment 1, pouch and cylindrical cells) and that their underlying production processes, as explained in recital (17) of the opening decision, are different.

(22) Hungary explained that the battery cell production process consists of three steps (recital (18) of the opening decision) and that individual process innovations to be supported by the aid measure covered all such three steps (recital (19) of the opening decision, which also lists the most relevant individual innovation processes according to Hungary).

(23) Hungary also explained that some of these novel elements would be protected by patents, but essentially Samsung SDI Group would protect most of the new production process through blackboxing (12). The ‘stacking’ of electrodes, to be introduced by Investment 2, was still under development and testing in […] at the time of notification, but was planned to be applied in mass production in […].

(24) Hungary considered that the above-mentioned innovative processes and equipment would be introduced in the EV battery and accumulator production sector for the first time worldwide (see recital (21) of the opening decision).

(25) In conclusion, Hungary considered that the new production process qualified as a ‘new process innovation’ within the meaning of paragraph 15 of the Regional Aid Guidelines applicable at the time of notification (13) (‘the RAG’). Hungary also explained that, following the investment at stake, Samsung SDI’s market shares are expected to increase from [7-10] % in 2019 to [20-25] % in 2023 at European level and from [7-10] % in 2019 to [10-13,5] % in 2023 at global level.

2.3.2.   

Eligible investment costs

(26) The notified total eligible investment costs for Investment 2 amount to HUF 380,628 billion, rounded (EUR 1,2 billion (14), rounded) in nominal value, which is HUF 376,540 billion, rounded (EUR 1,187 billion, rounded) in present value (15). The eligible costs comprise costs of buildings, machinery, and equipment (see tables 2 and 3 in recital (23) of the opening decision).

(27) Hungary confirmed that only new assets were accepted as eligible expenditure. The eligible costs for Investment 2 do not include any costs of immaterial assets or costs of leasing (see recital (155) of the opening decision).

24.   

Form of aid, aid granting authority and national legal basis

(28) The aid measure is a non-refundable cash grant and constitutes individually notifiable ad hoc aid, which falls outside the scope of exemption of an aid scheme which was exempted from the obligation of notification under Commission Regulation (EU) No 651/2014 (16).

(29) The national legal basis for notifying the aid measure is Governmental Decree 210/2014 on the use of the earmarked scheme for investment promotion. The aid granting authority is the Hungarian Ministry of Foreign Affairs and Trade.

2.5.   

Aid amount

(30) The notified aid amounts to HUF 34,891 billion, rounded (EUR 110 million, rounded) in nominal value and HUF 34,304 billion, rounded (EUR 108 million, rounded) in present value. Hungary awarded the aid, subject only to the Commission’s approval, on 27 December 2021.

2.6.   

Aid intensity and cumulation with other investment aid

(31) On the basis of the notified total eligible costs of HUF 380,628 billion, rounded (EUR 1 200,000 million, rounded) in nominal value, the discounted value of the total eligible expenditure is HUF 376,540 billion, rounded (EUR 1 187,111 million, rounded). Taking those figures as a basis, and applying the 35 % standard regional aid ceiling for large undertakings that were applicable in Göd under the regional aid map for the period 2017-2021, as well as the scaling-down rules (17) of paragraph 20(c), and paragraph 86 of the RAG, the Commission calculated that the maximum total regional aid permissible for Investment 2 is HUF 49,360 billion, rounded (EUR 155,616 million, rounded) in discounted value, equivalent to a maximum aid intensity of 13,11 %.

(32) The envisaged aid of HUF 34,304 billion, rounded (EUR 108 million, rounded) in present value, for proposed eligible expenditure of HUF 376,540 billion, rounded (EUR 1 187 million, rounded) in present value, corresponds to an aid intensity of 9,11 %, calculated on the basis of the notified total eligible costs.

(33) In its notification, Hungary declared that the financial support under the aid measure would not be combined with any other financial support that would be disbursed for the same eligible costs from any other local, regional, national or European Union source.

(34) Hungary confirms that neither the approved maximum aid amount in present value nor the approved aid intensity will be exceeded if the amount of eligible expenditure deviates from the estimated amount.

2.7.   

Own contribution

(35) Hungary confirmed that the aid beneficiary would provide a financial contribution of at least 25 % of the eligible costs, through its own resources or by external financing, in a form that does not contain any public financial support.

2.8.   

Maintenance of the assisted activity

(36) The aid measure is awarded under the condition that the beneficiary will maintain the investment in the assisted region for a minimum period of five years after its completion.

2.9.   

Grounds for initiating the formal investigation procedure

(37) In the opening decision, the Commission took the view that the planned investment qualified as a new process innovation and was therefore eligible for aid under the RAG. However, in both the opening and extension decisions, the Commission raised doubts in particular regarding the substantive incentive effect of the aid and the credibility of the alternative investment scenario in China, that Hungary presented (see section 3.4.1.4. of the opening decision and recital (3) of the extension decision).

(38) First, the Commission could not exclude the possibility that a combination of strategic factors such as the quickly expanding European market, the proximity to European customers, the risk of forced transfer of technology and the hostile political and economic climate in China vis-à-vis South Korean undertakings, would not have constituted overriding strategic considerations that would have led the company to locate its investment in Hungary in any event, even in the absence of aid.

(39) Second, the Commission took the preliminary view that the proposed regional aid was not crucial for a positive location decision in favour of Hungary. This was because the Commission had doubts as to the existence and, if any, the extent of the claimed gap of EUR 173 million in the net present value (‘NPV’) of Investment 2 compared to the counterfactual investment in China. The Commission noted that this gap was largely based on significant differences in investment costs between the two alternative investment locations, justified by Hungary and the beneficiary on the basis of a local sourcing policy (18). However, the Commission precisely expressed doubts regarding such a local sourcing policy, which it found was prima facie not realistic or credible. The Commission also questioned the existence of alleged public support in China for the location decision by the beneficiary, and the relevance of any such support, as internal documents from Samsung SDI Group on the location decision pointed to the aid from Hungary as a decisive factor for the location choice, while they did not point to public support from the Chinese authorities.

(40) As it had expressed doubts as to the counterfactual scenario and the NPV comparison, the Commission also raised doubts regarding the proportionality of the aid (see section 3.4.1.5. of the opening decision). In addition, the Commission raised doubts concerning the contribution of the aid to regional development (see section 3.4.1.1. of the opening decision) and the appropriateness of the form of the aid (see section 3.4.1.3. of the opening decision) (see also recital (4) of the extension decision).

(41) Finally, the Commission identified the possibility of undue negative effects resulting from the aid. In particular, it could not exclude the possibility that the investment was at least indirectly causal for a relocation of Samsung SDI Group’s battery pack production activities from Austria to Hungary, which would constitute a manifest negative effect on trade within the meaning of paragraph 122 of the RAG (see section 3.4.2.4. of the opening decision and recital (4) of the extension decision). In the extension decision, the Commission also considered that the aid might have had a counter-cohesion effect by attracting investments away from a less-developed Polish region, which would be eligible for regional aid pursuant to Article 107(3), point (a), TFEU (see recital (72) of that decision).

(42) For the above reasons, in both the opening and extension decisions, the Commission raised doubts as to the compatibility of the aid measure with the internal market.

3.   

COMMENTS FROM HUNGARY AND SAMSUNG SDI

(43) In support of its claim that Investment 2 may benefit from regional aid, Hungary invoked a ‘scenario 2’ situation in the meaning of paragraph 61 of the RAG, arguing that without the aid measure, the investment project would have been implemented in Xi’an (China), where the Samsung SDI Group co-owns, in a joint venture (19), an EV battery cell production facility and sufficient free land to host an extension of the plant (20).

(44) Hungary argued that Investment 2 qualified as a new process innovation and is therefore eligible for aid under the RAG. Hungary noted that the Commission agreed with this conclusion in its opening decision.

3.1.   

Contribution to regional development and territorial cohesion

(45) Hungary explained that the investment would contribute to the regional development of the Central-Hungary region (see recital (34) of the opening decision). In 2016, Pest county had a gross domestic product (‘GDP’) per capita of 54 % of the EU-28 average.

(46) In addition, Hungary listed a number of positive effects that the investment would generate for the development of the region, including the creation of 1 200 direct jobs (see recital (35) of the opening decision).

3.2.   

Incentive effect

(47) Hungary reiterated that Samsung SDI submitted the application for aid on 13 September 2017, while the works on Investment 2 started on 1 December 2017. Hungary took note that the Commission agreed that the formal incentive effect test was complied with in the opening decision.

(48) Hungary and Samsung SDI also argued that the aid complies with the substantive incentive effect test. In their replies to the opening decision, however, both Hungary and Samsung SDI withdrew from certain initial statements contained in the notification and provided new arguments and evidence that partially contradicted earlier statements.

3.2.1.   

Counterfactual

(49) The investment project in Göd and the counterfactual scenario in Xi’an presented in the notification documents included similar investments in terms of buildings, machinery, and equipment. In particular, Hungary confirmed that the type, capacity, and quantity of machinery and equipment would be identical in both investment scenarios. Despite this fact, the total capital expenditure related to the investment would be significantly higher in Hungary than in China (EUR 215,3 million higher in present value). While the costs of buildings are slightly lower in Hungary compared to those in China (since Samsung SDI could use an existing building for its investment project in Göd), the costs of machinery and equipment would be significantly higher in Hungary compared to China. According to information first obtained from Hungary on 16 October 2018, this difference is due to Samsung SDI Group’s local sourcing policy, according to which machinery and equipment would be fully bought in the EEA (if the investment were to take place in Hungary) or in China (if the investment were to take place in China).

(50) Hungary and Samsung SDI contested the Commission’s preliminary view (as expressed in the opening decision) questioning the very existence of such a full local sourcing policy. In particular, they explained that, for Investment 1, Samsung SDI had decided, in line with a restriction included in its local sourcing policy, to rely on experienced South Korean suppliers of machinery and equipment rather than on EEA suppliers, in order to minimise risks (21). Such a particular circumstance could not put into question an otherwise overall local sourcing policy.

(51) However, Hungary explained that, although Samsung SDI’s local sourcing counterfactual analysis in this case was based on a premise that reflected the Samsung SDI Group’s settled policy, Samsung SDI’s investment Committee’s

ex-ante

assessment ‘may have been (…) too presumptive regarding its local sourcing capabilities’ (22).

(52) Hungary confirmed that up until early 2019 – when about [35-50] % of Samsung SDI’s Investment 2 was ordered with suppliers – Samsung SDI had not used EEA suppliers (see recital (44) of the opening decision). Instead, the vast majority of the equipment and machinery acquired for the aid measure originated, according to Hungary, from South Korea. Hungary considered, however, that the Commission cannot rely on

ex-post

facts on actual sourcing to question the credibility of business assumptions at the time of the location decision.

(53) Hungary explained that the Samsung SDI Group had to resort to South Korean suppliers for the Hungarian site – and disregard its original plans presented in the notification documents – because its search for appropriate local suppliers in the EEA did not yield the expected results.

(54) According to Hungary, Samsung SDI’s search for EEA suppliers started as early as 2015 and was, as of May 2019, on-going. Except for bilateral contacts with five EEA producers of equipment in 2015, 2016 and 2017 (which ended fruitfully for two of them) (23), the search culminated essentially in a conference organised jointly by the Samsung SDI Group and the […] (24), which was held on 24 February 2017 in […] (i.e. eight months before Samsung SDI’s investment decision). According to Hungary, out of 24 conference participants (and five other companies, which were contacted bilaterally by the Samsung SDI Group after the conference), a majority (23) of potential EEA suppliers were dismissed on various grounds (25) before the final decision to invest in Hungary was made on 27 November 2017. After the investment decision, discussions continued with only six potential EEA suppliers, but which could only provide equipment for a small portion of Samsung SDI’s needs for Investment 2 in Hungary.

(55) According to Hungary, in China, the Samsung SDI Group had sufficient experience (26) and contacts with local suppliers and therefore it did not need – in the planning phase of the notified investment – to take specific actions to gather information on their technical capabilities and pricing of standard equipment.

(56) Hungary also explained that, for the equipment and machinery used to implement the new process innovation, the Samsung SDI Group did conduct technical reviews in China from 8 until 12 January 2017. The conclusions of these technical reviews were that, while the Chinese producers had made significant technological improvements compared to the situation as it existed 2-3 years before and had in the meantime achieved price competitiveness (as they were [5-30] % cheaper than the South Korean suppliers), they did not meet the Samsung SDI Group’s quality requirements.

(57) More generally, Hungary contested the Commission’s doubts regarding the reality of the Chinese counterfactual.

(58) First, Hungary denied the existence of a regionalised investment strategy (27), the result of which would have been that the investment project would have in any event taken place in Hungary. Hungary considers that such doubts are actually based on the Commission’s misinterpretation of the minutes of an internal meeting of Samsung SDI Group that was held on 21 September 2017, regarding the ‘Review for new production investment to EU&[…]’ (28), attended by several Samsung SDI Group Managers and its Chief Financial Officer. Hungary notes that, if such a strategy had existed, it would have meant that a similar investment would have been carried out in […], which was not the case.

(59) Second, Hungary and Samsung SDI explained that Samsung SDI had the means and the experience to protect adequately its intellectual property (‘IP’) rights in China. Samsung SDI explained that, at the time of the location decision, it already had significant experience in setting up production plants in China (the Tianjin factory dates back from 1996) and dealing with the protection of its IP rights. Specific measures, as illustrated by evidence produced by Samsung SDI, had already been taken to protect such rights, such as black boxing (see recital (23)), confidentiality agreements for employees and for Chinese shareholders and strict policies on access to the factory (such as security scans, control of electronic devices), which proved adequate in practice to prevent any risk of IP breaches. Samsung SDI provided additional information proving that the Chinese administration announced in October 2017 (i.e. before the investment decision) at the 19

th

national congress of the communist party legislation to reinforce the protection of IP rights. Hungary and Samsung SDI claim that, as a result, Samsung SDI Group would have had no reason to exclude a counterfactual investment in China on account of IP concerns.

(60) Third, Hungary and Samsung SDI contested the Commission’s preliminary doubts regarding an alleged hostile environment in China vis-à-vis South Korean companies. Samsung SDI confirmed that the Samsung SDI Group had good business relations with Chinese authorities, as confirmed by the aid offers it received in relation to this investment or to other investments. Samsung SDI explained that the alleged hostile environment concerned sales on the Chinese domestic market, which would have been irrelevant in the present case, as the production in this investment would have been dedicated to export markets. The doubts expressed by the Commission in that regard would therefore be unfounded. Samsung SDI also provided an internal presentation prepared by the Automotive and ESS Business Division’s planning team of Samsung SDI and presented at the meeting of 21 September 2017 (see recital (58)) (‘the internal report of 21 September 2017’); according to Samsung SDI, this document (a site-by-site qualitative comparison) shows that Samsung SDI Group gave a […] rating to its relationship with the Chinese authorities.

(61) Fourth, to disprove the general assumption that Investment 1 had been pursued in Hungary with an intention to ramp it up later, Hungary forwarded to the Commission an Excel table provided by Samsung SDI indicating that, in 2020, the Xi’an production facility only had two production lines (corresponding to the two original production lines established under Investment 1).

3.2.2.   

Recalculated local sourcing levels in both Xi’an (China) and Göd (Hungary) counterfactuals

3.2.2.1.   

Xi’an (China)

(62) Hungary and Samsung SDI challenged the Commission’s assumption that Samsung SDI Group would not have sourced any equipment locally in China if it had chosen Xi’an as the investment location. They explained that the […] and confidentiality issues that prima facie could affect the reliability of Chinese suppliers (29) could have been tackled by involving the Chinese authorities (30) (which would, however, only be possible if the investment were to take place in China), by providing technical support to the Chinese suppliers and dispatching SDI engineers from the South Korean Samsung SDI Headquarters to work closely with them (31) and by using specific contractual arrangements and joint development agreements with vendors (32). As regards confidentiality issues, Hungary also argued that, except for specific machines, there are no confidentiality concerns regarding suppliers of machinery and equipment used for the various steps in the battery production process. In its comments on the opening decision, Hungary also claimed that, based on Samsung SDI’s

ex-post

recalculations using data available at the time of the investment decision, a proportion of local sourcing of equipment and machinery equal to [31-35] % (33) of the eligible investment expenditure would have been realistic in Xi’an, China, at the time of the location decision.

(63) The baseline local sourcing rate of [31-35] % in China was based on the hypothesis that Samsung SDI Group would have sourced:

a.

about [15-20] % of the value of the equipment necessary for the counterfactual investment in Xi’an from Chinese suppliers from which Samsung SDI Group had already purchased similar equipment in China for its production plant in Tianjin. Samsung SDI in that regard also noted that, contrary to the Commission’s views expressed in the extension decision (recital (41)), the share of local sourcing achieved in its Tianjin plant ([25-30] % in 2015-2016, [15-20] % in 2018-2019) had not been reached after 20 years of operation, considering that the plant started manufacturing battery cells only in 2008 (and even 2016 for cylindrical cells);

b.

about [10-15] % of the value of the equipment necessary for the counterfactual investment in Xi’an from Chinese suppliers without a transaction record with Samsung SDI in China, but that had sold to other Chinese customers, including EV battery manufacturers, equipment based on the same technology (34) as used in the counterfactual investment project;

c.

about [3,5-5] % of the value of the equipment necessary for the counterfactual investment in Xi’an from Chinese suppliers without a transaction record with Samsung SDI in China, but that had sold to other Chinese customers, including EV battery manufacturers, equipment used in the same production processes (35) as those used in the counterfactual investment project.

(64) In its comments on the opening decision, Hungary presented two further arguments in support of its claim that a significant level of local sourcing in China was realistic at the time of the investment decision in November 2017.

(65) First, it contended that during 2015-2016 and 2018-2019 Samsung SDI Group had achieved a local sourcing ratio of machinery and equipment between [10-15] % and [25-30] % in its Tianjin plant (on average [25-30] % for 2015-2016 and [15-20] % in 2018-2019). Hungary considered that the Tianjin investment constituted a relevant benchmark for the notified investment project because the respective manufacturing lines have ‘similar characteristics’ and ‘rely on analogous machinery and equipment’. Second, Hungary presented evidence on the level of technological development in the Chinese equipment and machinery market in 2017 (for its summary, see recital (11) of the extension decision).

(66) Hungary and Samsung SDI also recalled that the search for local suppliers is a dynamic process that can evolve over time, thereby potentially leading to higher shares of local sourcing than the one secured at the time of the investment decision.

(67) Hungary and Samsung SDI also considered that, in the absence of local sourcing, the beneficiary would use its South Korean suppliers. In such case, additional cost elements would need to be taken into account. These include shipping costs of the machinery and equipment to the plant in China (estimated at [2-3] % of the costs of the respective machinery and equipment), a [18-22] % mark-up to take into account the involvement of Samsung SDI Group headquarters in South Korea (‘Samsung SDI HQ’) in the procurement process (36) and customs duties of 1,1 % to be paid for such machinery and equipment to be imported into China from South Korea.

3.2.2.2.   

Göd (Hungary)

(68) In its comments on the opening decision, Hungary considered that the Commission had ‘no ground to question that local sourcing was a credible ex-ante assumption for Samsung SDI at the time it decided to invest in Göd’. Hungary claimed that it was ‘irrelevant how the sourcing of equipment and machinery for Hungary played out ex post, compared to what Samsung had initially envisaged in line with its settled local sourcing policy’.

(69) Hungary explained that, contrary to the Commission’s preliminary views, Samsung SDI Group could not have sourced equipment from cheaper Chinese suppliers, because of […] (in particular […]). In any event, Hungary noted that since the assumed costs of sourcing from South Korean suppliers were virtually identical to those of sourcing in the EEA (barring the shipping costs, estimated at [3-4] % of the costs of the respective machinery and equipment), a local sourcing percentage lower than 100 % in Hungary would have a very limited impact on the

ex-ante

NPV outcome of the investment in Hungary.

(70) According to Hungary, sourcing the machinery and equipment through Samsung SDI HQ acting as an intermediary, could replace local sourcing procurement efforts by Samsung SDI whenever the latter proved unfeasible. In case of central sourcing through Samsung SDI HQ, shipping costs and a mark-up of [18-22] % would apply (while no customs duties would be payable).

(71) Following the subsequent exchange of correspondence between the Commission and Hungary, Hungary submitted a recalculated NPV gap in January 2023, in which the share of the sourcing of machinery and equipment from outside the EEA for the Göd plant at the end of 2021 was indicated at [94-99] %, while the share of sourcing from within the EEA was set at [1-1,5] %.

3.2.2.3.   

The [18-22] % mark-up on non-locally sourced equipment

(72) Hungary and Samsung SDI argue that, should the Commission reject Samsung SDI’s initial assumption of a full local sourcing, based on the Samsung SDI Group’s policy, the NPV calculations should include a [18-22] % mark-up (37). In line with actual experience of Samsung SDI, such a mark-up would be justified by the fact that the machinery and equipment would first be purchased from the South Korean producers by Samsung SDI HQ and then the respective equipment would be sold to Samsung SDI in Hungary or China, after applying a [18-22] % mark-up to the underlying cost basis (which would consist of the acquisition price and all other costs incurred by Samsung SDI HQ in relation to the purchase, delivery and setting up of the machinery and equipment at the local investment site up until mass production can be run on it). This [18-22] % mark-up corresponds to the nominal difference between Samsung SDI HQ’s purchasing prices and the resale prices based on invoices it charges to its local affiliates. It would be a reasonable and appropriate level for such intra-groups transactions, which not only include the purchase of the equipment but also additional services provided by Samsung SDI HQ. Hungary and Samsung SDI argue that such [18-22] % mark-up would constitute a market-conform price, as evidenced both by studies before the location decision in the case at hand and by studies produced to the Commission during the present investigation (38).

3.2.3.   

Public support for the proposed alternative location in Xi’an, China

(73) Although Hungary had initially indicated to the Commission that the beneficiary ‘didn’t get any subsidy offer’ from the Chinese government during the decision-making process (see recital (13) of the extension decision), Hungary subsequently distanced itself from this statement.

(74) Hungary and Samsung SDI subsequently claimed that a direct investment grant covering [15-20] % of the investment costs would have been available in China. They also claimed that if the investment had been carried out in China, it would probably have benefitted from a corporate income tax (CIT) rate of 15 % rather than the standard rate of 25 %.

(75) With regard to the grant, Hungary and Samsung SDI produced an unsigned, unstamped offer dated 23 February 2017 by the Chinese local government, which contained two options (see recital (16) of the extension decision). Hungary and Samsung SDI provided additional evidence, notably documenting the course of the negotiations between Samsung SDI Group and the Chinese local government from January to July 2017 (39) and an updated aid offer dated 23 May 2018. In light of this additional evidence, Hungary and Samsung SDI claim that the negotiations concerning a [15-20] % grant were serious and realistic; only the possibility of obtaining an even higher grant reaching [25-30] % of the costs of the investment was still unclear.

(76) With regard to a possible lower level of corporate income tax of 15 %, Hungary explained that Samsung SDI Group’s initial assumption of a CIT rate of 25 % in China was based on a rather conservative estimate. Hungary claimed that it would have been legitimate for Samsung SDI Group to take a much lower tax rate of 15 % in China as basis for its NPV calculations (see recitals (25) and (26) of the extension decision) (40).

(77) Hungary and the beneficiary explained that the internal report of 21 September 2017 shows that Samsung SDI Group was aware of the potential Chinese subsidy offer. According to Samsung SDI, the fact that such subsidies were not factored into the viability gap calculations submitted to SDI’s Investment Committee and to Samsung SDI’s Chief Executive Officer in the successive decision-making steps leading to the location and investment decision is not relevant: it is standard Samsung SDI Group policy to take into account aid only after the NPV calculations, as also evidenced by the fact that potential aid from Hungary was not taken into account either at the stage of the NPV calculations.

3.2.4.   

Introduction by Samsung SDI of a ‘discrete’ and a ‘probabilistic’ approach for a viability gap calculation

(78) Although Hungary disagreed with the Commission’s doubts with regard to NPV comparisons, it explained that it was prepared to compare a few other plausible viability scenarios, with the purpose of illustrating how the viability gap between Hungary and China would remain significant in these scenarios and that, as a result, the State aid offered to Samsung SDI by Hungary would maintain its incentive effect. More specifically, Samsung SDI recalculated the NPV gap following the opening decision, envisaging two approaches: a ‘discrete’ approach, entailing four different scenarios, and a ‘probabilistic’ approach, leading to three possible scenarios.

(79) Under the discrete approach, Samsung SDI calculated two NPV gaps without any local sourcing in China or Hungary, but taking into account either a local Chinese grant of [15-20] % of the investment, or a 15 % CIT rate. In both scenarios, Samsung SDI argues that the viability gap would remain comparable to, or in excess of, the original gap of EUR 173 million. A third NPV gap recalculated based on a [31-35] % rate of local sourcing for the investment in China (and on no local sourcing for the investment in Hungary), no subsidy and a 25 % CIT rate for the investment in China, would lead to an NPV gap that is lower than the original but on par with the amount of State aid proposed by Hungary. Finally, a fourth NPV gap recalculated based on a [25-30] % rate of local sourcing for the investment in China (and no local sourcing in Hungary), no subsidy and a 25 % CIT rate for the investment in China, would lead to an NPV gap that is lower than both the original NPV gap and the amount of State aid proposed by Hungary.

(80) Samsung SDI argued that the discrete methodology above did not accurately reflect the decision-making process leading to the investment. Instead, it considered that a probabilistic approach would be a more realistic reflection that takes into account all factors that were relevant to Samsung SDI when it decided on the location of the investment.

(81) The three ‘probabilistic scenarios’ proposed by Samsung SDI following the opening decision combine the three different factors (i.e. local sourcing of [25-30] % for the investment in China, a [15-20] % grant from the Chinese authorities, and a 15 % CIT rate) to which different probability coefficients are applied to reflect the uncertainty relating to the materialisation of these factors. The proposed probability coefficients are 100 % for the factor relating to the conservative [25-30] % local sourcing in China for all three scenarios, and alternatively 25 % or 0 % for the other two factors (in two scenarios) or both 25 % (in the third scenario) (41). Samsung explains that all three resulting recalculated NPV gaps are within the range of the original NPV gap, and all three are higher than the amount of State aid proposed by Hungary. In view of the above, Hungary argued that the incentive effect of the notified amount of aid would be confirmed.

(82) Hungary argued that the Commission should take into account such

ex-post

recalculation of NPV gaps, performed on the basis of historic data even though it was not considered by the recipient of aid at the moment it chose the investment location.

(83) In their comments on the extension decision, Hungary and Samsung SDI argued that an internal company document contemporaneous to the decision-making process provided by Samsung SDI supports the claim that the beneficiary applied an inherently probabilistic approach at the

ex-ante

stage of the investment assessment. In that regard, they referred, more precisely, to the internal report of 21 September 2017.

3.3.   

Proportionality

(84) Hungary explained that the aid measure sought to compensate – partially – the claimed cost disadvantage of EUR 173 million in order to attract the investment to an assisted region. Hungary confirmed that the Samsung SDI Group was prepared to bear the remaining cost disadvantage of EUR 65 million, as it considered that it would benefit from certain qualitative and strategic advantages of the Hungarian site.

(85) Hungary and Samsung SDI repeated that, even if the NPV gap would need to be recalculated (see section 3.2.4.), the aid amount, to the extent that it would remain within the limit of such a revised NPV gap, would remain proportionate.

3.4.   

Necessity and appropriateness

(86) Hungary explained that the need for aid in the area concerned, namely Central-Hungary, is confirmed by its status as an assisted area under Article 107(3), point (c), TFEU. The Central-Hungary region benefitted in the past from numerous non-aid and horizontal measures, which alone proved insufficient to address the regional handicaps of the area. Pest County has remained particularly under-developed compared to the rest of the Central-Hungary region.

(87) Hungary considered that, according to the Commission’s practice, a direct grant constitutes an appropriate aid instrument to achieve the desired objective, which is to attract the investment to Göd.

3.5.   

Undue negative effects on competition and trade

(88) First, Hungary noted that the same customers would be served irrespective of whether the EV battery cells are produced in China or in Hungary. Equal volumes would be produced and sold for the same prices, to supply the same geographical markets (i.e. the EEA and […]).

(89) Second, Hungary did not contest that the location search for the investment project had not only included Samsung SDI Group’s existing three battery cell plants in China, South Korea, and Hungary (as claimed in the notification and during the preliminary examination phase), but that also greenfield investment sites had been considered in China (Wuxi), Poland (Środa), Slovakia (Sereď), Hungary (Tatabánya) and the Czechia (Most Joseph). Hungary and Samsung SDI explained however that these other possibilities, which would have been greenfield investments, were excluded at an early stage in the decision-making process. Samsung SDI recalled that its Group’s capacity at that time was short of its sales target in Europe and […]; therefore, Samsung SDI Business Division’s planning team proposed an investment plan to increase that capacity as soon as possible. Greenfield investments thus would not have met such an objective. The planning team retained only two brownfield locations: Göd and Xi’an. The Commission’s interpretation of the internal documents relating to the meeting of 21 September 2017 (42) would therefore be incorrect.

(90) Third, with regard to a possible relocation within the EEA, Hungary first argued that the production activities in Austria (battery packs) and the aided activity in Hungary (battery cells) were materially and technologically different (the Commission itself had accepted that the production process for the new battery cells in Göd constituted a ‘new process innovation’ (see recitals (89) to (101) of the opening decision). Furthermore, the reduction of activities in Samsung SDI Group’s Austrian plant fitted into a re-orientation of that plant’s mission and did not constitute a partial closure within the meaning of the RAG. In any event, there would be no causal link between the reduction of activities in Austria and the aid in Hungary for the expansion of the Göd site.

4.   

COMMENTS FROM THE MINISTRY OF TRADE, INDUSTRY AND ENERGY OF SOUTH KOREA ON THE EXTENSION DECISION

(91) In its comments, South Korea acknowledged the efforts of the Commission to resolve the issues reported in its on-going formal investigation procedure. It also indicated that further investments of Samsung SDI in Europe might be expected and a positive decision taken by the Commission in this case would serve as a catalyst for Samsung SDI’s further investments.

(92) Hungary confirmed that it had no observations on the comments submitted by South Korea.

5.   

ASSESSMENT

5.1.   

Existence of aid

(93) For the reasons set out in the opening decision (recitals (71) to (76) thereof), the Commission considered that the aid measure constituted State aid within the meaning of Article 107(1) TFEU, as it was imputable to the State and granted through State resources, was selective, procured an economic advantage to Samsung SDI, was likely to affect trade between Member States, and distorted or threatened to distort competition. That assessment remains valid, hence the measure at hand constitutes State aid.

5.2.   

Lawfulness of the aid measure

(94) The Commission recalled in the opening decision (recitals (77) to (80) thereof) that if the regional investment aid amount to be granted exceeds the notification threshold laid down in Article 4(1), point (a), of Regulation (EU) No 651/2014, an aid is not covered by the exemption provided by that Regulation and has thus to be notified individually to the Commission. In this case, in the Göd area, the applicable threshold is EUR 26,25 million. Therefore, the envisaged aid measure of EUR 108 million could not be exempted from notification.

(95) In addition, according to paragraph 15 of the RAG, regional aid to investments of large undertakings in areas eligible for regional aid under Article 107(3), point (c) TFEU is considered compatible only if it is granted for initial investments that create new economic activities in these areas, or for the diversification of existing establishments into new products or new process innovations. Aid for these activities is individually notifiable unless it supports initial investments that create new economic activities in these areas. The obligation to individually notify aid for the diversification of existing establishments into new products or new process innovations applies independently of the aid amount envisaged.

(96) In the opening decision, the Commission concluded (in recital (79) of that decision) that, by notifying the awarding of such aid subject to Commission approval, Hungary respected their obligations under Article 108(3) TFEU (see also recital (30) above). That assessment remains valid.

5.3.   

Compatibility of the aid measure

(97) By way of derogation from the general prohibition of State aid laid down in Article 107(1) TFEU, aid may be declared compatible if it can benefit from one of the derogations set out in the TFEU. As the objective of the aid measure is to promote regional development in an area designated as eligible for regional aid in accordance with Article 107(3), point (c), TFEU, the Commission will assess the compatibility of the aid measure in light of that provision, as interpreted in the RAG.

(98) Hungary awarded the aid, subject only to the Commission’s approval, on 27 December 2021 (recital (30)). Therefore, the aid is assessed under the RAG applicable from 1 July 2014 to 31 December 2021 (43).

(99) In line with the provisions of the RAG, the Commission first assessed in the opening decision (see recital (83) thereof) whether Investment 2, which is to be undertaken by a large undertaking in an area eligible for regional aid under Article 107(3), point (c) TFEU, and which does not concern an initial investment in favour of new economic activity within the meaning of paragraph 20(i) of the RAG, could be found eligible for regional aid, as Hungary claimed that its production process was based on a new process innovation. The Commission also found that Samsung SDI was not an undertaking in difficulty (recital (8) of the opening decision) (and has no contrary information to date). Finally, the Commission notes that Hungary committed to suspend the award and/or payment of the notified aid to the beneficiary if it has received earlier unlawful aid declared incompatible by a Commission decision until the beneficiary has repaid or paid into a separate account the unlawful and incompatible aid and the corresponding recovery interest. Therefore, paragraphs 18 and 19 of the RAG are complied with in the case at hand.

(100) The Commission accepted that the production process for the new battery cells in Göd qualified as ‘new process innovation’, and therefore consisted of an initial investment within the meaning of RAG (recital (37) above and recitals (89) to (101) of the opening decision). The Commission’s formal investigation did not identify any elements capable of calling that assessment into question. Consequently, the Commission’s conclusion remains valid for the purposes of this Decision.

(101) The Commission is therefore called upon to assess whether the aid measure meets the criteria laid down in Article 107(3), point (c), TFEU, i.e. whether it constitutes ‘aid to facilitate the development (…) of certain economic areas, where such aid does not adversely affect trading conditions to an extent contrary to the common interest’. That assessment is carried out by reference to the criteria laid down in the RAG.

5.3.1.   

Objective of promoting regional development

5.3.1.1.   

Real and sustained contribution to regional development

(102) The RAG require the Member State to prove in concrete terms the real and sustained contribution of the aided investment to the regional development of the target region. To that purpose, Section 3.2.2. of the RAG lists a number of indicators that Member States may use in order to demonstrate the regional contribution of individual investment aid.

(103) In the opening decision, the Commission confirmed that the Göd area was eligible for regional aid pursuant to Article 107(3), point (c), TFEU.

(104) To prove the real and sustained contribution, the Member State also had to show that the viability of the aid measure was demonstrated by a financial contribution of the aid beneficiary of at least 25 % of the eligible costs (44), provided through its own resources or by external financing, in a form that is free of any public financial support. In addition, the investment (the aided assets) had to be maintained in the area concerned for a minimum period of five years after completion of the investment (45). The Commission noted that those conditions were fulfilled (see recitals (31) and (32) of the opening decision).

(105) In the opening decision (see recital (105) thereof), the Commission raised doubts as to whether the investment would have all the beneficial effects described in the notification documents, as the local sourcing policy mentioned there did not seem to materialise, at least not for the acquisition of machinery and equipment used in the investment project.

(106) However, having assessed the overall evidence in the file, the Commission takes note that Hungary invoked several positive regional effects of the investment. Among others (see recital (35) of the opening decision), the aid measure was expected to create 1 200 new direct jobs. Due to its location, the investment is also likely to attract commuters from neighbouring regions that are assisted areas under Article 107(3), point (a) TFEU. The number of indirect jobs created as a result of the investment was expected to be around 200 in Hungary, and the construction works as such were expected to attract more than [30-60] different construction companies and more than [800-1 100] workers (recital (46)). The beneficiary was also planning to offer comprehensive training to improve both the general and specific skills of its workforce and to cooperate with Hungarian higher education institutes (therefore promoting knowledge spill-over and technology transfers). Hungary also referred to follow-on investments and research, development and innovation activity envisaged in the region. The Commission finds that there is no element in the file to challenge the reality of these expected benefits at the time of the conditional aid grant. With regard to its initial doubts in relation to local sourcing, the Commission notes that those doubts could not in any event affect the reality of the contribution of Investment 2 to the regional development of the area (as established above) but could only affect its scope.

(107) As a result, the Commission considers that it is sufficiently proven that the investment would provide a positive contribution to the development of the region.

5.3.1.2.   

Incentive effect

(108) According to Section 3.5. of the RAG, regional aid can only be found compatible with the internal market if it has an incentive effect. An incentive effect is present when the aid changes the behaviour of an undertaking in a way that it engages in additional activity contributing to the development of an area, in which it would not have engaged without the aid or would only have engaged in such activity in a restricted or different manner or in another location. The aid must not subsidise the costs of an activity that an undertaking would have incurred in any event and must not compensate for the normal business risk of an economic activity.

(109) As explained in recital (116) of the opening decision, the formal incentive effect requirement laid down in paragraphs 64, 65 and 69 of the RAG, namely that the works can start only after the application form for aid has been submitted, has been respected in case of Investment 2.

(110) However, as there may be many valid reasons for a company to locate its investment in a certain region, even without any aid being granted, the RAG also require the Commission to verify in detail that the aid is necessary to provide a substantive incentive effect for the investment. In this context the Member State is required to specify which scenario described in paragraph 61 of the RAG applies (investment decision – scenario 1; or location decision – scenario 2).

(111) Hungary notified the aid measure as falling within scenario 2 of paragraph 61 of the RAG. According to paragraph 61 of the RAG, scenario 2 is present where the aid gives an incentive to opt to locate a planned investment in the relevant area rather than elsewhere because it compensates for the net disadvantages and costs linked to a location in the area concerned. As set out in Section 3.5.2. of the RAG, in a scenario 2 case, the Member State must provide clear evidence that the aid has a real impact on the choice of the investment location. It thus places the burden of proof regarding the existence of an incentive effect on the Member State. To that end, the Member State must provide a comprehensive description of the counterfactual scenario in which no aid would be granted to the beneficiary by any public authority in the EEA (including in the chosen location). Paragraph 71 of the RAG indicates that for scenario 2 – which Hungary refers to in the present case – the Member State could prove the incentive effect of the aid by providing company documents showing that a comparison has been made between the costs and benefits of locating the investment in the assisted area concerned and those in the alternative area(s). Such documents should be contemporaneous with the decision-making process concerning the investment or the choice of its location. For that purpose, the Member State is invited to rely on official board documents, risk assessments, financial reports, internal business plans, expert opinions and other studies and documents that elaborate on various investment scenarios.

(112) The Commission must verify whether the counterfactual scenario presented by Hungary is realistic and credible.

(113) According to paragraph 68 of the RAG, a counterfactual scenario is credible if it is genuine and relates to the decision-making factors prevalent at the time of the decision by the beneficiary regarding the investment (46).

(114) In the present case, Hungary raised a number of claims and provided documentary evidence to support its position that Investment 2 satisfied the requirements of the RAG as a project falling with in scenario 2 of paragraph 61 of the RAG. It argued a net present value (NPV) viability gap of EUR 173 million compared to an alternative investment location in Xi’an, China. This gap resulted essentially from a significant difference in planned investment costs of the aid measure in the two alternative locations. Whilst the beneficiary would invest EUR 1,2 billion in nominal value (EUR 982 million in present value (47)) into buildings, equipment, and machinery in Hungary, the equivalent costs in China amounted to only EUR […] million in nominal value (EUR 766,7 million in present value). The difference of EU 215,3 million (in present value) was due exclusively to the cost of machinery and equipment, which was presented in the notification to be almost [35-50] % higher in Hungary compared to China.

(115) In the opening decision, the Commission noted that the EUR 215,3 million difference in investment costs was partially offset by the net advantages of the Hungarian investment (EUR 42,3 million), resulting in a final net viability gap of EUR 173 million between the locations in China and Hungary, to the disadvantage of the location in Hungary. The Commission also noted that Hungary claimed that the notified aid of EUR 108 million in present value partly compensated for the net disadvantage of EUR 173 million linked to the decision to locate the investment in Hungary as compared to China.

(116) As is apparent from the extension decision (notably recitals (5) and (6)), in reply to the opening decision, Hungary and Samsung SDI distanced themselves from certain statements made during the preliminary examination phase and presented new claims and evidence concerning the assumptions made by the beneficiary in the counterfactual scenario, which at least partially contradicted earlier statements and claims.

(117) In the opening decision and in the extension decision, the Commission expressed doubts concerning the credibility of the counterfactual scenario (an investment in China), as overriding strategic considerations would have led the company to locate its investment in Hungary in any event. It also expressed doubts on the NPV gap comparisons, which did not appear realistic and credible in light of the evidence available.

(118) In the following section, the Commission assesses each of the points on which it had expressed doubts in relation to the NPV gap and the location decision. These points can be grouped under three headings: the existence of overriding strategic considerations nullifying from the outset the credibility of the Chinese alternative location; the difference of machinery and equipment costs in the two alternative locations (and in particular the levels of expected local sourcing); and the alleged financial support by the Chinese authorities in favour of the Chinese location.

5.3.1.2.1.   The existence of overriding strategic considerations nullifying from the outset the credibility of the Chinese alternative location

(119) In the opening decision, the Commission raised doubts as to the credibility of the Chinese counterfactual, since it could not exclude that the investment would in any event have taken place in Hungary for a series of circumstances. Those circumstances were: (i) the hostile economic environment in China towards South Korean EV battery producers, (ii) the risk of forced transfer of technology, (iii) the unlikeliness of Samsung SDI having carried out Investment 1 in Göd without considering a ramp-up investment at a later stage, (iv) the advantages of a European location, such as the quickly expanding European market and the proximity to European customers (coupled with a regionalised investment strategy that Samsung SDI Group might, in the Commission’s preliminary view, have pursued).

(120) With regard to the hostile economic environment in China towards South Korean EV battery producers (see recital (139) of the opening decision), the Commission takes note that, as evidenced by internal documents contemporaneous to the decision-making process, and as argued in the notification, the planned alternative investment in Xi’an would have served as a basis for exports (it was not expected to generate sales on the Chinese market (48)). As a consequence, the Commission accepts that the fact that China had implemented, in 2016, measures that hindered the sales of South Korean EV battery producers on the domestic Chinese market does not appear to have created a hostile economic environment for the investment at stake, which was not intended to serve the Chinese market. In addition, Samsung SDI has submitted contemporaneous documents and publicly available information showing ongoing good business relationships between the Chinese authorities and Samsung SDI Group (in particular the subsidy offer from the Chinese authorities of 23 February 2017) and other South Korean EV battery producers. The Commission finally notes that no document in the case file supports the conclusion that Samsung SDI Group had regarded such hostile environment as a decisive factor at the time of the location decision. On the contrary, the internal report of 21 September 2017 shows that, shortly before the investment decision, Samsung SDI Group gave a […] rating to its relationship with the Chinese authorities – the same score as was given to the Hungarian authorities, and higher than the rate attributed to other European authorities. Therefore, the Commission considers that the investment project was not threatened by a potentially hostile economic environment in China and that, in any event, the business environment in China was regarded neither as negative nor as an overriding factor by the beneficiary at the time of the location decision.

(121) As regards the potential risk of forced transfer of technology in China (see recital (138) of the opening decision), Hungary and Samsung SDI provided additional information proving that the beneficiary had planned a number of measures in the counterfactual investment scenario in order to ensure that its IP rights would be sufficiently protected in China, based on its experience of operating plants there (recital (59)). Samsung SDI also explained that the Chinese administration announced in October 2017 (i.e. before the investment decision) legislation to reinforce the protection of IP (recital (59)). In light of these observations, the Commission considers that its initial doubts on a possible forced transfer of IP rights was not such as to exclude a counterfactual location in China.

(122) A further argument (see recital (140) of the opening decision) that raised questions as to the general credibility of the Chinese counterfactual investment scenario was that, in the Commission’s preliminary view, it was unlikely that Samsung SDI would have started a small cell production facility without aid, like the one in Göd (i.e. Investment 1), if it had not considered ramping up the investment at a later stage. This would imply that the intention to invest in and ramp up capacity in Göd already existed at the time when Investment 1 was pursued in 2016. However an internal Samsung SDI document showed that its original Investment 1 in Xi’an had not been ramped up: it has therefore been possible for a Samsung SDI Group production site not to be developed beyond the first investment stage if a location decision in favour of the competing site were made. Furthermore, the existence of an initial investment in Göd does not mean that there was no choice between Göd and Xi’an, as both sites had an existing investment that could be ramped up. More generally, the Commission acknowledges that even the existence of an implicit or explicit intention to ramp up a production site (such as the fact that a company bought a piece of land that is larger than that required for a first factory) does not necessarily mean that the new investment will be carried out in any event. In that regard, the Commission notes that there is nothing in the case file expressly showing that Samsung SDI had plans to ramp up the investment in Göd from the outset.

(123) Furthermore, the Commission generally considered in the opening decision (recital (143) of the opening decision) that a number of advantages of a European production site (for example, the quickly expanding European market and the proximity to European customers) may have constituted overriding strategic considerations that would have led Samsung SDI Group to locate its investment in Hungary in any event. This doubt was further developed in the extension decision (recital (57) of the extension decision) where the Commission considered that Samsung SDI Group may have followed a regionalised investment strategy leading it to invest in any event in production sites in Europe. However, even though one internal document of Samsung SDI Group (49) suggests that a diversification of productions sites in each continent ‘could be helpful’, the cautious tone of that document and the context of that statement (diversification was merely one of several parameters in the qualitative site-by-site comparison, alongside other considerations such as logistics flow by train between the possible Xi’an site and Europe) show that this factor did not constitute an overriding strategic consideration for Samsung SDI Group. Other contemporaneous internal documents of Samsung SDI Group merely state that such qualitative factors ‘should not [be] overlook[ed]’ (50). There is nothing else in the case file supporting the doubt raised in that regard and the existence of an overriding regionalised investment strategy at the time of the investment decision. In particular, none of the evidence concerning the fourth and fifth steps in the beneficiary’s decision-making process (recitals (68) and (69) of the opening decision) – which both followed the meeting of 21 September 2017 – point to any guidance in favour of regional diversification in the final location decision process. Furthermore, as noted by Hungary, if the proximity to customers were a key driver in Samsung SDI Group’s investment strategy at the time of the location decision, it would have started in parallel an EV battery production plant in […]. Internal company documents however show that, to bridge the supply gap identified in the […] market, only the two alternatives in Hungary and China were considered. The lack of such an overriding regionalised investment strategy is further revealed by the internal Samsung SDI documents provided by Hungary, according to which both the initial investment in Xi’an and the envisaged ramp up of the Xi’an site would have served foreign markets (Europe and […], not the domestic Chinese market) (51).

(124) In light of the foregoing considerations, the Commission considers that the strategic considerations identified above were not decisive, and/or not taken into account (as such or as being of an overriding significance) in the decision on the location of the investment project. Accordingly, Samsung SDI Group could in general realistically envisage such an investment in China.

5.3.1.2.2.   Differences in costs of machinery and local sourcing levels in Xi’an and Göd

5.3.1.2.2.1   Lack of credibility of the initial assumption of full local sourcing for equipment and machinery in both locations

(125) In the notification, Hungary justified the significant difference in investment costs for equipment and machinery by invoking Samsung SDI Group’s local sourcing policy, which in principle requires the group to buy the equipment and machinery and the other inputs necessary for the investment from local sources (see recitals (42) and (130) of the opening decision). In the opening decision, the Commission considered that the claim that Samsung SDI would source locally 100 % of its machinery and equipment in both scenarios considered (i.e. both if the investment would occur in Xi’an or in Göd) was not credible and realistic.

(126) As regards local sourcing of equipment for the Hungarian site (i.e. sourcing from EEA suppliers), the Commission noted (recital (47) of the opening decision) the limited steps undertaken by Samsung SDI to identify local suppliers (bilateral contacts with five EEA suppliers in 2015, 2016 and 2017, conference in February 2017) and highlighted the limited outcome of these steps, as 23 out of 29 EEA suppliers considered were dismissed for various grounds before the investment decision (while discussions were ongoing with only six suppliers at the time of the investment decision, who could merely provide equipment for a small portion of Samsung SDI’s investment in Hungary). The Commission also noted (recital (48) of the opening decision) that none of the equipment for Samsung SDI’s first investment (greenfield Investment 1) in Göd had been sourced from EEA suppliers. Samsung SDI, in line with a restriction in its local sourcing policy, did not try to use local sourcing at that time and thus did not develop contacts with local suppliers (see recital (50) and footnote 20 above). Hungary merely explained that the search for EEA suppliers is a continuous effort and that, ultimately, some equipment could be sourced from the identified suppliers. In the light of the above and of the absence of any indication, at the time of the investment decision, that EEA-based suppliers had been seriously considered for any significant part of the deliveries of machinery and equipment for the purposes of Investment 2, the Commission maintains its conclusion that Samsung SDI could not credibly assume that it could source 100 % of the equipment for the Hungarian plant locally. The Commission notes that, for Investment 2, Samsung SDI has in fact sourced almost none of the equipment from EEA suppliers (52).

(127) As regards local sourcing for the alternative investment in Xi’an, the Commission first notes that, at the time of the location decision, Samsung SDI Group knew that it only achieved a limited level of local sourcing for its plant in Tianjin manufacturing battery cells for mobile devices. According to evidence (submitted by Hungary on 3 November 2020, recital (63)(a) that could have been available at the time of the location decision, an average local sourcing of [25-30] % was achieved in 2015-2016 for the Tianjin plant (and that was the highest average local sourcing share recorded for that plant up to that point in time).

(128) Furthermore, the Commission notes that, at the time of the location decision, Samsung SDI Group considered that using Chinese suppliers could in general raise issues in terms of […] and confidentiality (recital (62)). Even though Hungary and Samsung SDI subsequently argued that these issues could be, or may have been, tackled by several means (see below sub-section 5.3.1.2.2.2.2.), the Commission considers that internal company documents would not have focussed on these issues if they could be solved in a timely matter, at no additional cost and in all cases, for all suppliers and machines.

(129) The Commission thus maintains that Samsung SDI Group could not credibly assume that it could source locally 100 % of the equipment for the Chinese plant. This is confirmed by the

ex-post

assessment of the level of local sourcing that could have been realistically assumed at the time of the investment decision: the highest local sourcing ratios submitted by Samsung SDI and Hungary do not exceed [40-45] % (53).

(130) On the basis of the above, the Commission concludes that the 100 % local sourcing policy both in Göd and in Xi’an was not credible and realistic at the time of the location decision. Therefore, the full local sourcing could not reasonably be part of Samsung SDI NPV comparisons. Consequently, the claimed investment costs differential between Hungary and China of EUR 237,5 million (in present value) – where the costs of machinery and equipment in Hungary were about [40-45] % higher than in China – cannot be justified.

(131) Considering that the Commission does not deny that the beneficiary took account of local sourcing in its NPV assessment, and that local sourcing is a relevant factor to be assessed, but only concludes that the assumption of a full local sourcing was not credible and unrealistic, it will assess the extent to which local sourcing could have been regarded as credible and realistic at the time of the location decision.

5.3.1.2.2.2.   Commission’s assessment of a realistic level of costs of machinery and equipment and of the share of local sourcing

5.3.1.2.2.2.1.   General considerations

Use of supplies from South Korea in the absence of local sourcing

(132) Hungary and Samsung SDI argued in their comments on the opening decision that, in light of Samsung SDI Group’s local sourcing policy, and in the absence of adequate local suppliers, Samsung SDI uses its usual trusted South Korean suppliers as a second best option. Even though the Commission acknowledged that point in recital (131) of the opening decision, it subsequently (see recital (45) of the extension decision) raised a doubt concerning why, if partial sourcing from Chinese suppliers were to be considered realistic, the machinery and equipment needs in Hungary should not be sourced from much cheaper Chinese suppliers (instead from South Korea), which would have lowered machinery and equipment costs in the Hungarian scenario and reduced the NPV gap between the two locations.

(133) Samsung SDI however explained in its comments on the extension decision that the timing issues (delivery delay) could only have been addressed by the involvement of the Chinese authorities (recitals (62) and (69)). Samsung SDI submitted that such involvement could only have been expected if the investment itself took place in China. The Commission accepts that it would not have been credible from Samsung SDI’s perspective to source machinery and equipment for the investment in Hungary from Chinese suppliers, as this would have exposed Samsung SDI to the […] issues that would have undone the prima facie short term price advantage. The Commission accordingly considers that, contrary to its initial assertions, to the extent that local sourcing was not realistic in either investment scenario (i.e. Hungary and China), Samsung SDI would likely have used its trusted South Korean suppliers.

Additional costs in the case of non-local sourcing: shipping costs and customs duties

(134) If, instead of being sourced locally, machinery and equipment for the investment was sourced from South Korean suppliers, the Commission considers, as also accepted by Hungary and Samsung SDI, that additional costs (shipping costs, customs duties, if any) would arise. The Commission notes that Hungary estimated the shipping costs (see recital (69) above) from South Korea to Hungary at [3-4] %, while the shipping costs from South Korea to China were estimated at [2-3] % of the respective machinery and equipment costs (recital (67) above). Additionally, the Commission also accepts that (as mentioned in recital (67) above) customs duties of 1,1 % would also apply on non-locally sourced equipment in the scenario in which the investment would take place in China, while no customs duties would be charged if the investment took place in Hungary (recital (70) above).

Additional costs in case of non-local sourcing: mark-up by Samsung SDI HQ

(135) As described in recitals (44) and (45) of the opening decision, the machinery and equipment used up to March 2019 for the investment in Göd (54) had not been purchased directly by the beneficiary from the South Korean suppliers. Instead, Samsung SDI HQ had acted as an intermediary, by buying the machinery and equipment in South Korea and reselling it to Samsung SDI in Hungary, after applying a [18-22] % increase (mark-up) to the underlying costs basis (55). Samsung SDI explained that this involvement of Samsung SDI HQ is in line with its local sourcing policy and allows in particular Samsung SDI HQ to directly communicate with the South Korean suppliers and encourage them to act swiftly if a problem arises in one of Samsung SDI Group’s production sites. According to Samsung SDI, Samsung SDI HQ would also have been involved in the case of non-local sourcing if the investment were to take place in China.

(136) The Commission expressed doubts on whether these intragroup transactions resulted in a market-conform price and whether they did not artificially inflate the eligible costs of the scenario in which the investment were to be located in Hungary, and thus the NPV gap (recital (136) of the opening decision and recital (47) of the extension decision).

(137) The Commission notes first that Hungary submitted with its comments on the opening decision an internal Samsung SDI document of 10 January 2017 showing that the [18-22] % mark-up was based on a study by an external consultant and the results of benchmarking other companies providing equipment. Hungary also clarified in its comments on the extension decision that the apparently lower values of [13-17] % (median value) and [14-18] % (for Samsung SDI), both mentioned by a report from Deloitte in December 2018 (on the basis of which the Commission had raised doubts), did not take into account additional costs of [3-4] % incurred by Samsung SDI HQ when purchasing equipment, which explains the difference with the nominal mark-up of [18-22] % used in the recalculation.

(138) Finally, Samsung SDI provided a further study from Copenhagen Economics, dated 13 April 2020 (56), which assessed whether the remuneration of intragroup transactions within the Samsung SDI Group was at arm’s length, using the OECD Guidelines methodology (57). This study concluded that [18-22] % falls in the interquartile range of comparable companies’ net cost plus margin (i.e. in between a median value of 16,8 % and a 3

rd

quartile value of 23,5 %) and thus confirmed that the mark-up applied by Samsung SDI HQ to intragroup transactions was at arm’s length.

(139) In light of the additional arguments and evidence provided by Hungary and Samsung SDI, the Commission considers that, to the extent that machinery and equipment for the investment would not have been sourced locally (with respect to both locations considered), it is realistic and credible to assume that such non-local sourcing would have been carried out through Samsung SDI HQ and would have involved a market-conform [18-22] % mark-up on the costs of such machinery and equipment.

Additional costs in case of local sourcing: procurement costs

(140) Because the [18-22] % mark-up covers in particular procurement costs supported by Samsung SDI HQ when it sources machinery and equipment on behalf of a subsidiary (in Hungary or China), the Commission considers that it would have been realistic to include corresponding procurement costs in case the equipment is sourced locally (directly by the subsidiary in Hungary or China). In its submission dated 1 July 2022, Samsung SDI specified that the local sourcing of its equipment in China and Hungary would indeed result in additional costs (corresponding in particular to testing and validating). Samsung SDI explained that these costs would (logically) be higher for novel equipment than for not-novel equipment (58). As, under the revised, more credible, assumptions (see below), not all equipment (and in particular none of the novel equipment) would have been sourced locally, the additional procurement costs to be applied are relatively limited. According to calculations provided by Samsung SDI, these costs would represent [3,5-5] % of the machinery cost. The Commission has no factual reason to raise doubt on this level of costs.

5.3.1.2.2.2.2.   Credible level of local sourcing of machinery and equipment in the counterfactual scenario

(141) As set out in recital (130) above, although the Commission concludes that the assumption that for either of the investment locations, the entirety of machinery and equipment would be locally sourced was not credible and realistic, the Commission agrees that the share of local sourcing is a relevant factor that Samsung SDI took into account at the time of the location decision. Therefore, the Commission needs to assess what share of local sourcing could be found credible and realistic for each of the location sites considered at the time of the location decision, based on the evidence available at that time. In the opening decision, the Commission preliminarily considered that a more realistic assumption – for both sites – would have been that Samsung SDI would source the respective equipment and machinery from its experienced and trusted existing suppliers in South Korea.

(142) As explained above (see sub-section 5.3.1.2.2.1.), the lack of credibility of (full) local sourcing in China was in particular based on the fact that using Chinese suppliers could in general raise issues in terms of (i) […], (ii) […] and (iii) confidentiality. The Commission acknowledges that Hungary and Samsung SDI convincingly explained (recital (62)) that Samsung SDI planned to tackle and could have tackled these issues, at least to some extent, by, respectively, (i) involving the Chinese authorities (which would, however, only be possible if the investment were to take place in China), (ii) dispatching SDI engineers from the South Korean Samsung SDI Headquarters to provide support to the Chinese suppliers; and (iii) using specific contractual arrangements and joint development agreements with vendors where it was needed (i.e. only for specific machines). The evidence in the file thus shows that, contrary to the preliminary finding made in recital (131) of the opening decision, some degree of local sourcing can be considered credible and realistic with regard to the counterfactual scenario.

(143) Furthermore, after the Commission had adopted the opening decision, Hungary and Samsung SDI submitted information showing that, as regards Samsung SDI Group’s Chinese plant located in Tianjin, substantial average levels of local sourcing had been achieved ([25-30] % in 2015-2016, [15-19] % in 2018-2019). In the extension decision (recital (41)), the Commission expressed a number of doubts, in particular as to the comparability of the two situations (the Tianjin plant and the planned plant in Xi’an). Hungary and Samsung SDI however provided explanations showing that the local sourcing achieved for the Tianjin plant was a relevant piece of information. They first recalled that the innovations in the process for the planned investment (alternative location in Xi’an) concern, as a general rule, the way in which the machinery and equipment is used and not the machinery and equipment themselves. As a consequence, the Commission acknowledges that the fact that the production process and the final products in the Tianjin plant would be less sophisticated is indeed not relevant to discard the experience drawn from the purchasing of machinery and equipment for the Tianjin plant (59). In addition, the fact that the figures for 2018-2019 could not have been known at the time of the location decision merely leads to the outcome that only the higher percentage of [25-30] % for 2015-2016 (concerning past data that could have been known at the time of the location decision) would be relevant. Hungary also clarified that these local sourcing figures were not achieved after nearly 20 years of operation of the plant (which may have meant that the Xi’an plant could only achieve such substantial levels of local sourcing after a similar time frame) as the plant only started to manufacture battery cells in 2008 (and even 2016 for cylindrical cells) (recital (63)(a)). Finally, Hungary confirmed that the local sourcing percentages submitted only included machinery and equipment and did not include in particular spare parts. On that basis, the doubts expressed in the extension decision can be lifted: the Commission acknowledges that the degree of local sourcing achieved in the Tianjin plant in 2015-2016 confirms that, at the time of the location decision, Samsung SDI Group could expect to achieve a significant level of local sourcing for the counterfactual investment in China.

(144) As regards more precisely the counterfactual investment in Xi’an, the level of local sourcing that could have been, in Samsung SDI’s view, credibly achieved was first calculated by Samsung SDI in its comments on the opening decision. For that purpose, Samsung SDI listed the equipment bought for the Göd plant up to that point (the investment was ongoing at the time) and assessed whether, for the counterfactual investment in Xi’an, each piece of equipment could have been sourced from a Chinese supplier. That calculation showed a ratio of [31-35] % between the value of the equipment that could have been sourced locally in China and the total value of the equipment bought at the time when the calculation was made (60). Samsung SDI also showed that the level of technological development in the Chinese machinery and equipment increased over time (see recital (65)), making it more likely to reach a higher level of local sourcing than in the past.

(145) In the extension decision (recitals (41) to (44)), the Commission expressed a number of doubts in that regard. In addition to the general doubts as to whether it would have been credible to rely on Chinese suppliers ([…] and confidentiality risks, already mentioned in the opening decision and addressed in recital (142)), the Commission considered notably that the sample used by Samsung SDI was not representative, in particular as it concerned mostly the first phase of the investment in Göd which was less innovative. Hungary however stressed that the sample represented more than [75-80] % of the total value of the machinery (including machines used in the second phase of the investment) (61) and that, in any event, both phases of the investment include new process innovation elements (62). Hungary also clarified that the stacking technology for phase 2 that was still under development at the time of the preliminary examination only accounts for a very small fraction (less than 5 %) of the total equipment costs and that, even for some machines used in the stacking process, Chinese suppliers could be envisaged.

(146) The Commission also noted that the data concerning the capabilities of certain Chinese suppliers, described in recitals (9)(b) and (10) of the extension decision, relied on evidence assembled

ex post

(recital (42) of the extension decision). However, as clarified by Hungary, most of the data concerned, even if compiled

ex-post

, was available to Samsung SDI Group at the time of the investment decision. While part of that data postdates the investment decision, Hungary clarified that only one of the six Chinese suppliers mentioned in recitals (9)(b) and (10) of the extension decision had never been contacted before the investment decision. As regards the other suppliers, meetings with those suppliers had started before the investment decision (even if they continued thereafter). Nevertheless, the Commission considers that at the time of the investment decision, Samsung SDI could reasonably expect that, if the Chinese location would have been chosen, it would have had further contacts or intensified contacts or made new contacts with potential Chinese suppliers of equipment for the Xi’an plant. This is particularly true since, as highlighted by Samsung SDI (see recital (11) of the extension decision), Chinese suppliers in the battery chain had experienced a positive evolution both in quantitative and qualitative terms, so that the possibility to identify adequate suppliers increased – more than linearly – as time passed. As a result, the Commission considers that the evidence in the case file sufficiently shows that Samsung SDI could have credibly considered a [31-35] % local sourcing rate if the investment were to take place in China.

(147) Finally, because the Commission considered in the extension decision that Samsung SDI Group’s experience concerning the local sourcing of machinery and equipment for its plant in Tianjin may not be comparable to the Xi’an investment, the Commission also implicitly raised doubts as to the specific component of the [31-35] % local sourcing rate corresponding to the local sourcing that could have been achieved for Xi’an using the actual suppliers of Samsung SDI Group for the Tianjin plant (component summarised in recital (9)(a) of the extension decision). Given that Hungary and Samsung SDI have duly explained that the two situations were comparable (recital (143)), the Commission no longer has reasons to harbour doubts on the reality of that component.

(148) In the light of the foregoing, the Commission concludes that a level of [31-35] % of local sourcing of machinery and equipment for the purposes of the counterfactual investment in China can be considered credible and realistic.

5.3.1.2.2.2.3.   Credible level of local sourcing of machinery and equipment for Investment 2 in Göd

(149) With regard to local sourcing (sourcing from EEA suppliers) of machinery and equipment in case of Investment 2 in Hungary, the Commission already concluded in recital (126) that an assumption of a 100 % share of local sourcing was not credible and was unrealistic. The Commission notes that, at the time of the location decision, Samsung SDI had limited information on what a realistic level of local sourcing of machinery and equipment for Investment 2 would be. In particular, Hungary confirmed (recital (50) above) that for Investment 1 Samsung SDI had not applied its local sourcing policy, as Investment 1 was its first battery cells investment within the EEA, for which it had decided to rely on experienced South Korean suppliers of machinery and equipment to minimise risks. Investment 1, therefore, could not provide a proper benchmark or inform Samsung SDI’s analysis concerning Investment 2.

(150) The Commission notes that Samsung SDI started a search for local suppliers in 2015. By the time of the location decision, Samsung SDI had discarded, for a variety of reasons, 23 out of 29 potential EEA-based suppliers that it had identified and was only continuing discussions with the remaining suppliers, who could only provide equipment for a small portion of Investment 2 (recitals (54), (126) and (146). Hence, the Commission concludes that only a marginal share of EEA-sourced machinery and equipment, if any, could be considered realistic and credible in the case of Investment 2.

(151) This finding is additionally corroborated by the fact that, for its plant in Göd, Samsung SDI had actually sourced only negligible amount of machinery and equipment in the EEA (approximately [1-1,5] % of the machinery and equipment, see recital (71)).

(152) The Commission therefore concludes that, at the time of the location decision, Samsung SDI should have considered that it would only marginally source locally if Investment 2 was to take place in Hungary. The Commission considers that it would have been realistic and credible to consider that only a marginal share of the machines and equipment used in Investment 2 would have been sourced locally from EEA suppliers. In this regard, since Hungary has itself pointed to a [1-1,5] % local sourcing share in its reply on recalculated NPV gap of January 2023 (recital (71)), which in addition corresponds to the actual level of local sourcing of machinery and equipment achieved for Investment 2, the Commission will apply that amount to quantify the abovementioned marginal local sourcing share.

5.3.1.2.2.3.   Summary

(153) On the basis of the local sourcing ratios determined in the two preceding sub-sections (5.3.1.2.2.2.2. and 5.3.1.2.2.2.3.), the Commission considers that it can calculate the credible equipment costs in the two locations in light of the general considerations laid down in the first sub-section (5.3.1.2.2.2.1.). This is done by applying, for the fraction of the machinery and equipment sourced locally, the additional local sourcing procurement costs and, for the fraction of the machinery and equipment sourced from South Korean suppliers through Samsung SDI HQ, the shipping costs, customs duties and the mark-up applied by Samsung SDI HQ.

Table 1

Local sourcing – Hungary’s notification and the Commission’s assessment

(154) The Commission therefore reflected the above changes in the NPV calculations provided by Hungary and Samsung SDI, keeping all other elements of the NPV on which the Commission had expressed no doubts unchanged. Notably, the Commission maintains the cost assumptions submitted by Hungary in the initial NPV calculations as regards the costs of sourcing machinery and equipment in South Korea (as explained in recitals (43) of the opening decision and as recorded in recital (69) above), i.e. it accepts that, apart from the additional costs summarised in Table 1, the costs of machinery and equipment were almost on par between South Korea and Hungary). As a consequence of the changes referred to in the preceding recitals, the NPV gap between the two locations is reduced from EUR 173 million to EUR 89,6 million (in present values).

5.3.1.2.3.   Public support for the alternative location in China

5.3.1.2.3.1.   Potential investment grant in China

(155) In its notification, Hungary confirmed that no grant offer had been formally made by the Chinese authorities for an investment in Xi’an (recital (13) of the extension decision). However, in reply to the opening decision, Hungary distanced itself from the statements contained in the notification, which they declared contained major errors, misunderstandings and erroneous information (recital (14) of the extension decision), and argued that Samsung SDI Group had held discussions with local Chinese authorities and that the possibility that the granting of such aid would materialise was by no means unrealistic.

(156) In the extension decision, the Commission expressed doubts both on the existence of an aid offer from the local Chinese authorities for the project at stake (see recitals 48-50 and 59-61 of that decision) and on whether such aid had been taken into account at the time of the location decision (see recitals (51) and (57) of that decision).

(157) In reply to the extension decision, Hungary and Samsung SDI objected to the two concerns raised by the Commission.

(158) As mentioned in recital (113) above, a counterfactual scenario is credible if it is genuine and relates to the decision-making factors prevalent at the time of the location decision. In the present case, independently of whether the Chinese local authorities had indeed provided a subsidy offer for the investment in Xi’an, the Commission considers that evidence contemporaneous with the decision-making does not support Hungary’s claim that such an offer was factored into that decision-making process at the time when the beneficiary adopted its decision on the location of the investment.

(159) Indeed, as noted in recital (51) of the extension decision, the Commission found that the Chinese aid grant offer was only mentioned in the internal report of 21 September 2017. The Commission noted, however, that the grant offer from the Chinese authorities had never actually been included in the NPV comparisons between the alternative investment scenarios. Hungary and Samsung SDI, however, objected that in the decision-making process of the beneficiary, the level of aid that may be received (in this case, either from China or from Hungary) is not included in the NPV calculations, but is taken into account in a subsequent step.

(160) The Commission notes that neither the potential subsidy from China nor aid from Hungary were included in the NPV calculations in the first comparison step of the decision-making process of the beneficiary. This is apparent from the internal report of 21 September 2017: both slides 8 and 9, relating to a quantified overview of, respectively, the Xi’an location and the Göd location, indicate that the overview does not include potential support from the local Chinese authorities or from the Hungarian authorities. However, it remains that, at the next steps of Samsung SDI’s decision-making process (steps 4 and 5 described in recitals (68) and (69) of the opening decision), the claimed aid offer from the Chinese authorities was not factored in to either the comparison of the two sites or in the final location decision and did not feature in supporting documents. Consequently, Hungary and Samsung SDI have not provided evidence to show that such offer from the Chinese authorities was considered in the location decision.

(161) By contrast, as it transpires from Samsung SDI’s internal company documents (63), the aid offer from the Hungarian authorities was considered in steps 4 and 5 of the decision-making process, i.e. was included in the final comparison (an internal document from 26 October 2017) and was presented as a key factor justifying the investment decision in Göd (on 27 November 2017) (see recital (59)).

(162) In conclusion, the Commission considers that Hungary and Samsung SDI have not demonstrated that the grant offer from the Chinese authorities, if any, had been regarded as a prevalent factor by the decision-makers and taken into account at the time of the location decision. Therefore, the aid offer by the local Chinese authorities cannot be considered as a credible and realistic element of the counterfactual scenario.

5.3.1.2.3.2.   Corporate income tax (CIT) in China of 15 % instead of 25 %

(163) In their replies to the opening decision, Hungary and Samsung SDI also introduced a new claim, according to which Samsung SDI Group may have benefited from a reduced rate of corporate income tax in China if its investment had been located in Xi’an, either on the basis of the Chinese policies for the Western regions of China, or based on the possibility that Samsung SDI Group could have benefited from the HNTE regime (64) (recitals (24) to (29) and (58) to (62) of the extension decision).

(164) In the extension decision, the Commission expressed doubts as to the credibility of such a tax reduction (given that, on the one hand, in 2017 the benefits of such reduction were unexpectedly withdrawn from the EV battery industry in China and, on the other hand, Samsung SDI did not appear to fulfil the criteria of the HNTE regime) and on whether such a reduced rate had indeed been considered by Samsung SDI at the time of the location decision.

(165) Independently of the claim that Samsung SDI Group could have credibly assumed that a reduced rate of corporate income tax would have been available to it for its investment in Xi’an, the Commission considers that no evidence in the case file supports the claim that Samsung SDI Group took such a reduction into account in its location decision. As in the case of the grant offer from the Chinese local authorities, at the final steps of Samsung SDI Group’s decision-making process, the claimed corporate income tax rate reduction was not factored into either the comparison of the two sites or the final location decision. Hungary and Samsung SDI have not provided any evidence to show that such reduction was considered in the location decision.

(166) By contrast, the internal documents mentioned in recital (161) show that the aid offer from the Hungarian authorities was considered in steps 4 and 5 of the decision-making process, i.e. was included in the final comparison (an internal document from 26 October 2017) and was presented as a key factor justifying the investment decision in Göd (on 27 November 2017).

(167) Therefore, the Commission considers that Hungary and Samsung SDI have not demonstrated that the claimed corporate income tax reduction, if any, had been regarded as a prevalent factor by the decision-makers and taken into account at the time of the location decision. Therefore, the corporate income tax reduction cannot be considered as a credible and realistic element of the counterfactual scenario.

5.3.1.2.4.   Discrete and probabilistic approaches to plausible NPV gap calculations

(168) Following the adoption of the opening decision, Samsung SDI (recital (34) of the extension decision) submitted several alternative approaches (probabilistic and discrete approaches, see recitals (79) and (81) above). The discrete approach seeks to recalculate the NPV gap between the two locations in four different scenarios (taking into account one or the other claimed support measures from the Chinese authorities, and revised local sourcing shares for both locations). However, by Samsung SDI’s own admission (see recital (80) above), that approach does not accurately reflect the decision-making process that led to the location decision. In addition, as noted in recitals (162) and (167) above, there is no evidence that either of the claimed support measures from the Chinese authorities were taken into account at the time of the location decision. Besides, none of the four different scenarios included all the cost elements assessed above in section 5.3.1.2., which, in the Commission’s view, affect the extent to which the NPV gap between the two locations can be found credible. Therefore, that approach is not capable of affecting the Commission’s conclusions regarding the presence and the extent of the incentive effect that the aid from the Hungarian authorities had on Samsung SDI Group’s decision-makers. As regards the probabilistic approaches, they were developed by Samsung SDI on the basis of an estimate of probability of occurrence of certain events (local sourcing in China, a grant from the Chinese authorities, reduced CIT rate in China) for the re-calculation of the NPV gap. All three probabilistic approaches resulted in a re-calculated NPV gap similar to that originally submitted by Hungary.

(169) The Commission noted in the extension decision that Samsung SDI did not provide any contemporaneous internal documents that would support the claim that such a probability-weighing approach was considered at the time of the location decision, let alone that the different values of such probability factors had been taken into account in that decision-making process.

(170) In order to prove that the probabilistic approach had support in contemporaneous documents, Hungary, in its comments to the extension decision, argued that it had already submitted contemporaneous documents to demonstrate that Samsung SDI’s decision-makers were aware of the subsidy offer from the Chinese authorities, as well as the reduced CIT rate potentially available in China. However, the awareness of those possibilities does not mean that the decision-makers actually considered them when taking the location decision, let alone that they assigned specific probabilities to such elements. Hungary also referred to the internal report of 21 September 2017. That document (which mentions a criterion concerning ‘consistency in tax and policies’ and the different possible levels of a grant from the Chinese authorities), however, does not prove that the investment decision taken on 27 November 2017 was based on a probabilistic approach. In addition, the internal report of 21 September 2017 does not mention any probability in respect of any of the factors to which Samsung SDI and Hungary

ex-post

suggest to apply the probabilistic approach.

(171) Throughout the multiple stages of the Commission’s investigation, Hungary and Samsung SDI have thus not provided evidence contemporaneous with the decision-making process that such a probabilistic approach had been considered at the time of the location decision. Rather, it appears that the probabilistic approach was first developed by Samsung SDI in its comments to the opening decision (recital (78)).

(172) Therefore, the Commission considers that Hungary has not been able to prove that a probabilistic approach was ever considered by Samsung SDI during the decision-making process on the location of the investment. Therefore, this probabilistic approach cannot be considered as a credible and realistic element taken into account in the comparison of the two locations for the investment.

5.3.1.2.5.   Conclusion on the incentive effect

(173) The Commission therefore concludes that Hungary has proven that the aid measure had formal and substantive incentive effects. However, with regard to the latter, the Commission considers that Hungary has not demonstrated that the assumption that in either of the locations considered, a 100 % local sourcing rate would be achieved, was credible and realistic at the time of the location decision. In the Commission’s view, only a lower share of local sourcing in both scenarios can be considered as credible and realistic, in light of the elements available to Samsung SDI Group’s decision-makers at the time when they chose the location for the investment. The Commission also concludes that Hungary has not established either that aid from the Chinese authorities or the possibility of the investment benefitting from a reduced CIT rate, even if these elements could be considered credible, were taken into account by Samsung SDI Group when taking the location decision. In light of these elements, the Commission concludes that Investment 2 in Göd would incur a cost disadvantage in comparison to an investment in the alternative location in China, and hence the aid measure can be considered to operate as an incentive, in that, in the absence of the aid measure, it would not have been sufficiently profitable for the beneficiary to make that investment in Göd. However, the aid measure can only be considered to have an incentive effect to the extent that the aid amount does not exceed the disadvantage of the Göd location in terms of NPV, compared to the counterfactual location, examined in the context in which the beneficiary decided on the location of the investment. The Commission has established that such NPV gap cannot be credibly quantified at EUR 173 million, as claimed by Hungary in its notification, but has found it credible up to the amount of EUR 89,6 million. Any aid above that amount would lack incentive effect for the part exceeding that cost disadvantage.

5.3.2.   

Need for State intervention

(174) According to Section 3.3. of the RAG, in order to assess whether State aid is necessary to achieve the objective of common interest, it is necessary first to diagnose the problem to be addressed. State aid should be targeted to situations where aid can bring about a material improvement that the market cannot deliver itself.

(175) As established in paragraph 49 of the RAG, State intervention is considered justified for the development of the areas included in the regional aid map. In accordance with Hungary’s regional aid map for the period 2017-2021 (see recital (10)), prevailing at the time when the aid measure was granted (subject only to the Commission’s approval) (recital (30)), Göd was eligible for regional aid pursuant to Article 107(3), point (c) TFEU, which is in line with Section 3.3. of the RAG. The Commission thus upholds its assessment that Hungary has demonstrated the need for State intervention in the case at hand.

5.3.3.   

Appropriateness

(176) As recalled in the opening decision, according to paragraph 50 of the RAG, the notified aid measure must be an appropriate policy instrument to address the policy objective concerned. The RAG underline that an aid measure will not be considered compatible if other less distortive policy instruments or other less distortive types of aid instruments are available. Section 3.4. of the RAG therefore introduces a two-pronged appropriateness test. Under the first appropriateness test, Member States have to identify the bottlenecks to regional development and the specific handicaps of firms operating in the target region, and to clarify to what extent bottlenecks to regional development could also be targeted by non-aid measures. Under the second appropriateness test, the Member State has to indicate why, in view of the individual merits of the case, the chosen form of regional investment aid is the best instrument to influence the investment or local decision.

(177) As regards the first prong of the test, the Commission accepted in the opening decision that State aid, and regional investment aid in particular, is an appropriate form of support to achieve the cohesion objective for the Göd area, an area eligible for regional aid in accordance with Article 107(3), point (c) TFEU. In the absence of any evidence to the contrary in the file, this finding remains valid.

(178) As regards the second prong of the test, the Commission also accepted that a direct grant constitutes in principle an appropriate aid instrument to bridge a viability gap (see recitals (113) and (114) of the opening decision). The Commission, however, expressed doubts as to the reality and extent of the viability gap and of the contribution to regional development. In light of the Commission’s conclusions on the reality and the extent of the viability gap and on the contribution of the aid measure to regional development, the Commission considers that the direct grant to Samsung SDI, if it is limited to the amount of the viability gap referred to in recital (173), is an appropriate aid instrument in this case.

(179) Therefore, the Commission concludes that the regional investment aid measure provided in the form of a direct grant (up to the actual viability gap) represents an appropriate form of support to achieve the regional development objective of the area concerned.

5.3.4.   

Proportionality of the aid amount

(180) According to Section 3.6. of the RAG, regional aid must be limited to the minimum needed to induce additional investment or activity in the area concerned. Therefore, the assessment of the proportionality of the aid amount can only be carried out once the incentive effect of the aid is confirmed. The analysis of the proportionality of regional aid for a given investment must be done on the date of the investment decision, that is to say, in the context in which the recipient undertaking decides on the location of its project (65).

(181) The assessment of the proportionality of aid is subject to a double test. First, as a general rule, notified individual aid will be considered to be limited to the minimum, if the aid amount corresponds to the net extra costs of implementing the investment in the area concerned, compared to the counterfactual scenario in the absence of aid. Pursuant to paragraph 80 of the RAG, in scenario 2 situations, the aid amount should not exceed the difference between the NPV of the investment in the alternative location and the NPV in the target area, taking into account all relevant costs and benefits. Second, aid to compensate the NPV difference must not exceed the maximum amount of aid that may be granted according to the regional aid ceiling applicable to the target region for an investment project of the given size (taking into account the scaling down rules laid down in paragraph 86 and 20(c) of the RAG).

(182) First, in its opening decision, the Commission expressed doubts as to the proportionality of the aid, considering the doubts regarding the viability gap calculations presented by Hungary in its notification. Indeed, the Commission questioned the credibility of the information submitted by Hungary and preliminarily considered that the NPV gap between the two alternative investment locations was unrealistic (see recital (148) of the opening decision).

(183) In the extension decision, the Commission expressed further doubts concerning the reality and credibility of the NPV gap calculation presented by Hungary.

(184) When assessing the incentive effect (section 5.3.1.2.5. above), the Commission concluded that the NPV gap as notified by Hungary is not based on credible information and does not appear to be realistic. For the reasons detailed in that section, the Commission concluded that an investment in Göd would not incur a cost disadvantage in comparison with an investment in the alternative location in China of EUR 173 million, as claimed by Hungary in its notification, but rather of EUR 89,6 million. In that regard, the notified aid amount of EUR 108 million is disproportionate. To remain proportionate to the NPV gap, the aid amount shall therefore not exceed EUR 89,6 million.

(185) Second, with regard to the aid intensity requirement, the Commission recalled in its opening decision that the applicable regional aid ceiling in the Göd area is 35 %. It also recalled that the maximum aid intensity for large investment projects must be scaled down using the mechanism set out in paragraph 20(c) of the RAG, on the basis of which the maximum allowable aid intensity reaches 13,11 % gross grant equivalent (GGE) for the aid measure. On that basis, the Commission concluded that the notified aid of EUR 108 million compared to the planned total eligible expenditure in present value for the aid measure of around HUF 376,540 billion (around EUR 1 187 million) corresponded to an aid intensity of 9,11 %, which was thus not higher than the regional aid ceilings corrected by the scaling down mechanism. Considering that the planned total eligible expenditure increased compared to the notified figures, and the amount of aid considered as compatible in this Decision is reduced to EUR 89,6 million, the resulting aid intensity is lower than 9,11 % and remains therefore below the regional aid ceilings.

(186) Finally, the Commission’s conclusion concerning the compliance with the criteria for the eligibility of costs laid down in the RAG, as recorded in recitals (155) and (156) of the opening decision (66), remain valid.

(187) Based on the above, the Commission concludes that the aid measure is proportionate, but solely to the extent that it does not exceed the NPV gap established by the Commission at the amount of EUR 89,6 million.

5.3.5.   

Avoidance of manifest undue negative effects on competition and trade

(188) The Commission further analysed whether the aid has manifest undue negative effects on competition and trade.

(189) Section 3.7.2. of the RAG lists a series of situations where the negative effects on trade and/or competition manifestly outweigh any positive effects that a regional aid measure might have.

5.3.5.1.   

No manifest negative effect on trade: The adjusted aid intensity ceiling is not exceeded

(190) According to paragraph 119 of the RAG, a manifest negative effect would exist where the proposed aid amount exceeds, compared to the eligible (standardised) investment expenditure, the maximum adjusted aid intensity ceiling that applies for a project of a given size.

(191) In the opening decision, the Commission noted that the eligible costs were established in accordance with the RAG (recital (153) of the opening decision). In the absence of any evidence to the contrary, this conclusion remains valid. The Commission also held that the maximum aid intensities were not exceeded (recital (157) of the opening decision). As mentioned in recital (185) above, this conclusion remains a fortiori valid with higher eligible costs and an amount of aid set at EUR 89,6 million, a level lower than the notified amount.

5.3.5.2.   

No manifest negative effect on competition: The aid does not create overcapacity in a market in absolute decline

(192) According to paragraph 120 of the RAG, a manifest negative effect arises also where investment aid creates capacity in a market in absolute decline, as such aid is likely to crowd out competitors, or to prevent low cost firms from entering, and risks weakening incentives for competitors to innovate. This would result in inefficient market structures, which are also harmful to consumers in the long run.

(193) Considering the expected growing market for battery cells (as well as battery modules and battery packs), both at EEA and global level, notably supported by evidence provided by Hungary (recital (172) of the opening decision), the Commission considers that, independently of the exact product and geographic market definition, the aid measure is unlikely to result in overcapacity in a declining market. In addition, in any event, given that, whatever the location of the plant (Göd or Xi’an), that production facility would have served the same clients in the same geographical areas (see recital (88)), the aid is not causal for the increased supply in the market (regardless of the geographic or product definition). In line with its conclusion in the opening decision, the Commission therefore considers that the aid measure complies with paragraph 120 of the RAG.

5.3.5.3.   

No manifest negative effect on trade: No counter-cohesion effect

(194) Paragraph 121 of the RAG prohibits an EEA region with a lower project-specific viability from participating in ‘subsidy races’ to the detriment of equally weak or worse-off regions (67).

(195) Although it initially did not express any doubts in that regard in the opening decision, the Commission did not exclude the risk of counter-cohesion effects in the extension decision, based on new evidence submitted by Hungary and Samsung SDI (see recital (72) of that decision). Indeed, while at the initial phase of the investigation Hungary had only referred to one alternative location outside the EEA (China), it became apparent from that new evidence that other potential locations in the EEA (Poland, Slovakia and the Czech Republic) had also been considered by Samsung SDI Group as potential locations for the investment. At recital (68) of the extension decision, the Commission accepted that the Czech and Slovak greenfield sites had been excluded by Samsung SDI Group at an earlier stage of the selection process, the Commission could not exclude the possibility that aid to the Hungarian site in Göd, eligible for regional aid pursuant to Article 107(3), point (c) TFEU, might have had a counter-cohesion effect by attracting an investment away from a less-developed Polish region, eligible for regional aid pursuant to Article 107(3), point (a) TFEU.

(196) Based on the arguments provided by Hungary and Samsung SDI in reply to the extension decision and on the evidence in the file, the Commission concludes that the possible alternative Polish site was excluded from the list of potential sites quite early in Samsung SDI Group’s selection process. As indicated in the extension decision (recital (69)), it is clear that the Polish location was not assessed under the next step of the selection process (‘detailed analysis of the final candidates’) involving a profit and loss simulation, timeline simulation and the potential for expansion. The Polish site, like other alternative sites, only went through a preliminary quantitative and qualitative assessment. The Polish site ranked far below other sites on the preliminary quantitative side. From the document referred to by the Commission in its extension decision, it appears that only brownfield sites were selected, and that greenfield sites like the Polish site were excluded (notably as they resulted in much longer construction works). Therefore, as is apparent from the internal report of 21 September 2017), only two sites were seriously considered for the next stage and considered in full detail, namely the target site in Göd and the Chinese site in Xi’an.

(197) In light of this, the Commission considers that, contrary to the assertions that it had raised in the extension decision, the aid is unlikely to have a counter-cohesion effect.

5.3.5.4.   

No manifest negative effect on trade: Closure of activities elsewhere – potential relocation from Austria

(198) Pursuant to paragraph 122 of the RAG, where the beneficiary has concrete plans to close down or actually closes down the same or a similar activity in another area in the EEA and relocates that activity to the target area, if there is a causal link between the aid and the relocation, this will constitute a negative effect that is unlikely to be compensated by any positive elements.

(199) Hungary and the aid beneficiary declared (see recital (33) of the opening decision) that the beneficiary, at group level, had not closed down the same or similar activity in the EEA in the two years preceding the application for aid, and did not have any concrete plans to do so within two years after completion of the investment.

(200) The Commission, however, raised doubts in that regard, in light of press reports which suggested that the Samsung SDI Group was reorganising its activities in the EU (recital (179) of the opening decision). Such reorganisation could lead to the relocation of the battery packs production activities of Samsung SDI Group’s subsidiary located in Austria to Hungary, and a loss of 100 jobs in the Austrian subsidiary.

(201) The Commission could not exclude that the aid for an investment into EV battery cell production in Göd was not at least indirectly causal for any closure or relocation of the same or similar activity within the EEA.

(202) Based on the additional submissions of Hungary and Samsung SDI and on the evidence in the file (recital (90)), the Commission takes note that the Austrian site was not closed down within the meaning of the RAG. In order to qualify as closing down for the purpose of paragraph 122 of the RAG, the activity must have closed down either in full or partially when this results in substantial job losses i.e. losses of at least 100 jobs or as a job reduction of at least 50 % of the workforce in the establishment on the date of the application (compared to the average employment in the establishment in either of the two years preceding the date of application (the ‘reference period’) (68). Hungary provided evidence that these conditions were not met in the present case. The scheduled loss of 140 lower education employees by 2021, compensated in part by an increase in the number of higher education workers, was expected to reduce the total number of employees to a number between 397 and 417, to be compared to the average employment of 401-470 in the reference period (between October 2015 and September 2017). Fewer than 100 jobs would thus be lost. In addition, given the total workforce of 401-470 in the reference period, the (gross) reduction of workforce by 140 employees would be below the 50 % threshold.

(203) Additionally, Hungary explained that there was no relocation from Austria to Hungary, as the battery cells produced in Göd are technologically different from the cells produced in the Austrian plant (see recital (90)).

(204) Furthermore, Hungary provided internal company documents showing that the reduction of activities in Samsung SDI’s Austrian plant was part of a re-orientation of that establishment (from production of battery packs into research & development and testing and quality control activities) following a decision by the customers (certain car manufacturers) (not by Samsung SDI) showing disinterest in the products produced in the Austrian plant. As a consequence, the Commission accepts that a causal link cannot be established between Investment 2 and the reorganisation of Samsung SDI Group’s Austrian plant.

(205) In view of those elements, the Commission has not been able to establish that the aid measure would be causal to any closure or relocation within the EEA.

5.3.5.5.   

Conclusion on the avoidance of manifest negative effects on competition and trade

(206) In view of the above, the Commission concludes that the aid has no manifest negative effect on competition or trade within the meaning of Section 3.7.2. of the RAG.

5.3.6.   

No other undue negative effects

(207) Apart from section 3.7.2. of the RAG, undue negative effects on competition that are to be taken into account in the remaining balancing test if the investment takes place in a scenario 2 context in two distinct geographic markets are identified in paragraphs 114 and 115 and 132 of the RAG. Such effects concern the creation or reinforcement of a dominant market position or the creation or reinforcement of overcapacities in an underperforming market (even if that market is not in absolute decline).

(208) First, the investment would have led, with or without aid, to the same volumes of battery cells being delivered to the EEA and worldwide market (as whether the plant would have been located in Xi’an or in Göd, the expected destination markets are the same, see recital (88)) so that the aid is not causal for such undue negative effects and it is not necessary to carry out these additional tests.

(209) In any event, the aid will not create or reinforce a dominant market position as Samsung SDI’s market shares (taking into account the investment at stake) were expected to increase from [7-10] % in 2019 to [20-25] % in 2023 at European level and from [7-10] % in 2019 to [10-13,5] % in 2023 at global level (see recital (25)). Nor will the aid create overcapacities in an underperforming market as the market – worldwide and at the EEA level – is growing (see recital (172) of the opening decision).

5.3.7.   

Balancing of positive and negative effects of the aid

(210) By triggering the location of the investment in an assisted region, the aid contributes to the regional development of Göd, which implies a series of positive effects for that region (see section 5.3.1.1.). Moreover, the assessment set out in the preceding sections showed that State intervention is needed, that the aid is appropriate and that the aid has an incentive effect, as long as it does not exceed the NPV gap corresponding to an amount of EUR 89,6 million. The Commission also concludes that State intervention is proportionate, to the extent it remains limited to that identified NPV gap. The assessment also showed that, despite the doubts initially raised, the aid has no manifest negative effect on competition and trade. In addition, as regards other negative effects on competition and trade, the aid will not create or reinforce a dominant market position nor overcapacities in an underperforming market.

(211) Therefore, the Commission concludes that the positive effects of the aid on the regional development, to the extent it does not exceed EUR 89,6 million, outweigh any negative effect.

5.3.8.   

Transparency

(212) In accordance with paragraph II.2 of the Transparency Communication from the Commission (69), Member States must ensure the publication on a comprehensive State aid website, at national or regional level, of a full text of the approved aid scheme or the individual aid granting decision and its implementing provisions, or a link to it; the identity of the granting authority or authorities; the identity of the individual beneficiaries, the form and amount of aid granted to each beneficiary, the date of granting, the type of undertaking (SME/large company), the region in which the beneficiary is located (at NUTS level II) and the principal economic sector in which the beneficiary has its activities (at NACE (nomenclature of economic activities) group level). Such information must be published after the decision to grant the aid has been taken, must be kept for at least 10 years and must be available to the general public without restrictions.

(213) The Commission notes that Hungary confirmed that all requirements concerning transparency set out in paragraph II.2 of the Transparency Communication will be respected (see recital (187) of the opening decision).

5.3.9.   

No relevant breach of EU law

(214) It does not result from the notification that the aid or the conditions attached to it, or the economic activities facilitated by the aid, could entail a violation of any relevant provision of EU law (70). In particular, the Commission has not sent a reasoned opinion to Hungary on a possible infringement of Union law that would bear a relation to this case. Furthermore, the Commission has not received any complaint or information that could suggest that the aid, the conditions attached to it or the economic activities facilitated by it might be contrary to relevant provisions of Union law.

5.3.10.   

Conclusion on compatibility of aid measure

(215) The Commission therefore concludes that the aid measure, to the extent it remains limited to a maximum amount of EUR 89,6 million, is compatible with the internal market pursuant to Article 107(3), point (c) TFEU.

6.   

CONCLUSION

(216) The Commission has accordingly decided that the aid measure is compatible with the internal market pursuant to Article 107(3), point (c) TFEU to the extent that its amount does not exceed EUR 89,6 million. The notified aid beyond that amount, up to EUR 108 million, is incompatible with the internal market.

HAS ADOPTED THIS DECISION:

Article 1

The State aid which Hungary is planning to implement in favour of Samsung SDI Magyarország Zrt, amounting to EUR 108 million (in present values), is compatible with the internal market up to an aid amount of EUR 89,6 million (in present values) within the meaning of Article 107(3), point (c), of the Treaty on the Functioning of the European Union, and is incompatible with the internal market for the remaining amount.

Implementation of the aid is accordingly authorised up to EUR 89,6 million (in present values). The remaining amount of the planned aid shall not be implemented.

Article 2

Hungary shall inform the Commission, within two months of notification of this Decision, of the measures taken to comply with it.

Article 3

This Decision is addressed to Hungary.

Done at Brussels, 28 February 2023.

For the Commission

Margrethe VESTAGER

Executive Vice-President

(1)  

OJ C 112, 3.4.2020, p. 12

.

(2)  Cf. footnote 1.

(3)  C(2021)4454 final, published on the Commission website on 17 January 2022.

(4)  State aid – Hungary – State aid SA.48556 (2019/C) (ex 2018/N) – Regional investment aid to Samsung SDI – Invitation to submit comments pursuant to Article 108(2) of the Treaty on the Functioning of the European Union (

OJ C 82, 18.2.2022, p. 21

).

(5)  SA.46346 (2016/N) (

OJ C 4, 6.1.2017, p. 1

). The Commission’s approval of that map was extended until the end of 2021 by Commission Decision of 7 October 2020 in case SA.58164 (2020/N) - Prolongation of the Hungarian regional aid map for the period 1.1.2021-31.12.2021 (

OJ C 430, 11.12.2020, p. 1

).

(6)  Confidential information.

(7)  Communication from the Commission — Guidelines on State aid for rescuing and restructuring non-financial undertakings in difficulty (

OJ C 249, 31.7.2014, p. 1

).

(8)  On 12 January 2023, Hungary informed the Commission that in 2022 Samsung SDI exceeded its initial targets, with a production capacity of around [6-9] million battery cells per month (equivalent to [18-21] GWh per year).

(9)  On 12 January 2023, Hungary informed the Commission that in 2022 [69,5-95] % of the production of the second plant in Göd (the investment at stake) served the EEA market (a higher percentage than anticipated).

(10)  In general, a module consists of multiple cells connected in series and/or parallel, encased in a mechanical structure. A battery pack is assembled by connecting multiple modules together in series or parallel with sensors and controllers, including battery management systems and thermal management systems, and then encased in a housing structure, as a final battery product designed specifically for each vehicle model.

(11)  In the first phase of Investment 2, Samsung SDI produced large metal can prismatic cells with […] electrodes, similar to the cells produced via the unaided Investment 1. However, changes in the production process steps and to the electrode materials of phase 1 would lead to a significantly improved performance in terms of driving range, speed to charge, and cost per kWh.

In the second phase of Investment 2, Samsung SDI implemented a significant innovation in the large format metal can prismatic battery cell structure approach (i.e. stacking electrodes instead of winding) which, together with further changes in process steps and input materials, would result in further improvements of the performance of the battery cells.

(12)  Blackboxing is a protection strategy for intellectual property, which prevents disclosure of business secret by limiting access to the full information regarding the potential innovation to only a few trusted employees. No patents applications are filed because already the publication of the subject of the invention to be patented is likely to divulge the nature of the invention to the competitors; furthermore, it is very difficult to prove violations of patent rights regarding process innovations as this requires access to the competitors’ production lines. Blackboxing means in practice that specific security measures need to be applied, such as prohibiting any recording within the manufacturing facilities, blocking mobile reception within the production facilities, enforcing strict entrance and exit rules, etc.

(13)  Guidelines on regional State aid for 2014-2020 (

OJ C 209, 23.7.2013, p. 1

).

(14)  Figures expressed in EUR are given in this decision on the basis of an exchange rate of EUR 1 = HUF 317,19, applicable on 16 May 2018, which is the date of the notification of the aid measure to the Commission.

(15)  The present values in this decision are calculated on the basis of a discounting rate of 1,09 % applicable at the time of the notification. Present values are discounted to the year 2018, which is the year of the notification.

(16)  Commission Regulation (EU) No 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty (

OJ L 187, 26.6.2014, p. 1

).

(17)  The adjusted aid amount means the maximum permissible aid amount for a large investment project, calculated according to the following formula: maximum aid amount = R x (50 + 0,50 x B + 0,34 x C), where R is the maximum aid intensity applicable in the area concerned, excluding the increased aid intensity for SMEs, B is the part of eligible costs between EUR 50 million and EUR 100 million and C is the part of eligible costs above EUR 100 million.

(18)  According to Hungary, which referred to Samsung SDI Group’s ‘Roadmap for the Establishment of a New Corporate Branch’ (July 2015), the local sourcing policy required the beneficiary to buy from local sources the equipment and machinery and the other inputs necessary for the investment.

(19)  Samsung SDI-ARN Power Battery Co. Ltd (SAPB) was established in 2014 and is a three-party joint venture of the Samsung SDI Group, Anging Ring New Group (a Chinese automotive parts supplier) and Xi’an Gaoke Group (a Chinese State-owned company). The Samsung SDI Group owns a 50 % equity share in the joint venture.

(20)  The Samsung SDI Group’s battery cell plant in Ulsan (South Korea) was excluded as an alternative investment location early in the decision-making process due to its high labour costs compared to those in Hungary or China.

(21)  Samsung SDI’s comments to the opening decision, paragraph 43 (‘[…]’).

(22)  Hungary’s comments to the opening decision, paragraphs 31 and 122.

(23)  Another EEA supplier was contacted in 2019 and was later on selected to provide equipment.

(24)  […].

(25)  […].

(26)  Given its existing EV battery manufacturing facility in Xi’an (built in 2015 to produce EV batteries).

(27)  To which the Commission referred in recitals (53) to (57) of the extension decision.

(28)  European Union/[…].

(29)  See recitals (50) and (131) of the opening decision ([…]), Samsung SDI’s comments to the extension decision ([…]), recital (43) of the opening decision and recital (44) of the extension decision (confidentiality).

(30)  Hungary’s reply to Q.5 of the request for information of 1 April 2019; Samsung SDI’s comments to the opening decision, paragraph 87, Samsung SDI’s comments to the extension decision, p. 3.

(31)  Samsung SDI’s comments to the opening decision, paragraph 87.

(32)  Hungary’s comments to the opening decision, paragraph 55.

(33)  In Samsung SDI’s comments of 3 May 2020 to the opening decision, a [30-40] % rate of local sourcing (baseline) was considered realistic (with a [25-30] % ratio as a conservative assumption). Upon questioning by the Commission, the baseline rate was corrected by Samsung SDI and Hungary, on 3 November 2020, to [31-35] %. Spare parts are not included in the various percentages of local sourcing.

(34)  In support of this claim, Hungary provided the names of four Chinese producers, summaries of the technical characteristics of their respective equipment, as well as summaries of meetings with them that took place in between 2014 and 2019.

(35)  In support of this claim, Hungary provided the names of two Chinese producers and summaries of meetings with them that took place in between 2014 and 2019.

(36)  Or, alternatively, the local machinery and equipment procurement costs if Samsung SDI HQ is not involved in the procurement process.

(37)  This [18-22] % mark-up was not included in the initial NPV calculations, since the latter were precisely based on the assumption of a 100 % local sourcing.

(38)  Deloitte study of December 2018 and Copenhagen Economics study of 13 April 2020.

(39)  Internal document ‘Xi’an Plant # 2 Investment Plan’ of 5 January 2017, summary of a ‘Business trip to China regarding Xi’an Plant 2 Investment’ of 13 January 2017, minutes of 18 February 2017 of a meeting between […] on 16 February 2017, presentation of 3 March 2017 prepared by Samsung SDI’s […] meeting on ‘[…]’, note ‘[…]’ of 14 March 2017, note ‘[…]’ of 20 March 2017, email of 21 July 2017 with a debriefing report of a meeting of 20 July 2017 with the local government.

(40)  Hungary submitted that Samsung SDI had also considered (Samsung SDI internal report dated 5 June 2017) an alternative possibility to benefit from a 15 % reduced corporate tax rate, under the ‘High and New Technology Enterprise’ (HNTE) regime (Notice of the Ministry of Science and Technology, the Ministry of Finance and the State Administration of Taxation on Revising and Issuing the Measures for the Administration of the Certification of High-tech Enterprises (2016)).

(41)  Samsung SDI added that a probability factor that would be slightly higher or lower than 25 % would not affect significantly the NPV gap and the aid amount.

(42)  As summarised in recitals 68 to 70 of the extension decision.

(43)  The RAG were prolonged until 31 December 2021 (see Communication from the Commission concerning the prolongation and the amendments of the Guidelines on Regional State Aid for 2014-2020, Guidelines on State Aid to Promote Risk Finance Investments, Guidelines on State Aid for Environmental Protection and Energy 2014-2020, Guidelines on State aid for rescuing and restructuring non-financial undertakings in difficulty, Communication on the Criteria for the Analysis of the Compatibility with the Internal Market of State Aid to Promote the Execution of Important Projects of Common European Interest, Communication from the Commission – Framework for State aid for research and development and innovation and Communication from the Commission to the Member States on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to short-term export-credit insurance,

OJ C 224, 8.7.2020, p. 2

). The new Guidelines on regional State aid (

OJ C 153, 29.4.2021, p. 1

) are only applicable to ‘notifiable regional aid awarded or intended to be awarded after 31 December 2021 ’ (paragraph 197 of the new guidelines).

(44)  See paragraph 38 of the RAG.

(45)  See paragraph 36 of the RAG.

(46)  See also, judgment of 15 December 2022, Veejaam and Espo, C-470/20, ECLI:EU:C:2022:981, paragraph 35.

(47)  The present values are discounted to the year 2018, which is the year when the aid was notified to the Commission.

(48)  In recital 18(c) of the extension decision, the Commission implicitly doubted that the planned Xi’an plant would be dedicated to export markets. This doubt resulted from the reference to a ‘Chinese policy risk’ in an internal document provided after the opening decision. However, Samsung SDI convincingly explained that this ‘policy risk’, in the context of the document at stake, only referred to the uncertainty on the actual level of aid that could be obtained in China, and not to an overall risk of protectionist measures on domestic sales. At the same time, the Commission notes that it is clear that the counterfactual investment in China would not have targeted the Chinese market (see minutes of the Business trip to China regarding Xi’an Plant 2 investment, 13 January 2017).

(49)  Minutes of an internal high-level Samsung SDI meeting on 21 September 2017, see recitals (20) and (53) of the extension decision.

(50)  Minutes of a meeting of the […] Investment Committee of 26 October 2017.

(51)  Hungary and Samsung SDI also convincingly explained the apparent discrepancies – highlighted in recital (49) of the extension decision – between the investment project described in reports documenting the subsidy negotiation with the Chinese local government and the counterfactual Chinese investment described in the notification. This is sufficient to alleviate the doubt – raised in recital (54) of the extension decision – that Samsung SDI envisaged in parallel an additional plant in China aimed at the Chinese domestic market, interpreted at the time by the Commission as a further indication of a possible regionalised investment strategy of Samsung SDI.

(52)  See sub-section 5.3.1.2.2.2.3. about the realistic level of local sourcing for the investment in Göd.

(53)  See sub-section 5.3.1.2.2.2.2. about the realistic level of local sourcing in China.

(54)  In March and May 2019 the aid measure’s implementation was already at an advanced stage (recital (52) above).

(55)  Such a mark-up would not have been relevant if the entirety of the machinery and equipment for the investment were to be sourced locally, since Samsung SDI Group’s subsidiaries would have made the purchases directly, with only limited involvement of Samsung SDI HQ.

(56)  Copenhagen Economics, SAMSUNG SDI – provision of equipment and machinery, Transfer pricing analysis, 13 April 2020 (Annex 36 to Samsung SDI’s comments submitted on 3 May 2020).

(57)  OECD (2017), OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2017, OECD Publishing, Paris.

(58)   ‘Not-novel equipment’ includes (i) equipment that has already been installed and fully validated once in an existing mass production line, even where such equipment has been slightly modified for an improvement, based on the already developed and installed equipment, or (ii) equipment that has been identified by the local subsidiary as adequate for use on any existing production line (even if not yet used).

(59)  This also addresses the concerns raised in recital (41) of the extension decision (about the possibility to take into account Chinese suppliers of the Tianjin plant for the Xi’an plant, as described in recitals (9)(a) and (40)(a) of the extension decision.

(60)  See recitals (9), (10) and (40) of the extension decision.

(61)  According to Hungary, approximately [45-50] % of the investment costs were assigned to phase 1 of the investment, see footnote (36) of the extension decision.

(62)  Hungary and Samsung SDI also repeatedly highlighted that the innovations in the process concern the processes themselves and not the machinery and equipment used so that the difference between the two phases is not decisive for the procurement of machinery and equipment.

(63)  Document ‘7.10. Samsung SDI Business Division Investment Committee meeting in October’ (October 2017). Document ‘7.12. Meeting minutes of Business Division’s Investment Committee’ (26 October 2017). Document ‘7.11. Samsung SDI Hungary Investment CEO Investment Committee in November’ (November 2017). Document ‘7.5. Meeting minutes of Investment Committee with CEO decision’ (27 November 2017). These documents were submitted on 16 May 2018 (as part of the notification by Hungary).

(64)  See footnote 39.

(65)  Judgment of 12 September 2017,

Bayerische Motoren Werke AG

v

Commission

, T-671/14, ECLI:EU:T:2017:599, paragraphs 115, 127 and 128, upheld on appeal in case C-654/17 P, ECLI:EU:C:2019:634. The Commission moreover notes that, in the present case, the location decision of 27 November 2017 was immediately followed by the start of works on 1 December 2017 (see recital (69) of the opening decision and recital (17) above).

(66)  In recitals (155) and (156) of the opening decision, the Commission found that the eligible costs had been established in accordance with the RAG: the acquired assets will be new, the initial investment concerns an initial investment in the form of a capacity extension and no leasing costs and no immaterial costs are taken into account.

(67)  According to paragraph 121 of the RAG, the counter-cohesion effect resulting from aid to the detriment of a weaker or similarly weak EEA region would constitute a negative element in the overall balancing test that is unlikely to be compensated by any positive elements, because it runs counter the very rationale of regional aid.

(68)  Commission Decision of 3 October 2016, case SA.44547 (2016/N) – Italy – LIP - Aid to STMicroelectronics S.r.l. (

OJ C 110, 7.4.2017, p. 1

), recital 106 and footnote 41.

(69)  Communication from the Commission amending the Communications from the Commission on EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks, on Guidelines on regional State aid for 2014-2020, on State aid for films and other audiovisual works, on Guidelines on State aid to promote risk finance investments and on Guidelines on State aid to airports and airlines (

OJ C 198, 27.6.2014, p. 30

).

(70)  See paragraph 28 of the RAG.

ELI: http://data.europa.eu/eli/dec/2024/2057/oj

ISSN 1977-0677 (electronic edition)