Commission, October 1, 2019, No M.9706
EUROPEAN COMMISSION
Decision
NOVELIS / ALERIS
COMMISSION DECISION of 1.10.2019
declaring a concentration to be compatible with the internal market and the EEA agreement
(Case M.9076 - NOVELIS / ALERIS)
(Text with EEA relevance) (Only the English text is authentic)
THE EUROPEAN COMMISSION,
Having regard to the Treaty on the Functioning of the European Union,
Having regard to the Agreement on the European Economic Area, and in particular Article 57 thereof,
Having regard to Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings1, and in particular Article 8(2) thereof,
Having regard to the Commission's decision of 25 March 2019 to initiate proceedings in this case,
Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission,
Having regard to the opinion of the Advisory Committee on Concentrations, Having regard to the final report of the Hearing Officer in this case, Whereas:
1. INTRODUCTION
(1) On 18 February 2019, the Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139/2004 (the ‘Merger Regulation’) by which Novelis Inc. (‘Novelis’, USA), a fully owned subsidiary of Hindalco Industries Limited (‘Hindalco’, India), acquires, within the meaning of Article 3(1)(b) of the Merger Regulation, sole control of the whole of Aleris Corporation (‘Aleris’, USA) by way of purchase of shares2 (hereinafter the ‘Transaction’). Novelis is designated hereinafter as the ‘Notifying Party’.3 Novelis and Aleris are designated hereinafter as the ‘Parties’. The entity resulting from the Transaction is designated hereinafter as the ‘Merged Entity’.
(2) Novelis is a global manufacturer of flat rolled aluminium4 products and a recycler of aluminium. The company operates 24 manufacturing facilities across North America, South America, Europe and Asia. Novelis’ parent company, Hindalco, is an India- based supplier of aluminium and copper.
(3) Aleris is a global manufacturer of flat rolled aluminium products. Aleris operates 13 production facilities in North America, Europe and Asia.
2. THE OPERATION AND THE CONCENTRATION
(4) Pursuant to an agreement signed on 26 July 2018, a wholly-owned subsidiary of Novelis will merge into Aleris, with Aleris surviving the merger as a wholly-owned indirect subsidiary of Novelis.
(5) The operation thus constitutes a concentration within the meaning of Article 3(1)(b) of the Merger Regulation.
3. UNION DIMENSION
(6) The combined aggregate worldwide turnover of the Parties is more than EUR 5 000 million (Novelis […]5; Aleris […]) and the aggregate Union-wide turnover of each of the Parties is more than EUR 250 million (Novelis EUR […]6; Aleris EUR […]).7 Neither of the Parties achieve more than two-thirds of their Union-wide turnover within one and the same Union Member State.
(7) The Transaction therefore has a Union dimension pursuant to Article 1(2) of the Merger Regulation.
4. THE PROCEDURE
(8) On 18 February 2019, the Notifying Party notified the Transaction to the Commission.
(9) During its initial (Phase I) investigation, the Commission reached out to a large number of market participants (mainly customers and competitors of the Parties), by requesting information through telephone calls and written requests for information pursuant to Article 11 of the Merger Regulation, including in the form of questionnaires.8
(10) In addition, the Commission sent several written requests for information to the Parties and reviewed internal documents of the Parties submitted at that stage.
(11) On 25 March 2019, based on its initial investigation, the Commission raised serious doubts as to the compatibility of the Transaction with the internal market and adopted a decision to initiate proceedings pursuant to Article 6(1)(c) of the Merger Regulation (the ‘Article 6(1)(c) Decision’).
(12) On 26 March 2019, the Commission provided non-confidential versions of certain key submissions of third parties collected during the Phase I investigation to the Notifying Party. On 28 March 2019, following a request from the Notifying Party, the Commission provided two further non-confidential versions of such submissions to the Notifying Party.
(13) On 4 April 2019, the Notifying Party submitted its written comments on the Article 6(1)(c) Decision (‘Reply to the Article 6(1)(c) Decision’).
(14) On 9 April 2019, following the Notifying Party's comments on the Article 6(1)(c) Decision, a State of Play meeting took place between the Commission and the Parties.
(15) On 11 April 2019, following a request from the Notifying Party, the time period set for the adoption of a final decision in relation to the Transaction pursuant to Article 10(3), first paragraph, of the Merger Regulation was extended by 20 working days pursuant to Article 10(3), second paragraph, of the same regulation.
(16) During its in-depth (Phase II) investigation, the Commission sent several requests for information to the Parties regarding various matters such as commercial strategy, capacity expansion plans and market data.
(17) In addition to collecting and analysing a substantial amount of information from the Parties (including internal documents and submissions), the Commission contacted a number of market participants (including customers and competitors of the Parties) and requested information from such third parties both through questionnaires9 pursuant to Article 11 of the Merger Regulation and telephone calls.
(18) On 13 May 2019, following the failure of the Parties to provide certain information it had requested pursuant to Art 11(2) of the Merger Regulation, the Commission adopted two decisions pursuant to Article 11(3) of the Merger Regulation, one addressed to the Notifying Party and another one to Aleris. The decisions requested the Parties to provide certain information as soon as possible and no later than 23 May 2019. Consequently, pursuant to Article 10(4) of the Merger Regulation and Article 9 of Commission Regulation No 802/200410 (‘Implementing Regulation’), the merger review time limit referred to in Article 10(3) of the Merger Regulation was suspended as from 7 May 2019, the working day following the date on which the Parties should have submitted complete responses to the relevant Art 11(2) requests for information. The referred time limit was suspended until 15 May 2019.
(19) On 20 June 2019 and following the results of the Phase II market investigation, a state of play meeting was held in order to inform the Notifying Party of the preliminary results of the Phase II market investigation and the scope of the preliminary concerns regarding which the Commission planned to issue a Statement of Objections.
(20) On 1 July 2019, the Commission adopted a Statement of Objections (‘SO’), which was sent to the Notifying Party on the same day. In the SO, the Commission set out the preliminary view that the Transaction would likely significantly impede effective competition in the internal market, within the meaning of Article 2 of the Merger Regulation, in relation to the production and supply of Aluminium ABS in the EEA due to (i) the creation or strengthening of a dominant market position in the relevant market; and (ii) horizontal non-coordinated effects resulting from the elimination of an important competitive constraint. The Commission’s preliminary conclusion was therefore that the notified concentration would be incompatible with the internal market and the functioning of the EEA Agreement.
(21) On 2 July 2019, the Notifying Party was granted access to the file. A data room was organised from 3 July to 9 July 2019 allowing the economic advisors of the Notifying Party to verify confidential information of a quantitative nature, which formed part of the Commission’s file. A non-confidential data room report (‘First Data Room Report’) was provided to the Notifying Party on 10 July 2019.11 The confidential report was taken to the Commission’s file on the same date.
(22) On 9 July 2019, the Notifying Party submitted an excel table identifying 286 passages from the file for which they requested less redacted versions. The Notifying Party was provided with replies of the Commission’s review of this request on 11 July 2019 (addressing 208 passages), 12 July (addressing further 26 passages), and 16 July 2019 (addressing further 39 passages). The reply concluding the Commission’s review and addressing the last 13 passages was provided to the Notifying Party on 17 July 2019. In the course of this review, 65 of the passages were completely or partially un-redacted.
(23) On 17 July 2019, the Notifying Party submitted their reply to the SO (the ‘Reply to the SO’).
(24) IG Metall and ABVV Metaal made an application to the Hearing Officer to be admitted as interested third persons in the proceedings, and they both were recognised as such by the Hearing Officer. They were provided with a non- confidential version of the SO. They both presented their views on the proposed Transaction at the oral hearing.
(25) On 23 July 2019, an oral hearing was held, upon request by the Notifying Party.
(26) On 26 July 2019, a state of play meeting was held, during which the Commission provided the Notifying Party with preliminary feedback following their Reply to the SO.
(27) The Notifying Party was granted subsequent access to the file on the same day.
(28) On 6 August 2019, a Letter of Facts setting forth evidence corroborating the objections set out in the SO – was sent to the Notifying Party. The Notifying Party submitted its comments on the Letter of Facts on 19 August 2019 (‘Reply to the Letter of Facts’).
(29) On 7 August 2019, the Notifying Party was granted subsequent access to the file. Another data room was organised from 8 August to 9 August 2019. A non- confidential data room report (Second Data Room Report) was provided to the Notifying Party on 12 August 2019.12 The confidential report was taken to the Commission’s file on the same date.
(30) On 9 August 2019, the Notifying Party submitted commitments pursuant to Article 8(2) of the Merger Regulation in order to address the competition concerns identified in the SO (the ‘Commitments of 9 August 2019’).
(31) On 13 August 2019, the Parties submitted revised commitments pursuant to Article 8(2) of the Merger Regulation in order to address the competition concerns identified in the SO (the ‘Commitments of 13 August 2019’).
(32) On 13 August 2019, the Commission launched a market test of the Commitments of 13 August 2019.
(33) On 2 September 2019, the Notifying Party was granted further access to file.
(34) On 3 September 2019, the Parties submitted revised commitments pursuant to Article 8(2) of the Merger Regulation in order to address the competition concerns identified in the SO (the ‘Final Commitments’).
(35) On 4 September 2019, the Commission sent a draft Article 8(2) decision to the Advisory Committee with the view of seeking the Committee’s opinion on it.
(36) The meeting of the Advisory Committee took place on 18 September 2019.
5. INTRODUCTION TO THE INDUSTRY AND PRODUCTS - ALUMINIUM FRPS
5.1. Production of metallic aluminium
(37) Aluminium is one of the most abundant elements in the Earth’s crust, and it is one of the most commonly used non-ferrous metals. Metallic aluminium13 can be produced either from aluminium-containing minerals (primary aluminium) or through recycling metallic aluminium (secondary aluminium).
(38) Most primary aluminium is produced from an ore called bauxite. Bauxite is an ore rich in aluminium minerals. It is obtained through mining at various locations across the world. To produce primary aluminium, bauxite is first refined into alumina (aluminium oxide) using a multi-stage process. Subsequently, alumina is further processed in reduction plants forming pure aluminium through an electrolytic process called smelting.14
(39) Secondary aluminium is produced by re-melting and re-converting used aluminium products or scraps generated during the manufacturing process of aluminium products.
(40) Once primary or secondary aluminium (or a mixture of them) is molten, certain alloying elements can be added to obtain the desired characteristics. The liquid aluminium can thereafter be cast into various forms, such as: (i) ingots/T-bars, for re- melting purposes; (ii) extrusion billets, supplied to extruders to produce aluminium extrusions; (iii) slabs, which are typically used by rolling mills to produce aluminium flat rolled products (‘Aluminium FRPs’); (iv) wire rod, used to make aluminium wire for applications such as electricity transmission or in the steel industry as a deoxidising material; and (v) foundry alloys, supplied to foundries for use in the machinery, tool and automobile industries.
(41) The Transaction concerns the manufacture and supply of Aluminium FRPs. Aluminium FRPs are produced in rolling mills, typically from slabs of aluminium alloys. Other types of aluminium products, such as extrusions, are thus not discussed further in this Decision.
5.2. Aluminium FRPs
(42) The manufacturing of an Aluminium FRP starts with an aluminium slab, which is processed mainly in three steps (Figure 1): (i) hot rolling, which involves the heating of the slab in a furnace before it is rolled and that reduces the slab’s thickness to a certain desired thickness depending on the type of Aluminium FRP being produced; (ii) cold rolling, which further reduces the thickness; and (iii) finishing, which can include a number of treatments (heat treatments, surface treatments, etc.). Rolled aluminium coils can be sold as such by the producer, or they can be cut to desired length and width.
(43) Aluminium FRPs can be used for various different end-uses, such as beverage cans, food cans, aluminium foil, construction applications and automotive applications.
(44) The activities of the Parties overlap in the production and supply of (i) Aluminium FRPs used to produce vehicle bodies in the automotive industry, also known as Aluminium Automotive Body Sheets (‘Aluminium ABS’), and (ii) Aluminium FRPs used for certain other applications (so-called ‘Standard FRPs’), including in its potential sub-segments for aluminium anodising sheet and aluminium sheet for compound tubes (also known as multi-layer tubes).
(45) More details regarding the various types of Aluminium FRPs, including Aluminium ABS and Standard FRPs are provided in Section 6.1.
5.3. Aluminium ABS
(46) Aluminium ABS are used in so-called body closures and in vehicles’ body structures. Body closures can be external or internal and include, for example, the bonnet (also called ‘hood’), doors, window frames, roof and boot of a car. The body structure is the core element of a car’s body. The car body connects all the different components; it houses the drivetrain as well as carries and protects the passengers.
(47) The so-called ‘body-in-white’ (‘BiW’) refers to the stage in automobile manufacturing in which the car body sheet material (including body closures and the body structure) has been assembled but before the components (such as engine, chassis, exterior and interior trim, seats and electronics) have been added to the body structure.
(48) Aluminium ABS (or ABS) is an expression commonly used in the automotive industry when referring to Aluminium FRP used for manufacturing the body-in- white.
(49) As explained in Section 5.1, certain alloying elements can be added to aluminium to achieve the desired characteristics of the final product. Depending on the main alloying element(s), certain alloy series can be distinguished. According to the Notifying Party, Aluminium ABS are almost exclusively made of the aluminium alloys series 5xxx and 6xxx.15
(50) 5xxx series (Al-Mg) are alloys in which magnesium is the principal alloying element. The 5xxx series are non-heat-treatable alloys. Alloys in this series possess moderate to high strength characteristics, as well as good weldablility and resistance to corrosion. 5xxx alloys are used for internal body closures (predominantly for the inner doors and inner bonnet) and for body structure applications.
(51) 6xxx series (Al-Mg-Si) are alloys in which magnesium and silicon are the principal alloying elements. The 6xxx series are versatile, heat treatable, highly formable, weldable and have moderately high strength coupled with excellent corrosion resistance. Due to heat-treatment by way of continuous annealing, the 6xxx series is stronger than the 5xxx series. 6xxx alloys are thus used for internal and external body closures (frequently for the bonnet, body side panels, wings and roof of a car) and for body structure applications.
(52) There are two production steps for Aluminium ABS that are additional to the production steps required for all Aluminium FRPs (such as casting, rolling and finishing, as described in recitals (40) and (42) of this Decision): (i) continuous annealing or batch annealing and (ii) pre-treatment.
(53) Aluminium FRP suppliers wishing to produce Aluminium ABS at an existing FRP production line need to include either batch annealing or continuous annealing capabilities. Annealing is typically carried out after rolling and before finishing to adjust the material properties, in particular to increase the strength of the material.
(54) In the production of Aluminium ABS, the annealing typically takes place on a continuous annealing line, also known as Continuous Annealing Solution Heat Treating ('CASH') (the finishing line is also referred to as Continuous Annealing Line with Pre-Treatment (‘CALP’).
(55) Annealing in a CASH line (see Figure 3) is a continuous process during which coils are joined to enable a continuous ribbon of aluminium sheet to run through the finishing process. CASH is the main production method used in Aluminium ABS production and is required for the production of 6xxx series. A CASH line is an asset configured to combine the heat treatment and pre-treatment processes. It is nonetheless possible for these processes to exist in a disaggregated system too (see Figure 3).
(56) The production of 5xxx series is at least to a certain extent possible using batch annealing. During the batch annealing process, the aluminium coils remain separate while they are heated, in batches, in specific furnaces to strengthen the material before passing through a pre-treatment line (see Figure 3).16 However, the production of 5xxx series through batch annealing is an option much less pursued in the industry as customers of Aluminium ABS typically prefer or even require continuous annealing treatment, due to possible quality issues.17 Further, the Parties in their internal considerations of market capacity regularly consider CASH line capacity as the relevant metric. Quality issues and the Parties’ internal consideration of capacity is further discussed in recitals (580) to (584) .
(57) A CASH line can interchangeably produce both 5xxx and 6xxx series. Nonetheless, there are certain differences between the production of 5xxx and 6xxx series in CASH lines. 6xxx series need to undergo additional procedures during production, and may therefore need to run through a CASH line more than one time, which leads to slower production compared to 5xxx series. 6xxx series also exhibit higher scrap rates during production.
(58) Conversely, there are certain similarities in the production of 6xxx alloys used for the exterior (‘6xxx-skin’) and 6xxx alloys used for the interior (‘6xxx structure’). Both 6xxx series run through the production line at a similar speed. One difference is that 6xxx-skin series have a higher scrap rate than 6xxx structure series during the production process. The production of 6xxx-skin series must ensure that all surface defects are removed. 6xxx–skin’s surface quality is also rougher, which requires more lubrication during production compared to 6xxx structure series. Despite these differences, 6xxx-skin and 6xxx structure series are produced with the same equipment. 5xxx alloys are only used for manufacturing non-visible vehicle components.
5.4. Procurement of Aluminium ABS
(59) The procurement of Aluminium ABS usually takes place through requests for quotations (‘RFQs’) and bidding procedures organised by automotive original equipment manufacturers (‘OEMs’). Sales of Aluminium ABS are also made to tier components suppliers that process Aluminium ABS for OEMs (‘Tier suppliers’) and to a smaller extent to distributors.
(60) Tier suppliers are those customers of Aluminium ABS manufacturers that apply further manufacturing steps on Aluminium ABS and sell the resulting products to OEMs.18
(61) The development and production of a vehicle by an OEM includes a number of subsequent steps from designing to actual production.19
(62) An OEM typically decides on the use of Aluminium ABS in a vehicle platform during the so-called design stage, which can start about five years before the start of production. At this stage, the OEM designs the vehicle according to various commercial, legislative, technical and industrial requirements.
(63) After the design stage, an OEM typically organises a bidding process and issues formal RFQs specifying the parts, alloys, and volumes desired and asking its qualified Aluminium ABS suppliers to quote.
(64) The Notifying Party has explained that whilst each OEM has a different purchasing strategy, OEMs usually organise separate tenders for each vehicle model. OEMs tend to split different components among suppliers after a tendering process. However, in some cases, OEMs may group parts together. This would for example take the form of grouping more attractive higher volume parts together with less attractive lower volume parts to ensure balance. Depending on their preferences, some OEMs may prefer to source certain components from the same supplier, whereas others may prefer to have more than one supplier for a given component.20
(65) In response to an RFQ, suppliers prepare and submit their offers. They typically offer a separate price quote for each component of the tender unless the OEM groups parts together.21
(66) European OEMs typically qualify several Aluminium ABS suppliers22 for each part or group of parts of the vehicle and issue RFQs to qualified suppliers only.23 A competitor explains that ‘[e]xclusive suppliers do still exist but to a limited extent as OEM[s] want to secure supply chain by qualifying multiple ABS producers’.24 Similarly, a customer explains that ‘the Company tries to avoid being too dependent from one supplier, and tries to balance the sources’.25
(67) Qualification or homologation of an Aluminium ABS supplier and of its products is a process that aims at ensuring that each Aluminium ABS product offered by a supplier in a tender meets certain desired technical characteristics.26 The qualification process of a certain product is valid only for a specific process route in a plant (specific hot rolling line, cold rolling line and CASH line)27 and may take up to 2 years, or, in some cases, longer. The qualification of the suppliers, which is additional to product qualifications, ensures that certain criteria in terms of, for example, financial stability and supply reliability are met by each supplier.
(68) However, the qualification of a supplier and of its products does not guarantee that a certain manufacturer will be chosen as a supplier. An OEM28 explained that, in selecting the supplier, ‘in addition to the price, the Company also considers other important factors related to the potential suppliers, including: their financial health, their strategy, their manufacturing capabilities and capacities’. And that ‘the Company considers the suppliers’ future available capacities, i.e. the capacities available when the aluminium FRP are expected to be manufactured and needed for the production of the respective vehicle’. Supply capacity, which in the present case is closely linked and can be approximated to CASH capacity,29 is a key requirement for Aluminium ABS suppliers to be able to compete for a customer. There is no point in competing in a given tender if an Aluminium ABS supplier has no capacity available to supply the customer.
(69) The tender process can involve a number of rounds of bidding and take some months to conclude. During a tender process, customers may provide feedback to Aluminium ABS suppliers on the price level they expect. Suppliers may then decide to discount their offer accordingly.30 The final decision is normally made no later than two years before the start of vehicle production. After the final round of offers, the OEM evaluates the bids received and nominates the selected supplier(s).
(70) Once the vehicle model is fully designed and prepared, it can be launched and put to series production. The total duration of the production run can vary but can typically be 5 to 7 years.
(71) The commercial relationship between the OEM and the winning Aluminium ABS supplier can be governed either by a specific supply agreement signed post tender or by the terms and conditions set out by both parties during the tender procedure. For example, […].31
(72) Typically, following an agreement, the Aluminium ABS supplier would deliver based on periodical purchase orders sent by the OEM.32 […].33
(73) Quantity deviations from the long-term strategic planning are possible. […].34
(74) Prices are typically composed by the London Metal Exchange (LME) price for primary aluminium,35 conversion revenues36 and other price components (related to, for example, metal freight to the supplier’s casting plant, transport to the OEM’s or the Tier supplier’s press shop, etc).
(75) The agreement between the OEM and the winning supplier typically specifies the conversion prices for the lifetime of a vehicle program, but further adjustments are not excluded.
(76) Supply agreements typically run for the estimated production run of the vehicle model. The termination of an agreement before the end of the agreed term or before the end of the renewal period can be possible in certain circumstances. However, if one of the parties terminates an agreement before it lapses, it can be obliged to compensate the other party. In the event of termination of a purchase order, OEMs can be liable for costs related to work in progress and raw materials acquired.37
(77) Agreements often include so called […] .38 […].39
(78) Supplies to Tier suppliers can take place on terms negotiated between an OEM and the Aluminium ABS supplier. However, Aluminium ABS suppliers and Tier suppliers also negotiate agreements independently from OEMs.40 According to the Parties, […].41 According to a competitor, ‘[t]o deliver ABS to Tier-suppliers the material also has to be qualified by the OEM’, therefore, ‘the Tier-supplier can purchase ABS only from a list of suppliers approved by the OEM’.42 Several Tier suppliers have confirmed that they receive Aluminium ABS only from suppliers that have been previously qualified by the OEM.43
(79) Agreements between Aluminium ABS and Tier suppliers are usually signed for a period of one to two years.44
(80) Aluminium ABS suppliers can also sell their Aluminium ABS to distributors. According to a competitor, ‘[t]oday some distributors also focus on automotive and are able to distribute all kind of automotive parts whereas skin and closures are the most demanding ones’ and ‘[a]n OEM typically decides to use distributors and purchase from them if the purchased quantity is below a certain amount’.45
5.5. Trends and industry requirements in Aluminium ABS
(81) Aluminium use in passenger cars has been growing over the past years. The amount of aluminium in an average car has increased from 50 kg in 1990 to about 150 kg today.46 Despite this, steel remains by far the most prevalent material in passenger cars.
(82) As explained by the Notifying Party, the increased use of aluminium has primarily been driven by more demanding carbon dioxide (CO2) emission standards worldwide. To meet these emission targets, OEMs are required to develop vehicles that are more fuel-efficient. The use of lighter materials, such as aluminium, plays an important role in this effort because it helps reduce the weight of the vehicle and thereby fuel consumption and CO2 emissions.47 This is generally referred to as ‘light weighting’ in the automotive industry.
(83) Therefore, as acknowledged by the Notifying Party, price is only one of many dimensions of competition between Aluminium ABS and steel. Aluminium has certain technical advantages over steel, such as its strength-to-weight ratio. Nevertheless, using steel has other advantages, including lower cost.48
(84) A customer explained that ‘[r]easons for the choice of aluminium are mainly light- weighting and, consequently, reduction of CO2-emissions as well as fuel consumption’.49 Another customer50 explained that ‘[u]sing aluminium ABS results in costs that are 2 to 3 times higher than by using steel ABS’ and that ‘[t]he choice of aluminium ABS is therefore not driven by costs, but by weight reduction requirements’. The same company considers ‘the reduction in the weight of cars as a means to comply with stricter CO2 emissions regulation’.
(85) The Notifying Party itself expects the overall aluminium content per vehicle to grow in the future.51
Figure 4 […]
[…]
(86) In the EEA, the growth of aluminium content in vehicles is driven by new CO2 emissions performance requirements for new passenger cars and light commercial vehicles in order to contribute to achieving the EU’s target of reducing its greenhouse gas emissions and the objectives of the Paris Agreement.52
(87) Union legislation sets mandatory emission reduction targets for new cars since 2009.53
(88) The 2009 Regulation on CO2 emissions standards establishes a target of 130 g CO2/km that applies since 2015 for the EU fleet-wide average emission of new passenger cars. In 2017, the average emissions level of the new cars registered in the EU was 118.5 g CO2/km. Since 2010, average emissions have decreased by 22 g CO2/km (15.5%).54
(89) From 1 January 2020, Union standards set an EU fleet-wide target of 95 g CO2/km for the average emissions of new passenger cars and an EU fleet-wide target of 147 g CO2/km for the average emissions of new light commercial vehicles registered in the EU. Stricter EU fleet-wide targets will apply from 1 January 2025.55 Therefore, the new EU standards set an EU fleet-wide target from 1 January 2020 that will be 35 g lower than the one that applies currently. As the Notifying Party explained,56 these CO2 emission targets are adjusted to the average weight of each OEM’s fleet. In particular, the heavier the fleet of an OEM, the less stringent the target. Nevertheless, although different targets apply to different OEMs, all of them have to comply with stricter and stricter CO2 emission limits.
(90) OEMs and competitors also expect Aluminium ABS demand to grow in the future because of stringent emission policies that incentivise the production of lighter cars.57 In this regard, one customer said that ‘[a]luminium is the material to choose in order to comply with the European policies to reduce fuel consumption, CO2 emissions’.58 Another customer explained that ‘the demand for aluminium ABS is forecasted to increase because the Company and most (if not all) of its OEM competitors are expected to increase aluminium ABS demand in the years to come, due to CO2 emission regulations’.59 A competitor said that demand for Aluminium ABS ‘is increasing by up to two-digit figure annually’ and that ‘[t]he growth is expected to continue in the short to medium term, due to the need to light-weight cars and lower their fuel consumption.’60 The same competitor explained that the increase in the demand of Aluminium ABS ‘is not driven by increasing total car production but rather driven by switching to aluminium from steel in car production especially by hang on parts (which are more stiffness driven than strength driven) for which there is – due to stiffness requirements - a physical thickness limit on steel’.
(91) In addition to the EEA, other jurisdictions such as the US and China are increasingly tightening fuel efficiency standards in response to increasing environmental problems. In the case of China, current standards are less strict than, for example, in the EEA, but future stricter standards respond, among other factors, to a heavy reliance on overseas energy and urban traffic congestion.61
Figure 5 […]
[…]
6. RELEVANT PRODUCT MARKETS
6.1. Categories of Aluminium FRPs
(92) Aluminium FRPs are a group of flat aluminium products that are used for a multitude of different applications. In previous decisions, the Commission has concluded that not all Aluminium FRPs belong to the same relevant product market due to supply- and demand-side considerations. In particular, the Commission has concluded that Aluminium FRPs used for certain applications constitute distinct product markets. These include: (i) beverage can bodies; (ii) beverage can ends; (iii) food cans; (iv) lithographic sheet; (v) aluminium foil; and (vi) automotive sheet62,63 In addition, the Commission has considered that Aluminium FRPs used for a number of other applications, such as for aerospace, may constitute distinct product markets but has left the question ultimately open.64
(93) Further, the Commission has concluded in previous decisions that a distinct product market exists for Standard FRPs. That market has been considered to include all Aluminium FRPs that do not constitute separate products markets (see recital (92)).65
(94) The present Decision concerns Aluminium ABS, which are used in the automotive industry to produce the BiW (see Section 5.3) and Standard FRPs, including its potential sub-segments aluminium anodising sheet and aluminium sheet for compound tubes.
6.2. Aluminium ABS
6.2.1. Aluminium ABS belongs to a market separate from other Aluminium FRPs
6.2.1.1. The Notifying Party’s view
(95) The Notifying Party submits that Aluminium ABS belongs to a relevant product market separate from other Aluminium FRPs.66
6.2.1.2. The Commission’s assessment
(96) The results of the market investigation do not call into question the previous practice discussed in Section 6.1 and the Notifying Party’s submission that Aluminium ABS belongs to a market separate from other Aluminium FRPs. Supply- and demand-side substitutability between Aluminium ABS and other types of Aluminium FRPs appear limited.
(97) From a supply-side perspective, the production of Aluminium ABS requires special equipment and knowhow compared to other types of Aluminium FRPs, as already explained in Section 5.3. In particular, the production of Aluminium ABS requires (i) (continuous) annealing and (ii) pre-treatment of the coil, which are additional steps to the other production steps (such as casting and rolling) needed for the production of all Aluminium FRPs.67
(98) Furthermore, the pattern of supply is different between Aluminium ABS and other Aluminium FRPs. In this respect, the Commission observes that only certain Aluminium FRP manufacturers also supply Aluminium ABS. For example, there are a number of manufacturers that compete with the Parties for more common types of Aluminium FRPs,68 but only few of them, namely Constellium, Hydro, AMAG, and, to some extent Profilglass and Alcoa/Ma’aden69 also manufacture Aluminium ABS (see Section 8.3.5).
(99) From a demand-side perspective, the results of the market investigation indicate that automotive OEMs and their Tier suppliers, which are also customers of the Parties, cannot substitute Aluminium ABS with other Aluminium FRP developed for other purposes. The specific requirements of automotive OEMs relate to, for example, the strength, formability as well as surface quality and finishing of the product. All these properties are typically specified by each OEM, and Aluminium ABS manufacturers need to develop specific products or customise their previously developed products for addressing these technical specifications.70 Both OEMs and Aluminium ABS manufacturers will have to undergo a lengthy series of tests (the so-called ‘homologation’ or ‘qualification’ process of a product and production chain), where OEMs ensure that the products conform to their specifications.71
(100) Finally, the Commission observes that Aluminium ABS suppliers such as the Parties are able to identify automotive customers and separate them from customer groups requiring other Aluminium FRPs, such as for example those active in aerospace, in construction and in other industrial segments.
(101) The manifest differences between these various customer groups are also reflected in the organisational structure of Aluminium ABS manufacturers. In the case of Novelis and Aleris, for example, each of them have dedicated business divisions within their industrial groups, which are dedicated exclusively to serving automotive customers with Aluminium ABS. […].72 None of these business divisions appear to be serving any other customer groups.
(102) The Commission further notes that trade between customer groups (that is, for example, between customers active in aerospace, or in automotive) and arbitrage by third parties is likely hampered by the customers’ specific requirements. Specific requirements exist for the products they source, as well as for the suppliers they source them from, and a strict qualification process is associated to the products supply chains, as well as to the suppliers (see recital (99)). This supports the Aluminium ABS suppliers’ ability to treat automotive customers as a distinct customer group.
6.2.1.3. Conclusion
(103) For the reasons set out in this Section 6.2 and considering all evidence available to it, the Commission concludes that Aluminium ABS belong to a relevant product market separate from other types of Aluminium FRPs.
6.2.2. Aluminium ABS and flat steel products used in automotive bodies belong to different markets
6.2.2.1. The Notifying Party’s view
(104) The Notifying Party submits that Aluminium ABS and steel products for similar automotive applications are substitutable and compete with each other and, hence, together constitute a single product market.73
(105) The Notifying Party argues that the definition of one single product market for both Aluminium ABS and flat steel products used in automotive bodies would be consistent with a Commission’s precedent, and would be justified by the competition that takes place during the design stage of a vehicle between Aluminium ABS and flat steel products used in automotive bodies. In particular, the Notifying Party disagrees with the Commission’s findings that CO2 emission regulations are the main drivers for OEMs to choose Aluminium ABS, and considers that such a choice is also the result of price competition between Aluminium ABS and steel products.74 According to the Notifying Party, the price difference between Aluminium ABS and flat steel products used in automotive bodies does not preclude price competition, because Aluminium ABS compete with steel products based on the ‘cost per kg- saved’.75
(106) The Notifying Party also submits that employing Aluminium ABS is only one of the alternatives that OEMs may pursue for reducing CO2 emissions. According to the Notifying Party, OEMs can reach their CO2 emission targets by using measures that are alternative to light-weighting in a vehicle body, and, in case they decide to lightweight the BiWs of their vehicles, weight reduction can also be achieved by using modern steel products76 or other aluminium products, such as extruded or cast products.77 In support of this argument, the Notifying Party provides two examples showing that comparable vehicle weights can be achieved with different levels of Aluminium ABS contents.78
(107) The Notifying Party also submits that the Commission’s SSNIP test is flawed and that on proper construction, it does not support the finding that Aluminium ABS belongs to a different product market than steel products for the same applications.79
(108) As regards the argument that there would be customer groups different from automotive OEMs which may not be able to switch to steel products, the Notifying Party submits that Tier suppliers and distributors do not generate demand for Aluminium ABS because OEMs negotiate with Aluminium ABS suppliers on their behalf, and therefore any potential demand-side substitutability related to these customers is not relevant.
(109) In support of its argument that Aluminium ABS and flat steel products used in automotive bodies belong to the same product market, the Notifying Party provides examples of vehicle parts that were previously manufactured using Aluminium ABS, and in more recent versions of the same vehicles are manufactured using steel products instead,80 as well as examples of benchmark exercises that Aluminium ABS manufacturers make versus steel products and vice-versa.81
6.2.2.2. The Commission’s precedents
(110) In the early Alcan/Pechiney (II) case, which is from 2003, the Commission noted that Aluminium ABS was ‘a new, nascent application’. While the Commission considered that Aluminium ABS and respective flat steel products seemed to – ‘for the time being’ – belong to the same relevant product market, 82 it nonetheless left the question open and assessed the effects of the remedy in that case on a pure Aluminium ABS market excluding steel products.83
(111) For completeness, in the recent Tata Steel/ThyssenKrupp/JV case, the Commission concluded that aluminium products do not belong to the same relevant product market with automotive hot-dip galvanised steel, the type of steel predominantly used in the construction of vehicle bodies.84
6.2.2.3. The Commission’s assessment
(112) For the following reasons, the Commission considers that, contrary to the Notifying Party's submission, Aluminium ABS and flat steel products used in automotive bodies are not in the same relevant product market. Overall, steel and aluminium have different physical and commercial characteristics. There is no supply-side substitutability and demand-side substitutability is limited at most.
(6.2.2.3.1) The definition of separate product markets for Aluminium ABS and flat steel products used in automotive bodies is not inconsistent with the Commission’s precedent
(113) In Alcan/Pechiney (II) the Commission stated that ‘for the time being’ – that is in 2003 – aluminium and steel ABS ‘seem[ed] to’ belong to the same relevant product market, but it did not reach a definitive conclusion on this, and ultimately left the product market definition open.
(114) In Alcan/Pechiney (II), the Commission concluded that ‘[…] even if automotive aluminium sheet were to constitute a separate product market, any competition problems would be solved by the remedies offered by Alcan in relation to other FRPs product markets’.85 Therefore, in that specific context, the Commission did not initiate proceedings pursuant to Article 6(1)(c) of the Merger Regulation and therefore did not have the opportunity to conduct an in-depth (Phase II) investigation of the market for Aluminium FRP.
(115) Moreover, as explained in Section 5.5, since 2003 substantial changes in the market conditions occurred, in terms of, for example, CO2 emission regulations and the related drivers for OEMs to employ Aluminium ABS. Therefore, the Commission’s findings in the present case are not inconsistent with the findings in Alcan/Pechiney (II), if the different market conditions are taken into account.
(116) With respect to the recent Tata Steel/ThyssenKrupp/JV decision, the Notifying Party points out that it has no insight into the arguments and evidence proffered in that decision.86 The Commission notes in this respect that the findings in the present case are not based on evidence obtained in case M.8713 – TataSteel/ThyssenKrupp/JV. The Commission merely notes, for completeness, that, as indicated in the public press release, the Commission deemed that a market for automotive hot-dip galvanised steel was a distinct one, and that the press release makes no reference to an overall automotive body sheet market that would consist of both steel and aluminium. To that effect, the Commission recalls that the press release titled ‘Mergers: Commission prohibits proposed merger between Tata Steel and ThyssenKrupp’ states that ‘[t]he Commission had serious concerns that the transaction as notified would have resulted in a reduced choice in suppliers and higher prices for European customers of […] automotive hot dip galvanised steel products, where the proposed merger would have eliminated an important competitor in a market where only a few suppliers can offer significant volumes of this steel’.87
(6.2.2.3.2) Aluminium ABS and the respective steel products are characterised by different conditions of supply and by a lack of supply-side substitutability
(117) The Commission observes that there is no supply-side substitutability between Aluminium ABS and flat steel products used in automotive bodies, and that the patterns of supply and conditions of competition are in general different.
(118) First, steel and aluminium are different metals, they have different physical characteristics and they require different equipment to manufacture at all stages of the production chain. Equipment used to produce either Aluminium ABS or flat steel products used in automotive bodies cannot thus be used to produce the other product.
(119) Second, patterns of supply between Aluminium ABS and flat steel products used in automotive bodies are different. In this respect, the Commission notes that, for example, the Parties manufacture and supply Aluminium ABS in the EEA, but neither of them manufactures or supplies flat steel products used in automotive bodies. The same is true for the main competitors of the Parties, namely Constellium, AMAG,88 and Hydro,89 as well as for recent entrants Alcoa/Ma’den90 and Profilglass.91
(120) As to steel, the type of steel predominantly used in the production of vehicles is (hot- dip galvanised) flat carbon steel. The main manufacturers and suppliers of that type of steel in the EEA include for example ArcelorMittal, ThyssenKrupp, Tata Steel, Voestalpine, Salzgitter and SSAB.92 None of these companies manufactures Aluminium ABS. […].
(121) Third, the Commission observes that the Parties' […] regularly benchmark themselves in particular against other Aluminium ABS suppliers. […].93 […].94 […],95 […].
(122) […];96 […].97
(123) […].
Figure 6 […]
[…]
(124) Fourth, Aluminium ABS and flat steel products used in automotive bodies are subject to very different trade defence instruments (‘TDIs’) affecting imports.98
(125) Currently, safeguard measures99 apply to 28 categories of steel products, including flat steel products for automotive applications, to limit the increase of imports to a level that is unlikely to cause serious injury to the European Union industry while ensuring that traditional trade flows are maintained and existing user and importing industry sufficiently supported.100 As also stressed in the recitals to the relevant safeguard measures regulation,101 these safeguard measures have been adopted due to the adoption of safeguard measures by other countries outside the EEA.
(126) Furthermore, definitive anti-dumping measures102 are imposed amongst others on certain corrosion resistant steel products originating from China, including hot-dip galvanised flat carbon steel used for automotive applications.103
(127) With regard to aluminium, definitive anti-dumping measures are currently imposed on certain aluminium foils104 and certain aluminium road wheels.105 However, there are no safeguard or anti-dumping measures currently affecting Aluminium ABS in particular. Although safeguard measures have been adopted by some countries outside the EEA with regard to certain aluminium products (as for certain steel products), the Commission has not currently imposed any safeguard measures on Aluminium ABS.
(128) Contrarily to the Notifying Party’s argument that TDIs are not relevant in terms of supply-side substitution because the relevant geographic market for the production and supply of Aluminium ABS is the EEA,106 the fact that imports of Aluminium ABS and flat steel products used in automotive bodies are subject to different TDIs suggests that the supply conditions in the EEA are also different. In particular, this confirms that also when assessing demand and supply conditions in view of potential reactions to external shocks in the supply of these products from outside the EEA, TDIs are targeted specifically at each of these product groups and no knock-on effects are expected.
(6.2.2.3.3) Demand-side: distinction between design- and production phase of a vehicle model
(129) While the Notifying Party mainly alleges demand-side substitutability on automotive applications for steel and aluminium, it is apparent that the alleged possibility of technically replacing parts made from two different materials does not result in finding that those two materials belong to the same relevant product market.
(130) In this respect, the Commission observes that two different situations need to be distinguished: (i) switching materials during the production phase of a vehicle model and (ii) switching materials during the design phase of a vehicle model.
(6.2.2.3.4) Demand-side: No substitutability during production phase
(131) As to switching materials during the production phase of a vehicle model, the results of the market investigation show that switching is in practice usually not possible or at least very difficult and costly. In practice, car manufacturers cannot substitute aluminium and steel during a car model’s production cycle (typically approximately 5–7 years).107 This relates to the fact that a vehicle model is designed with a particular material in mind and switching the material would often require both the re-design and re-testing of the vehicle, including crash tests depending on the component in question, as well as reworking vehicle production.
(132) To this effect, the clear majority of automotive customers responding to the market investigations stated that their ability to substitute Aluminium ABS for other materials during the production phase is at best ‘very limited’.108
(133) A major OEM explains: ‘A switch from a material to another (i.e. switching from aluminium to steel) has to be made before certification of the carline (e.g. crash, WLTP…) - - Switch between materials can happen during facelifts for design purpose mainly because a material offers more shaping flexibility. Any switch in material would also be hampered by the fact that the crash certification would have to be retaken’.109 That same OEM further clarifies that it ‘does not intend to switch back to steel after a price increase for aluminium ABS’. Another OEM further explains that during the production phase, a ‘change from aluminium to steel would mean major investment in the press shop (new dies), body shop (joining technology) and others. In addition it would not be possible anymore to implement closed-loop activities for aluminium which means an negative impact on product sustainability’.110 A major automotive OEM concurs: ‘Switching materials becomes very costly after the stamping tools have been manufactured and is only undertaken in exceptional circumstances.’111
(134) In practice, switching from Aluminium ABS to flat steel products used in automotive bodies and vice versa is a long-term decision that is taken at the design stage way before a tender procedure to source the material is launched. An OEM explains that a ‘change of the chosen material is easier during the engineering phase. Nevertheless, there is still some flexibility for switching before the start of production'.112
(135) For automotive manufacturers who are planning upcoming tenders that would be nominated in the next few years, the engineering has already been done and even if the production cycle has not started yet, it is thus unlikely that they can seamlessly switch from Aluminium ABS to flat steel products used in automotive bodies.
(136) The Notifying Party claims that the lack of demand-side substitutability at production phase is not relevant because prices for Aluminium ABS are defined before the production phase, and are fixed through a price formula that remains unchanged during the entire production phase.113 However, although Aluminium ABS prices cannot be increased during production phase, in principle, they could be reduced (see Sections 5.4, recital ). Therefore, contrarily to the Notifying Party’s claim, the lack of demand-side substitutability between Aluminium ABS and flat steel products used in automotive bodies at production phase is relevant for the purpose of defining a relevant product market for the production and supply of Aluminium ABS.
(6.2.2.3.5) Demand-side: Potential substitutability during design phase does not warrant finding of a combined market for Aluminium ABS and flat steel products used in automotive bodies
(137) Based on the results of the market investigation and the submissions of the Notifying Party, switching between Aluminium ABS and flat steel products used in automotive bodies is in principle possible during the design-phase of a vehicle. However, the Commission finds that such a possibility is not sufficient to substantiate a finding of a single product market for these materials, and that also in the design phase customers’ ability to switch is limited, even in the event of a price increase of aluminium ABS compared to steel products used in automotive bodies.
(138) First, the choice of employing Aluminium ABS instead of flat steel products used in automotive bodies is mainly driven by the need to comply with CO2 regulations, rather than their relative prices.
(139) An OEM typically chooses to employ Aluminium ABS for reducing the overall weight of its vehicles, and, ultimately to comply with CO2 emission regulations.114 As explained in Section 5.5, reducing the weight of a vehicle has the benefit of reducing its fuel consumption, and, consequently, the benefit of reducing its CO2 emissions. When aggregated for all the vehicles sold by a given OEM in the EEA, the reduction in average CO2 emissions allows that OEM to comply with CO2 emission regulations. […].115
(140) OEM respondents to the market investigation provided extensive evidence that the choice of employing Aluminium ABS is driven primarily by CO2 regulations. An OEM states that ‘[a]luminium is definitely lighter than steel, this supports the CO2 lightweight strategy’116. Elaborating further, the OEM says that ‘[t]he choice of aluminium ABS is […] not driven by costs, but by weight reduction requirements. The Company considers the reduction in the weight of cars as a means to comply with stricter emissions regulation'.117 Another OEM mentions that ‘[a]luminium is more and more used in our vehicles to shave off kilograms (light weighting), necessary to meet upcoming regulations’118 and further that ‘most of the time the choice of aluminium is highly related to CO2 concern’.119 Another major Aluminium ABS customer explains why Aluminium ABS is a material of choice with it being a ‘[w]eight saving driven decision’.120 Putting it straight forward, another OEM states that ‘aluminium is needed to meet CO2 emission targets’121 and that ‘[t]he demand of aluminium ABS is driven by the move in the industry to reduce emission by reducing car weight’.122 Another OEM also states that the ‘growth in aluminium FRP usage can be attributed to the light weighting need as well as increasing electrification, both in order to comply with the regulatory emissions limits’.123 A further OEM explains that ‘[l]ightweight is important because [the company] has to fulfil legislative requirements with regard to CO2-emissions and fuel consumption, and reducing the weight of a vehicle is major way for achieving these targets’.124 Phrasing it in clear terms, another OEM says that ‘[a]luminium is the material to choose in order to comply with the European policies to reduce fuel consumption, CO2 emissions'.125
(141) The Parties’ competitors seem to share the same view regarding the motivation of an OEM to choose Aluminium ABS over flat steel products used in automotive bodies. One of the Parties’ competitors, for example, in one of its earning calls stated that ‘[…] the increased aluminium usage is a secular trend for these markets. Aluminium’s favourable strength to weight ratio in comparison to steel, enable’s OEMs to lightweight vehicles, thereby increasing fuel efficiency and reducing CO2 and other emissions. Aluminium, also a superior energy absorption properties as compared to steel […]’.126 Another competitor to the Parties further states that ‘[t]he demand for aluminium ABS will further increase also because of the goal to reduce CO2 emissions and to further lightweight cars’ and further that ‘[i]n Europe, regulation drives the increased use of aluminium in the automotive sector to reduce weight and lower emissions’.127
(142) A Novelis’ internal document […],128 […].
Figure 7 […]
[…]
(143) Furthermore, the Notifying Party itself has publicly highlighted the importance of CO2 regulation for the adoption of Aluminium ABS. To this effect, the Vice President Automotive of Novelis Europe, Michael Hahne, explains in an industry publication Automotive World (May 2018) that ‘with CO2 emission standards set to tighten worldwide, it’s safe to say the main driver [for the increased use of aluminium] today is regulation’. He also highlights Europe as the region with the most stringent rules, as it ‘has set a target of 95 grams CO2 per kilometre by 2020’.129
(144) […].
Figure 8 […]
[…]
Source: Reply to request for information 18, ‘NOV-EU00292913.docx’, DocID1022-35912.
(145) In line with CO2 emission regulations being the main factor behind the choice of Aluminium ABS over flat steel products used in automotive bodies, OEMs seem to be adapting the quantity of Aluminium ABS (as opposed to cheaper steel) in the vehicle models they supply according to the CO2 emission regulations in place in the regions where the vehicles are to be sold.
(146) […].
Figure 9 […]
[…]
(147) […].
Figure 10 […]
[…]
(148) […]130 […]131 […].
(149) […].132 […].
(150) Another example of the influence of regulation on the adoption of Aluminium ABS is provided by […] releasing in 2014 its […] vehicle in the EU market and in the United States market with wings,133 doors, bonnet,134 roof and boot135 lid made of aluminium.136 The same car model does not have these components made of aluminium when produced for the Chinese market,137 where CO2 regulation is less stringent than in the EEA and in the United States.138
(151) In the Reply to the SO, the Notifying Party questions the correlation between CO2 emission regulations and Aluminium ABS demand. This is because, according to the Notifying Party, CO2 emission regulations apply equally to all OEMs, but some OEMs use in their fleets a much larger share of Aluminium ABS, compared to others.139 In support of its argument, the Notifying Party provided a graph representing, for each OEM active in the EEA, its Aluminium ABS purchasing share as well as the number of vehicles that each OEM produces every year.140
(152) However, the Commission finds that the evidence offered in rebuttal, on a proper construction, is instead consistent with its own arguments, as it demonstrates that OEMs which are more important customers of aluminium ABS are the ones more exposed to regulatory requirements (as they produce and sell heavier and therefore more CO2-emitting vehicles). Moreover, evidence from the files indicates that, in view of ever more tightening regulatory requirements, also the other OEMs expect that their consumption of Aluminium ABS will have to substantially increase in the near future in order to be able to comply with such requirements.
(153) […]. Although the emission regulation applies on the basis of the entire fleet, and not at individual vehicle level, as already explained in the SO,141 OEMs typically use more Aluminium ABS for manufacturing premium-segment vehicles, such as large saloons and SUVs mainly because i) there is a more pronounced need to reduce weight in these vehicle categories, and therefore more CO2 saving benefits; and ii) the higher prices of these vehicles allow to afford the increased cost of Aluminium, compared to steel. This is well-explained by an OEM that stated ‘in lower vehicle segments (for example, segments A, B and, to some extent, C), the share of aluminium is expected to be lower than in higher segments, as for example segments D and above. This trend is also related to the size itself of a vehicle: higher segments would lead to an excessive weight if the share of aluminium is modest, whereas smaller vehicles can be realised with less shares of aluminium’.142
(154) Concerning the Notifying Party’s argument that Figure 11 does not explain why different OEMs focusing on high-end vehicles consume different amounts of Aluminium ABS,143 the Commission refers to recitals (175)-(176), which explain why different options for reducing CO2 emissions are not equally available to all the OEMs, and therefore each of them decides to use different amounts of Aluminium ABS.
Figure 11 […]
[…]
Source: Reply to the SO, Figure 1.
(155) A document produced by a consultant for Novelis […].
Figure 12 […]
[…]
(156) In the second place, CO2 emission regulations are expected to become even stricter in the near future. As explained in Section 5.5, as of 1 January 2020 the limit for EU fleet-wide average emission of new passenger cars will be reduced by about 27% from the current regulation (i.e. from the current 130 g CO2/km to 95 g CO2/km).
(157) In view of this, and consistent with the Commission’s analysis, also the OEMs that typically produce a higher number of smaller vehicles (that is, for example, PSA, Fiat and Renault-Nissan) and now consume a relatively smaller share of Aluminium ABS compared to flat steel products for automotive bodies, are expected to increase the share of Aluminium ABS also in vehicles of small size. An OEM that responded to the market investigation stated ‘[t]raditionally the Company has used aluminium predominantly for vehicles of large segments, however, it is now also using it for the [small segment car model] and in the near future also smaller car segments (B segment) will have hoods made from aluminium ABS’.144 Another OEM indicated that ‘at the moment [emphasis added], aluminium is mainly used in closures for premium vehicles’,145 thus alluding to the fact that in the future also non-premium vehicles will contain more aluminium. Another OEM that to a large extent produces small- and medium-size vehicles stated that ‘[t]he Company is currently using approximately 10 – 15 kg of aluminium per vehicle on average. The next generation of the Company´s “[vehicle name]” will contain approximately 100 kg of aluminium per vehicle. The increased usage of aluminium is a general trend among OEMs because aluminium is one of the best candidates for securing light weighting’.146 147
(158) An internal document of Novelis […].
Figure 13 […]
[…]
(159) In the Reply to the Letter of Facts, the Notifying Party argues that light-weighting is an inferior way for OEMs to reduce CO2 emissions, compared to other technologies. The Notifying Party observes that, according to the EU CO2 emission regulation, the CO2 emission target of each OEM is adjusted to the average weight of its fleet in a way that OEMs producing heavier vehicles have less stringent CO2 emission targets. As a consequence, if an OEM reduces the weight of its fleet as a consequence of light-weighting, also the CO2 emission target becomes more stringent, therefore it is more incentivised to find alternative technologies for reducing CO2, while maintaining the same average weight.148
(160) However the claim of the Notifying Party, in essence, does not question the need of OEMs to reduce the weight of their fleets for meeting their obligations in terms of emission limits. The Notifying Party’s claim regards only how effective weight reduction is, and does not question the fact that limited or, often no alternative solution is available to OEMs. As explained in recital (173), the availability of alternative technologies for reducing CO2 emissions is rather limited and, in most cases, require a longer time-frame and additional costs, compared to lightweighting. Consistently, the majority of OEMs that replied to the market investigation consider that alternative technologies either would not be cost-effective or would not be timely available.
(161) In addition to considerations related to CO2 emissions and light-weighting in combustion engine vehicles, regulatory pressure to reduce CO2 seems to drive the adoption of Aluminium ABS adoption in a further, indirect way.
(162) Electric vehicles (‘EVs’) are a vehicle category that avoids the CO2 emissions of traditional vehicles with internal combustion engines. Therefore, one might consider that the penetration of EVs into the automotive market might reduce the need for OEMs to reduce the emissions of their traditional combustion vehicles, and consequently reduce their need to employ Aluminium ABS. […].149 […].150 […]151 […]152,153 […].154
(163) The Notifying Party claims that the influence that EVs are expected to have on the Aluminium ABS demand is unclear because, […]155, […].156
(164) In the Reply to the SO, the Notifying Party submitted a document […].157 […].
(165) Further evidence that a growing share of electric vehicles does not result in less demand for lightweight materials such as Aluminium ABS, is also included in third- party reports and studies on the matter. A McKinsey report states that ‘[w]hile share of lightweight materials could further decrease in BEVs, overall demand for lightweight could further increase until 2030’ and indicates a positive ‘[l]ightweight material demand’ trend for hybrid vehicles.158 The report also summarises that overall the trend to electrification will not reduce the demand for lightweight materials by stating that ‘[e]xpected increase of lightweight materials share and demand by 2030, driven though their use in ICEs and hybrids’, while also seeing a ‘[l]ikely decline of lightweight materials for BEVs in the next years, along with declining battery prices and increasing battery density’. In addition, and despite the foreseen electrification trend, material demand forecasts by third party providers continue to project Aluminium ABS content in cars and Aluminium ABS demand to increase. This is for example evidenced in the IHS Markit Forecast from August 2018, captioned in Figure 14 (see aluminium sheet’s growing figure), and the Notifying Party’s own projections as detailed, for example, in Figure 15.
(166) The Ducker report ‘Aluminium Content in European Cars’ from June 2019 also states that ‘Extrusions and sheet will win shares by 2025, mainly driven by Electrification components and Body Closures’ and further that ‘[e]lectrification components, Body Closures and Body Structure are expected to be the main growth areas by 2025’.159 While electrification components may be either made out of Aluminium ABS or other aluminium products (such as extruded products), this assessment shows that the aluminium content of Closures (very high share of Aluminium ABS), Structure (considerable share of Aluminium ABS) and electrification components (Aluminium ABS used for example in battery cases) are all expected to increase.
(167) Despite cost considerations, there are other reasons why OEMs are expected to continue to prefer, to some extent, Aluminium ABS in the future. […].160
(168) […],161 and this appears to be consistent with the expectations of many market participants, suggesting that EVs will contribute to the increase in demand for Aluminium ABS.
(169) An OEM explains that ‘[g]reater adoption of fully electric car platforms for example could see higher usage of aluminium ABS at the [Company’s group] generally’.162 Another major OEM ‘considers that aluminium ABS demand will increase due to more stringent emissions regulations, and to more widespread production of electric vehicles’.163 Another OEM ascribes the foreseen increase in Aluminium ABS demand to ‘the light weighting need as well as increasing electrification’.164 In anticipating potential own increased needs for Aluminium ABS, another OEM mentions ‘the proliferation of electric cars and the tendency to cleaner policies that incentive [sic] lighter cars’.165 Similarly, a competitor mentions that ‘[w]ith technology changing away from petrol-powered towards hybrid, all-electric or hydrogen powered cars, the demand of aluminium ABS from OEMs will increase.’166
(170) In addition to regulation compliance, OEMs’ decision to reduce the weight of their vehicles through Aluminium ABS leads to additional benefits, which include engine downsizing (which is translated into reduced costs and reduced fuel consumption and CO2 emissions), vehicle performance improvements, and, in some cases, specific design optimisations.167
(171) […].
Figure 15 […]
[…]
(172) Second, while the Notifying Party submits that alternative methods to light- weighting in the BiW exist for automotive OEMs to reach their emission targets, the results of the market investigation do not support a finding that such alternative methods are always easily available, or that they can immediately replace Aluminium ABS as the means of choice to reduce a vehicle’s weight.
(173) In the first place, the results of the market investigation suggest that OEMs are not able to pursue alternatives to the usage of Aluminium ABS for complying with CO2 emission regulations in a timely manner and without major additional costs. That is to say that OEMs are generally unable to ‘achieve their CO2 regulatory targets with steel’,168 by pursuing options for reducing CO2 emissions that are alternative to Aluminium ABS.
(174) The clear majority of the OEMs that expressed their view in the response to the market investigation consider that today’s benefits of Aluminium ABS in terms of fuel consumption and CO2 emissions cannot be replaced in a timely manner and without major additional costs by other alternatives, as for example, by improving the engine efficiency, by electrification, by improving aerodynamics and rolling resistance, etc.169 An OEM explains in this respect that ‘no economically viable material alternative to aluminium is available to OEMs for reducing its vehicles’ weight. Composite materials, for example, a too expensive option, while high strength steel can replace aluminium only to a limited extent’.170
(175) In the Reply to the SO, the Notifying Party compares two car models using little or no Aluminium ABS […] .171 According to the Notifying Party, this is a proof that OEMs do not require Aluminium ABS to achieve their weight saving targets, and, more in general, to comply with CO2 emission regulations. Similar claims are also made by the Notifying Party in its Reply to the Letter of Facts.172
(176) […].
Figure 16 […]
[…]
(177) Further, in achieving its CO2 emission target, an OEM might have strategic decisions leading to rely on Aluminium ABS more heavily for some vehicle models, and less for other models. […].
(178) […].173 […].
Figure 17 […]
[…]
(179) The Notifying Party argued that Aluminium ABS accounts for only a fraction of the total aluminium content in a car, and therefore other aluminium products, such as extruded and casted aluminium products, represent additional alternatives to OEMs for reducing the weight of their vehicles.174 […]175 […].
(180) […].176
Figure 18 […]
[…]
(181) More generally, contrarily to the Notifying Party’s claim,177 a future stricter CO2 emission regulation is likely to increase the rigidity of demand for Aluminium ABS178 and does not necessarily make alternative ways for reducing CO2 emissions more attractive to them.
(182) Further, while certain automotive OEMs have exploited for instance diesel engine technologies in an effort to reduce CO2 emissions, such technology has recently suffered setbacks due to the discoveries related to high NOx emissions. In an internal document, Novelis considers this as an opportunity, as ‘[d]iesel scrutiny increase[s] pressure to explore other means to reduce CO2 (e.g. lightweight)’.179
(183) Third, even if other technical means to achieve reductions of CO2 emissions in vehicles existed, they would not necessarily determine that Aluminium ABS would be immediately substitutable with steel to the extent that they would belong to the same relevant product market.
(184) In the first place, as a matter of reasoning, the argument of the Notifying Party is not that of an immediate and direct substitution between Aluminium ABS and flat steel products used in automotive bodies. In particular, the Notifying Party seems to acknowledge that, even in the event of an increase of the price of Aluminium ABS, the likelihood of OEMs switching to flat steel products for automotive bodies would be dependent on their possibility to find alternative technical means to reduce CO2 emissions.180 The availability of these alternatives would have to be assessed on a case-by-case basis, and, in most cases such an availability appears to be very limited. A large majority of the OEMs that responded to the market investigation consider that ‘[…] all potential alternatives [for reducing fuel consumption and CO2 emissions] either would not be sufficient, or will require major additional investments and time’.181
(185) […].182 […].
(186) However, nothwistanding the lack of availaible technologies explained in recital (184), the claim of the Notifying Party suggests that Aluminium ABS would not be in competition with flat steel products for automotive bodies, but rather with alternative technologies that are needed to achieve CO2 emission reduction, while maintaining the vehicles’ weight. Therefore, the substitutability of Aluminium ABS for flat steel products for automotive bodies would not be a result of a change in their comparative price, other things being equal, but it would be contingent on the availability and comparative cost of other alternatives for reducing CO2 emissions.
(187) In the second place, the need to find alternatives in a vehicle design in order to offset a price increase in aluminium by sourcing steel is also, on a proper construction, evidence of a limited demand-side substitutability rather than an argument supporting the ease of switch by customers.
(188) Fourth, consistently with the considerations in recitals (129) to (187) and as documented in Annex I – which is an integral part of the Decision – Section 2.1.1, there are significant price differences between Aluminium ABS and flat steel products used in automotive bodies, with Aluminium ABS being around […] than steel based on total price according to the Commission’s calculations based on data from the Notifying Party.183
(189) The difference in prices remains very large even after adjusting for differences in densities between the two materials: after adjusting for a light-weighting factor reflecting the fact that aluminium has a lower density than steel and therefore less aluminium is needed compared to steel to produce a given component, the adjusted price of Aluminium ABS is still […]% more expensive than flat steel products for automotive bodies. 184
(190) Even so, purchasers of Aluminium ABS have in the past shown a willingness to pay such significant premium because of the significant benefits (in terms of compliance with CO2 emissions regulations) brought about by aluminium’s material properties. This is consistent with the observation that switching between flat steel products for automotive bodies and Aluminium ABS is not predominantly driven by small but significant changes in relative prices, but by more structural properties of demand (including the desire to meet CO2 emission regulations).
(191) Contrarily to the Notifying Party’s claim that Aluminium ABS and flat steel products used in automotive bodies compete on price based on the kg-saved,185 the price difference indicates that OEMs are willing to pay a higher price for Aluminium ABS because their employment is an effective way for complying with CO2 emission regulations. Therefore, this price difference between Aluminium ABS and flat steel products used in automotive bodies does not support a finding of them belonging to the same relevant product market. While the market investigation suggests that the cost for an OEM of using Aluminium ABS instead of flat steel products used in automotive bodies (including high-strength steels) varies across different vehicle models and components, the price difference appears to be always significant.
(192) According to the Notifying Party’s own explanation,186 aluminium, as a metal, is up to about […] lighter than steel per the same volume (that is, it is […] less dense). Nonetheless, for certain applications, different amounts of material may be needed in order to achieve certain desired material properties (comparatively more aluminium may be needed in order to achieve the same strength as steel, or in order to undergo similar manufacturing processes). These factors altogether lead to weight saving by using Aluminium ABS rather than steel products that varies across different vehicle components and typically amount to about […]% weight saving for structural parts, […]% weight saving for doors, and […]% for bonnets (Figure 19).
Figure 19 […]
[…]
(193) Even after considering the reduced mass of aluminium compared to steel, a component manufactured with Aluminium ABS is typically substantially more expensive than the same component manufactured with steel products. According to the Notifying Party, even in the case of bonnets, where the penetration rate of aluminium is relatively high and higher than for other car parts, a part made from Aluminium ABS costs […]% more than the same part made of steel (that is […] EUR, compared to […]EUR in Figure 20). This substantial cost difference appears to corroborate the fact that the choice of employing Aluminium ABS is mainly driven by emission regulations, rather than the relative price difference between Aluminium ABS and flat steel products used in automotive bodies.
(194) Figure 20 also shows […].
(195) Therefore, contrarily to what the Notifying Party claims,187 the cost per kg-saved is not the metric driving competition between Aluminium ABS and flat steel products for automotive bodies. The cost per kg-saved is only responsible for prioritising the use of Aluminium ABS on a component instead of another. In the case of Figure 20, […].
Figure 20 […]
[…]
(196) Customers of Aluminium ABS confirmed in the market investigation that Aluminium ABS are more expensive than comparable steel parts. An OEM states that ‘[u]sing aluminium ABS results in costs that are 2 to 3 times higher than by using steel ABS’.188 Another OEM mentions that a ‘component made of aluminium in most cases appears to be more expensive than the same made component of steel’.189
(197) In line with the price difference, the market investigation also indicates that substitution of steel with aluminium has especially taken place in premium-segment- vehicles, such as large saloons and sport utility vehicles (‘SUVs’), where there is a more pronounced need to save weight (due to their higher weight, compared to, for example A-segment vehicles), and where the cost increase associated with Aluminium ABS does not penalise the overall final costs and profits.190
(198) With respect to premium-segment vehicles, and in addition to lightweight benefits, the choice of an OEM to use Aluminium ABS in its vehicles might also be driven by the premium status associated with the use of aluminium. To this effect, an OEM respondent to the market investigation considers that its customers purchasing premium-segment-vehicles value the presence of aluminium in the vehicles they purchase.191
(199) The role of aluminium in premium-segment vehicles is further reflected by studies of current and forecasts of future aluminium content in vehicles in the EU28 by Ducker.192 The pre-final findings of its ‘Aluminium Content in Cars (EU28)’ study, dated 9 May 2019 and prepared for the European Aluminium Association, point to an aluminium content in body closures of premium segment vehicles (E segment) of
62.3 kg on average in 2019.193 This is significantly more than in other segments, the average content being 14.7 kg in the D segment (that is, more than four times smaller than in the E segment) and 0.5 kg in the A segment (that is, more than 120 times smaller than in the E segment).194
(200) Fifth, when assessing the competition that takes place after the design of the vehicle (procurement and eventually production), the competitive dynamics of the procurement of Aluminium ABS show that, typically, once an OEM has made the engineering choice for the material to use in a given component (that is Aluminium ABS), only Aluminium ABS manufacturers will be requested to bid and compete for the supply. An OEM explains that ‘[t]he choice of materials has to be decided at concept freeze during design phase’.195 The OEM further states that '[a]t design phase, i.e. when the Company decides to employ FRP made of steel or of aluminium, the price considered is the forecasted market price. The final price at which the Company sources FRP, however, is fixed afterward, i.e. as a result of the tendering process’.196 Another OEM explains that ‘[o]nce the decision of employing aluminium ABS is taken, procurement is made through a tendering process’,197 again confirming that the decision for the employment of Aluminium ABS precedes the competitive interaction between suppliers in the tender.
(201) Therefore, competition between Aluminium ABS suppliers – rather than competition with other materials – will have a direct impact on price and other sourcing conditions of a component that an Automotive OEM has decided to manufacture from Aluminium ABS.
(202) Contrarily to the Notifying Party’s argument that there are several instances where competition between Aluminium ABS and flat steel products used in automotive bodies takes place at RFQ stage,198 the few examples provided by the Notifying Party are either not relevant for the RFQ stage (but rather for the engineering/design stage), or are rare exceptions, and do not represent the Parties’ standard course of business. […]199 […].
(203) Sixth, the Notifying Party claims that Aluminium ABS suppliers benchmark their product prices against corresponding steel products, and that Aluminium ABS need to be (price) competitive against steel.200 In particular, the Notifying Party claims that […]. The Commission considers that this benchmark exercise is not sufficient to show that Aluminium ABS and flat steel products used in automotive bodies belong to the same relevant product market, but rather is an indication that, to a certain extent, steel price may be an external constraint on the Aluminium ABS market, and that Aluminium ABS need to have similar or superior physical characteristics compared to flat steel products used in automotive bodies.
(204) With respect to costs, as explained in recitals (188)–(199), for OEMs Aluminium ABS are substantially more expensive than flat steel products for automotive body parts, with respect to both product costs and final cost (that is the cost of the products plus the costs of processing them for creating automotive bodies). This cost difference indicates that the competitive constraint of flat steel products for automotive bodies on Aluminium ABS is relatively limited.
(205) In the first place, the Commission does acknowledge that OEMs, in trying to achieve the required light-weighting for their vehicle fleets, compare the light-weighting characteristics of Aluminium ABS and flat steel products used in automotive bodies also in relation to costs (price-per-kg-saved). Nevertheless, i) OEMs are seen to have a need for Aluminium ABS with limited constraints from its relative cost to steel products (see recitals (147), (173)–(199)), which is always higher and ii), the pricing constraint exerted by flat steel products used in automotive bodies on Aluminium ABS at the procurement stage due to a comparison between the price-per-kg-saved characteristics of Aluminium ABS and flat steel products used in automotive bodies at the design stage, is limited for the following reasons:
(206) […]201 […],202 […].
(207) […].
(208) […].203 […].
(209) […]204 […].
(210) Therefore, the various documents cited by the Notifying Party as […]205 […].
(211) […].206 […],207 […]. Therefore, the pricing constraint exerted by flat steel products used in automotive bodies on Aluminium ABS appears to be limited.
(212) […].208 […],209 […]. (213) […].
(214) […].
Figure 21 […]
[…]
(215) […].210 […].211
Figure 22 […]
[…]
(216) Similarly […].
Figure 23 […]
[…]
(217) […].212 […].
Figure 24 […]
[…]
(218) […].
(219) A document titled ‘European BIW Material Forecast’ by IHS Markit from August 2018, which is captioned in Figure 25, shows that an increase in the net usage of ‘Aluminium Sheet’ for light vehicles in the EU is forecast from 0.5 billion of lbs in 2017 to 1.0 billion of lbs in 2027. While Mild Steel and High Strength Steel decrease in importance, AHSS, Ultra High Strength Steel (‘UHSS’) and GigaPascal Steel are projected to increase in importance. This underlines that certain recent innovations in steel are cannibalising other steel products and not Aluminium ABS, which is projected to continue to grow in importance.213
(220) This is further reflected in data submitted by the Notifying Party to the Commission.214 In projecting the ‘material shares’ from 2018 to 2023, the Notifying Party submits that […].
(221) […].
Figure 26 […]
[…]
(222) […].
Figure 27 […]
[…]
(223) […].
Figure 28 […]
[…]
(224) In the fourth place, the Parties appear to perceive Aluminium ABS to have superior weight-saving characteristics compared to steel, even when considering new steel product developments and innovation in steel. […].215
(225) […].
Figure 29 […]
[…]
(226) […].
Figure 30 […]
[…]
(227) Seventh, while the Notifying Party submits that OEMs can and have switched back to steel from aluminium, the results of the market investigation show that, if anything, there is an overall tendency of replacing a part of the components previously produced in steel with aluminium.216
(228) In the first place, when an overall vehicle fleet-level view is taken (as opposed to a model-by-model view), no observation of switching back from aluminium to steel can be observed, thus indicating that Aluminium ABS and flat steel products for automotive bodies are not in the same relevant product market from this perspective.217 This finding is not brought to question by the examples provided by the Notifying Party regarding OEMs switching back from aluminium to steel for some components in their vehicles.218 Indeed, the relevant product market for the production and supply of Aluminium ABS is not defined narrowly for single components, therefore the examples provided by the Notifying Party, which are limited to some components of some vehicle models, concern only a part of the Aluminium ABS market and do not take into account, for example, if other components of the same vehicle, or if other vehicles of the same OEM have been manufactured with an increased share of Aluminium ABS.
(229) Figure 31 shows that in the period 2012–2017, EEA sales of Aluminium ABS increased by more than 140% (from about 200 000 tonnes in 2012 to about 494 000 tonnes in 2017), which is by far larger than the growth of newly registered vehicles in the EU, which was about 25% (from about 12 million units to about 15 million units). 219 This marked growth indicates the continuous tendency of OEMs to switch from steel to aluminium. This trend occurred despite any possible variation of market conditions that occurred in the same period.
(230) […],220 […].221
(231) […].
Figure 32 […]
[…]
(232) Novelis’ view is further confirmed by OEMs that expressed their view during the market investigation. One OEM considers that ‘the use of aluminium ABS is expected to increase in the future’.222 Another car manufacturer states that ‘[t]he increased usage of aluminium is a general trend among OEMs because aluminium is one of the best candidates for securing light weighting’. 223
(233) The point that switch-backs to flat steel products for automotive bodies may take place for individual models, but not for overall fleets is further evidenced by the fact that despite the case of the switch back from aluminium to steel […] ,224 […].225 This corresponds with […]'s expectation of increasing purchases of Aluminium ABS until 2020.226
(234) […].227 […],228 […].
(235) The Notifying Party argues that, in the context of considering alleged ‘switches back’ from aluminium to steel, a fleet-level view is not appropriate because competition between Aluminium ABS and flat steel products for automotive bodies […].229 However, the Commission recalls that, as explained in Section 5.5, recitals (87)– (89), and as acknowledged on several occasions by the Notifying Party,230 regulations on CO2 emissions are referred to fleet average emissions. Individual examples of a switch back or relative aluminium penetration in a specific vehicle model are linked to OEMs’ strategy to break fleet targets down to individual emission targets through the calculation of cost thresholds for accepting weight reduction measures as explained in Figure 33, and are not a sign of a general trend.
Figure 33 […]
[…]
(236) The Commission also recalls that, as stated in recital (89), from 1 January 2020, EU standards set an EU fleet-wide target for passenger cars that will be 35 g lower than the one that applies currently.231 Stricter EU fleet-wide targets will apply from 1 January 2025 and 2030. These targets are defined as a percentage reduction from the 2020 starting points. For the average emissions of the new passenger car fleet, 15% reduction from 2025 on and 37.5% reduction from 2030 on will apply.
(237) […].
Figure 34 […]
[…]
(238) […] Although the Notifying Party provides examples of third party reports that might have reviewed downwards their growth estimates for some Aluminium ABS products, those reports do not contradict the finding that the demand of Aluminium ABS is expected to increase, due to an increased penetration of aluminium content in vehicles produced for the European market.232
Figure 35 […]
[…]
Figure 36 […]
[…]
(239) Eighth, the quantitative evidence available from Novelis’ data and estimates in the ordinary course of business do not support the Notifying Party’s claim that Aluminium ABS and steel products are part of the same relevant product market.
(240) As discussed in detail in Section 2.1.2 of Annex I, the Commission has used Novelis’ data on margins and information on how demand for Aluminium ABS responds to changes in the price of steel and Aluminium ABS to assess whether a hypothetical monopolist over Aluminium ABS would be able to profitably raise the price of Aluminium ABS by a small but significant amount, or whether competition exerted by steel would make such a price increase impossible (the ‘SSNIP test’).
(241) The Commission has used two standard methodologies to implement the SSNIP test: (i) by means of a critical loss analysis using critical elasticities of demand; and (ii) by means of a critical loss analysis using critical recapture ratios.
(242) In what follows, the Commission summarises the results of its analysis based on critical elasticities (see Section 2.1.2.1 of Annex I). A detailed discussion of the similar results obtained by expressing the SSNIP test in terms of critical recapture ratios is contained in Section 2.1.2.2 of Annex I.
(243) When presenting the critical demand elasticities beyond which steel would be in the same relevant market as Aluminium ABS, the Commission conservatively presented a range of values based on several scenarios: (1) 5% or 10% SSNIP, (2) profitable or profit-maximising version of the SSNIP test, (3) gross margin or EBITDA margin. The critical elasticities based on these scenarios range from […] to […], with an average of […].
(244) […].
(245) […]. Instead, as argued by the Commission based on the information available, a hypothetical monopolist on Aluminium ABS would be in a position to profitably raise prices of Aluminium ABS by a small but significant non-transitory amount.
(246) In response to certain arguments by the Notifying Party that the Commission considers the demand for Aluminium ABS to be insensitive to price,233 the Commission notes that its quantitative analysis does not imply that a price increase of Aluminium ABS would result in no switching from by OEMs from Aluminium ABS to the corresponding steel products. Nowhere in the SO or this Decision does the Commission make such a claim. The Commission’s quantitative (and qualitative) analysis simply indicates that in case of a small but significant non-transitory increase in price the switching from Aluminium ABS to the corresponding steel products would be limited and insufficient to consider Aluminium ABS and flat steel products used in automotive bodies to be in the same relevant market.
(247) Although the Notifying Party contests the methodology used by the Commission in conducting the SSNIP test, as explained in Section 2.1.3 of Annex I, the Commission discusses each of the Notifying Party’s critiques and shows that they are either unwarranted or do not undermine the robustness of the Commission’s conclusion that steel does not belong to the market for the production and supply of Aluminium ABS.
(6.2.2.3.6) Demand-side: Tier suppliers and distributors are not free to choose the material they use and do not represent a potential source of demand-side substitutability
(248) The Commission observes that OEMs are not the only customers of the Parties. In 2018, […]% of Novelis’ EEA sales, and […]% of Aleris’ EEA sales of Aluminium ABS were to Tier suppliers and distributors. For some Aluminium ABS products, […] can be even larger customers of the Parties’. For example, in 2018 Novelis […].234
(249) This is consistent with the fact that OEMs do not always perform in-house all the manufacturing steps that lead to vehicle components starting from Aluminium ABS. In some cases, Tier suppliers purchase Aluminium ABS, perform some automotive manufacturing steps such as stamping of Aluminium ABS,235 and sell the resulting intermediate product to OEMs. For these components, the Parties’ customers are the Tier suppliers, and not the OEMs. Similar is the case when an OEM or Tier supplier sources, typically relatively small volumes of Aluminium ABS from distributors.
(250) Nonetheless, Tier suppliers and distributors are bound to the choice of material made by the OEM. In fact, typically and as explained in Section 5.4, a Tier supplier or a distributor is even bound to those suppliers qualified by the OEM. A competitor explained in this respect that ‘[t]o deliver ABS to Tier-suppliers the material also has to be qualified by the OEM’, therefore, ‘the Tier-supplier can purchase ABS only from a list of suppliers approved by the OEM’.236 In line with that, several Tier suppliers have confirmed that they receive Aluminium ABS only from suppliers that have been previously qualified by the OEM.237
(251) Therefore, even if distributors and Tier suppliers negotiated independently supply terms with their suppliers, they cannot substitute Aluminium ABS with different materials – such as steel – because they have no influence over the respective OEMs’ decision in selecting their materials of choice.
(252) In line with this, all the Tier suppliers and distributors that took a view in the market investigation indicated their general inability to substitute aluminium with steel.238
(253) Therefore, the Commission concludes that Tier suppliers and distributors cannot decide on any possible substitution between Aluminium ABS and flat steel products for automotive bodies, which is independent from OEMs.
(254) The Notifying Party argues that Tier suppliers and distributors do not generate demand and therefore the lack of demand substitutability on their side is not relevant for the sake of defining a relevant product market.239 However, this argument is not in contradiction with the Commission’s conclusions in recital (253) that Tier suppliers and distributors do not represent a potential source of demand-side substitutability, which is independent from OEMs’ decisions.
(6.2.2.3.7) Conclusion
(255) For the reasons set out in this Section 6.2.2.3, and considering all evidence available to it, the Commission concludes that Aluminium ABS constitutes a distinct product market, separate from flat steel products used in automotive bodies.
6.2.3. Different Aluminium ABS products belong to the same differentiated product market
6.2.3.1. The Commission’s precedents
(256) In previous cases,240 the Commission did not consider possible differentiations of the market for the production and supply of Aluminium ABS.
6.2.3.2. The Notifying Party’s view
(257) The Notifying Party submits that a further segmentation of the Aluminium ABS product market is not justified because there is a high degree of supply-side substitutability across all Aluminium ABS alloys. The Notifying Party further holds that price differences between aluminium alloys are not informative of different product markets, due to different cost structures of the various alloys.241
6.2.3.3. The Commission’s assessment
(258) For the following reasons, the Commission considers that all Aluminium ABS products belong to the same, albeit differentiated, relevant product market.
(6.2.3.3.1) Demand-side substitutability
(259) The results of the market investigation and the Commission’s assessment suggest that demand-side substitutability is limited between different Automotive ABS products.
(260) In this respect, at least two possible segmentations appear to be commonly used by the Parties’ customers, namely by alloy (that is by chemical composition of the alloy), and by the OEMs’ end-use. With regard to the alloy segments, a commonly used segmentation in the market for the production and supply of Aluminium ABS is between 5xxx-series, 6xxx-series, and, occasionally, 7xxx-series alloys. Currently, it appears that the 6xxx-series are the most employed alloys by automotive OEMs, followed by the 5xxx-series, whereas the use of 7xxx-series alloys is currently not common among automotive OEMs, and some of them are considering their use in the future.242
(261) These segmentations are also reflected in the Parties’ internal documents.243 In practice, the ability of automotive customers to switch between products of these different segments appears limited.
(262) First, there are differences in the composition of Aluminium ABS belonging to different alloy series. These differences in the composition of the material are such that the final properties of Aluminium ABS of the different alloy series differ, in a way that makes them more or less suited for specific applications in the body of a vehicle.
(263) As explained in recitals (50) and (51), the main alloying elements are different in different alloy series. In 5xxx series (Al-Mg) magnesium is the principal alloying element, in 6xxx series (Al-Mg-Si) magnesium and silicon are the principal alloying elements, and in 7xxx series (Al-Zn) zinc is the principal alloying element. Each alloy series consists of a number of different alloys where additional alloying elements can be assed. For example, the 5182 alloy is a 5xxx-series alloy made of about 95% of aluminium, 4.5% of magnesium (the principal alloying element) and 0.35% of manganese.
(264) Consistently with these divergent technical features and applications, a large majority of the Parties’ customers who responded to the market investigation indicated that they are not able to substitute aluminium alloy series by one another.244 An OEM stated that ‘[t]he need of specific alloys prevents from substitution from one grade to another (e.g. from 5XXX to 6XXX)’,245 and further explained that the reason for such a lack of substitutability is due to ‘[d]ifferences of mechanical properties’.246
(265) In the sporadic cases where demand-side substitution could be technically possible, there might be reasons of a different nature that prevent substitution between different alloys. These seem to include at least scrap-handling requirements. That is, scrap made of different alloys is less valuable than scrap made of a particular alloy (series), and hence mixing different alloys can be detrimental to the customer who wishes to send the scrap back to the aluminium producer for re-melting. An OEM explains that ‘[i]n some applications 5XXX can replace 6XXX. However that is not done for scrap handling reasons. We cannot separate different scrap grades and if 5XXX and 6XXX is mixed the value of the scrap is significantly reduced’.247
(266) The lack of demand-side substitutability can be even more pronounced for Tier suppliers and distributors, which are bound by the specifications and requirements of their OEM customers, with little or no margin for providing different alloys.248
(267) Second, with respect to the segmentation by OEMs’ end-use, it appears that some Aluminium ABS products are specifically manufactured for specific uses in the vehicles. For example, a common differentiation that both the Parties and their customers make is between Aluminium ABS used for the outer parts (or ‘skin’) of a vehicle component, and those used for the inner parts. These two segments of Aluminium ABS usually cannot be substituted by one another. This is because Aluminium ABS with skin quality are usually suitable for manufacturing vehicle parts that are visible and therefore have more stringent requirements in terms of surface quality.249
(268) Another example of segmentation by OEM’s end-use is by the functional use of the Aluminium ABS in the vehicle. For example, Aluminium ABS for crash management are specifically designed for so-called ‘shock absorption’, which increases the passengers’ or pedestrians’ safety in case of a vehicle crash. Due to the specific requirements of these Aluminium ABS products, OEMs can hardly substitute them with other Aluminium ABS products that have not been developed for this specific purpose.
(269) Other examples of end-use segmentation include the formability of the product as well as its physical dimensions (for example width).
(6.2.3.3.2) Supply-side substitutability
(270) From a supply-side perspective, the market investigation suggests that there is a certain degree of supply-side substitutability between different Aluminium ABS products, which leads to consider that the various Aluminium ABS products belong to the same, albeit differentiated, relevant product market.
(271) In particular, it appears that there is notable substitutability between products in the 6xxx alloy series and that suppliers equipped with a continuous annealing CASH-line can typically produce all main grades, including both 6xxx and 5xxx series. However, the same is not the case for a supplier who only has a batch annealing facility and can only produce certain 5xxx series products but not 6xxx series. Moreover, there is evidence that the market conditions for the different products within Aluminium ABS can be somewhat different.
(272) First, as explained in Section 5.4, manufacturing of Aluminium ABS requires annealing facilities. For the production of 6xxx series, continuous annealing is required. Continuous annealing can also be used to produce 5xxx series, though that alloy series – but not 6xxx series – can in some cases be produced with batch annealing as well.
(273) A continuous annealing CASH-line can thus interchangeably produce both 5xxx and 6xxx series (and 7xxx series). Nonetheless, there are certain differences between the production of 5xxx and 6xxx series in CASH lines. 6xxx series need to undergo additional procedures during production, and may therefore need to run through a CASH line more than one time, which leads to slower production compared to 5xxx series. 6xxx series also exhibit higher scrap rates during production. Despite these differences, 5xxx and 6xxx series can be produced with the same equipment, a CASH line.
(274) Further, looking at different end-applications, the same CASH-line can be used to produce 6xxx series for both structural (non-exposed) car parts as well as to exposed (‘6xxx-skin’) parts. Both 6xxx series run through the production line at a similar speed; however, there can be differences in the scrap rate (production of 6xxx-skin can produce more scrap) and the production of 6xxx-skin must ensure that the surface quality is particularly high as surface defects could remain visible in the final vehicle.250 Despite these differences, 6xxx-skin and 6xxx structure series can be and are produced with the same equipment.
(275) Second, a supplier who does not have a CASH line cannot manufacture all Automotive ABS products. As the production of 6xxx series requires a CASH-line, manufacturers that do not have a CASH manufacturing line are obliged to limit their products to some 5xxx-series alloys, which in some limited cases can also be manufactured without a CASH manufacturing line.251 These manufacturers are not able to switch from 5xxx-series alloys to 6xxx-series alloys ‘in the short term without incurring significant additional costs […] in response to small and permanent changes in relative prices [between 5xxx-series and 6xxx-series alloys]’.252
(276) Third, even in cases where a manufacturer can use the same manufacturing equipment for the various Aluminium ABS products sold to its customers, the market investigation indicates that supply-side substitution can be limited by the required level of industrial know-how. This relates to the fact that Aluminium ABS product differentiation takes place at several stages of the manufacturing process, and in each of these processes, specialised know-how is required for each of the possible differentiations. For example, an alloy composition is defined when a slab – the input material of Aluminium ABS – is casted. Other characteristics making Aluminium ABS suitable for, for example, high-formability, crash management, skin or inner use are defined further downstream in the process (hot rolling, cold rolling, heat treatments), and surface properties are, to some extent, defined at the finishing stage.
(277) Specific know-how and industrial experience are required for manufacturing a large part of Aluminium ABS products at the quality level required by the OEMs. These factors constitute barriers to entry to some competitors that therefore limit their manufacturing scope to less sophisticated Aluminium ABS products. One of the Parties’ competitors responding to the market investigation considers that ‘[f]rom a manufacturing point of view there are huge differences between the 5000 series (“5xxx”) and 6000 series (“6xxx”) alloys’,253 and that ‘[m]anufacturing 6xxx alloys presents several barriers and requires detailed know-how and industrial experience’.254 The same view is shared by other Aluminium ABS manufacturers that took a view in the market investigation.255
(278) […].256 […].257
(279) Fourth, even if manufacturers are able to produce various different alloy series, their relative cost and competitiveness positions can be different. […]258 […].
Figure 37 […]
[…]
(280) Fifth, it appears that there are observable differences in terms of prices and margins for different Aluminium ABS segments. These differences suggest the presence of more demanding manufacturing steps for some aluminium ABS products, and possibly higher barriers to entry leading to different competitive conditions among the various Aluminium ABS segments.
(281) As explained in Section 8.3.5.1, for the purpose of conducting a market reconstruction, the Commission gathered the conversion prices at which the main manufacturers active in the EEA sell their Aluminium ABS products. […].
Table 1 […]
[…].
(282) […],259 […]260 […].
(283) […],261 […],262 […].
Figure 38 […]
[…]
(284) From a supply-side perspective, despite the various differences between some Aluminium ABS products, the Commission considers that the existing level of supply-side substituitability, although not perfect, justifies the conclusion that the various segments of Aluminium ABS belong to the same market,263 ‘[s]upply-side substitutability may also be taken into account when defining markets in those situations in which its effects are equivalent to those of demand substitution in terms of effectiveness and immediacy. This means that suppliers are able to switch production to the relevant products and market them in the short term […] without incurring significant additional costs or risks in response to small and permanent changes in relative prices. When these conditions are met, the additional production that is put on the market will have a disciplinary effect on the competitive behaviour of the companies involved. Such an impact in terms of effectiveness and immediacy is equivalent to the demand substitution effect’.
(285) In the present case, as explained in recitals (270)-(275), certain manufacturers can switch production between all the various types of Aluminium ABS products without incurring significant additional costs or risks. For the case of the 6xxx and 5xxx alloys, for example, in the case of a price increase of the 5xxx alloys, those manufacturers that produce both 5xxx and 6xxx alloys might potentially employ their entire production capacity for producing 5xxx alloys only. Therefore they represent a disciplinary deterrent for those manufacturers that can produce only 5xxx alloys in increasing their prices.
(6.2.3.3.3) Conclusion
(286) For the reasons set out in this Section 6.2.3.3, and considering all evidence available to it, the Commission concludes that the market the production and supply of Aluminium ABS is differentiated across different market segments, in particular with respect to different alloy-series where demand-side substitutability is limited but a level of supply-side substitutability exists. Nevertheless, although the market for the production and supply of Aluminium ABS is differentiated, the Commission concludes that all types of Aluminium ABS belong to the same relevant product market.
6.3. Standard FRPs
6.3.1. The Notifying Party’s view
6.3.1.1. Standard FRPs
(287) The Notifying Party submits that the market for the production and supply of Standard FRPs as identified by the Commission in previous cases includes standard and painted aluminium sheet of different gauges for a variety of end-uses such as building and construction, industrial, consumer products, electronics, packaging, transportation, floor heating, and others.264 According to the Notifying Party, Standard FRPs also include aluminium anodising sheet and aluminium sheet for compound tubes due to the high degree of supply-side substitution.265
(288) The Notifying Party considers that all rolling mills have the capability to make different Standard FRPs and that producing Standard FRPs typically only requires configuring the hot or cold mill in a different way, changing the alloy composition, or adding finishing capabilities. For this reason, according to the Notifying Party, every aluminium producer is able to produce the full range of Standard FRPs with minimal additional investment, and will opt to make one of the different types of Standard FRPs based on factors such as their rolling capacity and the volumes and profits they expect to obtain from selling the product in question.266
(6.3.1.1.1) Aluminium anodising sheet
(289) Aluminium anodising sheet is produced by treating aluminium sheet using an electro-chemical process known as anodising. During this process, the metal’s surface is modified to create a hard and transparent protective layer of aluminium oxide that makes it a durable product, which is scratch and corrosion resistant.
(290) The Notifying Party submits that neither of the Parties, nor any of their competitors, have in-house anodising capabilities.267 Therefore, they either sell ‘anodising quality sheet’ (aluminium sheet suitable for anodising) to customers that have it anodised (internally or by a third party) or sell pre-anodised aluminium sheet (sheet that has already been anodised), for which they outsource the anodising process to a third party. For the purpose of this Decision, aluminium anodising sheet refers to both anodising quality sheet and pre-anodised sheet.
(291) The Notifying Party argues that the production of anodising quality sheet does not require dedicated equipment and that it undergoes the same production steps as any Standard FRP, that is hot rolling, cold rolling and finishing. According to the Notifying Party, Aluminium FRP manufacturers not currently making aluminium anodising quality sheet and wishing to enter this product category could quickly do so by making small adjustments in the chemical composition of their products, and running tests to obtain a product of the required anodising qualities to meet customer demands.268 According to the Notifying Party, the investment is not different from the one required from an Aluminium FRP producer that wishes to add a new Standard FRP product category; and the knowhow that is required to make these adjustments would exist in all significant Aluminium FRP producers that have long- standing expertise in the production of Aluminium FRP.
(292) The Notifying Party submits that aluminium anodising sheet is mainly used in exterior building applications269 such as facades or other visible components of a building.270 According to the Notifying Party, other materials such as steel, vinyl, or titanium, and other Aluminium FRPs such as standard sheet with a painted (or coated) surface or anodised aluminium composite panels (ACP) could be used for the cladding of building facades alternatively to aluminium anodising sheet.271 In particular, the Notifying Party submits that, for architectural applications, anodising sheet competes head-to-head with ACP, which have the same appearance as aluminium anodised sheet because the top layer is aluminium that has been anodised.272
(6.3.1.1.2) Aluminium sheet for compound tubes
(293) Compound tubes are primarily used in domestic sanitary and drinking water installations, central heating and floor heating in the building and construction sector, for transportation of gas and special liquids as well as in other industrial applications.
(294) Compound tubes are made out of multiple layers: for example inner polythene layer, glue, aluminium on glue, glue again and polythene again. They are manufactured using thin strips of Aluminium FRP. The production process of compound tubes consists in first, bending the aluminium strips, then welding them in a continuous process and finishing them by adding plastic layers.
(295) According to the Notifying Party, other materials such as copper, plastic, or steel can be used instead of aluminium in compound tubes.
(296) The Notifying Party argues that the production of aluminium sheet for compound tubes relies on industry-standard manufacturing equipment. In addition to hot rolling, cold rolling and slitting (cutting into narrower coils), sheet for compound tubes passes a degreasing line to remove the oil from the rolling process by way of a thermal or chemical bath. According to the Notifying Party, slitting and degreasing equipment is commonly available and the knowhow involved in the production of sheet for compound tubes is present in most large Aluminium FRP manufacturers.273
(297) For Aluminium FRP suppliers not currently producing aluminium sheet for compound tubes, the Notifying Party argues that they could easily and quickly gain the relevant know-how, which relates mostly to process and quality control, by investing time in developing and testing the material and adjusting their quality control procedures to the specificities of the sheet.274
6.3.2. The Commission’s precedents
(298) As explained in recital (93), previous Commission decisions in the aluminium sector have found that Standard FRPs constitute a distinct product market.275 That market has been considered to include all Aluminium FRPs that do not constitute separate distinct products markets within the field of Aluminium FRP, including standard sheets, plates, foil stock, among other Aluminium FRPs. For these Standard FRPs, the Commission considered that there exists a certain degree of supply-side substitutability, as the market investigation had indicated that aluminium producers are able to produce the full range of Standard FRPs, switching production between the different types within a short period of time and without incurring significant additional costs.
(299) The Commission has not analysed the supply of aluminium anodising sheet or aluminium sheet for compound tubes as separate distinct markets in its previous decisions in the aluminium sector.
6.3.3. The Commission’s assessment
6.3.3.1. Standard FRPs
(300) The market investigation has confirmed the Commission’s findings in previous decisions that there is a distinct market for the production and supply of Standard FRPs. This market would contain all Aluminium FRPs which do not form a distinct separate product market within the field of Aluminium FRPs.
(301) From a demand-side perspective, substitution appears to be limited within the different types of Aluminium FRPs. However, from a supply-side perspective, the results of the market investigation suggest that Aluminium FRP producers are able to produce a number of commoditised products and to easily switch production between them within a short period of time.
(6.3.3.1.1) Aluminium anodising sheet
(302) On the basis of the market investigation, there are indications that the manufacture and supply of aluminium anodising sheet constitutes a distinct product market within the field of Aluminium FRPs and therefore, could be considered separately from the market for the production and supply of Standard FRP.
(303) First, from a demand-side perspective, although it seems that there is a certain degree of substitutability between aluminium anodising sheet and standard sheet with painted (or coated) surface as argued by the Notifying Party, such substitutability is limited. A customer explained: ‘coated sheets are cheap alternatives, but the coating cannot be compared with a[n] anodized surface (pure metallic surface, UV-resistant, corrosion and weather resistant, no susceptibility to filiform corrosion and a highly durable finish)’.276
(304) Second, in spite of a certain degree of supply-side substitutability with the production of other Aluminium FRP, the majority of competitors indicated that special know-how is required for the production of sheet of ‘anodising quality’, while some also suggested that even special production assets are required.
(305) Competitors explained277 for example that: ‘there is specific requirement for this kind of product [anodising quality sheet]’; ‘proper endowment is necessary in order to produce aluminium anodising quality sheets’; ‘from talking to competition, we know that not everybody has the assets’; ‘anodising quality requires special knowhow for casting (alloy & casting process) as well as rolling’; ‘specific assets and knowhow are required to meet customer demands’; ‘it [the production of aluminium anodising sheet] is question of Know How’; and ‘for anodising quality is a special hot rolled coil and a special rolling mill necessary’.
(306) Competitors suggesting that there is no need for specific equipment however consider that having the right know-how and procedures in place is required and may thus reduce supply-side substitutability.
(307) In particular, a competitor stated that ‘a company already present in the supply of aluminium flat rolled products would not require specific equipment or huge investments to produce anodizing quality sheet (…) However, it would be necessary for such supplier to invest in smaller investments like e.g. new filters and to adapt the production process (casting, scalping and rolling). Also, to produce anodizing quality sheet you need to treat the materials more carefully. It is necessary to follow a stricter production process and e.g. cast slower, which in practical terms results in higher costs. Know-how and experience are of high importance. That is why some who try to enter this market are not capable to produce high anodizing quality sheet or anodizing quality sheet immediately.’278
(308) Another competitor explained that ‘very specific know-how is required for the manufacture of anodizing quality sheet’ and that ‘the manufacture of anodizing quality sheet requires much more attention than other [Aluminium] FRP’.279 A further competitor stated that ‘it is a big challenge to produce anodising quality sheet as it requires mastering a very complicated engineering process’.280 Another competitor explained that ‘aluminium flat rolled products must be produced in a certain way in order to be suitable for anodising, including a very product-specific manufacturing know-how throughout each of the steps of the manufacturing process: casting, hot rolling, cooling, cold rolling, etc.’281
(309) Third, although the results of the market investigation are not clear as to how much time and capital is required to get the know-how necessary to start producing anodising quality sheet, some competitors suggested that it can last up to 5 years.282 One competitor explained in this regard that ‘the process to be able to switch production to anodizing quality and pre-anodized material can cost around EUR 200 000 pear year taking into account the investment in personnel and trials needed, as well as the cost of paying a third party for the anodization’.283
(310) Fourth, most competitors indicated that the investment required to produce aluminium anodising quality sheet is not similar to the investment required to add any other new Standard FRP category to the production mix of a production plant.284
(311) Fifth, the market investigation revealed that customers typically require previous certification/qualification of their suppliers of aluminium anodising sheet,285 whereas this is typically not the case for all Standard FRP. The qualification process can last up to two years.286
(312) The Commission therefore finds that there are indications suggesting that the market for the production and supply of aluminium anodising sheet may constitute a distinct relevant product market, separate from Standard FRP products. However, for the purposes of the present Decision, the precise market definition can be left open, because the Transaction would not significantly impede effective competition in the supply of aluminium anodising sheet under whatever product market definition.
(6.3.3.1.2) Aluminium sheet for compound tubes
(313) On the basis of the results of the market investigation, there are indications that the manufacture and supply of aluminium sheet for compound tubes constitutes a distinct product market within the field of Aluminium FRPs and therefore, could be considered separately from the market for the production and supply of Standard FRP.
(314) First, from a demand-side perspective, substitution appears to be limited with other Aluminium FRPs.
(315) Second, from a supply-side perspective, the production of aluminium sheet for compound tubes requires specific finishing capabilities, that is a slitting line and a degreasing line. Although the results of the market investigation show that Aluminium FRP producers typically have slitting capabilities, not all suppliers have a degreasing line.287 Moreover, the majority of competitors responding to the market investigation indicated that special know-how is required for the production of aluminium sheet for compound tubes, while some also suggested that even special assets are required.288
(316) Similarly to aluminium anodising sheet, competitors explained289 that ‘there is specific requirement for this kind of product [aluminium sheet for compound tubes]’; ‘proper endowment is necessary in order to produce aluminium anodising quality sheets’; and ‘from talking to competition, we know that not everybody has the assets’.
(317) A competitor explained that ‘it is generally true that a company already present in the supply of aluminium FRPs would not require any special equipment to start producing sheet for compound tubes (…) there is, however, a learning curve during which suppliers have to develop and acquire the required know-how to produce sheet for compound tubes’.290 Another competitor not currently supplying aluminium sheet for compound tubes further explained that ‘as for anodizing quality sheet, suppliers need very specific know-how to produce sheet for compound tubes’.291
(318) Third, although the results of the market investigation are not clear as to how much time and capital is required to get the finishing lines necessary to start producing aluminium sheet for compound tubes, some competitors suggested that it can last more than two years and cost more than EUR 5 million.292 With regard to know-how, some competitors suggested that it can last more than two years and take more than EUR 10 million.293
(319) Fourth, the market investigation has revealed that customers typically require previous certification/qualification of their suppliers of aluminium sheet for compound tubes.294 The qualification process can last from three months to two years depending on the customer.295 A competitor explained in this respect that ‘sheet for compound tubes constitute a precisely defined product’ and that ‘there exists a homologation process between the supplier and the producers of the tubes (…) Moreover, there also is a homologation process as regards the production of the tubes for which the Company cannot provide further information’.296
(320) Fifth, the results of the market investigation show that customers cannot substitute in a cost effective way aluminium by copper, steel or plastic in the production of compound tubes.297
(321) The Commission therefore finds that there are indications suggesting that the market for the production and supply of aluminium sheet for compound tubes may constitute a distinct relevant product market, separate from Standard FRP products. However, for the purposes of the present Decision, the precise market definition can be left open, because the Transaction would not significantly impede effective competition in the supply of aluminium sheet for compound tubes under whatever product market definition.
6.4. Conclusions on the relevant product markets
(322) For the reasons set out in this Section 6, and considering all evidence available to it, the Commission concludes that the production and supply of Aluminium ABS constitutes a distinct product market, separate from the production and supply of other types of Aluminium FRPs and from steel products. The Commission also considers that all types of Aluminium ABS belong to the same relevant product market but it nonetheless concludes that the market is differentiated, in particular with respect to alloy series and OEMs’ final use, and the differentiation is taken into account in assessing the competitive effects of the Transaction.
(323) The Commission further concludes that, for the reasons set out in this Section 6, and considering all evidence available to it, the production and supply of Standard FRPs constitute a distinct product market within the field of Aluminium FRPs. While there are indications that the production and supply of aluminium anodising sheet and aluminium sheet for compound tubes may constitute separate relevant product markets distinct from the production and supply of other Aluminium FRPs, the exact product market definition can be left open as, in any event, the Transaction would not significantly impede effective competition for these products under whatever product market definition.
7. RELEVANT GEOGRAPHIC MARKET
(324) The Commission recalls that, as explained in its Market Definition Notice,298 a relevant geographic market is the geographic area in which the undertakings concerned are involved in the supply and demand of products or services, in which the conditions of competition are sufficiently homogeneous and which can be distinguished from neighbouring areas because the conditions of competition are appreciably different in those area.
(325) In its assessment of the relevant geographic market, the Commission takes into account various factors, including:
(a) demand characteristics, including preferences for regional suppliers and need for a local presence;
(b) current geographic patterns of purchases;
(c) trade flows/patterns of shipments;
(d) barriers associated with trade across areas; and
(e) views of customers and competitors.299
7.1. Aluminium ABS
7.1.1. The Notifying Party’s view
(326) The Notifying Party initially submitted that the market for the production and supply of Aluminium ABS is at least EEA-wide, if not global and that it should at least also comprise the neighbouring regions such as Commonwealth of Independent States,
Russia, Eastern Europe, Turkey and the Middle East.300
(327) According to the Notifying Party301, there are strong indications that the Aluminium ABS market is not limited to Europe and neighbouring regions. The Notifying Party argues that OEMs operate globally to supply cars across the world and that the supply chain of Aluminium ABS producers has become increasingly global as well. Furthermore, the Notifying Party argues that transport costs for Aluminium ABS are not significant, typically accounting for […]% within Europe and for no more than […]% for deliveries from or to the US and China.
(328) According to the Notifying Party, both EU imports and EU exports of Aluminium FRPs, including automotive products, are significant; and imports of Aluminium ABS from outside the EEA exert additional competitive pressure on the EEA-based ABS suppliers.302
(329) The Notifying Party argues that the following overseas Aluminium ABS producers currently supply OEMs’ EEA-based plants:
(a) Alcoa/Ma’aden supplies […] from its facility in Saudi Arabia.
(b) Chinese producers such as Nanshan currently supply up to […] kt303 of Aluminium ABS per year into the EEA.
(c) More specifically, […] qualified Nanshan in 2017 to supply its European plants. […] recently qualified Nanshan also for 6xxx-skin series globally.
(d) Also, Kobe Steel, which produces Aluminium ABS through its Chinese subsidiary Kobelco, has already started importing Aluminium ABS into the EEA.304
(330) According to the Notifying Party, import duties into the EEA do typically not exceed 6%.305
(331) However, in the Reply to the Letter of Facts, the Notifying Party agreed to the EEA wide market definition by confirming that the Notifying Party does ‘not contest the SO’s conclusions on geographic market definition.’306
7.1.2. The Commission’s precedents
(332) In previous decisions, the Commission has generally considered that Aluminium FRP markets are at least the EEA plus Switzerland and possibly, for certain categories, even wider in scope.307
(333) Regarding flat rolled products for the automotive industry, in previous decisions the Commission has ultimately left open the exact geographic market definition.308
7.1.3. The Commission’s assessment
(334) The market investigation in the present case has shown that the relevant geographic market for the production and supply of Aluminium ABS is EEA-wide.309
(335) First, the range of available suppliers differs across world regions, whereby besides Novelis, Aleris and Constellium having CASH line equipped plants in Europe and the US, the rest of the supply is accounted for by players established in each of the three global regions where Aluminium ABS production assets are located as shown in the map captioned below in Figure 39. Moreover as evidenced in this map, there are no Aluminium ABS production sites in regions such as CIS, South America, Africa or Australia. The Middle East production consists exclusively of Alcoa/Ma’aden in Saudi Arabia, which regularly supplies into the EEA and is taken into account in the competitive assessment.
(336) While the Commission acknowledges the existence of trade flows between different world regions, these are nonetheless limited and producers sell the majority of their products in the region or regions in which their production sites are located. It appears from the Parties’ best estimates that trade flows are relatively small as reproduced below (Figure 40 and Figure 41). Moreover, trade flows between Europe and North America consist only in exports from Europe to North America and trade flows between Europe and China result in Europe being a net exporter to China.
Figure 40 […]
[…]
Figure 41 […]
[…]
(337) This trade flow pattern is directly linked to the price difference between global regions, as described below.
(338) Significant price differences appear to exist between the three production regions EEA, the US and Asia as confirmed by customers and competitors. An OEM explained that ‘[p]rice is significantly higher in the US’.310 Another competitor explained that ‘[p]rice level is higher in the US because of the higher metal premium and conversion margins are higher’ and that ‘[t]he US is a market separate from the EEA and hence the prices are not equalised between the two continents’.311 A customer explained that Europe, the US and China ‘are separate because they differ in price levels and capacities. In terms of price levels (“PL”), EU has the lowest PL while US has the highest PL. On the other hand, China has a mid PL’.312 Another customer states that ‘[t]he different price levels observed between different regions in the world limit competition. For example, it does not make sense for the Company to import aluminium ABS from North America because of the higher price there’.313
(339) […],314 […].
Figure 42 […]
[…]
(340) […].315
(341) Second, the sourcing pattern is confirmed by EEA customers responding to the market investigation referred to their preference for sourcing material from EEA- established players. A large majority of the automotive customers consider that the maximum distance that Aluminium ABS can be economically transported is within the EEA.316 This points to customers having a regional sourcing strategy as explained by an OEM: ‘[t]he Company is sourcing regionally’, ‘[t]herefore, for the EEA it is sourcing from suppliers within the EEA’.317 A European OEM explained that it has ‘[c]urrent minimal supply from outside the EEA’.318 This is in line with another automotive customer statement: ‘The Company is sourcing aluminium ABS regionally: Aluminium ABS for the EEA is purchased within the EEA; Aluminium ABS for the USA is purchased within the USA, etc.’.319 Another OEM does not see changes in these geographic demand pattern in the future and states that ‘also imports into the EEA are expected to be limited’.320 European customers’ sourcing patterns are indicative of regional competition and of different competition conditions in different world regions.
(342) Third, beside the EEA established suppliers, which account for a predominant share of EEA demand, the amount of imports alleged by the Notifying Party is limited. Most of the imports into the EEA originate from Switzerland. This is explained by the fact that Novelis has a manufacturing plant in Sierre (Switzerland), which serves the EEA market for the production and supply of Aluminium ABS and that is included in the competitive assessment. Moreover, the capacity of Alcoa/Ma’aden, considering its relative geographic proximity and regular supplies into the EEA, is also included within capacity shares to have a more robust assessment of the Parties’ position in the relevant market. The imports from China are listed in Figure 41 and correspond exclusively to Nanshan showing that in 2018 only […] were imported to the EEA compared to […] tonnes of sales in the EEA as detailed in Table 3. Moreover, Nanshan imports volumes are not expected to be more substantial in the foreseeable future growing only to […] tonnes in 2023. Kobelco is not listed in Figure 41 among those importing into the EEA currently or until 2023.
(343) Fourth, the market investigation also points at specific technical reasons that make it difficult, particularly for Aluminium ABS for high quality alloys, to be sourced from a long distance. The majority of the automotive customers considers that there are barriers making the import of Aluminium ABS into the EEA difficult.321 This is due, in particular, to the process of natural aging322 of Aluminium ABS.
(344) The aging of some alloys (specifically 6xxx-skin series) appears to be a significant constraint for long distance shipping and a hindrance to imports into the EEA. For the sake of completeness, it has to be noted that Alcoa/Ma’aden is currently not affected by this aging constraint as it is importing 5xxx alloys from Saudi Arabia,323 which are not subject to aging contrary to 6xxx alloys.
(345) Numerous customers of Aluminium ABS consider that aging is a barrier to imports into the EEA. The reasons suggested in the market investigation being the ‘6-month durability of the skin Aluminum grades’,324 ‘[s]helf life of some specifications’,325 ‘[a]geing of heat-treated alloys’,326 and ‘the age hardening (Shelf life) of aluminium’.327 One customer explained that ‘[b]esides costs and logistics, the main reason for sourcing regionally, is the aging process of the 6xxx alloys, which need to be used for manufacturing automotive components within approximately 6 months from the day they are manufactured’.328 Another customer stated that ‘[i]mporting aluminium ABS from Europe to US may affect the sheet’s formability. Formability is limited by the time it takes to transport the product and the change in temperature that may occur in the course of importation’.329 Finally, another customer explained that ‘[t]he age hardening (Shelf life) of aluminium restricts the ability on most alloys. To overcome this airfreight would be required for longer distances and this would be economically prohibitive’.330
(346) Suppliers also confirm that the aging process is a constraint to intercontinental shipping. A competitor of the Parties stated that ‘[t]here are other constraints linked to intercontinental shipping as 6000 alloys specifically must be used within 180 days of its production not to lose some of its properties’.331 Another competitor explained that ‘[o]ne factor limiting the transportability of ABS is that 6xxx series aluminium alloys harden even at room temperatures. In practice, such alloys need to be consumed in the final manufacturing within a certain period from the production of the aluminium sheet. Therefore, long transport distances and the associated time is a hindrance’.332
(347) Contrary to the Parties’ claim that ‘Transport costs for ABS are not significant, typically accounting for […]% within Europe and no more than […]% for deliveries from or to the US and China’, a customer of Aluminium ABS explained that ‘[l]ocation is a key factor when sourcing aluminium ABS. Often suppliers supply regionally given relatively high transportation costs’.333
(348) […].334
(349) In addition to the foregoing, imports of rolled aluminium of a thickness exceeding
0.2 mm, including Aluminium ABS, from a third country into the EEA are subject to an import duty of 7.5%.335 This may constitute and additional constraint for customers to source from outside the EEA. When coming from Switzerland, such imports are not subject to such import duty.336
(350) For all the above-mentioned reasons, the Commission reaches the conclusion that competition takes place on an EEA level as regards the production and supply of Aluminium ABS. The results of the investigation in the present case show that there are significant obstacles for EEA-based customers to source from outside the EEA and constraints for intercontinental supply, rendering an EEA-based supplier highly preferable. Therefore, the Commission does not accept the argument initially submitted by the Notifying Party according to which the geographic scope of the Aluminium ABS market should include the EEA, CIS, Russia, Eastern Europe, Turkey and the Middle East.
7.2. Standard FRPs, aluminium anodising sheet and aluminium sheet for compound tubes
7.2.1. The Notifying Party’s view
(351) The Notifying Party submits that the geographic scope of Standard FRPs, including aluminium anodising sheet and aluminium sheet for compound tubes, covers at least the EEA.337
7.2.2. The Commission’s previous view
(352) In previous decisions, the question whether the geographic market for the production and supply of Standard FRP is EEA-wide or wider has been left open.338
(353) The Commission has not specifically analysed the supply of aluminium anodising sheet or aluminium sheet for compound tubes in its previous decisions in the aluminium sector.
7.2.3. The Commission’s assessment
(354) The results of the Commission’s investigation of the present case support the view that the market for the production and supply of Standard FRP has an EEA-wide dimension.
(355) In particular, during the Commission's market investigation, most of competitors and customers stated that the supply and demand of Standard FRPs are EEA-wide and that prices of Standard FRP are similar throughout the entire EEA.339
(356) Similar views were expressed by market participants with regard to aluminium anodising sheet and aluminium sheet for compound tubes, which irrespective of whether they are defined as a separate relevant product market, from a geographic perspective appear to follow largely similar competitive dynamics as other types of (standard) aluminium FRP.340
(357) In light of the above, for the purposes of the present Decision, the Commission considers that the geographic scope of the market for the production and supply of Standard FRPs, and of the putative product markets for aluminium anodising sheet and aluminium sheet for compound tubes, are EEA-wide.
7.3. Conclusions on the relevant geographic markets
(358) For the reasons set out in this Section 7, and considering all evidence available to it, the Commission concludes that the market for the production and supply of Aluminium ABS is EEA-wide in scope. The assessment will include Switzerland for the supply part as production is located in the immediate vicinity of EEA thereby avoiding high transportation cost or aging constraint and products are not subject to import duties341. The demand part is not affected by this inclusion as there is no consumption of Aluminium ABS in Switzerland.
(359) The Commission further considers that, for the reasons set out in this Section 7, and considering all evidence available to it, the market for the production and supply of Standard FRPs is EEA-wide in scope. The production and supply of aluminium anodising sheet and of aluminium sheet for compound tubes, if treated as separate relevant product markets, would also be EEA-wide in scope.
8. COMPETITIVE ASSESSMENT
8.1. Introduction
(360) The Commission assesses in the following sections the impact of the Transaction on the relevant EEA markets for the production and supply of Aluminium ABS. As regards other products concerned, the Commission explains that the market for the production and supply of Standard FRPs would not be horizontally affected and further presents its assessment based on the narrowest plausible definition of separate markets for aluminium anodising sheet and aluminium sheet for compound tubes.
8.2. Framework of the competitive assessment
8.2.1. Legal framework
(361) Under Articles 2(2) and (3) of the Merger Regulation, the Commission must assess whether a proposed concentration would significantly impede effective competition in the internal market or in a substantial part of it, in particular as a result of the creation or strengthening of a dominant position.
(362) In this respect, a merger may entail horizontal and/or non-horizontal effects. Non- horizontal effects are those deriving from a concentration where the undertakings concerned are active in different relevant markets.
(363) As regards the assessment of horizontal effects, the Commission guidelines on the assessment of horizontal mergers342 (the ‘Horizontal Merger Guidelines’) distinguish, in addition to the creation or strengthening of a dominant position, between two main ways in which mergers between actual or potential competitors on the same relevant market may significantly impede effective competition, namely non-coordinated and coordinated effects. Non-coordinated effects may significantly impede effective competition if a merger weakens important competitive constraints between rival firms in the market, who consequently would have increased market power, without resorting to coordinated behaviour. In that regard, the Horizontal Merger Guidelines consider not only the direct loss of competition between the merging firms, but also the reduction in competitive pressure on non-merging firms in the same market that could be brought about by a merger.
(364) The Horizontal Merger Guidelines list a number of factors, which may influence whether or not significant non-coordinated effects are likely to result from a merger. For instance, these include (but are not limited to) large market shares of the merging firms, the fact that the merging firms may be close competitors, limited possibilities for customers to switch suppliers, the existence of capacity constraints, or the fact that a merger may eliminate an important competitive force. Not all of the factors indicated in the Horizontal Merger Guidelines as potentially relevant to the analysis of non-coordinated effects need to be present to make significant non-coordinated effects likely. Moreover, the list of factors is not exhaustive.
(365) The Horizontal Merger Guidelines describe the creation or the strengthening of a dominant position as a primary form of competitive harm. Further, reference is drawn in the Horizontal Merger Guidelines to Council Regulation No 4064/89, defining dominance as ‘a situation where one or more undertakings wield economic power which would enable them to prevent effective competition from being maintained in the relevant market by giving them the opportunity to act to a considerable extent independently of their competitors, their customers and, ultimately, of consumers’.343
(366) Very large market shares – 50 % or more – may in themselves be evidence of the existence of a dominant market position. However, smaller competitors may act as a sufficient constraining influence if, for example, they have the ability and incentive to increase their supplies. A merger involving a firm whose market share will remain below 50 % after the merger may also raise competition concerns in view of other factors such as the strength and number of competitors, the presence of capacity constraints or the extent to which the products of the merging parties are close substitutes.344
(367) A merger giving rise to non-coordinated effects, such as to the removal of an important competitive constraint on a seller,345 would significantly impede effective competition by creating or strengthening the dominant position of a single firm. This firm would typically have an appreciably larger market share than the next competitor post-merger.346
8.2.2. Competition in basic industries characterised by capacity constraints
(368) The Commission considers that in a basic industry characterised by capacity constraints unilateral effects may arise through at least two channels. First, given the available capacities in the market, a single merged entity may compete less aggressively on price post-merger for two reasons: (i) before the transaction the two independent merging parties were not concerned about cannibalising each other’s sales through price competition, but this changes once the combined capacity is controlled by a single entity; (ii) after the transaction the rivals of the merged entity control less capacity than the rival capacity faced by the merged entity pre- transaction. Second, merging producers may also compete less aggressively on capacity expansions post-transaction, as they will take account of the negative effect that new capacity in the market has on the sales of the respective merging partner. Otherwise stated, pre-transaction each merging party only took into account the negative impact that new capacity would have on its own sales (via the decrease in overall market price due to the additional capacity) but did not take into account the negative effect on the sales of the other merging party. This is internalised post– transaction, which results in a loss of competition on the market.
(369) Whether these theories of harm apply in a concrete case depends significantly on two important factors: (i) the level of the Parties’ market position (as characterised by their market shares and capacity shares) and (ii) the extent of spare capacities in the market (in particular those held by rivals).
(370) All else equal, anti-competitive effects are more likely if merging parties control a large part of the market after the transaction. This is because the effect of the internalisation of price and capacity competition between merging suppliers is particularly large if the parties represent a large part of the market. Conversely, if the merging parties are only minor players in a market, then a transaction would be unlikely to lead to significant anti-competitive effects, since the parties’ impact on market prices and capacity levels would then be too small to cause an appreciable effect.
(371) Moreover, anti-competitive effects are also more likely to result when the extent of spare capacities in the market (in particular those held by rivals) is moderate relative to projected demand. This is because the Merged Entity will possess appreciable market power post-Transaction when its own supply is necessary to ensure that the entire market demand is served. Instead, if rivals’ capacity in the market is significantly larger than projected demand, then rival firms may continue to exert significant competitive pressure on the Merged Entity post-Transaction, in an effort to fill their under-utilised production facilities.
8.2.3. Relevant characteristics of bidding markets
(372) Bidding markets are markets where purchasing is conducted through open, competitive tenders organised by the buyer of the input. The principal purpose of organizing a tender is to allow the buyer to make informed purchasing choices, guided by the information revealed during the tendering process, and to spur competition between rival sellers in order to obtain a good price. Bidding markets therefore permit purchasers to organise their purchasing in a transparent and competitive way, with the aim of getting the best deal possible.
(373) While bidding markets are often an efficient form of organising a competitive process, this does not imply that also the outcome of that competitive process will be efficient or devoid of the exercise of market power. On the contrary, if there are only few credible bidders or if bidders differ in their ability to serve the purchaser in question (for example, due to the existence of product differentiation or capacity constraints), then the outcome of the bidding process will not be very competitive. Sellers will then be able to earn supra-competitive profits, just as would be the case in any other market with imperfect competition. In the competition analysis of bidding markets, it is therefore important to distinguish the organizational form of a purchasing process from its competitive mechanism and eventual outcome.
(374) As the Commission explained in Case M.7278 – GE/Alstom,347 in markets characterised by tendering, the general mechanism through which a merger can influence competitive outcomes is similar to what occurs in mergers in other industries where firms compete on price. That is, a merger internalises the competitive pressure that two firms exercised on each other prior to the transaction and can lead each of the remaining firms to bid less aggressively post-merger due to the removal of a competitive constraint. The precise mechanism through which a merger can influence bids and the indicia of potential unilateral effects depend on how the tendering process is set up and on the information available to bidders. The main driving factors of the impact of a merger, however, are the same factors that drive competitive effects in any other market with price competition (for example, the number of rival suppliers, their competitive position in the market, the closeness of competition of the merging parties, their pre-merger market power, the existence of capacity constraints and so forth).
(375) In particular, there is no presumption in bidding markets that only a few bidders (even as low as two bidders) are sufficient to generate a competitive outcome. Such an extreme result could theoretically only hold under conditions of perfect competition, where suppliers sell identical products, have identical costs, have effectively unlimited capacity and there is perfect transparency about all competitively relevant information (for example, with respect to the cost of rival bidders). In such a situation, equilibrium margins on incremental sales would be zero already pre-merger, as competition is so intense that no profits could be earned on competitive sales.
(376) However, virtually no real world market is characterised by such extreme textbook conditions. In addition, indeed, the theoretical result of perfectly competitive markets no longer holds if firms offer differentiated products, compete based on limited capacities or if markets are otherwise imperfect.
(377) In the present case, the firms in the market at issue are differentiated in their competitive positioning, particularly as regards the alloy series they manufacture and sell (See Section 6.2.3.3), and their ability to serve incremental customers (see Sections 8.3.5 and 8.3.7). As to the latter crucial element firms must compete by investing in costly capacity extension, which require significant capital expenditures (see Sections 8.3.7 and 8.3.10.4, and recital (838)). Consequently, the leading players have established a strong position in the market on the back of past investments in capacity whereas spare capacity in the industry overall is projected to be limited (see Sections 8.3.7 and recital (838)). As a result, the Transaction can generate significant non-coordinated effects. The fundamental constraining role played by manufacturing capacities and their indication of market power held by firms is not affected by the bidding nature of the market. In bidding markets characterised by capacity constraints, in which customers’ (that is OEMs) ability to play one firm against the other is limited by the amount of capacity available to serve their demand.348
(378) Moreover, when analysing bidding markets, moreover, it is important to note that prices are individually negotiated with each customer and, therefore, suppliers can often engage in appreciable price discrimination across customers. In particular, a bid submitted to a customer in one specific tender does not have to be offered on similar terms to other customers in other tenders. The existence of such individualised pricing means that the price effects of a merger may be targeted at a particular subset of customers, for example, those that are more likely to substitute between the merging parties absent the merger. In other words, whereas a price increase across all customers may not be profitable (given that too many customers would be able to substitute away from the merging parties), a price increase for a specific subset of customers may be so, which increases the potential risk for the creation of anti- competitive effects through a merger.
(379) When prospective suppliers form and submit bids in a context where there is uncertainty over competing bids (for example about the quality and costs of rival offerings or the customer’s preferences), the pricing incentives of competing firms in bidding markets resemble those at work in ordinary markets with differentiated offers. In particular, if there is uncertainty on the required price level of the winning bid, each firm faces a trade-off between the probability of winning a tender and the margin earned in case of a successful bid. In other words, a higher bid would reduce the probability of winning the tender but would at the same time increase the margin earned if the bid is successful. This trade-off is qualitatively equivalent to the standard trade-off between quantity sold and price earned in other product markets. Each bidder therefore chooses its optimal bid to optimise the trade-off between bexpected sales and price, and thereby maximises its expected profits. Pricing incentives and the related incentives to exploit market power in bidding markets are therefore similar to those at work in standard pricing of differentiated products.
(380) In addition, the incentives to increase the price offered in bids following a horizontal merger in bidding markets characterised by uncertainty over competing bids are similar to those at work in ordinary markets with differentiated products. The primary difference is that the diversion of sales between competing firms should be understood in terms of expected sales (the probability of winning the tender) rather than actual sales. As noted above, the incentives of the merging firms to increase prices are thus determined by essentially the same factors that determine competitive effects in other markets (for example, measured by indicators such as diversion ratios, market shares, pre-merger margins, and the extent of spare capacity).
(381) Finally, in situations characterised by uncertainty on the quality, cost and spare capacity of rivals and on the customer evaluation for each of the products offered, the competitive constraint faced by each bidder is additionally influenced by the ex-ante probability that rival bidders may make more attractive offers on the basis of a superior competitive positioning and thus win the tender. Therefore, the number of credible bidders typically increases the ex-ante probability that the buyer will prefer a rival offer, and may thus increase the competitive constraint on any given bidder. As a consequence, it is not only the runner-up that may represent an appreciable competitive constraint on the winning bidder, and a decrease in the number of remaining bidders due to the merger may result in a reduction of the competitive constraint faced by the merged entity.
(382) With respect to the relevance of market shares in a bidding market, as explained in Section 8.3.4.3, the Commission considers that also in a market where most of the sales are made through bidding, market shares are informative of market power. In particular, market shares are informative of the capability of suppliers to successfully place volumes and satisfy customer requirements, and this ability is independent from the question of whether sales are made through bidding or not. While, considered alone, they may not fully reflect the extent of the competitive constraints in a prospective analysis, by being reflexion of actual market interaction they are however a fundamental starting point in the analysis of such competitive interaction.
8.3. Aluminium ABS
8.3.1. Introduction
(383) The activities of the Parties overlap in the production and supply of Aluminium ABS in the EEA.
(384) Both Parties are active in 5xxx and 6xxx alloys, while especially Aleris focuses in particular on 6xxx alloys. The Parties also both supply OEMs as well as Tier suppliers and distributors, and overlap at a significant number of customers.
8.3.2. The Notifying Party’s views
(385) The Notifying Party is of the view that the Transaction would not give rise to non- coordinated effects on the EEA market for the production and supply of Aluminium ABS.349 In particular, the Notifying Party argues the following:
(386) Flat rolled steel products exert a strong competitive constraint on Aluminium ABS. With around 90% of automotive body sheets made currently from steel in the EEA, Aluminium ABS is under constant pressure from steel, which is less expensive. OEMs adopt Aluminium ABS because of light-weighting needs, but as steel products become more and more advanced (and lighter), OEMs are ‘switching back’ certain parts to steel. Aluminium ABS is therefore exposed to competition from steel and Aluminium ABS suppliers are constrained in their pricing, because OEMs would revert back to steel in case of significant Aluminium ABS price increases.350
(387) The Parties’ combined share of supply of Aluminium ABS overestimates their competitive significance. This is in particular because the Parties’ shares are in large part driven by their respective largest customers, […].351
(388) The Aluminium ABS market is a bidding market and competition therefore restarts at every OEM tender.352 Overall market shares are therefore of limited informative value and the market position of suppliers is dependent on the outcome of a limited number of highly competitive and infrequent tenders.353
(389) The Parties are not close competitors, mainly because they focus on different OEMs and view different Aluminium ABS suppliers as their main competitors.354 Generally, closeness of competition is not to be seen as a particularly meaningful indicator for largely commoditised products.355
(390) There is and will continue to be excess capacity in the EEA market for the production and supply of Aluminium ABS. This results in significant price pressure,356 as competitors to the Parties are competing to fill their (recently expanded) lines.357
(391) Further, competitors to the Parties can accommodate the entire upcoming Aluminium ABS demand until at least 2024.358 By contrast, […].359
(392) With Constellium, Hydro, AMAG, Profilglass and Ma’aden, at least five competitors of the Parties will remain and compete with the Merged Entity.360 As the number of credible bidders is a relevant metric in bidding markets, this will ensure continued competition post-Transaction.361
(393) Imports and the threat of entry act as another competitive constraint on the Parties.362 In particular barriers to entry, while existent, have in the past not prevented new suppliers from entering the market and competing with the Parties.363 Further, OEMs are turning to non-EEA Aluminium ABS suppliers.364
(394) OEM customers have significant buyer power. Aluminium ABS demand is concentrated in a limited number of strong global OEMs that employ aggressive tactics to keep prices down.365
8.3.3. The Commission’s assessment
(395) In the following Sections 8.3.4 to 8.3.12 the above arguments by the Notifying Party are addressed.
(396) Based on the results of the market investigation and all the evidence available to it, the Commission finds that the Parties’ arguments cannot be upheld and that concerns arise for Aluminium ABS.
8.3.4. Market structure and market share metrics
8.3.4.1. Brief overview of Aluminium ABS manufacturers active in Europe
(397) The following section introduces the Aluminium ABS suppliers, other than the Parties, that are active in the EEA.
(8.3.4.1.1) Constellium
(398) Constellium is a global manufacturer of aluminium rolled and extruded products headquartered in Amsterdam (the Netherlands). Constellium was formed in 2011 through the spinoff of the Engineered Aluminium Products business unit from Rio Tinto, an international mining group.
(399) Constellium owns and operates more than 25 manufacturing sites in Europe, North & Central America and China, of which 15 production facilities are in Europe.366 In the EEA, Constellium produces Aluminium ABS in France (Neuf-Brisach plant) and in Germany (Singen plant). Constellium supplies both 5xxx and 6xxx Aluminium ABS series and is currently a qualified supplier for a majority of automotive OEMs in the EEA.
(400) In 2018, Constellium’s revenue was around EUR 5.7 billion.367
(8.3.4.1.2) Hydro
(401) Headquartered in Oslo, Norway, Hydro is a fully integrated aluminium company with operations covering all major activities along the aluminium value chain from bauxite extraction and alumina refining to extruded and Aluminium FRP. Hydro’s production of Aluminium FRP is exclusively located in Europe. In total, Hydro has five rolling mills, out of which two are located in Norway. Its automotive production is concentrated within their Grevenbroich (Germany) and Hamburg (Germany) plants.
(402) Hydro supplies both 5xxx and 6xxx Aluminium ABS series and its key customers for Aluminium ABS include […].368
(403) In 2018, Hydro’s revenue was around EUR 16.3 billion.369
(8.3.4.1.3) Austria Metal AG (AMAG)
(404) AMAG is a manufacturer of primary aluminium, premium cast and Aluminium FRP headquartered in Ranshofen, Austria, where it produces its Aluminium ABS in the EEA. AMAG produces 5xxx and 6xxx series.370
(405) In 2018, AMAG’s revenue was around EUR 1.1 billion.371
(8.3.4.1.4) Profilglass
(406) Profilglass is an Italian supplier of Aluminium FRP. Profilglass’ production plant is located in Bellocchi di Fano (Italy).
(407) Profilglass currently supplies Aluminium ABS made of 5xxx alloys. It has recently developed a CASH line that is expected to be fully ready in two years’ time after undergoing the necessary homologation process with the automotive OEMs. In addition to the new CASH line, Profilglass is also building a new cold mill.372 Profilglass’ customers are mostly located in Italy and Germany.
(8.3.4.1.5) Ma’aden Aluminium (‘Alcoa/Ma’aden’)
(408) Ma'aden Aluminium (‘Ma'aden’) was established in 1997 by the Saudi government. In 2008, 50% of Ma’aden’s shares were floated on the Saudi Stock Exchange. Ma’aden explores, develops and produces a diverse portfolio of mineral assets, including gold, zinc, phosphate, aluminium, and industrial minerals.
(409) In 2009, Ma’aden and Alcoa Inc. (‘Alcoa’), an American producer of primary aluminium, established Alcoa/Ma’aden, a vertically integrated Aluminium FRP supplier, by way of a joint venture.373
(410) In 2012, Alcoa/Ma’aden started the construction of the Ma’aden Alcoa Automotive Body Sheet Project as part of a multi-billion dollar investment374 that led to the addition of an automotive rolling mill at Ras Al Khair (Saudi Arabia). Alcoa/Ma’aden started supplying European OEMs in 2017. […] qualified Alcoa/Ma’aden in 2017 for 5xxx Aluminium ABS.375
(8.3.4.1.6) Aluminium ABS manufacturers with limited or no presence in the EEA
(411) In addition to the manufacturers described in the above recitals, there are some manufacturers not active in the EEA market for the production and supply of Aluminium ABS that have been or are being qualified by European OEMs.
(412) According to the Notifying Party, […] qualified Nanshan in 2017 for 5xxx series for its European plants and recently for 6xxx-skin series globally. The information provided by the Notifying Party suggests that, according to its best estimates, in 2018, Nanshan imported only 5 000 tonnes into the EEA (recital (336)).
(413) Furthermore, according to the Notifying Party, Kobe Steel, which produces Aluminium ABS through its Chinese subsidiary Kobelco, has some imports of Aluminium ABS into the EEA. However, in the information provided by the Notifying Party with regard to imports from China, there is no reference to past or future imports from Kobelco into the EEA (see Figure 41).
8.3.4.2. The Parties’ activities
(414) The Parties are global manufacturers of Aluminium FRP. Novelis is also a recycler of aluminium.
(415) According to the Notifying Party, in the EEA Novelis supplies Aluminium ABS, Standard FRP, aluminium beverage and food can and aluminium foil. Aleris supplies Aluminium ABS, Standard FRP, aluminium for aerospace applications and heat exchangers (Figure 43).
(416) Therefore, according to the Parties, in the EEA their activities only overlap in the manufacture and supply of Aluminium ABS and Standard FRP.
8.3.4.3. Market and capacity share metrics
(417) As explained in the Horizontal Merger Guidelines, ‘market shares and concentration levels provide useful first indication of the market structure and of competitive importance of both the merging parties and their competitors’. 376
(418) The choice of the relevant share measure depends on the circumstances of the specific industry in question. Moreover, different share measures may have different advantages and shortcomings in indicating market power. It may therefore be useful to analyse a combination of different share measures as complementary first indicators for market power. As explained further below, in this case, and consistently with its previous practice,377 the Commission considers two share measures to be relevant indicators for market power: (i) market shares on the basis of sales, which are expressed both in volume and in value; (ii) shares of capacities for the production of a particular product.
(419) First, market shares provide an important first measure of the relative positions of the different producers as suppliers to third parties.
(420) They are a first indicator of whether firms in a given market may possess market power.
(421) The Notifying Party, due to the bidding nature of the market described in Section 8.2.3, submits that market shares (based on sales) may not fully reflect actual market power in the Aluminium ABS industry. This is because market shares reflect the results of relatively large competitive tenders organised by OEMs in previous years.378
(422) The Commission recalls that in general it can be considered that the larger the market share, the more likely a firm is to possess market power.379 Therefore, market shares based on sales are informative of the capability of suppliers to successfully place volumes and satisfy customer requirements, and, in particular for market shares expressed in value, they are also indicative of the ability of suppliers to obtain sales of higher value alloys.
(423) If market shares in sales are considered alone, they may not fully reflect the extent of the competitive constraints in a prospective analysis. However, consistent with the Horizontal Merger Guidelines,380 they are a fundamental starting point in the analysis of such competitive interaction because they are a reflexion of actual market interaction.
(424) The fact that market shares expressed in sales are indicative of market power is also corroborated by the fact that the Parties themselves include market shares in their metrics for assessing their market performance vis-à-vis their competitors.381
(425) Second, in view of the characteristics of the industry as described in Sections 8.2.2 and 8.2.3, and in consideration of the fact that supply capacity constrains the capability of suppliers to serve customers, competition in the Aluminium ABS industry is largely driven by supply capacity,382 which in the present case is closely linked and can be approximated to CASH capacity, subject to the adjustments that will further be discussed in the assessment of spare capacities and in Annex I, Section 2.3.
(426) The importance of capacity as a parameter indicative of market power is also reflected by how Aluminium ABS manufacturers assess the market. The Parties take into account their capacity and the capacity of their competitors for assessing the competitive conditions of the market and for making their strategic decisions in their ordinary courses of business (see Section 8.3.10.1).
(427) Therefore, market shares alone may not be a suitable indicator of market power because in the market for the production and supply of Aluminium ABS the demand is growing and a manufacturer that recently expanded its capacity might have a relatively low level of sales, but could be in a position of potentially growing its market shares in the near future. Therefore, the market power of a manufacturer that has a low level of sales, but a large capacity, might have a market power that is higher than what is suggested by only its market share.
(428) Capacity shares provide a direct indication of production capabilities at EEA level, which the Commission regards as an important driver of competitive dynamics among EEA suppliers. The Commission considers therefore that capacity shares are an appropriate and informative structural metric of market power in the Aluminium ABS industry.
(429) Moreover, capacity shares, and particularly changes thereof, reflect lasting changes in the structure of a market as well as the magnitude of such structural change. In the market for the production and supply of Aluminium ABS at hand, which is characterised by limited spare capacity (see Section 8.3.7), over time market shares tend to align to capacity shares and are less subject to fluctuations, which are otherwise typical for bidding markets.
(430) In conclusion, and in line with previous decisions in the Aluminium ABS industry, in order to accurately capture the full market power of the Aluminium ABS suppliers, the Commission also considers capacity shares.383
(431) The Commission also observes that the combination of market shares and capacity shares as metrics for assessing market power is suitable for assessing the Parties’ market power, also with respect to possible first mover advantages that Novelis and Aleris have toward, respectively, […].384 This is because while market shares are suitable for assessing market power as of today and in the recent past years, capacity shares are more suited for assessing market power in the years to come, irrespectively of past sales (as explained Section 8.3.4.4, recital (465), the Parties’ capacity shares nevertheless have to be regarded as a conservative indicator of their actual market power).
(432) With respect to the Notifying Party’s argument that spare capacity is more representative of market power, compared to capacity (and therefore capacity shares do not provide indication of market power), the Commission explains in detail in Annex I, Section 2.5.3 that looking at uncommitted capacity only distorts the analysis because the extent of capacity utilisations is cyclical and, while over a given time span a supplier may be more (or less) constrained, as soon as contracts come up for renegotiation this capacity goes back into the market. For the avoidance of doubt, as explained in Section 8.3.7, and more specifically in recital (608), the Commission considers that spare capacity, and in particular the spare capacity of the Parties’ rivals, are important metrics for assessing the ability and the incentives that the Parties’ rivals have in increasing their supply in case of a price increase of the Merged Entity.
(433) Further, as explained in each relevant section of the present Decision, the Commission considers other quantitative and qualitative indicators to the extent that they might better reflect at least some aspect of market power in the supply of Aluminium ABS. As also indicated in the Horizontal Merger Guidelines (paragraphs 28-38), other factors include the elimination of the competitive force exerted by Aleris, the likely limited reaction of competitors, the limited nature of buyer power and the low level of spare capacity.
(434) The Notifying Party also argues that the reliance on market shares and capacity shares as metrics indicative of market power is not consistent with the Commission’s past practice.385 However, as explained in the following, a review of each of the cases cited by the Notifying Party reveals that this approach is fully consistent with past practice.
(435) As noted by the Commission in Mahle Behr / Delphi Thermal Systems Business (paragraph 29 of the decision), in order to obtain an accurate view of the parties' strength in bidding markets, ‘it is important to assess market shares over a significant period, covering both past and future’, which is precisely the Commission’s approach to the present case. In addition, the uncertainty associated with future market shares requires taking into account capacity shares as a second share metric (Section 8.3.4.4).
(436) The Notifying Party’s own estimates show the combined market share of the Merged Entity in volume to be consistently above [50-60]%, specifically [70-80]% in 2015 and [60-70]% in 2018, then [50-60]% in 2019 and [50-60]% in 2023. Thus, when assessing market shares over a significant period, covering both past and future, the Merged Entity’s market share would consistently account for over half of the market.
(437) Further to this, the Commission draws attention to a number of factual differences between the decisions cited by the Notifying Party where the Commission cleared mergers in the automotive sector, and the present case.
(438) In Mahle Behr / Delphi Thermal Systems Business, the Commission found evidence for the existence of spare capacity in the market as well as an ability to increase capacity in the short-term on the part of the merging parties' main competitors.386 However, in the present case, the Commission has found that spare capacity is limited (see Section 8.3.7) and that that capacity expansion requires relevant investment and time (see recitals (879)–(881) regarding the investment, and recital (884), mutati mutandis, for the time required for expanding capacity). Further, other than in the present case, where OEMs buyer power is limited (see Section 8.3.11), the Commission found OEMs to have bargaining power.387
(439) In the present case, capacity constraints play a significant role in the market and capacity cannot or is not likely to be expanded ‘in a timely fashion’, unlike in JCI / Automotive business of Keiper Recaro Group (paragraph 95, 108 and 115 of the decision) where it was explained that, although there did not seem to be a significant overcapacity in the market, capacity did not seem to be a significant barrier to entry or expansion as competitors confirmed that they could expand capacity in a timely fashion if awarded a contract. Furthermore, unlike in the present case, in JCI / Automotive business of Keiper Recaro Group (paragraph 108 of the decision), the market investigation showed that a significant majority of suppliers stated that they bid for requests for quotes or tenders even in cases where they do not have sufficient spare capacity given that a significant volume is to be awarded.
(440) In JCI / Automotive business of Keiper Recaro Group (paragraph 103 of the decision), the market investigation also confirmed that OEMs themselves could take steps to counter attempts by the merged entity to increase prices following the merger. In particular, OEMs accounting for a significant majority of car production bcould either switch to an alternative supplier and/or sponsor the entry of a new supplier and/or even start supplying recliners themselves. With respect to sponsoring entry in particular, some extra-EEA suppliers appeared to be ready to supply the EEA but found it difficult to establish a significant footprint due mainly to reasons that were within the control of OEMs to change (e.g. historical supply relationships, lack of experience in applying to tenders, etc.).
(441) In the present case, as explained in recital (901), there is a relevant amount of know- how that a new entrant needs to acquire, which leads to a long time-to-market for new entrants, particularly for 6xxx-alloys. The sponsorship of an OEM cannot reduce this long time-to-market, and the only beneficial effect of an OEM sponsorship would be the facilitation in selling some Aluminium ABS, once the required quality is achieved. Therefore, in the present case, the sponsorship of an OEM is less effective in terms of facilitating the entrance of a new manufacturer in the market. In addition, the market investigation has indicated that OEMs do not expect new entrants to the Aluminium ABS market in the EEA in the next three years (see recital (902)).
(442) Furthermore, […],388 the Commission points out that this ability alone would not be sufficient to demonstrate the OEMs’ ability to convince or compel the Merged Entity or its competitors to actually increase their capacity, in particular given their limited incentives or ability to do so (see Sections 8.3.9 and 8.3.10).
(443) In the present case there is no significant level of spare capacity in the industry (see Section 8.3.7) and the target is not ‘a supplier in decline, suffering from a lack of focus’, unlike in Johnson Controls / Robert Bosch / Delphi SLI (paragraph 17 of the decision), where it was deemed that a significant level of spare capacity would have to be up to 30%, which significantly exceeds the level of spare capacity found in the present case.389 In Johnson Controls / Robert Bosch / Delphi SLI (same paragraph) the market investigation also found that, unlike in the present case, suppliers could be switched with relative ease and, new suppliers could be introduced. Therefore it was considered that competitive pressure was effectively put upon the established players.
(444) Although the Notifying Party claims that in U-Shin / Valeo Cam the assessment of market shares did not play a significant role due to the bidding charateristics of the market (paragraph 39 of the decision), there are several differences in the present case that have to be highlighted.
(445) As opposed to U-Shin / Valeo Cam (paragraph 37 of the decision), in the present case the market shares of the Notifying Party and of its rivals are not likely to fluctuate to a significant extent because, although market shares depend on the number and volume of successful bids, these depend in turn on the competitors’ ability to make a credible offer, that is their available capacity. Since no significant expansions are currently foreseen and spare capacities of competitors are limited, the Notifying Party’s market shares will be rather stable and, in the long run, largely converge to its capacity share (see for example recital (500)).
(446) In line with this, due to Novelis’ position in the market, the loss of a tender or even a number of tenders could not dramatically affect the Merged Entity’s market share. Furthermore, while in U-Shin / Valeo Cam the market investigation confirmed that there were no significant barriers to entry (paragraph 48 of the decision) in the present case, the Commission found that barriers to entry are significant (Section 8.3.10.4).
(447) In ITW / EF&C (paragraph 31 of the decision), the Commission found that there were a number of credible competitors (17 for fasteners and 6 for PRVs), whereas in the present case the number of credible competitors able to supply the range of relevant products other than the Parties is limited to three (recital (973)). Also, in ITW / EF&C (Tables 1 and 2), it is shown that the combined market shares of the parties were much lower than in the present case (less than 40%) and that only on a narrower individual product market definition would they exceed 50-60%.
(448) Unlike in the present case, in Magna / New Venture Gear (paragraph 63 of the decision), the Commission found that there were ‘no significant capacity constraints on entry or growth’ and that past volatility in market shares was evidence of the possibility of entering the market, in, Magna / New Venture Gear (paragraph 45 of the decision). Furthermore, in Magna / New Venture Gear (paragraph 50 of the decision), the Commission found that OEMs had countervailing buyer power, which could induce new entry into the market. Also, unlike in the present case, in Magna / New Venture Gear (paragraphs 54, 61 and 62 of the decision), the market investigation revealed that potential competitors could participate in bids.
(449) Unlike in the present case, in Volkswagen / MAN (paragraph 26 of the decision), the Commission found that suppliers had ‘spare capacity and a fully developed network for sales and after sales services across the EEA countries, allowing for each of the competitors to expand in case of a price increase in any of the countries’. Furthermore, in Volkswagen / MAN (paragraph 121 of the decision), the Commission found the market to be very narrow, something that is no longer true for aluminium ABS, because OEMs are employing more and more Aluminium ABS.
8.3.4.4. Data sources and computations of market shares
(450) The Notifying Party submitted EEA market shares expressed in sales volume, and capacity shares, of both the Parties and of their rivals.
(451) Both market shares and capacity shares provided by the Notifying Party cover the period 2015 to 2018, as well as a prediction until 2023.
(452) The Notifying Party argues that the market share data they provided for the period beyond 2018 are ‘reliable predictions of future distribution of supplies among the different players’ because they are based on sales nominated by OEMs.390 Nominated sales are those sales expected to occur as a result of a tender. As explained in Section 5.4, after a tender is complete, the OEM that organised the tender nominates the selected supplier and concludes an agreement where, among other terms, the supply period and the expected quantities to be supplied are agreed.
(453) The Notifying Party estimates that these nominated values represent […]% of the total sales in 2019, […]% of the total sales in 2020, and […]% of the total sales in 2022.391
(454) The Commission considers that, as explained in Annex I Sections 2.2.1 and 2.2.2, market shares calculated by the Notifying Party for future years are affected by a large degree of uncertainty. The Commission therefore concludes that the reliability of these shares is low, particularly for year 2020 and beyond. The uncertainty is due mainly to two reasons.
(455) First, nominated sales might differ from the sales that will occur in the future. As explained in recital (73), during the course of the years, the quantities supplied to an OEM might deviate from those originally nominated. As one OEMs explained, ‘[i]n the tender the Company does not commit on volumes with the suppliers’ and also explained that as a result of the tender, it ‘provides the suppliers with their best forecast of aluminium ABS demand, based on the program details and the expected sales of the vehicle’.392 Supply contracts often foresee the possibility that OEMs review annually their planned Aluminium ABS consumption, and the related purchased quantities are reviewed accordingly (see recital (73)).
(456) In a submission to the Commission,393 the Notifying Party explained that OEMs in the past years modified, or investigated opportunities for modifying, their long-term supply contracts with the Parties. The Notifying Party also provided evidence of such modifications, as explained below.
(457) In 2019, […].
(458) In 2017, […].
(459) In 2018, […].
(460) Captioned in Figure 44, […]:
Figure 44 […]
[…]
(461) […]:
Figure 45 […]
[…]
(462) In the Reply to the Letter of Facts, the Notifying Party itself states that ‘the Parties have never contested that nominated sales may differ from actual sales’.394
(463) Second, according to the Notifying Party, in 2019 and 2020 nominated sales represent, respectively, […]% and […]% of the total expected sales in the EEA. These values leave uncertainties for market shares value of up to, respectively, […]%-points and […]%-points, and it would be very difficult to predict which manufacturers might win the remaining bids. For example, an Aluminium ABS manufacturer that was not very successful in previous bids, and therefore has to utilise its spare capacity, might have an incentive to submit more competitive bids and obtain a large share of the non-nominated sales. This uncertainty becomes particularly severe beyond 2019, when the fraction of nominated sales over the expected demand becomes smaller.
(464) Therefore, the Commission considers that, with respect to market shares, the most reliable evidence on the current position of the Parties and of their rivals are based on data of 2018 or in any event at most up to 2020. Nevertheless, the Commission bconsiders that market share estimates for future years may provide qualitative indications of the expected evolution of market shares for future years. It therefore considers that market share estimates for future years provide for an indication of how the market is likely to trend as opposed to an indication (in absolute terms) of market shares.
(465) With respect to capacities, data are less affected by uncertainties,395 therefore future capacity shares are more reliable than future market shares. As explained in Section 8.3.10.4, due to the considerable amount of money and time required for expanding Aluminium ABS capacity, capacities are more stable over time, and, when variations occur, these can be reliably predicted by both the Notifying Party and by each manufacturer that provided data for the market reconstruction. Indeed in the response to the Commission’s RFI for the market reconstruction competitors did not only indicate for future years the capacity currently installed but also the amounts of capacity that they are planning to add over the time period up to 2023 (included), given the current level of demand. While changes to forecasted capacity developments can occur (e.g. if demand grows beyond current projections), an assessment of future years’ capacity is a more reliable (in the sense of ‘less uncertain’) and stable metric than sales shares.
(466) In the present case, it is important to note that capacity shares ought to be considered the the most conservative indicators of the Parties’ future competitive position in the market. An exclusive look at capacity shares would assume that all competitors, even recent entrants, will be equally successful at filling their capactities. However, although new entrants might have an incentive to bid at lower price for filling their capacities, the Parties are established suppliers with a very good track-record of filling their lines and delivering to OEMs high quality and high grade products, such as the 6xxx and 6xxx-skin products, which represent a large part of the Aluminium ABS market, exceeding [60-70]% of the EEA demand (see Figure 66).
(467) Further, since capacity is a measure intrinsically related to volumes, an exclusive focus on capacity numbers would not take into account the fact that historically and at present the Parties’ market share based on value is larger than the market share based on volume, suggesting that the Parties focus on the higher end of the market or in any event are able to extract more value from OEMs. Looking at capacities only would underrepresent the Parties’ competitive position, as they are clearly particularly successful in higher value segments.
(468) In addition to the data provided by the Notifying Party, the Commission gathered sales and capacity data from all the Aluminium ABS manufacturers considered by the Notifying Party in its submission of market shares. In particular, the Commission obtained sales data expressed in volume and in value for the period 2015–2018, nominated sales until 2023, and sales forecasts for the years 2019–2023, which are nominated sales, plus forecasts of future sales (not yet nominated). The Commission also gathered data on manufacturing capacity of each of the manufacturers for the period 2016–2023.
8.3.5.4. Even when accounting for developments claimed by the Notifying Party, the Merged Entity would hold significant sales and capacity shares with important increments
(497) The present section demonstrates that, based on the evidence available to the Commission, the Merged Entity would hold significant sales and capacity shares with important increments, even when the market developments claimed by the Notifying Party are accounted for.
(8.3.5.4.1) Expected market shares
(498) The Notifying Party claims that the high market shares of the Parties are a legacy of their first mover advantages and they are going to decrease in the future,405 because of the recent capacity expansion or market entry of the Parties’ competitors.406 As a proof of their claim, in addition to the historic trend of the market shares from 2015 to 2018, the Notifying Party estimated market shares in volume for the period 2019- 2023 (Table 8).
(499) As explained in Section 8.3.4.4, the Commission considers that market shares for the year 2020 and beyond (which are based on nominated sales) cannot be considered a reliable metric. The lack of reliability of market shares beyond the year 2020 is confirmed also by the data reported in Table 8: the reported total nominated sales decrease year after year from 2018, whereas the sales of Aluminium ABS are expected to increase in the years to come (see, for example Annex I, Section 2.5.1), which confirms that the total nominated sales reported by the Notifying Party only represents a small and decreasing percentage of the total market in the years to come.
(500) Nonetheless, if, as the Notifying Party claims, estimates of future market shares based on nominated sales are deemed to be representative of the shares of supplies among the various manufacturers,407 the Notifying Party’s estimates would indicate that the Parties’ combined market share in volume for the period 2019–2023 would remain very high and above [50-60]%, with increments of [10–20]%-points. Noticeably, the expected reduction of the Parties combined market share in volume in the period 2019–2023 is relatively limited and the combined market share is expected to fluctuate from [50-60]% in 2019 to [50-60]% in 2023, and, in the period 2019-2020 are expected to increase by [0-5]%-point.
(501) The reason for this expected modest reduction of the combined market share appears to be due to the fact that in the period 2015–2018 the Parties’ combined market share decreased from [70-80]% in 2015 to [60-70]% in 2018 (Table 2), which is a much more pronounced reduction than what is expected by the Notifying Party for the period 2019-2023. This indicates that the expected reduction of the Parties’ combined market shares, due to the capacity expansion of the Parties’ rivals, already took place, to a large extent, during the period 2015-2018.
(518) […].
(519) […]411 […].
(520) […].
Figure 46 […]
[…]
(521) […],412 […].
(522) Fourth, the majority of the Parties’ customers consider that Novelis benefits from certain market strengths that give it a competitive advantage compared to its rivals with respect to Aluminium ABS.413
(523) An OEM identified Novelis as the technology leader and remarked its strong technical capabilities: ‘Novelis would be regarded as the technology leader in this area’.414 Another OEM expressed a similar view and also remarked its strength in terms of innovation and in offering competitive products: ‘NOVELIS being one of the leaders in the Automotive FRP market, it is a supplier that can offer strong technical support, innovation, competitive solutions’.415 The emphasis on innovation is shared by another customer, who considers the ‘[…] exclusive[ly] rights to produce own developed alloys for FRP for automotive end-uses’ one of Novelis’ competitive advantages over its competitors.416 A further OEM remarked that ‘Novelis has […] excellent quality and delivery precision’,417 while another one remarked that, in addition to superior capabilities, ‘Novelis has better […] footprint with respect to the competitors’.418
(524) Another competitive advantage of Novelis over its competitors acknowledged by one Tier automotive supplier is its ‘biggest product portfolio’,419 which qualifies Novelis as ‘[…] the leader in product and develop[ment]’.420
(525) […].
Figure 47 […]
[…]
8.3.5.6. Conclusion
(526) For the reasons set out in this Section 8.3.5, and considering all evidence available to it, the Commission concludes that the Transaction results in very large combined market shares and high combined capacity shares, indicative of a dominant market position in the EEA market for the production and supply of Aluminium ABS. Novelis is able to act at least to some extent independently of its customers and competitors already prior to the Transaction, and the Transaction would consolidate that ability for the Merged Entity. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.6. By controlling large capacities, the Merged Entity would be a pivotal player and enjoy significant market power
(527) In terms of capacity, the Merged Entity would, according to the Notifying Party’s data, have a capacity share of [40-50]% in 2018 and of [40-50]% in 2023 (as detailed in Table 6 and Table 10 in Section 8.3.5). The Commission’s market reconstruction puts the Merged Entity’s capacity share clearly above [50-60]% in 2018, specifically at [50-60]%, and close to [50-60]% in 2023, specifically at [40-50]% (as detailed in Table 7 and Table 11).
(528) In order to further investigate to which degree the Merged Entity’s high combined capacity post-Transaction would lead to market power, the Commission analysed the extent to which the capacity of the Merged Entity’s competitors would post-Transaction be able to cover the entire market demand.421
(529) Indeed, it is well known from the economics literature, and consistent with the Commission’s case practice, that in markets with capacity constraints, pivotal firms enjoy an appreciable degree of market power.422 This is because even in a worst-case scenario, where rivals successfully win orders filling their entire capacity, the pivotal supplier would nonetheless be de facto the only supplier for the remaining part of demand that cannot be served by rivals. Pivotal suppliers are therefore in a position to exercise an appreciable degree of pricing power in the market, being aware that the market (that is, customers) are dependent on their supply.
(530) Small suppliers have a strong incentive to undercut competitors because if they fail to do so they risk ending up with no sales (as their competitors can fully cover the entire market demand). To the contrary, pivotal suppliers (those who face some degree of residual demand that cannot be covered by competitors) face a trade-off between pricing aggressively to capture some of the demand for which they face competition from competitors and keep prices high to exploit the portion of (residual) demand that cannot be covered by rivals. The larger the portion of residual demand faced by the incumbent supplier, the larger the amount of demand for which the incumbent knows it is de facto the only supplier and therefore the larger the incentive to keep prices high and avoid undercutting competitors.
(531) The degree of market power exercised by a pivotal supplier depends on its degree of pivotality (that is, on the extent to which rivals are insufficient to cover total market demand). A merger may therefore cause anti-competitive effects by making a supplier pivotal that previously was not or by conferring to a supplier that was already pivotal even more control over indispensable production facilities.
(532) The Commission’s pivotality calculations, based on the figures for market shares, total demand and capacity suggest the following:
(a) First, as displayed in Table 22–24 of Annex I, Novelis is already pivotal pre- Transaction. That is, it faces significant residual demand that cannot be covered by its rivals. This indicates that Novelis already has an appreciable degree of market power prior to the Transaction, […].
(b) Second, Novelis would become even more pivotal post-Transaction. That is, it would face even more demand (an additional […] tonnes circa, corresponding to [10-20]% of demand in 2019) for which it would be de facto the only supplier.
(533) These conclusions are particularly strong when the CASH capacity from the Commission’s market reconstruction adjusted for sales to non-EEA and non-ABS customers are used (Figure 48). The Commission considers this to be the most appropriate measure for capacity available to ABS customers in the EEA. However, as shown in Annex I, the conclusions hold also when considering the figures provided by the Notifying Party or the figures from the Commission’s market reconstruction, not adjusted for sales to non-EEA and non-ABS customers.
Figure 48 […]
[…]
Source: See Annex I.
(534) In the Reply to the SO and the Reply to the Letter of Facts, the Notifying Party fundamentally disagrees with the Commission's analysis and claims that based on uncommitted capacity and non-nominated demand rivals do have enough capacity to cover upcoming demand up to at least 2023.
(535) Concretely, the evidence on pivotality based on uncommitted capacity and non- nominated demand submitted by the Notifying Party shows the following:
(a) First, Table 3 and Figure 5 of the CRA capacity report show that (based on the Parties’ data) rivals have enough uncommitted capacity to cover the total outstanding non-nominated production that is expected to take place up to […].
(b) Second, Table 4 of the CRA capacity report shows that rivals also have enough uncommitted capacity to cover the […] production volume of tenders that are currently indicated as ‘open’ in the bidding data of Novelis.
(536) This complex discussion is covered in detail in Section 2.5.3 of Annex I, where the Commission sets out the reasons why the approach proposed by the Notifying Party gives a distorted view of competition in the market for the production and supply of Aluminium ABS. At a high level, the Commission notes the following considerations.
(537) First, the methodology of the Notifying Party is flawed and fails to show that rivals of Novelis have enough capacity to cover upcoming demand up to at […].
(538) As regards point (a), the methodology proposed by the Notifying Party covers the open demand for production taking place between […]. This cannot be taken as the basis for assessing the relevant capacities for tenders that will be nominated in the near future (let alone those that will take place further in the future).
(539) Production occurring over the […] period is mostly about competition that occurred in the past and therefore focussing on uncommitted capacity and non-nominated demand for production occurring over the […] period is not very informative about competition (tender processes) that will take place over the same period (for tenders nominated during the next five years – from […] – production will take place over […]).
(540) For example, tenders nominated in 2020 (that is, six months from now), for which SOP would be in 2022, since SOP is typically at least two years after nomination, would lead to a production until 2027–29 under a normal 5–7 year production cycle. In other words, even for very current tenders, the Notifying Party’s proposed methodology (which covers open capacities […]) would address only a small part of the production cycle. Tenders that are two years out in the future are instead effectively not covered at all by Notifying Party’s methodology.
(541) In relation to point (b), the Commission considers that the approach is in principle more appropriate but the methodology suffers from the flaw that Novelis’ bidding data does not comprise the entirety of upcoming tenders, as it can only account for the volumes relating to those tenders that are open at present and, as such, recorded already in the bidding data of Novelis. The only upcoming tenders registered in Novelis’ bidding data […].
(542) […].
(543) […].
(544) Third, if the approach proposed by the Notifying were correct, the Notifying Party’s conclusion that not even the largest firm in the market (Novelis) is pivotal would suggest that tender outcomes should be rather competitive (as each competitor knows that by pricing above costs it will likely be undercut by a rival willing to better utilise its spare capacity). […].
(545) […].
(546) The Commission agrees that there cannot be price effects on contracts already signed.423 Indeed, the Commission did not claim in the SO (nor does it in this Decision) that the Transaction would lead to price effects for contracts already signed nor that the main harm from the Transaction would be price increases for production taking place in the next five years. This production is indeed mostly the result of competition that occurred in the past.424 The Commission’s main concern is the outcome of the tenders that will take place in the next five years (as contracts continuously come to expiry and have to be re-negotiated). If the Transaction occurred in a spot market and was therefore capable of affecting even the prices of production taking place immediately after the Transaction, the Commission’s concerns would have simply been even more pronounced.
(547) In conclusion, the Parties would not only control a large share of the market (in terms sales and capacity) after the Transaction but would also face competitors with limited capacity in light of the level of market demand. The inability of competitors’ overall capacity to cover the whole market demand means that Novelis has maket power already before the Transaction. Moreover, by acquiring the capacity of Aleris, Novelis would face even less rival capacity after the Transaction. This reduced competitive constraint on Novelis would allow the merged entity to increase prices after the Transaction without facing competitors’ ability to readily expand supply.
(548) These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.7.Spare capacities in the market are low
8.3.7.1. Current spare capacities are low
(549) The Notifying Party submits that ‘[t]here is significant aluminium ABS excess capacity in the EEA’.425 Specifically, the Notifying Party submits spare capacity in 2018 to be […].426
(550) The Commission however finds that the Notifying Party has significantly overestimated current spare capacity. The Commission, in undertaking a market reconstruction involving all competitors to the Parties in the EEA market for the production and supply of Aluminium ABS, found the EEA spare CASH capacity in 2018 to be less than 160 kilotonnes.427 This amounts to only between 10 and 20% of total 2018 EEA CASH capacity (as visualised in Figure 49).
(551) The Commission further found that the Merged Entity accounted for more than […].428 […].429 This means that the other ‘three established suppliers with a proven track-record of meeting demands of OEMs (Constellium, Hydro and AMAG)’430 – which are the Parties’ main competitors – together with the remaining two competitors and recent entrants, Alcoa/Ma’aden and Profilglass (both currently almost exclusively supplying 5xxx grades), only accounted for […]. This further means that the demand that could be readily contested by these companies through their spare capacity in reaction to a price increase by the Merged Entity is very limited, and considerably lower than the Notifying Party estimated.
Figure 49: […]431 432
[…]
(552) An assessment of the development of CASH capacity in the EEA market for the production and supply of Aluminium ABS, and of the Parties’ own view on capacity in internal documents further supports the Commission’s finding that current spare capacities are limited and indicates that spare capacities will be even lower in coming years. This evidence is presented in Section 8.3.7.2.
(553) Section 8.3.7.3 presents evidence on customer and competitor views on the limited nature of current and future capacity, and of the impact this has on opportunities to switch and on prices.
8.3.7.2. Even when accounting for recent and planned capacity expansions, spare capacities are limited and have been overestimated by the Notifying Party
(8.3.7.2.1) Capacity expansions relevant for Automotive ABS occur mainly in the form of addition of CASH lines
(554) As explained in Section 5.3, the production of Aluminium ABS generally requires a CASH line.
(562) Constellium announced a new CASH line at its plant in Neuf-Brisach, France, which the Notifying Party expects to have an Aluminium ABS capacity of […] tonnes by 2020.
(563) Hydro is constructing a new CASH line at its Grevenbroich plant in Germany. The Notifying Party expects it to reach an Aluminium ABS capacity of […] tonnes by 2020.
(564) AMAG is in the process of ramping up a CASH line recently installed at its Ranshofen site in Austria, which the Notifying Party expects to reach an Aluminium ABS capacity of […] tonnes by 2020.436
(565) Profilglass, a recent entrant, is ramping up its recently built CASH line at its Fano plant in Italy and the Notifying Party expects it to reach an Aluminium ABS capacity of […] tonnes by 2020.
(566) Alcoa/Ma’aden, which is also a recent entrant, is ramping up its CASH line at its Ras Az Zawr site in Saudi-Arabia and the Notifying Party expects it to reach an Aluminium ABS capacity of […] tonnes by 2020.
(567) The Parties, in contrast, have not added new CASH lines since 2016 (Novelis constructed its ‘LYNX’ line in 2015 and it is also considered the ‘[f]irst mover in capacity expansions’437). However, both Parties are set to increase their Aluminium ABS capacity by measures undertaken on their existing lines. These increases will however lag far behind those of the Parties’ competitors.438
(568) However, and in contrast to Novelis’ actions with respect to capacity expansions since 2016, when considering capacity developments prior to those described in recitals (561) to (567), Novelis does not appear to be a ‘first mover’, but rather to be reacting to competitor initiatives and needs of a long term customer. The Notifying Party states in a slide presented during the Oral Hearing on 23 July 2019 that […] .439 […],440 […]. The Commission thus observes that three competitors to the Notifying Party (Constellium, Hydro and AMAG) likely preceded Novelis in announcing capacity expansions in 2013. […],441 […]442 […]. The decision by the market leader Novelis in 2013 to increase its capacity by means of investing in a new line thus happened in the context of preceding expansion announcements by competitors and in relation to the need to accommodate demand of a major long-term customer.
(569) The Notifying Party claims that as Novelis […] .443 However, the Commission observes that a single year (that is 2013) cannot be taken as representative of capacity expansion behaviour. If a longer-term view is taken, as explaineded in recitals (561)–(567), and more in particular in recital (568), Novelis appears to have expanded its capacity less than its rivals.
(570) The Commission also observes that Novelis’ expansion decision in 2013 can be seen, in addition to being a means for accommodating increased […] demand, also in the context of the preceeding expansion announcements by competitors. That a market leader like Novelis reacts to capacity expansions by competitors, which at the time are significantly trailing in terms of capacity shares, by adding significantly more capacity than them, is fully in line with economic theory, according to which it is in the interest of a market leader not to proactively expand overall market capacity, but rather to react to competitors' actions. That Novelis in its ordinary course of business […].
(8.3.7.2.3) Current CASH line spare capacity is limited and will further decrease in coming years
(571) […].444
(572) For the reasons set out below, the Commission however finds that, in line with its results from the market reconstruction (presented in Section 8.3.7.1), current spare capacities are low and will likely decline further.
(573) […].
(574) […].
Figure 51 […]
[…]
(575) Second, when considering the Notifying Party’s submission on the projected balance of supply and demand for Aluminium ABS in the EEA, captioned in Figure 52, it also is apparent that, as demand is continuously rising, spare capacity does not […] , but rather […]. This […] of spare capacity […] occurs in spite of overall capacity still […] over the next years. The […] in overall capacity projected by the Notifying Party is largely due to […].
Figure 52 […]
[…]
(576) The Notifying Party’s claim that spare capacity will increase in the coming years is […].445
(577) As a further illustration, with regard to the supposed increasing overcapacity, a major customer observes that, while capacity is forecasted to increase to catch up with increasing demand, ‘[s]uppliers are currently delaying their investments until demand has grown enough, and thus avoiding creating manufacturing overcapacity’.446
(578) Third, the view that there is allegedly ‘significant, current and forecast, aluminium ABS excess capacity’447 is evidently not shared by other suppliers, as all those surveyed in an industry ‘Sentiment Indicator’ in April 2019 indicated the capacity situation to be ‘satisfactory’.448
(579) Fourth, finishing methods not involving a CASH line cannot be considered as adding readily available capacity to the EEA market for the production and supply of Aluminium ABS.
(580) The Notifying Party holds that the supply-demand balance represented in Figure 52 […] .449 […],450 that Aleris is currently […]451 […].452
(581) However, a majority of automotive customers responding to the market investigation disagree with the statement ‘that a CASH manufacturing line can be substituted by other manufacturing processes (for example batch annealing)’ while preserving quality, consistency, and cost-effectiveness. One major OEM explains that ‘[b]atch annealing has been superseded by the use of a CASH line which delivers more consistent material properties throughout the coil’.453 […].454 […]. In addition, a competitor to the Parties states with respect to automotive customers’ preferences or requirements for continuous annealing or batch annealing that ‘[t]here is no preference as long as material is within customer specification. But majority of material is with passivation455 and therefore most suppliers have to use a Cash-line with integrated surface treatment’.456 Further, in replying to the question whether automotive customers require specific annealing processes, one OEM replies ‘Continuous (industry standard)’.457 While some others say they have no requirement with respect to the annealing process, one major OEM points out that ‘[m]aterial specification need to be met and demonstrated to be consistent. Production methods are not driven by the customer, the requirements of the end product need to meet the specification’.458 This customer is the same that has in the past rejected batch annealed products due to their inferior quality,459 suggesting that while not all OEMs may have a requirement for a specific annealing process, their end-product quality specifications in effect make batch annealed products unsuitable. While OEMs also reject products that were run through a CASH line if they do not meet their quality requirements, they do not do so for the reason of associating inferior quality with the annealing process as such, but rather due to specific quality control issues to be solved by the supplier in question.
(582) […].460 […].461 […].462 […].
(583) The document and in particular the tab […] .463 […].
(584) The argument made by the Notifying Party, that in case of tight capacity, customers could be supplied with batch annealed 5xxx in order to free up CASH line capacity for the production of 6xxx products, would thus imply that customers would be worse off as they would be receiving inferior quality products. The fact that a limited volume of batch annealed products is currently supplied by some suppliers, does not, contrary to what the Notifying Party suggests,464 mean that more customers would be willing to accept batch annealed volumes.
(585) Fifth, it is important to note that the Commission’s analysis of spare capacity based on CASH line capacity likely overestimates spare capacity readily available for automotive customers. This is in part due to CASH lines being potentially used for production of products other than Aluminium ABS. One competitor in this context notes that it ‘does not have a CASH line exclusively dedicated to automotive, and each CASH line produces a mix of products of ABS, aerospace and other products’.465 It can thus not be expected that all of spare CASH capacity is readily available to meet demand from Aluminium ABS customers.
(586) In the Reply to the SO, the Notifying Party states that […] .466 […],467 […].
Figure 53 […]
[…]
(587) The Commission further observes that a competitor to the Parties, which states that each of its ‘CASH line[s] produces a mix of products of ABS, aerospace and other products’,468 also explains that ‘not fulfilling its commitments, particularly with automotive and aerospace customers would lead to seriously harming its credibility and losing business in the future years’.469
(588) Sixth, given the projected increase in Aluminium ABS demand in the EEA market, the level of spare capacity is likely to decrease further. This holds in particular, as ‘in Europe, there has been no recent announcement of a new CALP line’.470 An internal Novelis document, captioned in Figure 54, even states that […] .471 This would imply that while demand is projected to grow further, significant increases in capacity (in the form of new CASH lines) would be unlikely to occur in the foreseeable future.
Figure 54 […]
[…]
(589) With respect to the document captioned in Figure 54, the Notifying Party states that […].472 The relevant question for effective competition in a market is […] whether customers have realistic alternatives to supply by the market leader (as further explained in Sections 8.3.6 and 8.3.7.3). Figure 54 also puts into question the Notifying Party’s claim […].473 As explained in Section 8.3.7.3, market participants expect capacity to tighten already before 2024, and yet Novelis […].
(590) […].
(591) Therefore, even when accounting for recent capacity expansions, spare capacities in the EEA market for the production and supply of Aluminium ABS are limited. This trend is set to continue and holds in particular when considering the Parties’ main competitors, limiting the demand which could be contested by them in case of a price increase by the Merged Entity.
8.3.7.3. Even when assessing immediately available spare capacity, this has been overestimated by the Notifying Party
(592) As detailed in Section 8.3.7.1, spare CASH capacity in the EEA market was limited for the year 2018 and will continue to be limited for the coming years.
(593) This means that customers of Aluminium ABS in the EEA market only have a limited ability to switch from their current suppliers to others. In particular, and in line with the pivotality analysis presented in Section 8.3.6, it means that competitors’ spare capacity would not suffice to accommodate the demand of the Merged Entity’s customers in case of a price increase.
(594) The Notifying Party submits that ‘there is no projected shortage of capacity for ABS’ and that the available capacity in the market does not constitute an obstacle to switching for customers, in particular for OEMs.474
(595) The Commission however finds that the immediately available spare capacity is limited and that it has been overestimated by the Notifying party. As a result, customers already today experience tight capacity situation and perceive limited immediately available capacity as an obstacle to switch suppliers, and are concerned of future capacity development.
(596) First, one of the main competitors to the Parties states that ‘[c]urrently the Company has only very little spare capacity on its CASH lines’.475 This suggests an inability to accommodate a significant amount of additional volumes on part of that competitor.
(597) Second, some customers allude to a tight capacity situation, which would imply a difficulty to switch significant volumes to alternative suppliers even for upcoming tenders. An OEM states that ‘[t]here is new capacity for aluminium ABS brought about by recent investments by Constellium and Hydro. Constellium and Hydro are building their own heat treatment line. However, these plants are already overbooked even prior to its [sic] full operations’ (emphasis added by the Commission).476 Another major OEM states that ‘[c]urrently available capacity in Europe as well as the amount of time needed to install new capacity result in general difficulties if a customer were to replace Aleris and Novelis as suppliers’,477 thus suggesting that switching of volumes away from the Parties is difficult due to the capacity situation in the market. A tier customer further states that ‘[i]n 2018, the rolling mills’ capacity to supply 6xxx alloys (outer skin) was tight. The Company also experienced a shortage of supply regarding some alloys for 6xxx inner parts of a car in terms of passivation and EDT. The Company is not aware of any overcapacity or overproduced products on the market’.478 While the situation has somewhat improved in 2019, ‘[t]he Company expects a gap between production capacity and demand around 2021 and 2022 because the demand is expected to increase faster than the capacity for aluminium ABS on the market’.
(598) Other customers, who did not reference a currently tight capacity situation, however also estimate capacity to become tight in coming years. An OEM states that ‘[a]vailibility will become a greater problem in the future, especially because aluminium body sheet demand will increase and capacity is not responding’.479 Another OEM mentions that ‘[t]he current capacities are not very large and there is only a modest overcapacity, which is expected to be filled in the years to come’.480 Another major OEM mentions that ‘[c]apacity is not very large in Europe’ and that in light of growing demand, ‘it is not clear when […] additional capacity will be available’.481 Further, another OEM states that it ‘does not currently experience capacity constraint in the EEA. However, the demand for aluminium ABS is forecasted to increase because the Company and most (if not all) of its OEM competitors are expected to increase aluminium ABS demand in the years to come, due to CO2 emission regulations. Therefore, capacity constraints are expected to occur in the next few years’.482
(599) […].
(600) […].483
Figure 55 […]
[…]
(601) […].484
(602) […].485 […].486 […].
Figure 56 […]
[…]
(603) […].487 […],488 […].
Figure 57 […]
[…]
(604) […].489 […],490 […].
(605) In another internal email […].491 This further suggests that available spare capacity in the market is limited, Novelis has a pivotal share of market capacity and the Notifying Party’s claim that […]492 only holds to a limited extend.
Figure 58 […]
[…]
(606) […].493 […].494 […]495 […] .
Figure 59 […]
[…]
(607) As evidenced in this Section 8.3.7.3, immediately available spare capacity appears to be limited. Therefore, as some automotive customers have already experienced tight capacity and others are expecting it in the coming years, their ability to switch to alternative suppliers is constrained, and thus their ability to avoid price increases following the Transaction is limited.
8.3.7.4. Conclusion
(608) For the reasons set out in this Section 8.3.7, and considering all evidence available to it, the Commission concludes that spare capacities in the EEA market for the production and supply of Aluminium ABS are limited, are likely to decrease further, and are thus limiting customers’ choices. These limitations on spare capacity, especially competitors’ spare capacity, result in further pricing power for the market leader Novelis, and in the fact that it is unlikely that competitors would have the ability to offset a price increase by the Merged Entity by readily increasing supply through spare capacity. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.8. The Transaction is likely to result in higher prices for Aluminium ABS
8.3.8.1. By controlling a large share of the output and of production capacity, the Merged Entity would enjoy significant pricing power
(609) As detailed in Section 8.3.5, the Merged Entity would control large shares of sales and of production capacity. The Merged Entity’s capacity share would be [40-50]% (according to the Commission’s market reconstruction [50-60]%) and its market share in volume [60-70]% (according to the Commission’s market reconstruction [50-60]%) in the whole Aluminium ABS product market in 2018.496
(610) Such high market shares, particularly in terms of capacity in a market characterised by little spare capacity, are in themselves indicative of a dominant market position497 and of the ability to increase prices.498 The term ‘increase prices’ can here also be understood to refer to an ability to maintain prices at a certain level, despite market conditions which would favour a price decrease. Smaller suppliers may face the need to lower prices in certain situations to win market share or out of fear to be left without any volumes. The same rationale does not hold for the market leader who controls a large share of the output and capacity. In addition, it may not be reasonable for the market leader to increase its own capacity, given the negative effect on market price this would entail. […].499
(611) […], as in the EEA market for the production and supply of Aluminium ABS, the Merged Entity would be a clear pivotal supplier. As explained in Section 8.3.6, this puts the Merged Entity in a position where it faces a substantial portion of demand for which […] it is de facto the only supplier and therefore it has an incentive to keep prices high and avoid undercutting competitors.
(612) Further, the rationale of a market leader controlling a large share of the output and of production capacity, and the resulting pricing power is clearly shown in the Novelis internal email discussed in recital 0. […]. The ability to act in this independent fashion from customers and competitors would be further strengthened by the Transaction and the addition of Aleris’ capacity to Novelis.
(613) […].500 […]. In this respect, the Commission notes that the Merged Entity’s ability to increase prices is to be seen irrespective of whether demand at a point in the future is lower than forecast, because the relevant comparison is between prices post- Transaction with a given demand level and prices absent the Transaction with the same given demand level. Moreover, the relevant issue for the assessment of the effects of a merger is not merely a temporal one, but rather a counterfactual one, i.e. an assessment of whether prices post-Transaction are likely to be higher than absent- the-Transaction. Thus, many factors may have an impact on prices (input costs, aluminium index price, demand evolution and forecasts) which is consistent with the fact that prices fluctuate. Against this background, the Commission finds that post- Transaction, all other things being equal, prices would likely be higher than absent the Transaction (rather than and as opposed to before the Transaction).
(614) The Notifying Party also claims that the revised reduced demand forecasts from Ducker 2019 […]. In this respect, the Commission points to Section 2.5.1 of the Annex where it is shown that Novelis is clearly pivotal even under a scenario in which the lowest available demand forecasts from Ducker 2019 are coupled with capacity estimates that do not exclude any exports, any sales to non-ABS customers and add the batch annealing capacity of a rival.
(615) The Notifying Party claims that OEMs buyer power and competitors’ abilities to offset price increases would offset the Parties attempts to raise prices post- Transaction.501 However this appears unlikely. While pre-Transaction Novelis is already in a pivotal position,502 the addition of Aleris’ capacity would mean that competitors would be even less able to cover the entire market demand with their capacity. Further, the residual demand faced by the Merged Entity would be even larger than by Novelis pre-Transaction, thus increasing its pricing power.
8.3.8.2. The elimination of Aleris negatively affects competitive interaction
(616) The Notifying Party submits that Aleris does not represent a significant competitive constraint over Novelis and that the Parties do not have strong competitive interaction:
(a) the analysis of the Parties’ bidding data shows limited competitive constraint as further developed in Section 7.2 of the Reply to the SO;
(b) the Parties do not focus on similar products, as further developed in Section 5.2 of the Reply to the SO;
(c) the Parties’ competitive interaction is inherently limited due to the limited customer overlap between the Parties, as further developed in Section 5.3 of the Reply to the SO.
(617) The Notifying Party further submits in the reply to the SO that […]. However theses alleged limitations are, in any case, already taken into account in the following competitive interaction assessment.
(8.3.8.2.1) Bidding interaction between the Parties is evidence of significant competitive constraint exerted by Aleris
(618) For the reasons explained in this section, the Commission considers that the elimination of Aleris as a potential bidder for new contracts negatively affects competitive interaction on the Aluminium ABS market.
(619) First, the market investigation indicated that the number of credible bidders in a tender largely affects the degree of competition taking place in the tender itself. Therefore the elimination of Aleris as a potential bidder through the Transaction will negatively affect the competitive interactions taking place on the Aluminium ABS market. A large majority of the OEMs having responded to the market investigation considers that the number of suppliers responding to a tender greatly influences competition among suppliers and the resulting price.503
(620) In its Reply to the Article 6(1)(c) Decision even the Notifying Party, when referring to the bidding nature of the Aluminium ABS market, stated that ‘[c]ompetition in a bidding market depends primarily on the number of credible bidders’.504
(621) Second, contrary to the Notifying Party’s claim, the bidding data […].505
(622) As a preliminary remark, the market for the production and supply of Aluminium ABS is a basic industry characterised by capacity constraints as set out in detail in Section 2.7 of Annex I, and further developed in section 8.3.7.
(623) In such a market, unless there is evidence that either the Parties products are intrinsically very differentiated or that the Parties current and future plans involve targeting different customer groups (neither of which is the case in the present case as further developed in recitals (608) to (635)), the loss ratios are likely to simply reflect differences in capacities506 as opposed to intrinsic differences in the products or strategies of the various Aluminium ABS suppliers.
(624) With this caveat in mind, the Commission engaged with the evidence on loss ratios provided by the Notifying Party and considers that the evidence suggests that Aleris is indeed a competitive constraint on Novelis:
(a) […].507 […];
(b) […];
(c) […].508
(625) […]509, […].
(a) The Commission notes that Novelis is by far the largest of the two Parties and therefore the relevant question is whether Aleris poses a competitive constraint on Novelis rather than whether Novelis poses a competitive constraint on Aleris. In relation to this question the Commission considers that the […] the conclusion that Aleris is a significant constraint on Novelis.
(b) The Commission considers that the loss ratio from Aleris to Novelis is affected by the fact that Novelis collaborates very closely with […], a customer that accounts for a significant part of the volumes won by Novelis. […].510 […].
(c) […].511
(626) The CRA reports submitted by the Notifying Party on 30 November 2018 and 19 May 2019 also present an econometric analysis of the impact of Aleris’ presence on Novelis’ prices and margins. […].512 […]513 […].514
(627) Third, absent the Transaction, Aleris would have exerted a competitive constraint on Novelis in upcoming tenders. […].515 […].516 […]517 […],518 […].
(628) […]519 […]520 […].
(629) […]521 […]522 […].
(630) […].523 […].524
(631) […].525 […].526 […].
(632) […].527 […].
Figure 60 […]
[…]
(633) In these above listed cases of impeding direct competition between the Parties at upcoming tenders (upcoming tenders as submitted by the Notifying Party), the Transaction would reduce the competitive interaction in these tenders as it would lead to the removal of Aleris as a credible actual or potential bidder. Further, these cases can be regarded as illustrative of the competitive interaction generally being lost for future upcoming tenders.
(634) The ability to act as a credible bidder is also influenced by the uncommitted available capacity528 of a supplier for the duration of the production of the vehicle in question. The Notifying Party submits that Aleris’ ability to act as a credible bidder and therefore to pose a significant competitive constraint on Novelis is limited due to Aleris’ limited uncommitted capacity and states that […] .529 The Notifying Party however also submits that Aleris will have significant available uncommitted capacity in the coming years. The Notifying Party expexts that, for 2019, Aleris will have […] tonnes, for 2020 […] tonnes and for 2021 […] tonnes […].530 This significant available uncommitted capacity would enable Aleris to compete for upcoming open tenders. The Notifying Party rejects this argument and holds that Aleris will in future have uncommitted capacity due to its […] .531 First both arguments appear to contradict each other and second the argument that Aleris has a less competitive portfolio does not seem in line with […]. It is in any case more reasonable to assume that the uncommitted capacity in coming years is simply due to current supply contracts and agreements coming to an end (a trend which is reflected in the Notifying Party’s submission not only for Aleris, but also for the other Aluminium ABS suppliers532). Furthermore it is reasonable to expect that Aleris will be successful at contracting out its uncommitted capacity, both in light of its past success in doing so and given the overall limited spare capacity in the market (as detailed in Section 8.3.7). Therefore, the Commission maintains that post- Transaction, there would be one less credible bidder with available uncommitted capacity – the competitive interaction at upcoming tenders can thus be expected to decrease.
(635) […].
(636) […].
Figure 61 […]
[…]
(637) […].533 […].534 […].
(638) […] itself perceived Novelis to be ‘not very competitive’ because Novelis ‘was offering relatively high prices’. This is while […] was of the impression that ‘Novelis has however been willing to get higher volumes’535, which would have resulted in a challenge to […]’s main supplier Aleris.
Figure 62 […]
[…]
(639) Overall, based on the evidence available from the bidding data the Commission considers that the Transaction would eliminate the significant competitive constraint posed by Novelis in the high-value market segments and in particular for a number of important OEMs. This element supports the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
(8.3.8.2.2) The Parties focus on the same product segment
(640) The Notifying Party further submits that Aluminium ABS is a largely commoditised product and that closeness of competition is thus of limited relevance for the competitive assessment.536
(641) In response, the Commission recalls that, as explained in Section 6.2.3., Aluminium ABS is a market differentiated along, for example, alloy series. With this background, even considering the Notifying Party’s view that closeness of competition is of limited relevance, the findings of the in-depth market investigation support the view that the Parties engage in competitive interaction. The Commission observes that the Parties are, prior to the Transaction, both focusing on similar segments within Aluminium ABS and serve common customers as further explained in recitals (662) to (666) . As described by the Notifying Party, 6xxx alloy products are key for car manufacturers as they are ‘versatile, heat treatable, highly formable, weldable and have moderately high strength coupled with excellent corrosion resistance. Due to heat-treatment by way of continuous annealing, the 6xxx series is stronger than the 5xxx series. 6xxx alloys are thus frequently used for hoods, structures, tail gates, body side panels, fenders, and roofs’.537 For the following reasons, the Commission considers that the Parties have strong competitive interaction in the higher value 6xxx skin segment.
(642) First, the Parties were first movers in the supply of Aluminium ABS, and were the first fully capable of producing the most demanding 6xxx skin products. In contrast Hydro, Constellium and AMAG entered the market later and, in 2015, they were still not able to supply the most demanding products such as cladded high forming 6xxx skin.538 This historic advantage allowed the parties to develop their expertise in this market segment.
(643) Second, the Parties’ focus on and success in 6xxx alloys, and more specifically 6xxx skin, is evidenced by their much stronger position on these high-value products than on lower-value 5xxx alloys (or the overall Aluminium ABS market). The relevant segment shares are presented in Table 12, Table 13 and Table 14 based on the Parties’ data539.
confirming their competitive interaction. Comparing the capacity share and the segment shares highlights the fact that both Novelis and Aleris are serving a much higher share of the 6xxx skin segment than what their capacity share would suggest as evidenced in Figure 63. In particular, dividing the segment share by the capacity share highlights the relative success to serve a specific segment at a given capacity level through a ratio. The ratios show that the Parties are both more present in the high value segment of the market than other competitors at a given level of capacity share. For example in 2018 Aleris was serving [20-30]% of the 6xxx skin segment with only [10-20]% of the overall market capacity (Figure 63). […], compared to […] for Constellium […] for Hydro and […] for AMAG. In the Reply to the Letter of Facts, the Notifying Party claims that ‘volume market shares that exceed capacity shares are simply the flipside of not having much spare capacity.’ 540 However, while the lack of spare capacity is a transient phase, the ratios seem to show low variation and do not seem to be impacted by any freeing of capacity at Aleris or the potential filling of competitors’ spare capacity in the foreseeable future.
Figure 63 […]
[…]
(648) In this context, the Commission recalls that, as explained in Section 6.2.3, 6xxx series alloys are more difficult to manufacture than 5xxx series alloys. Moreover, 6xxx skin series alloys, where the Parties’ competitive interaction seems strongest, can be even more demanding to manufacture than 6xxx structure series as explained in recital (267). Skin grades are used for the external parts of a car that are visible and so need to meet more stringent surface quality requirements. The Parties thus seem to mostly compete in these most demanding segments of the market.
(649) Third, competitors’ segment shares do not in general exhibit similar presence or focus in the 6xxx skin series as evidenced in Table 12, Table 13 and Table 14.
(650) Moreover, if nominated sales are considered as relevant, while the Parties’ combined nominated segment shares in the 6xxx grades are projected to remain relatively constant (around [50-60]% for 6xxx structure and around [60-70]% for 6xxx-skin) from 2019 to 2023, during the same period, in the 5xxx grades Novelis’ share is expected to decline (from [40-50]% to [20-30]%) and Aleris’ share is set to stay very low (at around [0-5]%). At the same time, other suppliers show somewhat different development. For example Constellium’s segment shares are expected to remain steady (around [20-30]% in 5xxx series and [20-30]% in 6xxx series) whereas Hydro’s segment shares are expected to increase significantly in 5xxx series (and remain steady in 6xxx series) during the same period.
(651) Therefore, the Parties’ segment shares support a finding that they have a similar focus on higher value 6xxx grades and that they are not concentrating on the 5xxx series. That is different from other market participants, which, as appears from Table 12, Table 13 and Table 14, which are based on information submitted by the Notifying Party, are expected to have a different market share development. The Parties’ focus on the higher value segment is confirmed by the results of the market reconstruction in Table 3 and Table 4 as the Parties’ combined market shares in value appear to be even higher than those in volume as further detailed in recital (476).
(652) […].
(653) […]541 […].542 […].
Figure 64 […]
[…]
(654) […]543 […]544 […]
Figure 65 […]
[…]
(655) […]545 […].
(656) […].
Figure 66 […]
[…]
(657) […].
(658) […].546 […].
(659) […].
Figure 67 […]
[…]
(660) […].
Figure 68 […]
[…]
(661) Aleris’ focus on other alloys rather than 5xxx alloys is also observed by a customer: ‘Aleris, as well as some of its competitors, appear to be usually reluctant to supply 5xxx series alloys’.547
(8.3.8.2.3) The Parties serve common customers in 6xxx skin segment
(662) Contrary to what the Notifying Party claims,548 the Parties have several customers in common which they supply with similar 6xxx grades of Aluminium ABS, and for which the Parties have strong competitive interactions. The high combined supply shares are particularly evident for the 6xxx-skin segment.
(663) Based on the Notifying Party’s own estimates for the 6xxx skin segment, […]:
(a) […];
(b) […];
(c) […].
(664) […]:
(a) […]
(b) […].
(c) […].
(665) […].
Figure 69 […]
[…]
(666) When considering the segment share, more granular data, at OEM level, shows that the Parties have strong competitive interaction at common customers by both focusing on 6xxx skin segment.
(a) While specifically assessing Audi in paragraph (338) of the Reply to the SO the Notifying Party claims that, […].
Figure 70 […]
[…]
(b) […].
(c) […].
(d) Novelis and Aleris are not only serving common customers, but rather focus on 6xxx skin segment within these customers. This emphases the competitive interaction that occurs between the Parties.
(8.3.8.2.4) Conclusion.
(667) For the reasons set out in this Section 8.3.8.2, and considering all evidence available to it, the Commission considers that the Transaction would, by eliminating a competitor to Novelis, negatively affect competitive interaction, at current customers and upcoming tenders, and in particular in the high-value 6xxx alloys and even more specifically in the 6xxx skin market segment. This element supports the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.8.3. Customers expect prices to be higher as a result of the Transaction
(668) Consistent with the assessment of the effects of the Transaction on prices, market participants expect that the Transaction will result in higher prices on Aluminium ABS.
(669) All but one of the Aluminium ABS customers expect that the Transaction will have an impact on their business.549 When asked to further elaborate on the expected impact some customers spontaneously mention price ‘[i]t could have a negative impact on prices’ or ‘[c]ommercial competition in the market will decrease’. When specifically asked about the price effect, a large majority expressed that the Transaction will have a negative effect on prices.550
(670) Besides indicating these concerns with respect to the overall impact of the Transaction or its effect on prices, automotive customers also provided detailed explanations as to why they are concerned, highlighting potential price increases, capacity constraints and an overall decline in competition.
(671) Automotive customers of Aluminium ABS seem to be concerned that the Transaction might have a negative effect on competition. One OEM stated that ‘the combination of the capacities may have a global negative effect on competition as it consolidates capacity medium to long term in a market place with limited competition currently.’551 Another OEM states that ‘[t]here will be less competition after the merger in the EEA region.’552
(672) Another OEM expressed the concern that, considering that Novelis’ prices are already the higherst, the Transaction may lead to upward alignment of price from the Merged Entity. The customer states that ‘[t]he transaction would lead to Novelis gaining an additional 200,000 tonnes from Aleris in the USA and another 100,000 tonnes in the EU. This development may lead to higher market prices and may lower the level of competition in the market. Novelis price being the highest, [the customer] is concerned that Aleris prices may align after the merger.’553
(673) Another automotive customer sees the decrease of alternative suppliers as a driver for price increase. This customer clearly links the number of alternative suppliers and the price level, stating: ‘The Company estimates that the proposed Transaction will affect price and availability of materials. In some alloy specifications, they might even be the sole supplier’.554 The same customer relies on precedents to confirm the direct link, already observed, between limited available suppliers and higher prices, stating that ‘[e]ven today, the level of competition is noticeably lower in supply situations with only 2 to 3 suppliers. The Company experiences that price levels are lower in situations where the required specification is met by all 5 suppliers compared to situations where a lower number of rolling mills are qualified. According to the Company, higher production cost for a certain specific alloy is not the sole reason for the price difference in these situations.’
(674) This concern is echoed by another automotive customer: ‘Price increase may occur due to the transaction trimming down the number of suppliers. [The customer] is not particularly glad about the transaction as it will limit their options from 5 to 4.’555
(675) With respect to both 6xxx structure and 6xxx skin alloys, an OEM expects the Transaction to have an effect on prices, explaining that ‘[t]hey will go up du (sic!) to less competition’.556 The OEM further explains that the Transaction may also have effects on manufacturing capacity for Aluminium ABS in the EEA, as while it is ‘[d]ifficult to say’, it ‘assume[s] that one company will expand less than two’.557
(676) Another OEM states that ‘the number of suppliers will be less in Europe. Less supply always lead (sic!) to higher price’.558
(677) Another OEM is linking the Transaction to potential capacity constraints leading to a price increase: ‘The Company is concerned that the increased concentration resulting from the merger would reduce the number of its qualified suppliers […]. This increased concentration might potentially result in capacity constraints in the EEA, which would lead to a price increase.’ 559 This anticipated price increase is seen by this OEM as likely to have an impact on the vehicles’ input cost: ‘In the view of a substantial increased use of aluminium ABS in its vehicle (see above), the Company considers that the effect of a price increase would have an important impact to the vehicles’ input costs.’
(678) Another OEM, in order to illustrate ‘the possible effect of manufacturing capacity constraints, […] took the example of the aluminium ABS market in the United States, where capacity constraints have been existing for several years. As a result, not only the price in the United States is remarkably higher than in other parts of the world, but also the buyer power of OEMs is substantially lower. OEMs often need to accept to pay capacity reservation fees to aluminium ABS manufacturers, and are not able to select their suppliers according to a tender process, due to the lack of manufacturing capacity’.560
8.3.8.4. Steel is not a sufficient constraint to offset price increases in Aluminium ABS
(679) The Notifying Party argues that, in case Aluminium ABS were considered to be part of a separate product market from flat steel products used in automotive bodies, steel would in any case constitute a powerful out-of-market competitive constraint.561
(680) For the reasons set out in Section 6.2.2.3, the Commission has concluded that Aluminium ABS and flat steel products used in automotive bodies are part of different product markets. In this Section, the Commission concludes that, post- Transaction, the competitive pressure from steel suppliers would not be enough to prevent the Merged Entity from increasing Aluminium ABS prices in the EEA post- Transaction, for the following reasons.
(681) First, as previously explained in Section 5.5, price is only one of the many dimensions that OEMs take into account when choosing either aluminium or steel for their new vehicle models. This is because, as acknowledged by the Notifying Party,562 compared to steel, aluminium has certain technical advantages, such as its strength-to-weight ratio, which helps reduce the weight of a vehicle and thereby fuel consumption and CO2 emissions. As described in Sections 5.5 and 6.2.2, instead of price considerations, the increased use of aluminium in vehicles in the past years has been primarily driven by more demanding CO2 emission standards worldwide and in particular in the EEA. For this reason alone, a price increase in Aluminium ABS would not be easily offset by steel, which is heavier (and thus leads to higher fuel consumption in vehicles and higher CO2 emissions).
(682) The increase in Aluminium ABS use despite being more costly than steel shows that customers are willing to pay a significant premium because of the benefits brought about by aluminium’s material properties. Accordingly, past switching from steel to aluminium in vehicles’ components (as well as the significant switching expected in the future) has been mostly driven by non-price factors such as CO2 emission regulations, and takes place despite Aluminium ABS’ higher price, as further explained in section 5.5.
(683) Second, the market investigation has shown that OEMs with a preference for aluminium (for its particular properties) would not easily switch to steel in response to changes in relative prices. In line with this, a customer openly said that it ‘does not intend to switch back to steel after a price increase for aluminium ABS’.563 The Commission thus observes that switching between steel and aluminium is not driven by changes in relative prices, but by more structural properties of demand (including the desire to meet CO2 emission regulations through light weighting) as further developed in Section 5.5 and in Annex I Section 2.1.2.
(684) Moreover this difficulty for OEMs to switch to steel in response to changes in relative prices is confirmed by numerous market participants, customers and suppliers. In order to illustrate this, such market participants point to the North American market as a comparator market and as an example of a market experiencing higher price than Europe due to capacity constraints as detailed in recitals (686)–(688).
(685) This would be a particularly pertinent illustration of how the direct link between limited capacity and price of Aluminium ABS would confirm that price increases in Aluminium ABS are not being offset by the out of market steel pressure as confirmed by the market investigation.
(686) A competitor states, ‘In the US, however, the capacities of local manufacturers are tight. Price level is higher in the US because of the higher metal premium and conversion margins are higher. The US is a market separate from the EEA and hence the prices are not equalised between the two continents.’564
(687) A customer states ‘Currently, in the US, OEMs have to fight for each ton of capacity and new lines not only take 18 months to build but also have a significant ramp up period. During this ramp up period there are usually quality problems and missing ramp up volumes have to be found from other suppliers.’565
(688) Another customer states ‘Price is significantly higher in the US due to the shortage of capacity at present due primarily to Ford’s decision to switch to aluminium for its F-150. This decision increased enormously the demand for aluminium ABS in the US and now there are three rolling mills in the US dedicated to that. Further US-OEM are following this lightweight strategy.’566
(689) […]567 […].
Figure 71 […]
[…]
Figure 72 […]
[…]
Figure 73 […]
[…]
(690) In view of the foregoing, considering the high combined market shares and the increase in market power of the Merged Entity post-Transactionvon the Aluminium ABS market, the out of market constraint from steel would have to be extremely high to offset the dominant position of the Merged Entity.
(691) Furthermore, Tier suppliers buying Aluminium ABS have production lines focused on aluminium rather than steel. A Tier supplier explained that in case of a price increase, it would not be able to switch to steel because its production lines cannot process steel, but only aluminium.568 This Tier supplier also explained that it ‘invested EUR […] million to enter the aluminium business because of increasing demand for aluminium, especially for ABS.’ Moreover, this supplier ‘does not consider OEMs to have the opportunity or choice to switch back to steel.’
(692) Third, contrary to the Notifying Party’s claim that OEMs can achieve their regulatory targets exclusively with steel, as a customer explained during the market investigation: ‘no economically viable material alternative to aluminium is available to OEMs for reducing its vehicles’ weight. Composite materials, for example, a too expensive option, while high strength steel can replace aluminium only to a limited extent’.569
(693) Fourth, competitors do not actively monitor the price of steel for the purpose of their pricing decisions towards Aluminium ABS.570
(694) […]571 […]572 […]573, […].
(695) Similarly, competitors do not actively monitor capacity of the steel market. […].
(696) […].
Figure 74 […]
[…]
(697) […].
Figure 75 […]
[…]
(698) The Commission acknowledges that there are components in a vehicle for which the value proposition of aluminium may be stronger (such as bonnets and wings)574 and that suppliers are typically chosen by OEMs on a component-by-component basis.575 For certain components the value proposition of aluminium is stronger because the lightweighting factor is larger, that is, for those components it is possible to reach particularly high efficiency in weight reduction compared to steel. However, as emissions regulation progresses, OEMs are further induced to switch (at least part) of other components too from aluminium to steel. Indeed the evidence resulting from the Commission’s investigation suggests that the aggregate Aluminium ABS demand is inelastic with respect to relative price changes between steel and aluminium, as further described in Annex 1, Section 2.1.2. The Commission’s findings are based on the Transaction’s likely effect on the overall market and not only on its effects on the components for which the value proposition of aluminium may be stronger or on vehicles in which aluminium has a high penetration.
(699) Indeed, as explained in this Section, the choice of aluminium bonnets or wings and the choice of aluminium body structure applications is primarily driven by the OEM’s desire to produce lighter vehicles to meet CO2 emission targets. Therefore, to the extent that the Parties can price differentiate between different components of a vehicle, this would only potentially magnify the broad competition concerns in this Section.
8.3.8.5. The impact of the Transaction would be immediate
(700) Aside of the dynamic effect of a medium to long term nature, which is considered in Section 8.3.9 in particular with a view on capacity expansion, the Commission considers that the Transaction’s effect on the competitive interaction in the market for the production and supply of Aluminium ABS and on customers in particular will be immediate.
(701) First, the Transaction will have an immediate impact by leading to the removal of the competitive constraint currently exercised by the Parties on each other and affect the competitive interaction at future tenders.
(702) In the first instance, it is important to stress that while the Notifying Party was not able to provide detailed information on competitors’ participation in tenders, in sectors characterised by very material capacity constraints as is the case for Aluminium ABS, not all competitors place a competitive bid in all tenders. Knowing that in any event they cannot win all tenders (due to capacity constraints), competitors tend to be selective in their participation and, even when frequently participating, tend to not always place very competitive bids. Therefore, in such a market the reduction of one important competitor in a group of only 4 or 5, which moreover results in the creation of a player controlling a significant share of sales and capacities in the market, may result in a very significant loss of competition on a tender by tender basis, which would likely lead to higher prices for the OEMs.
(703) This will have an immediate impact on new tenders open for bids as of the Transaction’s closing. In any such new tender, there will be one less potential competitive bidder. The participation of bidders making competitive bids is, however, according to customers directly related to achieving competitive prices; with less bidders participating in tenders, ‘the level of competition is noticeably lower’576 and market prices can be expected to increase as a result.
(704) This affects in particular automotive customers on which both Novelis and Aleris have a focus. The automotive customers listed in Section 8.3.8.2, in particular those listed […] in respect of which the Parties are engaged in significant competitive interaction for the 6xxx-skin segment, can be expected to experience a reduction in competitive interaction at their tenders. […].
Figure 76 […]
[…]
(705) […].
(706) […].577 […].
(707) Second, the impact of the Transaction will result in an immediate reduction of competitive interaction in cases where ongoing […]. In such cases, the Transaction will immediately reduce the number of competitors who could potentially challenge the competitiveness of an existing contract.
(708) Third, it appears that for certain customers of Automotive ABS, the impact of the Transaction may be even more direct than for others. Tier customers tend not to have contract durations that span five to seven years as in the case of OEMs. […].578
8.3.8.6. Conclusion
(709) For the reasons set out in this Section 8.3.8, and considering all evidence available to it, the Commission concludes that the Transaction is likely to result in higher prices for Aluminium ABS in the EEA. The Merged Entity would have significant pricing power due to its large share of output and production, while Aleris would no longer be a constraint on Novelis, in particular in bidding interactions. Further, the competitive pressure from steel suppliers would not be enough to prevent the Merged Entity from increasing Aluminium ABS prices, and market participants expect the Transaction to result in higher prices. The impact of the Transaction would be immediate. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.9. The Transaction would significantly reduce the Merged Entity’s incentives to increase capacity in the market
(710) Aside of the effects of the Transaction, in particular on price, outlined in Section 8.3.8, the Transaction would also have effects in the medium term and on a more lasting basis. In particular, the Commission considers that the Transaction would significantly reduce the Merged Entity’s incentives to increase capacity in the market.
8.3.9.1. Theoretical framework – the market leader has reduced incentives to increase market capacity
(711) In addition to the direct detrimental effect on price competition between suppliers of Aluminium ABS, based on existing (and planned) levels of capacity,579 the proposed Transaction is also likely to create more long-term competitive harm by stifling capacity competition. Indeed, the economic theory of capacity competition predicts580 that as the number of producers decreases, the total capacity that firms are willing to supply to the market will decrease, which causes upward pressure on prices. This softening effect on capacity competition is known to be particularly strong if, as in this case, the market shares of the merging parties are high. This is because the capacity choices of firms with a larger market have a comparatively larger impact on market prices.581 Moreover, when capacity extensions put pressure on market prices, larger producers will have to absorb a comparatively larger part of this pressure on profits. When deciding on when and to what extent to increase their capacity, firms with large market share will therefore be particularly careful in trading off the benefit in terms of organic growth potential with the downside relating to the effect that extra capacity has on the market price.
(712) The Notifying Party suggests that powerful OEMs are able to induce Novelis and Aleris (or one of their competitors) to build extra capacity if this is needed. In this respect, the Commisison notes that Aluminium ABS is normally built only when some contractual pre-commitment is given by an OEM to fill at least part of the new capacity for a given time period. In view of the heavy capital investments required for capacity extensions, securing such initial customers is important to mitigate the commercial risk associated with a plant extension. However, the existence of such pre-commitments does not undermine the ability of suppliers to decide for themselves how much capacity they will build, at what time, and which terms to require from customers to agree to a capacity extension at all. On the contrary, suppliers will unilaterally consider a variety of factors that have a strategic impact on their long-term profitability when deciding whether to build additional capacity. For instance, profit maximising suppliers will take into account factors such as (i) expected aggregate capacity in the industry going forward, (ii) projections of demand growth or decline for the industry, (iii) possible strategic responses of competitors triggered by own capacity decisions, and (iv) the future willingness to pay and substitution opportunities of potential customers.
(713) Such long-term strategic considerations are far more pertinent for capacity choices than (say) for short-term price decisions. The Commission notes, in particular, that newly constructed factory buildings and production lines have a substantially longer lifespan than the duration of typical contractual pre-commitments of launch customers. When deciding whether to expand capacity today, suppliers are therefore faced with a material risk of future imbalances between demand and supply, for example due to the fact that they have no certainty that the customer will buy again from them at the end of the production cycle and/or due to the highly cyclical nature of automotive demand (which implies that demand can severely and unexpectedly decline in a recession or accelerate in a boom).
(714) For the above reasons, suppliers will not simply let OEMs dictate to them, based on OEMs’ current short-run interests, how much capacity they will build for the long run. Instead, suppliers negotiate hard with OEMs about undertaking possible expansions and the commercial terms of potential launch contracts. While doing so, they take careful account of their own long-term strategic interests, including how capacity extensions will affect likely future prices in the market.
(715) The Notifying Party also notes that it would be in the interest of suppliers of Aluminium ABS to ensure that enough capacity is available in the market, as alternatively OEMs would be able to switch back to steel (or switch from steel to Aluminium ABS to a lower extent).
(716) In this respect, the Commission notes that as discussed in Section 6.2.2.3, OEMs with a preference for Aluminium ABS (especially those with preferences driven by exogenous factors such as CO2 emissions regulations) cannot easily switch between Aluminium ABS and flat steel products for automotive bodies. If the price of Aluminium ABS increased by a very large amount, there is no doubt that at some price level OEMs would start considering flat steel products for automotive bodies as an option. However, OEMs are unlikely to readily switch to flat steel products for automotive bodies in response to small but significant changes in relative prices.582
(717) Moreover, the Commission notes that in a context in which demand for Aluminium ABS is constantly growing, the idea that adding the capacity (and current sales) of Aleris to Novelis would create a supplier with lower incentives to engage in capacity expansions should not be interpreted as a concern that the Merged Entity would stop investing and no longer add capacity in the market. The market is growing significantly and the Commission considers that all Aluminium ABS suppliers will continue investing after the transaction in order to keep up with this growing demand. However, the Merged Entity’s incentives to expand after the Transaction would be significantly lower than the incentives that Novelis and Aleris would have absent the Transaction, in particular in relation to those investments aimed at growing organically at the expense of rival suppliers. In this respect, the Commission presents in Section 8.3.9.4 evidence that Novelis considers Aleris’ growth to overlap with its own organic growth plans.
(718) As detailed in Section 8.3.5, Novelis is the current EEA market leader in terms of Aluminium ABS manufacturing capacity, with a share of [30-40]% (according to the Commission’s market reconstruction [40-50]%) in 2018. The Merged Entity would be an even clearer market leader with a share of [40-50]% (according to the Commission’s market reconstruction [50-60]%). In 2023, the capacity share of the Merged Entity in the manufacturing of Aluminium ABS is still forecast to be high with [40-50]% (according to the Commission’s market reconstruction [40-50]%).583 This suggests that pre-Transaction Novelis and post-Transaction the Merged Entity would have a limited incentive to expand capacity. As the addition of Aleris’ capacity would post-Transaction strengthen Novelis’ capacity leadership, the Transaction therefore can be considered to lower Novelis’ incentives to expand capacity organically.
8.3.9.2. […]
(719) […].
(720) […].584
(721) […],585 […].586
(722) […].
Figure 77 […]
[…]
(723) […],587 […].
Figure 90 […]
[…]
(765) […].
Figure 91 […]
[…]
(766) […].
Figure 92 […]
[…]
Figure 93 […]
[…]
Figure 94 […]
[…]
(767) […].627
(8.3.9.3.3) […]
(768) […].
(769) […].
(770) […].
(771) […].
(772) […].
(773) […].
(774) […].
8.3.9.4. Novelis’ acquisition of Aleris reduces its incentives to increase capacity, and enables it to maintain its market leadership without facing detrimental decreasing pressure on prices
(775) As discussed in Section 8.3.9.1, Novelis’ acquisition of Aleris is likely to restrict capacity competition. It is also likely to further reduce Novelis’ incentives to increase capacity.
(776) The Commission considers that the following considerations by Novelis ought to be understood in this light. […],628 […].
Figure 95 […]
[…]
(777) […].629
(778) […].630
(8.3.9.4.1) Novelis’ incentive to grow inorganically
(779) […]631 […].632 […].633 […].
Figure 96 […]
[…]
(780) […].634
(781) Second, a further advantage of the Transaction for Novelis is the strengthening of its market power. Inorganic growth allows Novelis to strengthen its market leader position, to prevent a competitor from acquiring Aleris’ assets and to remove a competitor from the market, […],635 […].
Figure 97 […]
[…]
(782) […].
Figure 98 […]
[…]
(783) […].636 […].
(784) […].
(785) In the Reply to the SO, the Notifying Party states that in order to ‘capture as much’ of share from steel as possible, ‘suppliers need to build additional capacity’.637 Aluminium ABS suppliers therefore have an incentive to invest in additional capacity, which according to the Notifying Party is summarised in a quote from an internal doc […]. According to the Notifying Party, Aluminium ABS suppliers thus are said to have an incentive to avoid tight supply-demand situations in order not to lose Aluminium ABS share to flat steel products used in automotive bodies. The Notifying Party further underlines this claim by stating that […] .638 The Commission however holds that this is not an accurate representation of OEMs’ options and Aluminium ABS suppliers’ incentives, in particular not of the market leader Novelis (which prefers inorganic growth).
(786) First, OEMs’ ability to simply design vehicles ‘with HSS’ in the absence of sufficient Aluminium ABS capacity is limited. As discussed in recital (147), the Parties are aware that […].639 Further, if OEMs were to revert back to designing entire vehicle fleets with steel only instead of Aluminium ABS, they would need to achieve Aluminium ABS’ weight savings through other means. A clear majority of OEMs responsive to the market investigation however states that the benefits of Aluminium ABS in terms of fuel consumption and CO2 emission, could not be ‘replaced in a timely manner and without major additional costs by other alternatives’ such as improved engine efficiency, electrification, improved aerodynamics or rolling resistance.640
(787) […],641 […].
Figure 99 […]
[…]
(788) […].
Figure 100 […]
[…]
(789) […].
Figure 101 […]
[…]
(790) […].
Figure 102 […]
[…]
(791) While the Notifying Party holds that capacity expansion efforts since 2018 by Novelis, Constellium and Aleris in North America have the aim ‘to ensure that OEMs have sufficient capacity to design with aluminium ABS rather than HSS’,642 the fact that this is only occurring in 2018, a year in which capacity is already very tight, cannot be seen as evidence of suppliers avoiding supply crunches due to pressure from steel.643 Further, despite these anticipated capacity additions, the supply-demand balance in North America is projected to remain tight in the coming years.
(792) Therefore, the above captioned documents evidence the fact that, […] this does not entail a market-wide substitution of Aluminium ABS with flat steel products used in automotive bodies, […]. The negative effects of the Transaction on the Parties’ (and competitors’) incentives to expand capacity are thus unlikely to be counteracted by a competitive pressure from steel.
(793) In the Reply to the SO,644 the Notifying Party further states that internal documents evidence […]. The Commission, however, observes, that the acquisition of Aleris would contribute to maintaining or increasing Novelis' share at a number of customers, namely those that are common customers of both Novelis and Aleris. Further, while Aleris in 2018 […]. The acquisition of Aleris therefore also helps Novelis in accomplishing its formulated ambition.
(8.3.9.4.2) […]
(794) […].
(795) […].
(796) […].
Figure 103 […]
[…]
(797) […].
Figure 104 […]
[…]
(798) […].
(799) […].
(800) […].
Figure 105 […]
[…]
(801) […].645 […].646
(802) […].
Figure 106 […]
[…]
(803) […].
Figure 107 […]
[…]
(804) […].
(8.3.9.4.3) The lasting reduction of incentives to increase capacity would not be offset by behavioural undertakings to complete the Duffel CALP II project
(805) The Notifying Party submitted to the Commission on 25 July 2019 a concept for a potential remedy informally named ‘invest or divest’. This concept in essence foresaw a commitment by Novelis to build the Duffel CALP II line, thus expanding the Merged Entity’s capacity and overall market capacity. Should the Notifying Party fail to meet this commitment, the Duffel Plant would be divested.
(806) The Commission informally provided feedback to the Notifying Party, arguing that such a concept would not remove the overlap between the Parties and would increase the capacity share of the Merged Entity, and therefore not address the competition concerns of the Commission.
(807) In relation to this concept for a potential remedy by the Notifying Party, the Commission further observes that it cannot be regarded as evidence showing that Novelis does not have reduced incentives post-Transaction to increase capacity.
(808) First, the proposal to build CALP II was made as an informal remedy proposal to the Commission. It thus does not represent an ordinary course of business decision by the Merged Entity, but rather an attempt to address the Commission’s competition concerns.
(809) Second, as explained in this Section 8.3.9, both Parties were pre-Transaction independently considering capacity expansion. In other words, pre-Transaction, two projects for capacity expansion existed. The informal remedy concept would have led to the implementation of one of these projects, the other one nevertheless being lost.
(810) Third, consistently with economic theory, on a lasting basis the incentives of Novelis, even if it decided to enact one of the two expansion plans, would be lower than if it had to compete with an independent rival on a lasting basis. Otherwise stated, the Notifying Party’s concept for a potential remedy is in any event not future-proof.
(811) Finally, and as explained in Section 8.3.9.1, the Commission does not argue that the Transaction would result in the Merged Entity having no incentive to expand capacity anymore, but that this incentive would be reduced as compared to how it would be in the absence of the Transaction. In a growing market, circumstances naturally may reach the point where Novelis would decide to make an investment into capacity expansion. An implementation of one of two capacity expansion projects thus does not disprove this view of the Merged Entity’s and market leader’s overall incentives in relation to capacity expansions.
8.3.9.5. Capacity in Aluminium ABS is strictly correlated with pricing ability
(812) Economic theory suggests that in any market, by the law of supply and demand, for a given amount of demand, the lower the supply, the higher the price. […].
(813) […].
(814) […].647
(815) […].
Figure 108 […]
[…]
(816) […].648
(817) […].649
(818) […].
Figure 109 […]
[…]
(819) […].
(820) […].
(821) […].
Figure 110 […]
[…]
(822) […].650
(823) […].
(824) […].
Figure 111 […]
[…]
(825) […].
Figure 112 […]
[…]
(826) […].
(827) […].651
(828) […].652
Figure 113 […]
[…]
(829) […].
(830) […],653 […].
Figure 114 […]
[…]
(831) […].654
(832) […].
(833) […],655 […].
(834) In general, it is proper to assess decisions by the Parties on capacity, and in particular decisions on capacity increases (which would also increase overall market capacity) also in terms of their effect on price.
(835) As evidenced in 8.3.9.2 to 8.3.9.4, the […]. This reasoning is also shared by OEMs. The majority of respondents taking a view in the market investigation state that a 50 000 to 100 000 tonnes increase in Aluminium ABS manufacturing capacity in the EEA would have a positive effect.656 One major OEM in this context states that ‘[a]s the price for ABS is market driven an increase in the supply will end up in a lower price’. Another OEM agrees on the effect of such a market capacity increase: ‘Price will go down and availability higher’.657
(836) However, the inorganic growth pursued by Novelis in the form of the Transaction achieves a capacity increase for Novelis without the growth of overall market capacity, and without a negative effect on price. Similarly, the Merged Entity may be incentivised to delay any capacity expansions ([…]) given their anticipated effect on price.
8.3.9.6. Conclusion
(837) For the reasons set out in this Section 8.3.9, and considering all evidence available, the Commission concludes that the Transaction would have dynamic effects in the medium term and on a more lasting basis. As the market leader, Novelis has limited incentive to increase overall market capacity. Post-Transaction, its incentive to increase capacity would be further reduced. Novelis will have less incentive than an independent Aleris would have to expand Aleris’ capacity as this would have a negative effect on Novelis’ own business (as as it would lower the market price). Novelis would also have less incentive to expand its own capacity (such as implementing the Monza expansion) as the Transaction offers a preferable way of growing the business inorganically. The acquisition of Aleris would thus allow Novelis to maintain its market leadership without facing detrimental downward pressure on price due to a larger overall market capacity. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.10. Competitors are unlikely to offset price increases resulting from the Transaction
(838) The present section addresses potential competitor reaction which could offset a price increase resulting from the Transaction. According to the Horizontal Merger Guidelines, ‘When market conditions are such that the competitors of the merging parties are unlikely to increase their supply substantially if prices increase, the merging firms may have an incentive to reduce output below the combined pre- merger levels, thereby raising market prices’658 and ‘[s]uch output expansion is, in particular, unlikely when competitors face binding capacity constraints and the expansion of capacity is costly […]’.659
(839) For the reasons explained below, the Commission finds that competitors lack the ability and the incentive to react to a price increase in such manner capable of offseting the negative effects of the Transaction.
(840) The lack of ability to increase output in reaction to a price increase resulting from the Transaction is contingent to their limited spare capacities, as detailed in Sections
8.3.6 and 8.3.7. Competitors would also lack incentives to increase their output (as a result of a capacity increase), because they would benefit from the resulting price increase.
8.3.10.1. Manufacturing capacity is tightly monitored to ensure that it evolves in line with the expected demand and it does not negatively affect prices
(841) It is explained in Sections 8.3.6 and 8.3.7 that the Merged Entity’s competitors lack the ability to react to a price increase because their spare capacity is limited. For the reasons explained in the present section, the Commission also considers that such situation is unlikely to be addressed by the prospect of capacity expansions by competitors, because manufacturers of Aluminium ABS tightly monitor the existing manufacturing capacity to ensure that it evolves in line with the expected demand, and that it does not negatively affect prices.
(842) […].
(843) […] competitors appear to closely monitor the evolution of manufacturing capacity for Aluminium ABS in the EEA. It is likely that they make use of the resulting price prediction for making their decisions on capacity expansion.
(844) One of the main competitors of the Parties, namely Constellium, for example, has explicitly communicated that investments resulting in new capacity are carefully evaluated after monitoring the market conditions. In a recent earning call, the CEO of Constellium, in the context of Aluminium ABS for the European market stated: ‘[…] we continue to closely monitor the market. We have been and will remain prudent with our investments. As I have noted many times in the past and as Peter just said earlier, we will not make incremental investments without firm customer commitments’.660 In its response to the market investigation, Constellium stated: ‘The Company monitors potential capacity expansion projects by competitors by analysing press releases. Once it is known when a competitor is supposed to start the operation of a line and the line’s capacity, the Company can make assumptions about the ramp-up of the new line. […] This monitoring and analysis helps the Company to assess if investing in additional capacity would be advantageous for the Company itself’.661
(845) Similar to […] Constellium, both AMAG and Hydro analyse the available capacity on the market, and, based on their forecasted demand, attempt to forecast the effect of the demand-supply balance to the expected price of Aluminium ABS. AMAG, for example, stated that ‘[i]n terms of short-term planning for the automotive industry, the uncertainty how the next years will look like is very low’ and clearly stated that ‘[t]he capacity availability will influence the price level’.662 Hydro also routinely estimates demand and available supply and, specifically for the expected capacity, it stated that ‘[…] data about new capacities for aluminium FPR for ABS and rolled- products […] provides some information about the supply/demand situation’, and based on that, it currently forecasts a tight demand-supply in 2023-2024, and considers that its competitors might be evaluating future capacity expansions.663
(846) […].664 […].
(847) Therefore, the Commission considers that both the Parties and their competitors tightly monitor the industry capacity to ensure that it evolves in line with the expected demand and it does not negatively affect prices. The Commission also considers that both the Parties and their competitors, in the context of growing demand and limited spare capacity, appear to have the ability to make capacity investment decisions driven by the expected resulting market price of Aluminium ABS, rather than by increased demand only.
8.3.10.2. Novelis has formal, structural and ad-hoc links with remaining competitors, which already pre-Transaction soften competition, may have detrimental effects on price and make a countervailing reaction from competitors to a price increase more unlikely
(848) As explained in recital (840), in the case of a price increase from the Merged Entity post-Transaction, competitors would lack incentives to increase their output (as a result of a capacity increase). This is because competitors would also benefit from the resulting price increase. The present section explains that, contrarily to the Notifying Party’s argument that links between these Aluminium ABS suppliers do not ‘diminish the intensive competition that characterizes the ABS industry’,665 the market for the production and supply of Aluminium ABS is characterised by a number of links between manufacturers which already soften competition pre- Transaction and which may, post-Transaction, have detrimental effects on price, making a reaction from competitors more unlikely. This would de-facto, further decreases the incentive of the Merged Entity’s competitors to react to a post- Transaction price increase.
(849) Such a lack of incentive by competitors to expand their capacity in reaction to the Transaction is corroborated by feed-back from the Parties’ customers. In particular, automotive customers expect that the Transaction would affect Aluminium ABS suppliers’ incentives to increase capacity. An OEM states that ‘post-merger, the reduced competition among the remaining suppliers might reduce the incentives to adequately expand manufacturing capacity in the EEA’.666 Another OEM mentions that it ‘considers that the probability of capacity expansions will would (sic!) probably decrease after the merger between Novelis and Aleris because there will be only 4 potential suppliers left’.667
(850) As detailed in Section 8.3.4.1, the number of Aluminium ABS manufacturers supplying their products to the EEA is already limited pre-Transaction..
(851) First, manufactures of aluminium products in Europe, including Aluminium ABS manufacturers supplying products to the EEA, are organised in the European Aluminium Association as well as in further associations on a national level. At the level of the European Aluminium Association, manufacturers of Aluminium ABS are organised in the Automotive and Transport Market Group. The market group undertakes research projects ‘to identify what OEMs could be sourcing in the future’ and ‘tries to get a good view of the different products demanded by OEMs’. This research is done in collaborating ‘with external providers such as Drucker or CRU’.668
(852) The European Aluminium Association also produces every three months an aggregated figure of shipments of flat rolled aluminium products for automotive applications in Europe, compiled by data confidentially submitted by its members.669 Further, in the past the European Aluminium Association has also carried out among its members a survey ‘on ABS capacity for the Automotive & Transport group’, sharing the overall Aluminium ABS capacity figure compiled on that basis with its members.670
(853) […].671
(854) These platforms provide manufacturers of Aluminium ABS with opportunities to arrive at a shared understanding of the development of their industry, exchange on issues of shared understanding like Aluminium ABS adoption by OEMs and to some extend also share information on sales and capacity (in aggregated form).
(855) Second, Aluminium ABS manufacturers have structural links with each other. […]. (856) […].672
(857) […].673 […],674 […]. (858) […].
(859) […].675
(860) Third, aside from these formal or structural links, manufacturers of Aluminium ABS in Europe appear to also interact with each other via signalling.
(861) […].676 […].677 […].
(862) In the second instance, earnings calls provide an opportunity for Aluminium ABS manufacturers to potentially signal intentions to competitors. This is for example evidenced in Constellium’s first quarter earnings call in which it was stated that ‘[w]e have been and will remain prudent with our investments’.678 In the second quarter earnings call, Constellium again made a very similar statement about its own capacity intentions: ‘We have been and will remain prudent with our investment. As I have noted many times in the past, we will not make incremental investments without firm customer commitment and strong confidence in end market demand’.679 Novelis, in talking about potential expansions, states that ‘there’s nothing of significance on our end to report’.680
(863) […].681
(864) In the Reply to the SO, the Notifying Party states that a […] .682 […].683 […].684685 […].
(865) The evidence presented above in this Section 8.3.10.2 points to numerous formal, structural and ad-hoc links between manufacturers of Aluminium ABS in Europe. In particular Novelis has many of such links.
(866) These links already pre-Transaction are liable to soften competition, as they lead to shared interests among competitors in a certain market outcome. Post-Transaction these existing links will increase in their relative significance on the market. Further, certain means of cooperation between competitors (such as signalling) will be easier, due to the smaller number of players active in the market.
(867) […].686
8.3.10.3. Imports are not a sufficient competitive constraint
(868) The present section demonstrates that it appears that imports into the EEA are not a sufficient competitive constraint to the Merged Entity.
(869) First, as indicated in Section 7.1.3, the limited imports into the EEA originate mainly from Switzerland, and, to a smaller extent from Saudi Arabia, and their competitive constraint to the Merged Entity is very limited.
(870) According to the information submitted by the Notifying Party, the only manufacturing plant in Switzerland producing Aluminium ABS belongs to the Notifying Party itself,687 therefore imports from Switzerland do not represent a competitive constraint to the Merged Entity.
(871) Imports from Saudi Arabia are manufactured and effected by Alcoa/ Ma’aden.688 In 2018, these represented less than [0-5]% of the total sales in the EEA (Section 8.3.5.1, Table 2), and no significant increase in sales is expected until 2023 (Section 8.3.5.1, Table 8). In addition, Alcoa/Ma’aden has […] in the EEA, which is […],689 to which it sells only […] alloys,690 on which both Novelis and Aleris have very limited strategic focus (see Section 8.3.8.2).
(872) Second, contrary to the Notifying Party’s claim that Chinese suppliers pose competitive constraints to EEA Aluminium ABS suppliers,691 the market investigation indicates that this alleged competitive constraint, if existent at all, is limited, and is not expected to increase in the foreseeable future.
(873) As evidenced in Section 7.1.3 and in particular in Figure 41, Aluminium ABS imports from China are very limited, if existent at all. […].692 […].693
(874) In the Reply to the SO, the Notifying Party estimates that imports from China are ‘up to […] kt of aluminium ABS per year’ and submits that the Chinese market for the production and supply of Aluminium ABS is characterised by […].694
(875) The Commission considers that the Notifying Party’s arguments are not in contradiction to its analysis, particularly concerning the imports from China. This is because […] tonnes of Aluminium ABS are consistent with the amount considered in Figure 41 in Section 7.1.3, and, in any event, represent less than […]% of the overall demand in the EEA.695 Therefore import from China at present do not represent a material competitive constraint to the Parties.
(876) With respect to the Notifying Party’s estimation of the overcapacity in China concerning the Aluminium ABS market, the Notifying Party does not provide any proof or even indication that such an overcapacity might result in increased imports into the EEA. On the contrary, the market conditions highlighted in Section 7.1.3 suggest the Aluminium ABS manufacturers located outside the EEA (as for example, Chinese manufacturers) would favour other regions for exporting their Aluminium ABS products. This is because the price for Aluminium ABS in the EEA is the lowest, compared to, for example, North America and China itself.
(877) Based on the above, the Commission concludes that imports into the EEA are not a sufficient competitive constraint to the Merged Entity.
8.3.10.4. Barriers to entry are high – new entrants face significant difficulties in establishing themselves as credible alternatives
(878) For the reasons explained in the present Section, the Commission considers that barriers to entry to the EEA market for the production and supply of Aluminium ABS are high, and that new entrants face significant difficulties in establishing themselves as credible alternatives.
(879) First, manufacturing Aluminium ABS requires additional equipment compared to most other Aluminium FRPs. This equipment requires substantial investment and time for its installation and ramping-up to full capacity.
(880) The manufacturing equipment used for addressing most of the non-automotive customers cannot be used for addressing automotive customers in a timely manner and with minor increased costs.696 As explained in Section 5.3, in addition to the equipment needed for manufacturing more common Aluminium FRPs (that is, for example, equipment for hot rolling, cold rolling, cutting, slitting, stretch-benching, and so on), a CASH line is also required for meeting OEMs’ technical requirements (see Section 8.3.7.1 and recital (581) in particular for why alternative technological solutions are not substitutes for a CASH line).
(881) One respondent to the market investigation indicated that the installation of a CASH line with a typical capacity of 100 000 tonnes requires about EUR 100 million,697 which is an investment of the same magnitude indicated by the Notifying Party. 698 This level of investment is relevant for the Aluminium ABS market participants and might represent a barrier to enter for potential manufacturers.
(882) With respect to the time required for installing a CASH line, ramping it up to full capacity and qualifying the resulting products with OEMs, it appears that there is a significant difference between manufacturers that already manufacture Aluminium ABS and new entrants. While most of the competitors agree […],699 that is that building a new CASH line requires about two to three years,700 new entrants would require longer time to ramp-up the production line to full capacity, obtain the required level of product quality, and qualify its products and processes with the various OEMs they intend to supply. As to that regard, one competitor emphasised: ‘The qualification process for each part is valid only for a specific process route in a plant (Specific hot rolling machine, cold rolling machine and cash line)’.701
(883) The time required for a new entrant to ramp-up a new production line depends on what extent the new entrant is new to the market for the production and supply of Aluminium ABS in the EEA. At least three types of new entrants should be distinguished: i) manufacturers already present in the market for the production and supply of Aluminium ABS, but not in the EEA; ii) manufacturers present in other Aluminium FRP markets but not in the Aluminium ABS market; and iii) potential new entrants not present in any Aluminium FRP market.
(884) An Aluminium ABS manufacturer active outside the EEA, would be able to ramp-up a new CASH line without the need for significant additional time, compared to an Aluminium ABS manufacturer already active in the EEA and expanding its capacity. […].702 […],703 […].
(885) In the case of manufacturers active in other Aluminium FRP markets, ramping-up requires more time because of the lack of industrial know-how for operating the CASH line itself as well as the time required for qualifying the product and the company with the various OEMs. A new entrant to the Aluminium ABS market explained: ‘The homologation process with the car manufacturers places a challenge due to the requested high quality level as well as a certain internal organization the OEMs require ensuring that its suppliers work in a certain way. As a new entrant to the market, the Company will not only have to qualify its product but also the company itself. This is different from companies like Novelis which are already qualified both with its product and the company itself’.704. The company further explained that, in order to qualify a new CASH line, the ‘homologation process is expected to be completed in about 2 years’.705
(886) Another competitor of the Parties quantifies the time required for a new entrant to be fully operational to be about four to five years, and explains that: ‘[e]quipment lead- time is roughly 2 years, then one year is needed for qualification. If the new CALP line is being built by a competitor that is experienced and already has a CALP line, ramp up to almost full capacity is expected to take 2 years. However, in the case of a green field investment by a new entrant, it can take 4-5 years’.706
(887) A new entrant with no previous experience will require much more time to be able to ramp-up its manufacturing facilities. In addition to a CASH line, all the manufacturing facilities for casting, hot rolling, cold rolling, slitting, etc. will have to be both set-up and ramped-up. This might require a relevant amount of time, particularly for gaining the know-how required for manufacturing Aluminium ABS of a quality level acceptable for OEMs. A competitor of the Parties, referring to a potential new entrant that has no previous experience in manufacturing any Aluminium FRP, considers that ‘[w]ithout prior know-how, […] it would take more than 10 years to enter the market for aluminium ABS’.707
(888) Second, the market investigation indicates that manufacturing Aluminium ABS requires significant a amount of industrial know-how, which appears to be particularly relevant for the 6xxx-series alloys, and, more specifically for the 6xxx- skin products.
(889) One of the Parties’ competitors explained that ‘[f]rom a manufacturing point of view there are huge differences between the 5000 series (“5xxx”) and 6000 series (“6xxx”) alloys'.708[…] ‘Manufacturing 6xxx alloys presents several barriers and requires detailed know-how and industrial experience’.709 More specifically, the competitor also explains that ‘[i]n terms of know-how, the Company considers that it is necessary to have a lot of experience and know-how. In order to become a qualified supplier for an OEM or for a Tier-1-supplier, rolled products need to undergo a series of tests for fulfilling an OEM’s qualification criteria. In order to fulfil these criteria, a large experience is needed, for example in electro discharge texturing ("EDT") and in accurately setting-up rolling and annealing schedules. Therefore, besides equipment there is a need to gather enough knowledge and experience to produce high qualities and to enter the market as a supplier.’710
(890) Another competitor estimates that ‘the time frame to enter the 6xxx alloy market to be approximately 3 years (1 year to develop the required know-how plus 2 years to build and qualify a CASH line). However, this estimate applies to a company with market presence in the 5xxx alloys, and with already some tier 1 automotive supplier customers’.711
(891) A Chinese manufacturer which recently entered the Aluminium FRP market, but not the Aluminium ABS market, indicated that ‘ABS is very difficult to manufacture […] thus making the entrance to this market more difficult’. In addition, it explained that ‘[q]uality requirements, and therefore manufacturing difficulties, are particularly high for those alloys that are used for exterior purposes. These are typically 6xxx series alloys with particular quality requirements related to the fact that they will be visible to the final customers, thus they require almost zero defect when it comes to surface quality, on top of stringent mechanical properties and excellent formability’.712 The same manufacturer explained that the difficulties in entering the Aluminium ABS market is the same in every part of the world because most OEMs have international footprint and require the same quality level in every regions where they manufacture their vehicles.713
(892) Another competitor shares the same view, and, as an indication of the importance of know-how to the competitive process in the market for the production and supply of Aluminium ABS, states that ‘[…] companies like Novelis and Aleris, […] accumulated substantial know-how in the last years’.714 The same competitor also stated that it monitors its competitors’ know-how by reviewing their patents and publications.715 […].716
(893) Third, OEMs’ procurement processes might represent an additional barrier to enter for new manufactures. As explained in Section 5.4, for each vehicle in production, typical Aluminium ABS supply contracts have a duration covering the entire vehicle life-time (that is, typically, five years). During this period of time, OEMs rarely change suppliers (although they have the possibility to). Therefore, contrarily to a market where customers purchase their yearly need every year, in the Aluminium ABS market only a part of the demand is available on a yearly basis because tenders cover a much longer period. Therefore, a new entrant that wants to acquire market shares will have to wait until new tenders related to new vehicles are organised by the various OEMs.717 This waiting period would further increase the time needed for a new entrant to gain market shares.
(894) […].
(895) […]. 718
(896) […].719
(897) […].720 […].
Figure 115 […]
[…]
(898) […],721 […].
(899) […].722 […].723 […].
(900) […].724 […].
(901) In its Reply to the SO, the Notifying Party considers that barriers to entry can be circumvented by OEM sponsorship.725 The Commission considers however that, in the market for the production and supply of Aluminium ABS, the so-called sponsorship from an OEM could provide only a partial mitigation to the high barriers to entry and would by no means allow for their circumvention. An OEM sponsorship would help a new entrant in securing some initial contracts, which might mitigate the barriers represented by the initial investments. However, an OEM could do very little, for example, to reduce the time that is required for a new entrant to acquire the required know-how and to be able to significantly contrain the Parties. The two new entrants cited by the Notifying Party, Profilglass and Alcoa/Ma’aden, indeed do not appear to represent a significant competitive constraint to the Parties as explained in recitals (894)-(900).
(902) The Notifying Party also argues that in its internal documents it […].726 However, despite the email exchange cited by the Notifying Party which is the sole internal document cited by the Notifying Party regarding […], the market investigation did not corroborate this argument: OEMs indicated that they do not expect new entrants to the Aluminium ABS market in the EEA in the next three years.727
8.3.10.5.Conclusion
(903) For the reasons set out in this Section 8.3.10, and considering all evidence available to it, the Commission concludes that competitors are unlikely to offset price increases resulting from the Transaction. Competitors’ spare capacity appears to be limited and it is tightly monitored to move in line with demand. Links existing between manufacturers of Aluminium ABS in the EEA, and in particular between Novelis and its competitors, reduce the likelihood of a reaction from competitors. Further, neither imports nor potential new entrants can be considered as sufficient competitive constraints. In this context, the Merged Entity would have an incentive and the capability to reduce the output below the combined pre-Transaction levels and thereby raise market prices. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.11. The negative effects of the Transaction are likely not countervailed by buyer power
(904) Buyer power refers to the bargaining strength that the buyer has vis-à-vis the seller in commercial negotiations due to its size, its commercial significance to the seller and its ability to switch to alternative suppliers.728
(905) The Notifying Party claims that […].729 […].730
(906) For the reasons set out below, the Commission however considers the buyer power of automotive customers of Aluminium ABS suppliers to be limited, primarily due to the limited spare capacity available in the market and various barriers to switching. Further, the Transaction would reduce any residual customer buyer power there is. Thus, customers would not be able to avoid price increases by the merged entity or by other suppliers of Aluminium ABS in the EEA.
8.3.11.1. Industry concentration is high and spare capacities are low
(907) As detailed in Section 8.3.5, concentration in the EEA market for the production and supply of Aluminium ABS is already high pre-Transaction. Only five suppliers (Novelis, Aleris, Constellium, Hydro and AMAG) can be considered as full-portfolio suppliers; and even some of these suppliers face limitations with respect to specific specifications (as described in Section 8.3.11.2). Novelis pre-Transaction and the Merged Entity post-Transaction would be the clear market leader with very high market shares. Such a limited number of suppliers and the resulting high level of supply side concentration are indicative of customers having limited potential options to switch their demand in case of a price increase by the Merged Entity.
(908) The Notifying Party however submits ‘that market shares do not reflect market power in markets where customers have buyer power’731 and that the Aluminium ABS market is characterised by an ‘extraordinary degree of buy side concentration’.732
(909) Nonetheless, in the following the Commission demonstrates that concentration in the EEA market for the production and supply of Aluminium ABS is in fact higher at the upstream supplier level than at the downstream customer level. This difference in concentration suggests that a supplier is more important for a customer than a customer is for a supplier. Further, evidence is presented highlighting that both the Notifying Party’s market power vis-à-vis certain customers and the limited availability of spare capacity further restrict customers’ buyer power.
(910) First, the Notifying Party submits that currently ‘[a]round […]% of total aluminium ABS demand in the EEA is concentrated in the hands of only four OEMs, […]. Only two OEMs, […], account for about […]% of total demand. ABS suppliers are thus highly dependent on a few customers. OEMs use this dependency to exercise buyer power’.733
(911) In the Reply to the SO, the Notifying Party states that the Merged Entity would face significant buy-side concentration.734 The Commission recalls that post-Transaction, the largest two Aluminium ABS suppliers would account for [80-90]% of total sales in volume in 2018 and for [70-80]% in 2023.735 The Notifying Party claims that currently the two largest OEMs, […] account for [60-70]% of demand. According to the Notifying Party’s projection captioned in Figure 116, in 2025 their share of demand would be [40-50]%. Thus, looking at the two largest suppliers and customers, supply-side concentration is currently higher and will remain higher in the future.
(912) When considering the four largest Aluminium ABS suppliers pre-Transaction (Novelis, Constellium, Hydro, Aleris), their combined share of total sales would be [90-100]% in 2018. In 2023, the largest four suppliers would pre-Transaction have a combined share of total sales of [90-100]%. Post-Transaction, the largest four suppliers in 2023 (the Merged Entity, Constellium, Hydro, AMAG) would have a share of 97%. The Notifying Party states that currently the four largest customers of Aluminium ABS, […] account for [80-90]% of demand. In 2025 these four OEMs would account for [80-90]% of total demand.736 Thus looking at the four largest suppliers and customers, supply-side concentration is currently higher and will remain higher in the future.
Figure 116 […]
[…]
(913) The decrease in relative importance of the traditionally large Aluminium ABS customers is not due to a decrease in their Aluminium ABS demand. Rather, it is due to the increase in demand from other customers, such as […]. As new demand is emerging, the bargaining position of the traditional customers is likely lowered.
(914) Second, the limited number of instances in which OEMs switch their demand to alternative suppliers during the lifetime of a vehicle is further evidence of OEMs’ limited buyer power. In the Reply to the SO, the Notifying Party in postulating that OEMs have ‘significant bargaining power […] vis-à-vis ABS suppliers’ states that it has […] .737 […].738 […].739
(915) […].740 The Commission however notes that other factors explaining the lack of switching of supply are likely to include factors related to limited buyer power, such as limited available uncommitted capacity or the small number of qualified and capable suppliers. In any case, the instances described by the Notifying Party of OEMs shifting volumes away from the Parties during the lifetime of a vehicle largely […].
(916) Further in the Reply to the SO, the Notifying Party claims that ‘[s]uppliers are in a take-it-or-leave it situation when negotiating with OEMs and supplying products’.741 […]742 […].743
(917) Third, the Notifying Party’s claim that OEMs have substantial buyer power is further put into question by […] and the position of non-OEM customers (such as Tier suppliers) in the market. […].
(918) […].
Figure 117 […]
[…]
(919) […].744 […].
(920) […],745 […].746
(921) In describing the expected impact of the Transaction on its business, JLR stated in reply to the market investigation that ‘[c]ommercial competition in the market will decrease, potentially further growing Novelis current capacity dominance’747 and further ‘[n]o overall capacity impact, however capacity concentration within one company needs checking further. % of EU supply with one supplier’.748
(922) […].
Figure 118 […]
[…]
(923) […],749 […].750 […].751
(924) […].752 […].753 […],754 […].755 […].
Figure 119 […]
[…]
(925) […].756 […].757
(926) […].
Figure 120 […]
[…]
(927) […]758 759 […].760 […].
(928) […].761 […].762
(929) […]. In the Reply to the SO, the Notifying Party claims that because ‘around […]% of Novelis’ current sales are […][they are] thus irrelevant to the competitive analysis’.763 The Notifying Party however also submits that […].764 Further, […]. This evidences that Novelis has experienced and expects competition for volumes which are contracted and at a customer at which it is pivotal for certain volumes. Novelis’ current share of sales to […] is thus an outcome of a competitive process and a result of Novelis’ relative market power. Furthermore, volumes contracted for coming years may also be contested.
(930) In the second instance, it is at this point also important to recall that not all of the Parties’ customers are OEMs. Distributors and Tier suppliers also purchase Aluminium ABS products from the Parties (as detailed in Section 5.4). One of these distributor customers reports that based on OEMs’ requirements, it is ‘allowed to purchase only from qualified suppliers’.765 It does not organise the qualification process itself, but is nevertheless bound by the qualification agreed between the OEM and the Aluminium ABS manufacturer. This shows that such distributor customers lack buyer power.
(931) Fourth, in the Reply to the SO, the Notifying Party states that […] and concludes that the Transaction therefore […].766 In reaction to this argument,767 the Commission observes that the Transaction will have a negative impact on […] ability to source Aluminium ABS from alternative suppliers to Novelis by removing Aleris as a potential supplier, based on the following.
(932) […].
Figure 121 […]
[…]
(933) […].
Figure 122 […]
[…]
(934) […].768 The Commission does not dispute that Aleris’ current capacity situation would make Aleris’ entry at […] difficult, however given Aleris’ continued exploration of the option to increase its capacity (as discussed in Section 8.3.9.3), absent the Transaction, there would be an opportunity for this to change.
(935) […].769 […].770 […].
(936) […].771
Figure 123 […]
[…]
(937) […]772 […].
Figure 124 […]
[…]
(938) The above evidence shows that Aleris is deemed capable to in the future qualify for certain parts at […] and that Aleris has already considered undertaking this qualification. The Transaction will thus remove a potential competitor for […] volumes from the market and in turn further increase Novelis’ market power at […].
(939) Furthermore, and considering the evidence presented in recitals (917) to (938), the Commission concludes that Novelis' current share at […] is both the result of its pivotal position at […] for a certain share of […] demand […] as well as the result of Novelis competitive interaction with competitors for another share of […] demand (with Ma'aden and Hydro in particular, but also with Constellium and Aleris, which so far have not successfully entered at […]).
(940) […].
(941) […].773 […].774 […].
Figure 125 […]
[…]
(942) […].
Figure 126 […]
[…]
(943) Sixth, as discussed in Section 8.3.6, Novelis is already pre-Transaction a pivotal player in the market and would be even more so post-Transaction. This means that it faces some residual demand which cannot be covered by its competitors. In other words, for a certain amount of their overall demand, automotive Aluminium ABS customers only have Novelis (or post-Transaction the Merged Entity) available as a potential supplier. This is a source of market power for Novelis and limits customers’ buyer power.
(944) Seventh, the limited available spare capacity in the EEA market for the production and supply of Aluminium ABS further limits customers’ buyer power, as it restricts competitors’ ability to serve customers with additional volumes and, hence, hampers customers’ ability to shift significant volumes to alternative suppliers were the merged entity to raise prices.
(945) As detailed in Section 8.3.7, and contrary to the Notifying Party’s claim, spare capacity in the EEA market for the production and supply of Aluminium ABS is limited. In 2018, spare CASH capacity in the EEA was less than […] tonnes, with the Notifying Party controlling […]. 775 In line with that, a competitor to the Parties explicitly states in the market investigation that it currently has little available uncommitted capacity and could not take on a significant number of additional volumes. The competitor mentions that ‘[c]urrently the Company has only very little spare capacity on its CASH lines’.776
(946) Further, and as explained in Section 8.3.7.2, not the entire CASH line capacity can be considered readily available for automotive customers, as some is currently used for production of for example aerospace products.
(947) The limited spare capacity makes it difficult for Aluminium ABS customers to avoid price increases by shifting volumes to alternative suppliers. It thus limits their countervailing buyer power. Novelis, the current market leader, holds […] of the spare CASH capacity. […]. This is a further source for Novelis’ market power.
8.3.11.2. While customers qualify multiple suppliers, already pre-Transaction, customers face difficulties in switching significant volumes
(948) A source of countervailing buyer power would be the ability ofa customer to credibly threaten to resort, within a reasonable timeframe, to alternative sources of supply should the supplier decide to increase prices.777
(949) The Commission notes that Aluminium ABS customers’ ability to resort to alternative sources of supply in case of a price increase is, aside of the limited availability of spare capacity discussed in Section 8.3.11.1, limited.
(950) First, the ability of customers to threaten to switch their supply is limited by the number of suppliers of Aluminium ABS in the EEA. As discussed in Section 8.3.4.1, there are currently five Aluminium ABS suppliers in the EEA capable of supplying the range of alloys of relevance for the automotive industry (5xxx and 6xxx). Post- Transaction, this number will decrease to four.
(951) This is particularly relevant due to the multi-sourcing exercised by many automotive OEMs. These customers are cautious in not being too exposed or reliant on a single supplier. Sourcing (both for a given vehicle model and on an Aluminium ABS procurement level overall) often occurs from three or more suppliers. In a market with this characteristic, market shares of suppliers tend to underestimate their real market power as customers rely on receiving at least two competitive offers, as opposed to only one in an industry without multi-sourcing. In the Reply to the SO, the Notifying Party claims that ‘multi-sourcing practices will not be affected by the continued availability of at least four suppliers’.778 This claim, however, is based on a misconception of what multi-sourcing really means. Multi-sourcing is, other than suggested by the Notifying Party779 not the practice of sourcing from two suppliers which are the Parties, but rather the practice of sourcing from a number (two or more) suppliers in order to avoid too great an exposure to individual suppliers. Such a practice will become more difficult when the number of credible bidders declines (in particular in light of the barriers associated with sponsoring entry of potential new suppliers, as detailed in Section 8.3.10.4). Further, should some OEMs want to shift supplies away from the Merged Entity as a result of the Transaction, the relative market power of competitors would be strengthened and OEMs would, as a result of the Transaction, be exposed to stronger suppliers.
(952) Second, not all suppliers of Aluminium ABS in the EEA appear to have a complete product portfolio with all products currently being supplied to automotive customers.
(953) […],780 […].
Figure 127 […]
[…]
(954) In the second instance, some customers […] mention […] limitations in terms of suppliers’ product portfolios. One tier customer even notes that ‘rarely all big 5 are available to supply some special alloys according to an OEM’s requirements’ and that in some cases this could result in only ‘one remaining supplier for certain specifications’.781 The customer provides the examples captioned in Figure 128 and Figure 129 of certain alloy requirements by two automotive OEMs for which currently only at most three Aluminium ABS manufacturers are capable of supply, while post-Transaction in some cases only two would remain.
(955) Further, in indicating which products by other suppliers are equivalent to certain Novelis products it has qualified, one major OEM states that for certain products only two or three products overall are equivalent and qualified.782 The same customer states that ‘on some products/alloys, only 2 to 3 suppliers, including Aleris, are active’.783
(956) Such instances in which not all suppliers are providing certain specifications further limit customers’ options for switching their volumes to alternative suppliers in case of price increases.
(957) Third, customers’ qualification requirements can result in further difficulties for switching.
(958) In the first instance, the general industry standards appear to be very broad and rather a baseline than a set of specifications customers exclusively depend on for their procurement choices. One major customer explains that ‘[i]n terms of technical specifications, the industrial standards in place are very general in nature. Their technical specifications are too wide for automotive industry requirements. Therefore, [the Company] tests and qualifies all the various alloys before they are purchased’.784 Another customer confirms that ‘[e]ach part of the car has different requirements and grades. According to strength, formability and joining’.785
(959) In the second instance, the qualification exercises undertaken by OEMs can take a significant amount of time, particularly where a certain product of a supplier had not been qualified already. This results in additional time being required for switching supply to alternative suppliers. One OEM ‘considers that the qualification process takes about one year’.786 Another OEM’s ‘qualification process can take 4-5 months for alloys that are already known to [the OEM] (e.g. the 5xxx alloys) and up to 2 years for new alloys (this would be the case, for example, for the 7xxx alloys, which are new to the automotive industry)’.787
(960) In the third instance, the time required for qualification may become relevant in cases where currently not all of the suppliers that are supplying a particular customer are also qualified for all the product and/or alloy segments that customer is using. On the question of high-formability alloys, a majority of OEMs that expressed an opinion state that only some of their qualified suppliers of Aluminium ABS are also qualified for high-formability alloys.788
(961) […].
Figure 130 […]
[…]
(962) In the Reply to the SO, the Notifying Party claims that ‘ample evidence of past customer switching’ confirms the absence of barriers to switching.789 The Commission however observes that this concerns instances of switching among suppliers that are qualified and thus is not informative about qualification (and the lack thereof) as a barrier to switching. Furthermore, the examples of switching listed in paragraph 512 of the Reply to the SO are not an indication that by switching suppliers ‘OEMs have sufficient ability to counteract any attempt by the merged entity to start acting independently from them and its competitors’.790 This is evidenced by the fact that, while the switching examples include numerous examples of […].791 In addition, OEMs only can switch away from Novelis or the Merged Entity to the extent competitors have available open capacity. As discussed in Section 8.3.7.3, this spare capacity is limited.
(963) In the Reply to the Letter of Facts, the Notifying Party claims that as qualification is a fundamental part of the market for the production and supply of Aluminium ABS and therefore non-qualification of a supplier cannot be regarded as a barrier to switching. Further the Notifying Party provides examples of OEMs that have qualified new suppliers in the past years ([…]).792 The Commission, however, maintains that, in practice, a lack of qualification can be a barrier to switching for OEMs as, in certain tender situations, OEMs ability to source all parts from all suppliers is limited by the fact that not all are qualified (and qualification, even in a case where a supplier has the technical capability to meet requirements, requires time and costs).
(964) Fourth, even if customers have multiple Aluminium ABS manufacturers qualified, they may encounter difficulties in switching supply between them once a specific tender has been awarded to one supplier.
(965) In the first instance, supply agreements between OEMs and Aluminium ABS suppliers often contain competitiveness clauses as explained in Section 5.4. However, the Notifying Party submits that, to the best if its knowledge, […].793 […].794 All in all, OEMs may face difficulties in switching supplier during the lifetime of a contract despite having qualified several suppliers, when such ‘competitiveness rights’ are not available or if, despite the existence of such rights, the other qualified suppliers are not able to address the request, for example due to lack of spare capacity.
(966) In the second instance, customers may be faced with a situation in which not all qualified suppliers have sufficient available capacity for the tender in question. While a majority of OEMs that responded to the market investigation submit that, over the last three years, they did not have to exclude suppliers due to manufacturing capacity constraints, one major OEM submits that it always had to do so.795 As discussed in 8.3.7.3, available spare capacity is limited. In particular OEMs with demand for substantial Aluminium ABS volumes would find it difficult to switch entirely away from current suppliers. In light of this one major OEM states that ‘[c]urrently available capacity in Europe as well as the amount of time needed to install new capacity result in general difficulties if a customer were to replace Aleris and Novelis as suppliers’.796 This statement directly contradicts the Notifying Party’s claim in the Reply to the SO that post-Transaction ‘there is no indication […] that four credible bidders is insufficient to allow OEMs to switch their supplier in the case of a price increase’,797 as the customer would not be able to avoid a price increase by the Merged Entity due to the limited available capacity of competitors. Another OEM states that ‘in the recent past, during 1 tender a supplier did not quoted (sic!) 5xxx alloys due to limited capacity (6xxx alloys more profitable for aluminium producers)’.798 This is further evidence of automotive customers’ inability to rely on (competitive) bids by all qualified suppliers in all their tenders.
(967) In the third instance, a need for new stamping tools by the OEM may lead to a cost associated with switching even among qualified suppliers. One OEM states that a factor hindering the switch of suppliers from one tender to the next ‘is a new tool (stamping tool) required’.799 Another OEM also confirms that adjusting ‘press tools of our press shop […] to a supplier’s material’ can lead to costs.800 These costs can constitute a hindrance to switching even among qualified suppliers.
(968) Fifth, existing long relationships between customer and supplier can also create an obstacle for the customer to switch significant volumes to other suppliers. As one major OEM explains, ‘[d]ue to the long relationship between Aleris and the Company, Aleris has a number of products that can be easily customised to comply with the Company’s specifications’ and ‘[t]he Company and Aleris develop products very closely together’.801 In case the customer in question would have to switch significant volumes away from Aleris, these benefits of the close relationship would be lost.
(969) Further, a majority of respondents to the market investigation state that, when selecting a supplier of Aluminium ABS, they consider that past relationships influence the selection.802
(970) Sixth, and to recall Sections 6.2.3.3 and 8.3.8.4, flat steel products are not a viable option for current customers of Aluminium ABS suppliers to switch significant volumes to.
8.3.11.3. Post-Transaction, customers would face even more difficulties in switching to alternative suppliers
(971) It is not sufficient that buyer power exists prior to the merger. This is because a merger of two suppliers may reduce buyer power if it thereby removes a credible alternative.803
(972) Post-Transaction, customers would have one less credible supplier available and thus would be even more restricted in their options of switching volumes in case of a price increase.
(973) First, post-Transaction, the number of Aluminium ABS suppliers in the EEA capable of supplying a comprehensive range of Aluminium ABS products of relevance to the automotive industry would decrease to only four. This decrease in the number of suppliers and the resulting higher concentration on the side of supply in the EEA-market for the production and supply of Aluminium ABS would further reduce customers’ buyer power.
(974) Second, as explained in Section 8.3.7, spare capacity in the market is low. Post- Transaction, the spare capacity held by the Merged Entity’s competitors would be limited (overall spare CASH capacity was […] tonnes in 2018, […] Novelis pre- merger and by the Merged Entity post-merger). In addition, the Parties together supplied approximately […] kt804 of Aluminium ABS to their customers in 2018. Already pre-merger automotive customers do not have the ability to switch significant volumes to competitors of Novelis. By capturing the sales of Aleris, the Merged Entity would post-merger have even more captive customers, that is control more quantities supplied to customers. Thus, in relative terms there would be post- Transaction higher volumes (those supplied by the Merged Entity) which would be directly affected by any price increase by the Merged Entity, while there would at the same time only be very limited available spare capacity. The customers’ buyer power would thus further decrease.
(975) Third, as discussed in Section 5.4, […]. The loss of competition through the proposed Transaction would thus not be confined to upcoming tenders, but also further limit the already limited competitive interaction on existing supply relationships. The loss of a competitor in the market for the production and supply of Aluminium ABS thus results in a further limitation of the possibility for automotive customers to receive lower prices. The market power of suppliers would further increase as the acquisition of Aleris by Novelis removes one supplier from the market who could have potentially offered better terms and thus challenged volumes already awarded to Novelis.
(976) Fourth, customers for which, pre-Transaction, Novelis and Aleris both had significant shares of supply or both shared a strategic focus on, would see their buyer power decrease. These customers would post-Transaction lose the competitive interaction currently present between Novelis and Aleris.
8.3.11.4.Conclusion
(977) For the reasons set out in this Section 8.3.11, and considering all evidence available to it, the Commission concludes that EEA Aluminium ABS customers likely already pre-Transaction have limited buyer power, and that any residual buyer power they do currently have would be further reduced due to the Transaction. Specifically, EEA Aluminium ABS customers would not be able to switch significant volumes away from the Merged Entity and would thus not be able to avoid price increases by the Merged Entity. These elements support the conclusion that the Transaction would lead to the creation or strengthening of a dominant position.
8.3.12. Conclusion
(978) Considering all evidence available to it, and in light of the considerations explained in this Section 8.3, the Commission considers that the Transaction would significantly impede effective competition in relation to the market for the production and supply of Aluminium ABS in the EEA because the Transaction would create or strengthen a dominant market position in the relevant market. In any event, the Transaction would also give rise to horizontal non-coordinated effects in relation to the production and supply of Aluminium ABS in the EEA resulting from the elimination of an important competitive constraint.
8.4. Standard FRPs
(979) Within Standard FRPs, the activities of the Parties overlap in the supply of aluminium standard sheet for a variety of end-uses.
(980) In the EEA, Novelis supplies a number of Standard FRPs, including aluminium standard and painted sheet, for a variety of end-uses such as gutters, roofing, ceilings, roof windows, floor heating, roller shutters, sun-breakers, lighting, solar power technologies, wind turbines, energy storage, transportation, packaging, electronics and home appliances.805
(981) In the EEA, Aleris supplies aluminium standard sheet of different gauges for a variety of end uses including building and construction, industrial, consumer products, electronics, packaging, transportation, floor heating, and many others.806
(982) The Notifying Party argues that the Transaction does not raise any competition concerns with respect to Standard FRPs as the Parties’ combined market shares in the EEA market for the production and supply of Standard FRPs are small, they are not close competitors and will continue facing strong competition from multiple actual and potential players.
(983) The Commission observes that, on a putative overall market for the production and supply of Standard FRPs, the combination of the Parties' activities would not result in an affected market within the meaning of the Form CO. Based on the Parties’ best estimates, the market share of the Merged Entity in the EEA market for the production and supply of Standard FRPs would be limited to [5-10]%807 with Novelis accounting for approximately [5-10]% and Aleris [0-5]%.
(984) The Commission also notes that a number of other players would still compete with the Merged Entity in the EEA market for the production and supply of Standard FRPs post-Transaction, including Hydro ([10-20]%), Arconic ([10-20]%), Constellium ([0-5]%), Amcor ([0-5]%), AMAG ([0-5]%), Elval ([0-5]%) and others.808
(985) For these reasons, the Commission considers that the Transaction would not impede effective competition in the supply of Standard FRPs in the EEA. In order to exclude concerns, however, the Commission has further assessed the effects of the Transaction on the narrower plausible relevant markets for aluminium anodising sheet in the EEA and aluminium sheet for compound tubes in the EEA. The Commission's assessment is presented in the below sections.
8.5. Aluminium anodising sheet
8.5.1. Introduction
(986) In the EEA, the Parties’ activities overlap in the supply of aluminium anodising sheet. Although neither Party has in-house anodising capabilities, both Parties either manufacture and supply anodising quality sheet to customers that have it anodised (internally or by a third party) or they supply pre-anodised aluminium sheet, for which they outsource the anodising process to a third party.
8.5.2. The Notifying Party’s views
(987) The Notifying Party argues that, were the production and supply of aluminium anodising sheet in the EEA to constitute a distinct market, the Transaction would not raise any competition concerns with respect to that market for a number of reasons.809
(988) First, on the demand side, the Notifying Party argues that aluminium anodising sheet competes with various other materials for use in building facades, including steel, vinyl, and titanium, as well as with other kinds of aluminium sheet, including standard sheet with a painted or coated surface, and ACP. Furthermore, according to the Notifying Party, customers increasingly also use standard aluminium sheet and have it anodised for use in their building facades.
(989) Second, on the supply side, the Notifying Party argues that producers can easily expand production given the very low demand for these products (limited to approximately […] annually). Moreover, according to the Notifying Party, other Aluminium FRP suppliers can adjust the production process and start producing aluminium anodising sheet without requiring additional equipment as the anodising can be outsourced to external providers.
(990) Lastly, the Notifying Party argues that the Parties’ modest combined share of supply, which furthermore is trending downwards, does not give rise to any concerns as the Merged Entity would continue competing against over 15 players such as Hydro, AMAG, Constellium, Slim, Arconic, Profilglass or Aludium, each of which has the capacity and ability to expand their output.810
8.5.3. The Commission’s assessment
(991) The Commission has assessed a putative distinct market for the production and supply of anodising sheet in the EEA, excluding any other materials such as steel, vinyl or titanium. In its assessment, the Commisssion has taken into account particularly the following factors.
(992) First, the Parties’ combined market shares remain modest. Based on the Parties’ best estimates, their combined market share in the EEA market for the production and supply of aluminium anodising sheet would be [30-40]% in 2016, [20-30]% in 2017 and [20-30]% in 2018, with Novelis accounting for approximately [10-20]% and Aleris [10-20]% in 2018. Therefore, post-Transaction the market share of the Merged Entity would remain modest at [20-30]%.
(993) Second, a number of other competitors remain in the market. Based on information supplied by the Notifying Party, such competitors include for example Hydro ([20- 30]%), AMAG ([10-20]%), Constellium ([5-10]%), Slim ([5-10]%) and Elval ([5-10]%)811.
(994) In line with this, the broad majority of customers consider that they have adequate alternatives to source aluminium anodising sheet.812 A customer notes that ‘[t]he demand of These products ist [sic] lower than the offer of such Sheets – So there is still a good Competition for our Needs’. Another customer concurs: ‘Yes, there are enough alternative suppliers to choose from’.813
(995) Third, recent entry into the market has taken place and barriers to entry, to the extent present, have not prevented such entry. In this respect, the Commission notes that while some respondents have referred to barriers to entry into the market,814 market participants have also explained that some EEA suppliers, such as AMAG, have recently entered the market.815
(996) Fourth, […], the Commission’s investigation has shown that there are a number of capable actual and potential competitors in this narrow putative segment.816
(997) In line with this, the Commission observes that even if one considered the potential narrower segment for aluminium anodising sheet for architectural applications that require colour uniformity, based on the Parties’ best estimates, the market share of the Merged Entity would remain at [20-30]%,817 that is close to the figure for their overall combined market share in aluminium anodising sheets.
8.5.4. Conclusion
(998) For the reasons set out in Section 8.5.3 above, the Commission considers that the Transaction would not significantly impede effective competition in the putative market for the production and supply of aluminium anodising sheet in the EEA.
8.6. Aluminium sheet for compound tubes
8.6.1. Introduction
(999) In the EEA, the Parties’ activities overlap in the supply of aluminium sheet for compound tubes.
8.6.2. The Notifying Party’s views
(1000) The Notifying Party argues that, were the production and supply of aluminium sheet for compound tubes in the EEA to constitute a distinct market, the Transaction would not raise any competition concerns with respect to that market for a number of reasons.818
(1001) First, on the demand side, the Notifying Party argues that aluminium sheet for compound tubes competes with other materials such as copper, plastic, or steel, and that, in Western Europe, plastic, copper and aluminium is used in roughly equal measures in compound tubes.
(1002) Second, on the supply side, the Notifying Party argues that producers can easily expand production given the very low demand for these products (limited to approximately […]). Moreover, according to the Notifying Party, other Aluminium FRP suppliers not currently manufacturing aluminium sheet for compound tubes can normally adjust the production process to start producing sheet for this product category without making significant investments in equipment.
(1003) Lastly, the Notifying Party argues that the Parties combined market shares are modest and that the Merged Entity would continue to compete against several strong competitors including Madar, Elval, Arconic, Hydro, and Constellium.819
8.6.3. The Commission’s assessment
(1004) The Commission has assessed a putative distinct market for the production and supply of aluminium sheet for compound tubes in the EEA, excluding any other materials such as plastic, copper and steel. In its assessment, the Commisssion has taken into account particularly the following factors.
(1005) First, while the Parties would achieve not insignificant combined market shares, the combined increment brought about by the Transaction is limited. Based on the Parties’ best estimates, their combined market share in the EEA market for the production and supply of aluminium sheet for compound tubes would be [40-50]% in 2016, [40-50]% in 2017 and [30-40]% in 2018. The market shares are primarily due to Aleris’ market position with Novelis accounting for approximately [0-5]% and Aleris [30-40]% in 2018. The increment brought about by the Transaction would thus be limited.
(1006) Second, the overlap between the Parties has been reducing significantly in the past years: For example, according to the Parties’ best estimates, Novelis’ market share […] from [5-10]% in 2017 to [0-5]% in 2018. In line with this, internal documents of the Parties point at the loss of Novelis’ market share in recent years in the EEA.
(1007) Third, a number of competitors would remain in the market. Based on information provided by the Notifying Party, such competitors include for example Madar ([20- 30]%), Elval ([10-20]%), Hydro ([10-20]%) and Arconic ([5-10]%)820.
(1008) In line with this a clear majority of the customers taking a position indicated that they have adequate alternative suppliers.821 A customer indicates in this respect that sourcing is possible from ‘at least 6/7 different suppliers’.822
(1009) Fourth, while the Commission’s investigation did not reveal any recent entry into the market, at least one Aluminium FRP supplier not currently active in the production and supply of aluminium sheet for compound tubes has explained that it could enter the market in the short term.823 In addition, one of the established competitors has recently increased its capacity and has ‘plans to grow in the market for aluminium sheet for compound tubes’.824
(1010) Fifth, it appears that customers can in practice change suppliers despite qualification requirements. The Commission notes in this respect that a customer explained in the market investigation that, when it learned about the envisaged merger between Novelis and Aleris in 2018, it decided to qualify a third supplier and intends to source from such new supplier already in 2019.825
8.6.4. Conclusion
(1011) For the reasons set out in Section 8.6.3 above, the Commission considers that the Transaction would not significantly impede effective competition in the putative market for the production and supply of aluminium sheet for compound tubes in the EEA.
9. EFFICIENCIES
(1012) The Commission considers substantiated efficiency claims in its overall assessment of a proposed Transaction. It may decide that, as a consequence of the efficiencies that the merger brings about, there are no grounds for declaring the merger incompatible with the internal market. This is the case when the Commission is in a position to conclude on the basis of sufficient evidence that the efficiencies generated by the merger are likely to enhance the ability and incentive of the merged entity to act pro-competitively for the benefit of consumers, thereby counteracting the adverse effects on competition which the merger might otherwise have.826 Most of the information, allowing the Commission to assess whether the merger will bring about the sort of efficiencies that would enable it to clear a merger, is solely in the possession of the merging parties. It is, therefore, incumbent upon the notifying parties to provide in due time all the relevant information necessary to demonstrate that the claimed efficiencies are merger-specific and likely to be realised. Similarly, it is for the notifying parties to show to what extent the efficiencies are likely to counteract any adverse effects on competition that might otherwise result from the merger, and therefore benefit consumers.827
(1013) The Commission therefore considers positive effects of efficiencies that benefit consumers as part of its overall assessment of the Transaction, provided the efficiencies are substantiated and satisfy the following three cumulative criteria:
(a) Efficiencies have to benefit consumers in the sense that they should be substantial and timely and should, in principle, benefit consumers in those relevant markets where it is otherwise likely that competition concerns would occur;828
(b) Efficiencies have to be a direct consequence of the concentration and cannot be achieved to a similar extent by less anticompetitive alternatives;829
(c) Efficiencies have to be verifiable such that the Commission can be reasonably certain that the efficiencies are likely to materialise, and be substantial enough to counteract a merger’s potential harm to consumers.830
(1014) The Notifying Party describes the efficiencies it expects to realise through the Transaction in Section 9 of the Form CO. Novelis submits that it expects ‘approximately USD […] of run-rate cost synergies’.831 These consist of […] from […].
(1015) The Commission notes that no explanation of how and whether these synergies would benefit customers and consumers, or result in pro-competitive effects more generally, is provided by the Notifying Party in the Form CO.
(1016) The Notifying Party further, in a letter dated 4 June 2019 and addressed to the Commissioner for Competition, stated that the Transaction ‘will make Aleris and the aluminium industry more competitive, bring benefits to our customers and the ultimate consumers, and contribute to the EU’s ley policy goals of promoting the circular economy, reducing emissions, and securing skilled employment in the EU’.832 The Notifying Party considers that the Transaction would, in particular, allow an increase in aluminium recycling activities in Europe, more investments into research and development by the Merged Entity and an increase in ‘the aluminium body sheet capacity available to automotive manufacturers in Europe’. This will lead to ‘cost savings, innovation and operational benefits [that] will be passed on to […] automotive customers and ultimately the end-consumers’.
(1017) On the basis of these submissions, the Commission notes that the Notifying Party submits that the Transaction would entail a number of benefits and efficiencies, but did not substantiate its arguments in such a way that would allow for a comprehensive assessment under Section VII of the Horizontal Merger Guidelines. In particular, the submissions do not lay out in detail benefits to consumers, merger specificity and the absence of less anticompetitive alternatives, or evidence in support of the efficiency claims. The Commission therefore considers that a full engagement with the Notifying Party’s efficiency claims, and a comprehensive assessment of these, is not possible.
10. CONCLUSION ON THE COMPATIBILITY OF THE NOTIFIED TRANSACTION WITH THE INTERNAL MARKET
(1018) For the reasons set out above in Section 8, the Commission concludes that the Transaction, as notified, would significantly impede effective competition in the EEA market for the production and supply of Aluminium ABS because the Transaction would create or strengthen a dominant market position in the relevant market. In any event, the Transaction would also give rise to horizontal non- coordinated effects in relation to the production and supply of Aluminium ABS in the EEA resulting from the elimination of an important competitive constraint.
(1019) As set out in Section 8.3.5, the Merged Entity would have a very large market share in terms of sales and capacity ([…] [50-60]% in 2018) in the EEA market for the production and supply of Aluminium ABS. As explained in the Horizontal Merger Guidelines and in Section 8.2.1, very large market shares – 50% or more – may in themselves be evidence of the existence of a dominant market position.
(1020) While the Commission’s market reconstruction projects that the Merged Entity’s capacity share will be […] [50-60]% for certain coming years (from 2021 onwards), the Commission nevertheless considers the Transaction to lead to the creation or strengthening of a dominant market position of Novelis for the following reasons:
(a) The Merged Entity’s market shares in volume will remain […] [50-60]%, also in the coming years;
(b) The Merged Entity will only face competition from a limited number of competitors and will have a market and capacity share that is at least […] as large as that of its next competitor, also in the coming years;
(c) Competitors to the Merged Entity have limited spare capacity and the Merged Entity is a pivotal player in the market;
(d) Furthermore, the Transaction would likely result in higher prices in the relevant market and reduce the Merged Entity’s incentives of adding capacity to the market. In addition, competitors would be unlikely to offset price increases resulting from the Transaction and, to the extent that there would be any buyer power, it would be insufficient to offset the effects of the Transaction.
(1021) In any event, the Transaction would also give rise to horizontal non-coordinated effects due to the removal of Aleris as an important competitive constraint on Novelis.
(1022) On this basis, the Commission finds that the Transaction, as notified, would significantly impede effective competition in the internal market within the meaning of Article 2(3) of the Merger Regulation and Article 57 of the EEA Agreement.
11. COMMITMENTS
11.1. Introduction
(1023) The Notifying Party did not submit commitments during the Phase I investigation. (1024) In order to render the Transaction compatible with the internal market in relation to
the market for the production and supply of Aluminium ABS in the EEA, the
Notifying Party submitted commitments on 9 August 2019 (the ‘Commitments of 9 August 2019’), pursuant to Article 8(2) of the Merger Regulation.
(1025) The Commitments of 9 August 2019 were not market tested and the Commission provided feedback to the Notifying Party to that effect.
(1026) The Parties submitted revised commitments on 13 August 2019 (the ‘Commitments of 13 August 2019’).
(1027) While the Commitments of 13 August 2019 were submitted later than 65 working days after proceedings were initiated (that is on day 67), the Commission exceptionally decided to market test them on 13 August 2019, deeming them to be a substantial improvement compared to the Commitments of 9 August 2019.
(1028) On 28 August 2019, and following the market test, the Commission provided feedback to the Parties.
(1029) The Parties submitted revised commitments on 3 September 2019 (the ‘Final Commitments).
11.2. Analytical framework
(1030) The following principles from the Merger Regulation and the Commission’s Notice on Remedies acceptable under Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004 (‘Remedies Notice’)833 apply where parties to a concentration offer commitments with a view of rendering a concentration compatible with the internal market.
(1031) Where a concentration raises competition concerns in that it could significantly impede effective competition, in particular as a result of the creation or strengthening of a dominant position, the parties may seek to modify the concentration in order to resolve the competition concerns and thereby gain clearance of their concentration.834
(1032) The Commission only has power to accept commitments that are capable of rendering the concentration compatible with the internal market so that they will prevent a significant impediment to effective competition in all relevant markets where competition concerns were identified.835
(1033) To that end, the commitments have to eliminate the competition concerns entirely and have to be comprehensive and effective from all points of view.836 In assessing whether proposed commitments are likely to eliminate all competition concerns, the Commission considers all relevant factors including inter alia the type, scale and scope of the commitments, judged by reference to the structure and particular characteristics of the market in which those concerns arise, including the position of the parties and other participants on the market.837
(1034) Moreover, commitments must be capable of being implemented effectively within a short period of time.838 In case of implementation risks and implementation uncertainties for instance related to third party consents, it is incumbent on the parties to remove such uncertainties.839
(1035) Where a proposed concentration threatens to significantly impede effective competition, the most effective way to maintain effective competition, apart from prohibition of the concentration, is to create the conditions for the emergence of a new competitive entity or for the strengthening of existing competitors via divestitures by the merging parties.840
(1036) The divested activities must consist of a viable business that, if operated by a suitable purchaser, can compete effectively with the merged entity on a lasting basis and that is divested as a going concern. The business must include all the assets which contribute to its current operation or which are necessary to ensure its viability and competitiveness and all personnel which are currently employed or which are necessary to ensure the business’ viability and competitiveness.841
(1037) Personnel and assets which are currently shared between the business to be divested and other businesses of the parties, but which contribute to the operation of the business or which are necessary to ensure its viability and competitiveness, must also be included. Otherwise, the viability and competitiveness of the business to be divested would be endangered.842
(1038) Normally, a viable business is a business than can operate on a stand-alone basis, which means independently of the merging parties as regards the production and supply of input materials or other forms of cooperation other than during a transitory period.843
(1039) The business to be divested has to be viable as such. Therefore, the resources of a possible or even presumed future purchaser are not taken into account by the Commission at the stage of assessing the remedy. The situation is different if already during the procedure a sale and purchase agreement with a specific purchaser is concluded whose resources can be taken into account at the time of the assessment of the commitment844 (the so-called ‘fix-it-first’ remedy).
(1040) The intended effect of the divestiture will only be achieved if and once the business is transferred to a suitable purchaser in whose hands it will become an active competitive force in the market. The potential of a business to attract a suitable purchaser is an important element of the Commission’s assessment of the appropriateness of the proposed commitment.845
11.3. The Commitments of 9 August 2019
11.3.1. Description of the Commitments of 9 August 2019
(1041) The Notifying Party formally submitted commitments on 9 August 2019 which consisted of the divestiture of Aleris’ (i) Aluminium ABS Business, and (ii) Other FRPs Business at Aleris’ plant in Duffel (together the ‘Divestment Business of 9 August 2019’) to an independent purchaser with proven expertise in the aluminium and/or in the flat rolled products sector, subject to approval by the Commission. The Commitments of 9 August 2019 further include a commitment to, at the purchaser’s option, make available to the purchaser capex funding for three investment projects at the Duffel Plant (see recitals (1047)–(1049)).
(1042) The Divestment Business of 9 August 2019846 comprises all of Aleris’ Aluminium ABS and other FRP tangible assets located in Duffel, Belgium (the ‘Duffel Plant’). The Duffel Plant currently produces Aluminium ABS and a number of other FRPs (for example for compound tubes, building facades, floor heating and other applications).
(1043) The Divestment Business of 9 August 2019 further includes intangible assets such as intellectual property necessary to operate the Divestment Business of 9 August 2019 as well as licences, permits and authorisations for its benefit. It further comprises contracts, leases and commitments, including supply agreements as well as lists of customers and customer contracts and purchase orders. It also includes Personnel and key personnel necessary to operate and ensure the viability of the Divestment Business of 9 August 2019. All ABS-related EEA R&D activities also form part of the Divestment Business of 9 August 2019. The Divestment Business of 9 August 2019 does not include any assets, contracts or staff related to the production and sale of non-ABS FRPs by any Aleris’ and Novelis’ plants other than the Duffel Plant.
(1044) The Divestment Business of 9 August 2019 is an integrated business that covers the whole production chain from the casthouse upstream to rolling and finishing lines downstream. […].
(1045) The Divestment Business of 9 August 2019 does not include any of the assets located at other Aleriss plants than the Duffel Plant. Instead, to ensure that the current production flows can be maintained, the Commitments of 9 August 2019 provide, at the option of the purchaser and for a transitional period of up to three years, that the Merged Entity will enter into supply agreements on terms and conditions equivalent to those at present afforded to the Divestment Business of 9 August 2019 for (i) […], (ii) […] (iii) […]. These are to cover the volumes required by the Duffel Plant according to existing business plans or volume projections. Further, also at the option of the purchaser and for a transitional period of up to three years, the Merged Entity will enter into contractual agreements with the purchaser for the provision of services such as HR, order handling, sales handling, IT and logistics.
(1046) The Commitments of 9 August 2019 include three further contractual agreements between the Merged Entity and the purchaser. For three years, the Duffel Plant is to process intermediate aerospace and heat exchanger products for the Merged Entity, with the aerospace processing to be done at a reasonable cost basis. In addition, the Commitments of 9 August 2019, also for three years, include an agreement to ship scrap between the Duffel, Koblenz and Voerde plants according to existing business plans and projections.
(1047) In the Commitments of 9 August 2019, the Notifying Party further undertakes to make available at the purchaser’s option capex funding in an escrow account of the quantity sufficient for […].
(1048) The capex provision is conditional on the purchaser demonstrating that the capex will be used to fund the three investment projects and that it is used within three years. The Commission shall waive the capex commitment if the purchaser has the assets or access to the inputs that make the capex funding unnecessary, provided that this does not affect the viability and competitiveness of the Divestment Business of 9 August 2019.
(1049) The Commitments of 9 August 2019 do not specify the amount of the capex funding.
11.3.2. The Notifying Party’s Arguments
(1050) According to the Notifying Party,847 the Commitments of 9 August 2019 resolve any competition concerns by completely removing the overlap between Novelis and Aleris in the EEA market for the production and supply of Aluminium ABS and by enabling the timely entry of a new supplier or the significant expansion of an existing supplier in the Aluminium ABS market, through the acquisition of the Divestment Business of 9 August 2019. The Notifying Party submits that therefore the Commitments of 9 August 2019 are capable of rendering the Transaction compatible with the internal market.
(1051) The Divestment Business of 9 August 2019 includes all assets, contracts and personnel necessary to ensure the viability and competitiveness of the Aluminium ABS Business. The Notifying Party considers that the scope of the Divestment Business of 9 August 2019 is broader than required and proportionate to address the competition concerns because the Commitments of 9 August 2019 remove the EEA overlap entirely, and also include all assets, contracts and personnel necessary for the operation of the Other FRPs Business currently located in and served by the Duffel Plant, even though the Commission has not identified competition concerns in any FRP markets besides Aluminium ABS.
(1052) The Notifying Party submits that the Commitments of 9 August 2019 would completely remove the overlap in Aluminium ABS between Novelis and Aleris, both in terms of sales and capacity.848 Further the Notifying Party argues that the purchaser would be able to take over Aleris’ capabilities and market position in the EEA market for the production and supply of Aluminium ABS and exert the same competitive constraint that exists in the market pre-Transaction.849
(1053) The Notifying Party further submits that the Commitments of 9 August 2019 are comprehensive and effective because the Divestment Business of 9 August 2019 constitutes a viable and competitive Aluminium ABS business that is already active in the market.850 According to the Notifying Party this is supported by the fact that (i) the Divestment Business of 9 August 2019 is a comprehensive package in that it includes all necessary assets, (ii) the purchaser will have immediate access to customers, (iii) the purchaser will benefit from experienced staff, (iv) the purchaser will be perceived by customers as a reliable supplier, and (v) the Divestment Business of 9 August 2019 is an attractive business opportunity for the purchaser.851
(1054) With respect to the exclusion of production assets located at the Koblenz plant and at the Voerde plant from the Divestment Business of 9 August 2019, the Notifying Party submits that these plants currently supply limited quantities of input for Aluminium ABS production by the Duffel Plant, which are of limited importance and unnecessary to ensure the viability of the Aluminium ABS business.
11.3.3. The Commission’s Assessment of the Commitments of 9 August 2019
(1055) The Commitments of 9 August 2019 were clearly insufficient, independently of the potential outcome of a market test, to entirely and effectively solve the competition concerns identified by the Commission and therefore, the Commission decided not to market test the Commitments of 9 August 2019. In particular, the Commitments of 9 August 2019 would not have ensured that the Divestment Business of 9 August 2019 would have constituted a viable and competitive business that would have been independent of the Merged Entity. This is for the following reasons.
(1056) First, in line with paragraph 32 of the Remedies Notice (see recital (1038)), a viable business should be able to operate on a standalone basis, independently of the merging parties in particular as regards the supply of input materials or other forms of cooperation, other than during a transitory period.
(1057) The Commission notes in this respect that the Divestment Business of 9 August 2019 as such constitutes an integrated business that, at the Duffel Plant, covers the whole production chain from casthouses to finishing of the final products. Nonetheless, while the Duffel plant is capable of and in practice produces most of its upstream inputs (e.g. slabs and hot rolled coils), for a part of such inputs, the Duffel Plant relies on other Aleris plants.
(1058) Currently, the Duffel Plant receives three key inputs from other Aleris plants: […] . (1059) In terms of upstream production of slabs, the Duffel Plant has the capability and
capacity to cast the majority of the slab requirements ([…]% in 2018852) of its downstream operations. The remaining part is either sourced from […].
(1060) […].
(1061) […].
(1062) Therefore, the Commission notes that while the Duffel Plant is capable of producing internally the majority of its upstream input needs, it is at present not fully independent of other Aleris plants but relies on them for a not insignificant part of its upstream supplies. In addition to capacity limitations at the Duffel casthouse, the deficiencies relate to the ability to produce the widest input materials that the Duffel Plant uses in its production of Aluminium ABS.
(1063) […].
Figure 131 […]
[…]
(1064) In light of this, it is necessary to address the capability and capacity deficiencies of the Duffel Plant’s upstream supply equipment in order to ensure the independence of the Divestment Business of 9 August 2019 from the Merged Entity, and to guarantee its viability and competitiveness. This applies not only to the potential immediate transitionary period but going forward thereafter.
(1065) Second, to address the deficiencies in upstream supply at the Duffel Plant, Novelis committed, on the one hand, to provide transitionary supply agreements and, on the other hand, capex funding in an escrow to develop the upstream production equipment at the Duffel Plant.
(1066) As to transitional supply of the key input materials, Novelis committed to provide, at the request of the purchaser, transitional agreements for the supply of input materials the Duffel Plant currently receives from Aleris’ other EEA plants, […]. For a period of up to three years and at the option of the purchaser, these would continue to be supplied by the Merged Entity to the Divestment Business of 9 August 2019. As these would essentially cover everything that is currently supplied to the Duffel Plant from Aleris’ other plants and would maintain the present supply chains, the immediate needs of the Divestment Business of 9 August 2019 would be covered, including both in terms of volume supplies and ability to fulfil customer requirements.
(1067) Further, to replicate Aleris’ capabilities with respect to the upstream input materials by the end of the transitional supply agreements, capex funding in escrow was included, at the option of the purchaser, in the commitments in order to […] .
(1068) As explained in recitals (1055) to (1064), addressing the upstream supply question is required to ensure the independence of the Divestment Business from the Merged Entity and to guarantee its viability and competitiveness, also following any transitional period.
(1069) However, under the Commitments of 9 August 2019, the capex funding for the three investment projects at the Duffel Plant is made available by Novelis only ‘at the Purchaser’s option’. In the Commission's view, this results in significant uncertainty on the purchaser carrying out such investments and cutting ties with the Merged Entity. It could also not be excluded that the Notifying Party would be in a position to select potential purchasers primarily on the basis of them committing to not request the capex funding but continuing to rely on supplies from the Merged Entity.
(1070) Third, the capex funding for the three investment projects at the Duffel Plant is not quantified. This does not allow the Commission to assess the adequacy of the funding to be ultimately made available by Novelis for the three projects, including making it impossible for the Commission to verify the amounts through a market test. Further, in terms of implementation of this commitment, the amount to be placed in the escrow account would remain uncertain at the moment of the Commission’s decision.
(1071) Fourth, the proposed casthouse modification at the Duffel Plant under the capex funding only concerns the modification of the casthouse capability, but not of its capacity. […].
(1072) Enabling the Duffel […] is necessary to, on the one hand, ensure that the Divestment Business of 9 August 2019 is not reliant on the Merged Entity for such volumes, and on the other hand to ensure that the reliance of the Divestment Business of 9 August 2019 […] not increased compared to the pre-Transaction situation. […] the competitiveness of the Divestment Business of 9 August 2019, due to the double- marginalisation associated with the […] purchased from the merchant market.
(1073) Fifth, the tolling agreement for the processing of intermediate aerospace products included in the Divestment Business of 9 August 2019 principally for the benefit of the Merged Entity is foreseen on a reasonable cost basis. In order to ensure the viability of the Divestment Business of 9 August 2019, the Commission considers that the agreement should also include a margin for the Divestment Business of 9 August 2019, as this is common practice in the industry when services are provided to third parties on an arms-length basis.
11.3.4. Conclusion on the Commitments of 9 August 2019
(1074) On the basis of all the above considerations in this Section 11.3, the Commission concludes that the Commitments of 9 August 2019 are not comprehensive and effective from all points of view, and involve significant uncertainties as to the viability and competitiveness of the Divestment Business of 9 August 2019. The Commission therefore decided that the Commitments of 9 August 2019 could not be market tested.
11.4. The Commitments of 13 August 2019
11.4.1. Description of the Commitments of 13 August 2019
(1075) The Parties submitted revised commitments on 13 August 2019. These commitments again consist of the divestiture of Aleris’ (i) Aluminium ABS Business, and (ii) Other FRPs Business at Aleris’ Duffel Plant (together the ‘Divestment Business of 13 August 2019’ or the ‘Divestment Business’) to an independent purchaser with proven expertise in the aluminium and/or in the flat rolled products sector, subject to approval by the Commission.
(1076) The Divestment Business of 13 August 2019 includes the same tangible and intangible assets, as well as licences, permits and authorisations, supply and customer contracts as the Divestment Business of 9 August 2019. The Commitments of 13 August 2019 however include the following improvements.
(1077) First, the capex funding in an escrow account funding is not to be made available at the purchaser’s option but included in principle – though giving the Commission the possibility to remove the obligation if the purchaser candidate were to demonstrate that it either could itself provide the inputs or otherwise have independent access to them in a way that ensures the viability and competitiveness of the Divestment Business.
(1078) Second, the amount of the capex escrow is specified. In particular, the escrow is quantified as follows: […].
(1079) Third, this modification of the Duffel Plant casthouse explicitly includes a modification to enable it to accommodate […].
(1080) The Commitments of 13 August 2019, contrary to the Commitments of 9 August 2019, define the contractual agreements for the processing of intermediate aerospace and heat exchanger materials by the Divestment Business for the Merged Entity at cost plus reasonable margin basis.
11.4.2.The Notifying Party’s Arguments
(1081) The Notifying Party again submits853 that the Commitments of 13 August 2019 resolve any competition concerns by completely removing the overlap between Novelis and Aleris in the EEA market for the production and supply of Aluminium ABS and by enabling the timely entry of a new supplier or the significant expansion of an existing supplier in the Aluminium ABS market, through the acquisition of the Divestment Business of 13 August 2019.
(1082) The Notifying Party highlights that it commits to make available to the purchaser capex funding to fund the three investment projects detailed in recital (1077).854
(1083) The Notifying Party submits that therefore the Commitments of 13 August 2019 are capable of rendering the Transaction compatible with the internal market.855
11.4.3. Results of the market test on the Commitments of 13 August 2019
(1084) The results of the market test were overall positive. In particular, both customers and competitors considered that the Commitments of 13 August 2019 were in principle capable of addressing the competition concerns the Commission had identified in Aluminium ABS. A competitor to the Parties states that a ‘divestiture of the entire ABS business of Aleris […] should be sufficient to eliminate any Commission’s [sic] concerns with respect to ABS sold to automotive customers in the EEA’.856 A customer of the Parties concurs and states that ‘the divestiture of these assets can restore competition’ in the present case.857
(1085) A majority of respondents that expressed an opinion stated that they consider the Commitments of 13 August to be sufficient in scale and scope to ensure both the immediate viability and competitiveness, and the independence of the Divestment Business from the Parties.858
(1086) Further, a majority of the customers that expressed an opinion indicated that, under the Commitments of 13 August 2019, they would likely continue sourcing from the Divestment Business the current types of products and volumes going forward.859
(1087) Nonetheless, some market participants – though not all or a majority – noted that there were certain points where the Commitments of 13 August 2019 should be improved with a view of ensuring the viability and competitiveness of the Divestment Business. In particular respondents to the market test mentioned that the foreseen duration of the transitional supply agreements may not be sufficient, in particular in view of the time required to implement certain capex investments to install or modify assets at the Duffel Plant.860 Both customers and competitors further pointed to a need for the supply from the Merged Entity to the Divestment Business in the context of the transitional supply agreements to be prioritised over production and supply internal to the Merged Entity.861 Respondents further noted a need for limited clarifications, including for example R&D assets to be transferred to the Divestment Business.
11.4.4. The Commission’s Assessment of the Commitments of 13 August 2019
(1088) In assessing the suitability of the Commitments of 13 August 2019 in relation to Aluminium ABS, the Commission takes into account the following factors.
(1089) First, the Commitments of 13 August 2019 remove all overlap in Aluminium ABS brought about by the Transaction.
(1090) The Commitments of 13 August 2019 include the divestment of Aleris’ whole Aluminium ABS business in the EEA, including among others the Duffel Plant. The Duffel Plant is Aleris’ only EEA plant that has a CALP line and that supplies Aluminium ABS to customers in the EEA. The divestment would cover the full overlap both in terms of Aluminium ABS sales and capacity in the EEA that the Transaction would bring about.
(1091) Second, the Commitments of 13 August 2019 constitute an integrated business that is at inception already active at all relevant levels of the production chain from casting of slabs to rolling and downstream finishing of the final products. All of these functions are already at present perfomed at the Duffel Plant.
(1092) Third, as explained in recitals (1055) to (1064), the Duffel Plant relies on supplies from other Aleris plants for a minor part of its upstream supply needs. The Commission further recalls that addressing the upstream supply deficiencies is required to ensure the independence of the Divestment Business from the Merged Entity and to guarantee its viability and competitiveness.
(1093) As already included in the Commitments of 9 August 2019, the Commitments of 13 August include transitionary supply arrangements. As explained in recital (1066), the transitional supply arrangements would, at the option of the purchaser of the Divestment Business, essentially cover everything that is currently supplied to the Duffel Plant from Aleris’ other plants and would maintain the present supply chains. This would cover the immediate needs of the Divesment Business, including both in terms of volume supplies and also with a view of being able to fulfil customer requirements.
(1094) Nonetheless, as explained in recital (1084), it was raised in the market test that (i) the Commitments of 13 August 2019 are not adequately clear as to whether the Merged Entity would be clearly required to prioritise input supplies to the Divestment Business and that (ii) the transitionary agreements may be too limited in duration. In light of these, the Commission considers the following.
(1095) In the first place and with respect to the prioritisation question, the Commission recalls that the purpose of the commitments is to ensure that the Divestment Business can operate in the market as a viable and competitive force, also in competition with the Merged Entity. The Commission further notes that, the Divestment Business, under a new ownership, may be at its most vulnerable immediately after the divestment and before it has built its internal capacity to provide all upstream inputs. Therefore, it is necessary that the Merged Entity would prioritise supplies to the Divestment Business, for example in situations of production disruptions or capacity bottlenecks.
(1096) In the second place and with respect to the duration of the transitionary agreements, the Commision recalls that the purpose of the commitments is, on the one hand, to ensure that the Divestment Business has all necessary inputs required to be a viable and competitive market participant from the beginning and, on the other hand, that the Divestment Business is an independent actor not dependent on the Merged Entity. Therefore, the duration of the transitionary agreements is also a question of balancing between these two aims.
(1097) In this respect, the Commission recalls that the transitionary supply arrangements foreseen in the Commitments of 13 August 2019 have a duration of a maximum of three years after divestment, at the option of the purchaser. While many respondents in the market test indicated that three years is an adequate duration, some respondents noted that more time is needed taking into account not only the investment itself but the qualification requirements of customers.862 Therefore, in light of the replies to the market test and the specificities of the case, for example the fact that customers may require qualification of products and production chains, it should be made possible for the transitional agreements to be extended for a limited period at the request of the remedy purchaser. Nonetheless, in order to balance this extension with the aim of ensuring the independence of the Divestment Business, the extension should be made subject to guarantees that the extension is only triggered in duly substantiated situations.
(1098) Therefore, the Commission considers that the Commitments of 13 August 2019 in principle adequately address the need for transitionary supply arrangements, subject to (i) clarifying that the Merged Entity shall prioritise supplies to the Divestment Business over any other production and (ii) providing for the possibility of a limited extension to the duration of the transitionary agreements in duly substantiated situations.
(1099) Fourth, in order to ensure the long-term independence, viability and competitiveness of the Divestment Business, the Commitments of 13 August 2019 provide for capex funding in escrow to […].
(1100) In its assessment of the suitability and adequacy of the capex escrow commitment, the Commssion has in particular considered the following points.
(1101) In the first place, the Commission recalls that the Duffel Plant is already vertically integrated throughout the relevant production chain. At the downstream level, in the production of Aluminium ABS, the Duffel Plant is self-sufficient and already covers the entire overlap brought about by the Transaction. The same applies to the immediate input into the production of Aluminium ABS, cold rolled coils. Further, at the upstream levels (slab casting, scalping of slabs and hot rolling), the Duffel Plant also already produces internally the majority of its needs.
(1102) Therefore, the deficiencies are limited to a minority of the upstream input needs. For the majority of its operations, the Duffel Plant and the Divestment Business are self- sufficient to begin with and the capex escrow investment mainly aims at ensuring full independence of the Divestment Business from the Merged Entity.
(1103) In the second place, the capex escrow aims at expanding specific capabilities of the Duffel Plant at the upstream production levels. As the Duffel Plant already has production at each of the production chain steps concerned, it already has knowledge, capabilities and capacities in these fields. The capex escrow commitment does in this respect not aim at creating totally new capabilities but rather providing limited further capacity and the ability to supply wide upstream inputs.
(1104) In particular, […]. (1105) […].
(1106) In the third place, unlike the Commitments of 9 August 2019, the Commitments of 13 August 2019 provide that the capex escrow is in principle included in the Divestment Business and is no longer provided at the option of the remedy purchaser. The Commission considers that this better ensures that the remedy purchaser is incentivised to make the investments required, as they are (i) in any event included in the divestment package, (ii) likely accounted for in the valuation of the Divestment Business when sold and (iii) the purchaser can only draw the escrow, for which they have likely already paid for, if they make the said investments.
(1107) This finding is not inconsistent with a clause in the Commitments of 13 August 2019 that provides that the Commission can waive this requirement if the remedy purchaser demonstrates that it can either supply the necessary upstream inputs internally or otherwise has adequate access to them in a way that does not jeopardise the viability and competitiveness of the Divestment Business.
(1108) In the fourth place, […].863
(1109) […].
Figure 132 […]
[…]
(1110) In the fifth place, the Commission observes that the Commitments of 13 August 2019 provide a purchaser criteria whereby the potential purchaser of the Commitments of 13 August 2019 shall be an industrial actor. The Commission considers that this provides additional certainty that the purchaser is likely to be able to undertake the capex investments successfully.
(1111) In the sixth place, unlike the Commitments of 9 August 2019, the Commitments of 13 August 2019 quantify the capex escrow funding. […]. The replies to the market test do not credibly call the figures into question.864
(1112) In the seventh place, the results of the market test do not call the efficiency and suitability of the capex arrangement into question.865 While the majority of respondents – in particular customers – considered themselved not to have adequate information to reply to the question, the results of the market test did not reveal any fundamental problems in the concept.866 Moreover, as explained in recital (1086), the majority of customers taking a position considered that they could continue purchasing from the Divestment Business under the arrangments foreseen in the Commitments of 13 August 2019.
(1113) Nonetheless, as explained in recital (1084), the Commission notes that in the market test it was indicated that the period, three years, in which the capex funds would need to be drawn may be too short.
(1114) In this respect, the Commision recalls that the purpose of the commitments is that the Divestment Business is independent of the Merged Entity and that it can operate viably and competitively in the market, also in competition with the Merged Entity. Further, the Commission recalls that, as explained in recital (1096), the aim of the capex escrow is to provide funding for the Divestment Business to ensure its complete independence. From this perspective, the investments for which the capex is provided should take place as soon as possible after the divestment takes place.
(1115) Therefore, it is not appropriate to extend the period unnecessarily. However, in light of the comments in the market test, it is in order to provide some flexibility in case of investments that have been initiated in due course but that cannot have finalised by the end of the three-year period.
(1116) In light of the above, the Commission considers that the Commitments of 13 August 2019 in principle adequately address the need for ensuring the complete independence of the Divestment Business going forward through the provision of a limited capex escrow arrangement aimed at ensuring the full independence through investments in limited upstream production capabilities and capacities, subject to providing flexibility in the capex use period in case of investments that have been initiated in due course but that cannot have finalised by the end of the three-year period.
(1117) Fifth, as explained in Section 11.4.3, the results of the market test were overall positive. Nonetheless, as explained in that Section, market participants raised some limited comments discussed therein and in this Section 11.4.4.
(1118) Sixth, the Commission identified, also in light of the results of the market test, certain technical clarification needs in the commitments text.
11.4.5. Conclusion on the Commitments of 13 August 2019
(1119) Considering all evidence available to it, and for the reasons explained in Section 11.4.4, the Commission concludes that the Commitments of 13 August 2019 are not adequate and suitable to fully remove the competition concerns the Commission has identified. This is in particular due to considerations related to (i) the duration of the transitionary agreements, (ii) prioritisation of supplies to the Divestment Business over other production by the Merged Entity when it comes to the transitionary supply agreements, (iii) the duration of the capex use period and (iv) the need to provide certain technical clarifications in the commitments text.
11.5. The Final Commitments
(1120) The Parties submitted revised commitments on 3 September 2019 (the ‘Final Commitments’). The Final Commitments include a limited number of changes compared to the Commitments of 13 August 2019, in order to, following the replies to the market test, remove from the Commitments of 13 August 2019 the remaining uncertainties with respect to the Divestment Business’ viability and competitiveness.
(1121) For the following reasons, the Commisssion considers that the Final Commitments adequately address the issues identified in the Commitments of 13 August 2019.
(1122) First, the supply agreements for the supply of […], at the option of the purchaser and for a transitional period of up to three years, can be extended for up to two additional years in duly justified situations and subject to a substantiated request by the remedy purchaser and the opinion of the Monitoring Trustee. During this extension, the supply from the Merged Entity to the Divestment Business will be on a cost plus reasonable margin basis.
(1123) The Commission considers that this will ensure that the Divestment Business has sufficient time to replicate the respective inputs (in particular by installing or modifying assets at the Duffel Plant with the capex funding) and to achieve customer qualification for these new or modified assets where needed. At the same time, it adequately balances with the need to ensure the independence of the Divestment Business from the Merged Entity by limiting the extension to two years, subjecting the extension to conditions and providing it at terms (cost plus reasonable margin) that are less advantageous to the remedy purchaser than those during the initial three- year term (at cost), thereby removing the possibility and incentives for the remedy purchaser to unduly prolong the transitionary supply arrangements.
(1124) Second, in the context of these transitional supply agreements, the supply to the Divestment Business will be prioritised over the supply to the Merged Entity. The Commission considers that this provides additional assurance that the supply from the Merged Entity to the Divestment Business will be maintained, also in cases where there are production irregularities or a conflict between the Merged Entity’s internal needs and the supply obligation to the Divestment Business.
(1125) Third, the capex use period is provided with flexibility whereby the drawing of funds is in certain situations possible even after the three-year period in case of investments that have duly been initiated within the three-year period.
(1126) Fourth, a number of clarification changes were made to the text of the Commitments of 13 August 2019.
11.6. Conclusion on commitments
(1127) For those reasons, the Commission considers that the Final Commitments are suitable and sufficient to eliminate the significant impediment to effective competition to which the Transaction would give rise and the Final Commitments therefore render it compatible with the internal market and the EEA Agreement.
12. CONDITIONS AND OBLIGATIONS
(1128) Pursuant to the second subparagraph of Article 8(2) of the Merger Regulation, the Commission may attach to its decision conditions and obligations intended to ensure that the undertakings concerned comply with the commitments they have entered into vis-à-vis the Commission with a view to rendering the concentration compatible with the internal market.
(1129) The fulfilment of the measure that gives rise to the structural change of the market is a condition, whereas the implementing steps which are necessary to achieve this result are generally obligations on the parties. Where a condition is not fulfilled, the Commission’s decision declaring the concentration compatible with the internal market is no longer applicable. Where the undertakings concerned commit a breach of an obligation, the Commission may revoke the clearance decision in accordance with Article 8(6) of the Merger Regulation. The undertakings concerned may also be subject to fines and periodic penalty payments under Articles 14(2) and 15(1) of the Merger Regulation.
(1130) In accordance with the basic distinction described in Recital (1129) as regards conditions and obligations, this Decision should be made conditional on the full compliance by the Notifying Party with the Section B (including the Schedule and Annexes to the Schedule) and the Final Commitments and all other Sections should be obligations within the meaning of Article 8(2) of the Merger Regulation. The full text of the Final Commitments is attached as an Annex 2 to this Decision and forms an integral part thereof.
HAS ADOPTED THIS DECISION:
Article 1
The notified operation whereby Novelis acquires sole control of Aleris within the meaning of Article 3(1)(b) of the Merger Regulation is hereby declared compatible with the internal market and the EEA Agreement.
Article 2
Article 1 is subject to compliance with the conditions set out in Section B of Annex 2.
Article 3
Novelis shall comply with the obligations set out in the Sections of Annex 2 not referred to in Article 2.
Article 4
This Decision is addressed to:
Novelis Inc.
3560 Lenox Road, Suite 2000
Atlanta, Georgia,
United States of America
Done at Brussels, 1.10.2019
ANNEX I: QUANTITATIVE ANALYSIS
1. INTRODUCTION
(1) This Annex to the Decision focuses on the economic analysis of the Transaction in the market for Aluminium Automotive Body Sheets (‘Aluminium ABS’).
(2) As presented in the main body of the Decision1, the Commission considers that in a basic industry characterised by capacity constraints unilateral effects may arise through at least two channels. First, given the available capacities in the market, a single Merged Entity may compete less aggressively on price post-merger. While before the Transaction the two independent merging parties did not care about cannibalising each other’s sales through price competition, this changes after the Transaction, once the combined capacity is controlled by a single entity. Moreover, after the Transaction the rivals of the Merged Entity control less capacity than the rival capacity faced by the Merged Entity pre-Transaction. Second, merging producers may also compete less aggressively on capacity expansions post-Transaction, as they will take account of the negative effect that new capacity in the market has on the sales of the respective merging partner. Otherwise stated, pre-Transaction each merging party only took into account the negative impact that new capacity would have on its own sales (via the decrease in overall market price due to the additional capacity) but did not take into account the negative effect on the sales of the other merging party. This is internalised after the Transaction.
(3) Whether these theories of harm apply in a concrete case depends significantly on two important factors: (i) the level of the Parties’ market position (as characterised by their market shares and capacity shares)2 and (ii) the extent of spare capacities in the market (in particular those held by rivals3).
(4) All else equal, anti-competitive effects are more likely if merging parties control a large part of the market after the transaction. This is because the effect of the internalisation of price and capacity competition between merging suppliers is particularly large if the merging parties represent a large part of the market. Conversely, if the merging parties are only minor players in a market, then a transaction would be unlikely to lead to significant anti-competitive effects, since the parties’ impact on market prices and capacity levels would then be too small to cause an appreciable effect.
(5) Moreover, anti-competitive effects are also more likely to arise when the extent of spare capacities in the market (in particular those held by rivals) is moderate relative to projected demand. In particular, the Merged Entity will possess appreciable market power post- Transaction if its own supply is necessary to ensure that the entire market demand is served. Instead, in particular where rivals’ capacity in the market is significantly larger than projected demand (so that rivals’ capacities can cover the sales of the Merged Entity), rival firms may continue to exert significant competitive pressure on the Merged Entity post-transaction, in an effort to fill their under-utilised production facilities.4
(6) In the current Transaction, the Parties will hold a large share of the market and projections of demand and capacity suggest that rivals will not have enough capacity to defeat unilateral effects.
(7) In support of the theories of harm presented above, this Annex presents:
(a) An analysis of the quantitative evidence available for product market definition: The evidence supports the view that Aluminium ABS is a separate product market from steel.
(b) An analysis of the sales, production and capacity data provided by the Notifying Party:5 The evidence suggests that the Parties have a very high combined market share and a very high combined capacity share, with a significant increment generated by the Transaction.
(c) The results of the market reconstruction work undertaken by the Commission using data from all major suppliers of Aluminium ABS:6 The evidence suggests that the Parties have an very high combined market share and a very high combined capacity share, with a significant increment generated by the Transaction. The data from the market reconstruction also shows that the figures on sales and capacity provided by the Notifying Party for the upcoming years tend to under-estimate the combined position of the Merged Entity.
(d) The results of the Commission’s pivotality analysis: This analysis looks at the balance between supply (capacity) and demand in the market for Aluminium ABS and shows that rivals do not have enough capacity to offset any price increase generated by the Transaction, as the Parties are pivotal for the entire market demand to be served.
(e) An economic analysis of the impact of the Transaction on capacity competition: Economic theory and the evidence available from internal documents support the view that the Transaction is not only likely to stifle direct price competition, but will also lead to significant unilateral effects relating to the reduction of the Parties’ incentives to expand capacity.
(f) An assessment of the bidding analyses submitted by Novelis and Aleris (win-loss analysis and econometric analysis):7 Contrary to the Notifying Party’s claims, the available evidence does not suggest that the Parties are distant competitors and do not exert significant competitive constraints on each other.
(8) This Annex also contains an assessment of economic arguments presented by the Notifying Party in the Form CO,8 the Reply to the Article 6(1)(c) Decision,9 the Reply to the SO10, and the Reply to the Letter of Facts.11
(9) The analysis presented in this Annex is based on the latest version of the data received from the Notifying Party. With respect to sales and capacity data, the Commission refers to the data packages submitted to the Commission in response to request for information 36. With respect to bidding data, the Commission refers to the data underlying the CRA submission of 19 May 2019.
2. ALUMINIUM ABS
2.1. Product market definition
2.1.1. Product and price differences
(10) First, as discussed in detail in Section 6.2.2.3 of the Decision, the Commission considers that product and price characteristics support the conclusion that steel does not belong to the same relevant product market as Aluminium ABS. There is no supply side substitutability between Aluminium ABS and steel products used in automotive bodies and the patterns of supply and conditions of competition are in general different.
(11) Second, as also discussed in Section 6.2.2.3 of the Decision, for several OEMs the choice of Aluminium ABS over steel is not primarily driven by considerations of price, but rather by an increasingly stringent regulation of CO2 emissions. Accordingly, past switching from steel to Aluminium ABS (as well as the significant future expected switching12) has been mostly due to non-price factors such as emissions regulation. As emissions regulations become more stringent, the demand for aluminium becomes increasingly rigid (that is less response to prices) due to the fact that OEMs’ demand for aluminium is generated by a regulatory requirement as opposed to how competitive the price of Aluminium ABS is relative to steel. A hypothetical monopolist over Aluminium ABS would therefore have very substantial market power, since customers are in large part dependent on Aluminium ABS for reasons other than price.
(12) The idea that OEMs’ purchases of Aluminium ABS are mainly driven by regulatory considerations as opposed to commercial considerations is also consistent with the fact that the share of Aluminium ABS in a car has been historically very low and its share is projected to remain low in the future.13 This is because OEMs would prefer to use steel, which is and has constantly been much cheaper than Aluminium ABS, but increasingly stringent regulations require them to increase the portion of the car that is based on a lighter (and significantly more costly) material.
Figure 1: […]14
[…]
(13) Third, and consistent with the point above, the Commission notes that there is a very significant difference between the price of steel and the price of Aluminium ABS, with Aluminium ABS being around […] more expensive than steel based on total price (see Table 1).15
many other cases), substantial differences in price for the same use reflect limited exchangeability on the part of customers.
(20) In conclusion, the Commission considers that product and price characteristics support the conclusion that steel does not belong to the relevant market for Aluminium ABS.
2.1.2. SSNIP test
(21) In addition to the evidence on product characteristics and pricing, the Commission has also considered whether the quantitative evidence contained in Novelis’ bidding data and internal documents might support the Notifying Party’s claim that steel belongs to the same product market as Aluminium ABS. This is not the case. Instead, the available quantitative evidence supports the view that steel is not in the same product market as Aluminium ABS.
(22) Specifically, the Commission has used Novelis’ data on margins and information on how demand for Aluminium ABS responds to changes in the price of steel and Aluminium ABS to assess whether a hypothetical monopolist over Aluminium ABS would be able to profitably raise the price of Aluminium ABS by a small but significant amount, or whether competition exerted by steel would make such a price increase impossible (the ‘SSNIP test’).
(23) Two standard ways of assessing the SSNIP test in antitrust practice are (i) the critical loss analysis using elasticities of demand and (ii) critical loss analysis using recapture ratios. In what follows, the Commission has considered both approaches.
2.1.2.1. Critical loss analysis using elasticities of demand
(24) A SSNIP test assesses whether a hypothetical monopolist in the candidate market (here: Aluminium ABS) would be capable of profitably raising prices by a small but significant non- transitory amount. The Notifying Party’s contends that steel is in the same market as aluminium and therefore the competitive constraint exercised by steel would stop a hypothetical monopolist from raising prices of Aluminium ABS.20
(25) In this section, the Commission first considers a computation of the SSNIP test based on critical elasticities of demand. This approach calculates the elasticity of demand beyond which the market for Aluminium ABS would be too narrow to constitute a relevant market. This critical elasticity is then compared to an estimate of the actual elasticity of demand of Aluminium ABS to consider whether additional products, such as steel, have to be included in the relevant product market.21
(26) In antitrust practice, two variants of the SSNIP test have been used, depending on whether the hypothetical monopolist could profitably raise prices by a SSNIP (the so-called ‘breakeven’ version of the test) or whether it would engage in such a price increase (the so-called ‘profit maximisation’ version of the test).22 In what follows, the Commission considers both variants.
(27) With linear demand, a candidate market such as Aluminium ABS would be defined too narrowly under the ‘breakeven’ version of the SSNIP if and only if the elasticity of demand of the candidate market satisfies
(33) The Commission has compared the above threshold elasticities to estimates of the elasticity of Aluminium ABS based on data from Novelis’ internal documents. Specifically, Figure 3 shows an internal estimate of Novelis of how the prospective demand for Aluminium ABS would change in response to a shift in the relative prices of steel and Aluminium ABS.31 Concretely, Novelis considers a […]. Using these figures, the Commission has calculated the elasticity of demand for Aluminium ABS that is implicit in Novelis’ estimates as […].32 In other words, Novelis’ own predictions about demand sensitivity are substantially below the threshold elasticities displayed in Tables 2 and 3. On the basis of Novelis’ own data, the SSNIP test therefore suggests that steel is not part of the relevant market for Aluminium ABS. Instead, as argued by the Commission, a hypothetical monopolist on Aluminium ABS would be in a position to profitably raise prices of Aluminium ABS by a small but significant non-transitory amount.
Figure 3: […]33
[…]
2.1.2.2. Critical loss analysis using recapture ratios
(34) To crosscheck the result of the SSNIP test, the Commission has also considered the second commonly used methodology for quantitative market definition, so-called critical loss analysis using recapture ratios.34 Again, there is a ‘breakeven’ and a ‘profit maximization’ variant of the test.
(35) With linear demand, critical loss analysis predicts that a hypothetical monopolist would not be capable of profitably raising prices in the candidate market by a SSNIP35 if and only if
where d𝑀 denotes the so-called ‘recapture ratio’ of the candidate market, whereas 𝑚 and 𝑋 continue to denote the variable cost margin and the SSNIP as before.
(36) Similarly, in the ‘profit maximization’ variant of the test, a hypothetical monopolist would not find it in its interest to raise prices by a SSNIP if and only if
(37) As explained in the U.S. Horizontal Merger Guidelines, the ‘recapture ratio’ d𝑀 of a candidate market measures ‘the percentage of sales lost by one product in the candidate market, when its price alone rises, that is recaptured by other products in the candidate market’.36 It is useful to illustrate this concept with an example. For example, suppose that a producer of Aluminium ABS (say, Novelis) unilaterally raises its prices. Because of this price increase, Novelis will lose some of its customers, who will switch to competing producers (either to rival Aluminium ABS producers or, potentially, to producers of steel). For example, if 70% of switching customers substitute towards other Aluminium ABS producers, whereas 30% of switchers
competitive interaction between steel and Aluminium ABS. Indeed, producers of Aluminium ABS do try to divert a larger proportion of car parts from steel towards Aluminium ABS. However, as the above figures show, delineating a product market that is wider than Aluminium ABS would require an implausibly low substitution between competing producers of Aluminium ABS and an implausibly high substitution between producers of Aluminium ABS and steel. The Commission therefore concludes that, also from the perspective of critical loss analysis using recapture ratios, Aluminium ABS does not appear to be an overly narrow product market definition, as alleged by the Notifying Party. On the contrary, the Commission’s analysis suggests that a hypothetical monopolist over Aluminium ABS could profitably increase prices of Aluminium ABS by 5-10%.
2.1.3. The Commission's response to the Notifying Party’s arguments
(43) The following sub-sections summarise the critiques raised by the Notifying Party in relation to the Commission’s SSNIP test. In these sub-sections, the Commission discusses each of these critiques and shows that they are either unwarranted or do not undermine the robustness of the Commission’s conclusion that steel does not belong to the market for Aluminium ABS.
2.1.3.1. Using the economic margins proposed by the Notifying Party, even the elasticity estimated by the Notifying Party suggests that Aluminium ABS is the relevant market
(44) When presenting the critical demand elasticities beyond which steel would be in the same relevant market as Aluminium ABS, the Commission conservatively presented a range of values based on several scenarios: (1) 5% or 10% SSNIP, (2) profitable or profit-maximising version of the SSNIP test, (3) gross margin or EBITDA margin.40 The average critical elasticity based on these scenarios is […], as discussed in Section 2.1.2.1.
(45) In the Reply to the quantitative part of the Letter of Facts, the Notifying Party claims that (1) ‘since substitution between aluminium and steel ABS occurs over a longer time horizon it is more relevant to look at EBITDA rather than shorter term gross margins’ and (2) that the Commission should have used the average industry margins for Aluminium ABS and by using […].
(46) The Commission considers that lower comments, which advocate for the use of lower margins compared to those used by the Commission, strengthen the conclusion that steel does not belong to the same market as Aluminium ABS.
(47) As an illustration, the Commission notes that if only the EBITDA margin is considered (as opposed to basing the calculations also on the gross margin), based on Table 2 and Table 3 the elasticity of demand for Aluminium ABS ranges between […] and […], with an average of […].41 Under the assumption that the EBITDA margin is the relevant economic margin in this case, since ‘substitution between aluminium and steel ABS occurs over a longer time horizon’, even the implausibly high elasticity of […] estimated by the Notifying Party42 would fail to support the conclusion that steel is in the same market as Aluminium ABS.43
(48) Therefore, the Commission considers that under the assumption that the relevant economic margins in this case are the EBITDA ones, both the Notifying Party’s and the Commission’s elasticity estimates support the view that steel does not belong to the same market as
when using the EBITDA margin as advocated by the Notifying Party, the average critical elasticity is […].58 Therefore, according to the calculations of the Notifying Party, for steel to be in the market for Aluminium ABS the price increase would have to start in 2023.
2.1.3.3. The Notifying Party’s elasticity estimates are unrealistic and inconsistent with the available evidence on substitution between Aluminium ABS and steel
(65) Independently of the methodological points discussed in Section 2.1.3.4, the Commission considers that the Notifying Party’s elasticity estimates are unrealistic and inconsistent with the available evidences on substitution between Aluminium ABS and steel.
(66) As noted in Section 2.1.2.2, the group elasticity of a candidate market towards outside goods is directly related to the recapture ratio of that market.59 In the present transaction, the recapture ratio of Aluminium ABS (d𝑀) measures the proportion of switchers that would substitute to other Aluminium ABS producers rather than steel in response to a unilateral price increase of their Aluminium ABS supplier. Conversely, the diversion to steel 1 − d𝑀 measures the proportion of customers that would switch to a steel producer rather than other Aluminium ABS producers.
(67) The recapture ratio of Aluminium ABS is therefore a measure of closeness of competition between Aluminium ABS and steel. If Aluminium ABS producers mostly compete with other Aluminium ABS producers, then one will find a high recapture ratio of Aluminium ABS (high d𝑀) and a low diversion to steel. Conversely, if Aluminium ABS producers compete more closely with steel producers than with other Aluminium ABS producers, then one would find a low recapture ratio of Aluminium ABS (low d𝑀) and a high diversion to steel.
(68) As explained in Section 2.1.2.2, the elasticity of Aluminium ABS e is related to the recapture ratio of Aluminium ABS through the formula d𝑀 = 1 − 𝑚e, where 𝑚 denotes the margin of Aluminium ABS production. This relation implies that the higher the elasticity of Aluminium ABS, the lower the recapture ratio of Aluminium ABS. This is intuitive, as it simply means that the stronger the competitive pressure exerted by steel, the higher will be the proportion of OEMs that would switch to steel rather than alternative Aluminium ABS if their Aluminium ABS supplier raises prices.
(69) Using the above formula, one can then verify, as a sensibility check, which recapture ratios a given elasticity estimate would imply. In the SO, the Commission showed that its elasticity estimate of […] implies a recapture ratio of Aluminium ABS of […].60 Accordingly, the diversion to steel is […]. This is a sensible estimate, as it implies, as one would expect, that Aluminium ABS producers more closely compete with other Aluminium ABS producers than with steel producers, especially in light of the reasons set out below.
(70) By contrast, CRA’s mean elasticity estimate of […] implies a recapture ratio of Aluminium ABS of […] and thus a diversion to steel of […]. The Notifying Party’s elasticity estimate therefore implies that Aluminium ABS producers compete at least as much, if not significantly more, with steel producers than with other Aluminium ABS producers. This implication, however, is grossly at odds with the facts of the case, corroborating the conclusion that the Notifying Party’s elasticity estimates are inflated and unreliable.
(71) First, not even the Notifying Party itself denies that Aluminium ABS manufacturers are closer competitors to each other than they are to steel producers. Rather, the Notifying Party argues that Aluminium ABS producers are also constrained by steel. After all, Aluminium ABS producers compete directly with each other, offering the same material in ‘Aluminium ABS only’ competitions. […].
(72) […].61 […].62
(73) Second, the strong demand for Aluminium ABS is not mainly driven by considerations of relative pricing between steel and Aluminium ABS, but by the impact of environmental regulations. Accordingly, […].63 […].64 […].65 66
(74) Third, according to Novelis’ own assessments, the competitive conditions among Aluminium ABS producers have a substantial influence on prices. For instance:
(a) […].67
(b) […]68 […].69 […].70
(75) The views of other market participants also reflect the fact that the number of competing Aluminium ABS rivals matters (‘Even today, the level of competition is noticeably lower in supply situations with only 2 to 3 [Aluminium ABS] suppliers’).71
(76) If steel producers were as close competitors to Aluminium ABS producers as other Aluminium ABS competitors, one would not expect any of this. Because if the Notifying Party’s proposed market delineation of steel plus Aluminium ABS was correct, no Aluminium ABS producer would hold a market share of more than 2%.72 The number and capacity of such minor competitors could then not have more than a trivial impact on market prices. According to Novelis’ key executives, however, the opposite is the case.73
(77) Fourth, contrary to the Notifying Party’s claims, […].74 […].75 […]. Steel makers, by contrast, have suffered from low margins during the same period.76 If steel were indeed an imminent constraint on Aluminium ABS, one would expect arbitrage between Aluminium ABS and steel to prevent Aluminium ABS producers from charging significant mark-ups. This, however, is not the case.
(78) In light of the above, the Commission maintains its conclusion that the available evidence strongly suggests that a hypothetical monopolist of Aluminium ABS would be able to profitably increase Aluminium ABS prices by a small but significant amount. The Notifying Party’s estimate of the elasticity of Aluminium ABS, by contrast, is inflated, implausible and inconsistent with the facts of the case.
2.1.3.4. […]
(79) […].77 […]78 […]. (80) […].
(81) Second, the Notifying Party claims that the Commission’s analysis ignores significant market developments that occurred since 2017 (the document used by the Commission was produced in October 2017). In the Reply to the quantitative part of the Letter of Facts, the Notifying Party argues that significant switching back to steel has taken place.79 First, the Commission notes that the lower demand the Notifying Party refers to is not necessarily ‘lower demand for Aluminium ABS compared to the previous year due to switching to steel’ but simply a forecasting error, that is, ‘lower demand growth for Aluminium ABS compared to previous expectations’. While the Notifying Party labels as ‘switching back to steel’ these lower demand forecasts, the Commission notes that lower Aluminium ABS demand than previously expected could be due to a simple delay in the process of switching from steel to Aluminium ABS or due to lower car production forecasts.80 Second, the Commission notes that while the Notifying Party provided a few examples of instances in which components have been switched from Aluminium ABS to steel, it has not provided any evidence that for a given OEM at fleet level there has been in the past a net switch back to steel. Given that emissions regulation is at fleet level, it is plausible that an OEM may decide to switch for one car model some volumes back to steel and compensate by increasing the volumes of Aluminium ABS for another model.
(82) In relation to the point above on the Commission’s analysis being out of date, the Commission further notes that demand for Aluminium ABS is, if anything, likely to become more rigid (i.e. to have a lower elasticity) in the future, as discussed in more detail in Appendix II.
2.1.3.5. The Commission does not ignore competition from steel in its economic analysis
(83) In the Reply to Annex I of the SO and the Letter of Facts, the Notifying Party argues that even if one concludes that steel is not part of the relevant market, ‘the Commission should have still taken steel into account in its competitive assessment’ (i.e., as an out-of-market constraint). In particular, the Notifying Party asserts that the Commission ignores competition from steel when assessing market shares, bidding data and capacity competition.
(84) The Commission considers that this criticism is unfounded.
(85) First, with respect to the assessment of market shares, the Commission considers that it cannot attribute market shares to steel manufacturers when steel is not part of the relevant market.
(86) Of course, the Commission agrees that the competitive effects of a merger are not uniquely determined by the degree of concentration within the relevant market, but also by the elasticity of demand towards outside products. However, the purpose of market definition is precisely to determine the perimeter of products across which market shares have the most explanatory power, taking into account the elasticity towards other products. Therefore, if the competitive constraint exerted by steel were sufficient to prevent price increases in Aluminium ABS, then the elasticity of demand for Aluminium ABS would be large enough so that steel must be considered part of the relevant market. That, however, is not the case.
(87) Second, also with respect to the Notifying Party’s bidding analysis, the Commission disagrees with the assertion that it wrongly ignored competition from steel. The Notifying Party had submitted a bidding analysis (based on bidding data from Novelis and Aleris) to support its claim that competitive overlaps between the Parties are moderate, indicating a lack of closeness of competition between them. The SO rejected this conclusion, among other things because competitive overlaps are in fact not moderate according to the data and therefore not supporting of a view that Novelis and Aleris are distant competitors.81
(88) In the Reply to Annex I of the SO, the Notifying Party then argued that the Commission should have taken the possibility of switching to steel into account when drawing this conclusion. The Commission, however, cannot take alleged switches to steel into account that do not exist in the data the Notifying Party volunteered to submit to show lack of significant competitive interaction. Indeed, the submitted data only contains examples where the Parties won or lost competition against other Aluminium ABS producers.82 If anything, the Commission considers that the evidence from the bidding data reinforces the Commission’s claim that at the bidding stage there is no competitive interaction between Aluminium ABS and steel. Therefore, the Commission maintains its conclusion that (1) the Notifying Party’s bidding analyses fail to support the claim that the Parties do not exert a significant competitive constraint on each other and (2) no adjustments to the bidding analysis are required to account for the out of market effect of steel.
(89) Third, the Commission also rejects the claim that it ignored competition from steel in its analysis of capacity competition. Contrary to the Notifying Party’s assertion, the Commission does not claim that this Transaction will bring Aluminium ABS’ growth to a standstill. As emphasised repeatedly by the Commission, the market for Aluminium ABS benefits considerably from stringent regulatory obligations that force OEMs to meet increasingly more stringent emission targets.
(90) In the Commission’s view the suppliers of Aluminium ABS will continue to have an incentive to expand capacity to ensure that there is sufficient capacity in the market to be able to comfortably cover OEMs’ demand. However, this is far below the level of capacity that would be needed to obtain a competitive outcome on every tender. In this respect, the Commission considers that the Merged Entity’s incentive to engage in capacity competition will be less pronounced after the Transaction than it would be absent the Transaction. […].83 […].
(91) The Notifying Party also notes that the Transaction may actually increase the Merged Entity’s incentives to invest in capacity expansion as the merger entity is more likely to appropriate the benefits of capacity expansions. This would be because after the Transaction each of the Merging Parties internalises the fact that improving the supply-demand balance makes OEMs more likely to use Aluminium ABS as a material rather than steel. In this respect, as noted in the paragraph above, the Commission notes that in its view nothing changes after the Transaction compared to absent the Transaction in the sense that in both scenarios Aluminium ABS suppliers will likely have an incentive to ensure that the capacity in the market for Aluminium ABS is sufficient to cover the requirements of the OEMs. What the Transaction changes is how much larger the capacity in the Aluminium ABS market is compared to demand. In a well-functioning market, Aluminium ABS suppliers not only invest to serve the additional demand coming in the market but also invest in order to attract market share from rivals. This ‘duplication of capacity expansions’ is eliminated between the Parties after the Transaction.
(98) These market shares at more granular level suggest the following.
(99) First, Aleris' presence is very limited in the 5xxx segment. […].
(100) Second, both Parties’ share is particularly high ([60-70]% in 2018), […], for the 6xxx skin segment. As noted, this is a segment for which production is more complex […]. Conversely, the Parties’ rivals (for example, Hydro and AMAG) are projected to be comparatively less present in 6xxx skin than they are projected to be in 5xxx and 6xxx structure. Hydro, in particular, appears to be significantly more present in 5xxx than in 6xxx skin.
(101) Therefore, segment-level market shares suggest that the Parties are, if anything, particularly close competitors. Indeed, they both appear to have a comparative strength in providing the most sophisticated products […].
of this exercise is (i) to validate the market share estimates provided by the Notifying Party and (ii) to expand on the information provided by the Parties. In particular, the Commission has also calculated market shares based on value (in addition to volume) and based on the total forecasted future sales per supplier (as opposed to the nominated sales only).
(103) As discussed in Section 2.2.1, the most recent year with complete information on market shares is 2018. Whereas data on nominated sales is still reasonably complete up to at most 2020, projections on market shares beyond 2020 suffer from the fact that nominations for these years are still largely incomplete. When assessing the expected future market position of the Parties, considerable caution should therefore be applied when interpreting market shares calculated on the basis of volumes (or values) nominated to date.
(104) In view of the inevitable incompleteness of nominated sales data, the Commission has also assessed projected future market shares based on total forecasted sales volume (or value), as provided by each of the suppliers of Aluminium ABS. This approach provides an alternative measure of expected future sales. As other future projections, also this measure has limitations, however, as market shares based on suppliers’ expectations are by definition less certain than market shares based on actual realised sales.
(105) The Commission also notes that the market shares from its reconstruction are based on sales figures collected at an earlier point in time than the sales figures underlying the Parties’ own estimates of market shares provided in response to request for information 36 and presented in Section 2.2.1. This notwithstanding, both sets of market shares are based on internally consistent figures (that is, sales data from all market participants are collected or estimated at the same point in time).
(106) Table 10 to Table 13 present market shares based on nominated volumes, total forecasted volumes, nominated value and total forecasted value respectively. Based on the results of this market reconstruction, the Commission considers the following.
(107) First, with respect to nominated volumes, the market reconstruction confirms that the Parties’ combined market shares are very high and will remain very high for the foreseeable future. In the most recent complete year 2018, the Parties hold a combined share of [50-60]% overall, [40-50]% for 5xxx, [60-70]% for 6xxx, [60-70]% for 6xxx skin, [60-70]% for 6xxx structure. Even in 2023, the combined share of the Parties is still [40-50]% overall, [20-30]% for 5xxx87, [50-60]% for 6xxx, [50-60]% for 6xxx skin, [50-60]% for 6xxx structure. The market reconstruction, moreover, suggests that AMAG’s currently limited presence in the market is not expected to materially improve in the future (with shares ranging from 0-5% across all alloys in 2023). This contrasts with the estimates of the Parties in the Form CO, which place AMAG at a [5-10]% share in 2023 across alloys.
(108) Second, based on total forecasted volumes, the Parties’ future position is expected to be even stronger than the already high market shares based on nominated volumes suggest. More specifically, for 2023 the combined share of the Parties is expected to be [50-60]% overall, [30- 40]% for 5xxx88, [60-70]% for 6xxx, [60-70]% for 6xxx skin, [50-60]% for 6xxx structure. Moreover, also this part of the market reconstruction suggests that AMAG’s currently limited presence in the market is not expected to materially improve in the future, especially in the 6xxx skin segment (with shares ranging around 0-5%). Again, this contrasts with the estimates of the Parties in the Form CO, which place AMAG at a [5-10]% share in 2023 across alloys.
(109) Third, also the market reconstruction based on nominated value results in shares that are higher than the already very high volume shares provided by the Parties. Specifically, for 2018, the Parties’ combined is [60-70]% overall, [40-50]% for 5xxx, [70-80]% for 6xxx, [70-80]% for 6xxx skin, [60-70]% for 6xxx structure. For 2023, the Parties’ combined share is [40-50]% overall, [20-30]% for 5xxx89, [50-60]% for 6xxx, [50-60]% for 6xxx skin, [50-60]% for 6xxx structure. Given the particular competitive advantage of the Parties in […], value-based market shares are likely to be a particularly good measure of the competitive positioning of the Parties. Finally, the reconstruction based on nominated value also confirms the conclusion that AMAG’s currently limited presence in the market is not expected to materially improve in the future (with shares again ranging from 0-5% across alloys in 2023).
(110) Fourth, the market reconstruction based on total forecasted value suggests that the combined share of the Parties in the future could be even higher than suggested by the currently nominated sales. Specifically, for 2023, the Parties’ combined is [50-60]% overall, [30-40]% for 5xxx, [60-70]% for 6xxx, [70-80]% for 6xxx skin, [50-60]% for 6xxx structure).
2.2.3. The Commission's response to the Notifying Party’s arguments
(111) In response to the Article 6(1)(c) Decision, the Notifying Party made the following arguments. First, in a dynamic market such as Aluminium ABS with new entry and capacity expansions by rivals, market shares based on recent tenders better reflect the competitive dynamics than shares based on legacy volumes. The Parties therefore put forward a new methodology for calculating market shares in their the Response to the Article 6(1)(c) Decision. This methodology is based on the quotation year of tenders as opposed to the year of production. For instance, if Novelis won a tender in 2016 for the production of 300kt between 2018 and 2020, then according to the new methodology, the entire lifetime volumes of this project (300kt) are attributed to the quotation year (2016), instead of attributing them pro rata to the years of production (2018 to 2020).
(112) As shown in the below table, based on tenders with quotation date in 2016, 2017 and 2018, the share of Aleris (across all alloys) is respectively [20-30]%, [5-10]%, [0-5]%. The Parties argue that this shows that the actual competitive position of Aleris is not as significant as production- based market shares might seem to suggest.
(113) Second, although the Notifying Party does not contest the Commission’s concern that future nominated volumes only capture a fraction of total future demand,90 the Notifying Party considers that market shares based on nominated volumes are nonetheless a more accurate proxy of future shares than total forecasted sales, due to the unavoidably subjective nature of forecasts by the various suppliers of Aluminium ABS. Specifically, basing market shares on total forecasted sales by each OEM can lead to the following biases:
(a) Suppliers’ expected future sales are based on individual assumptions about future vehicle production volumes. As these assumptions may vary from supplier to supplier for a given program, aggregating these data to generate market shares therefore mixing different assumptions on vehicle production volumes.
(b) There may be double-counting of volumes if two suppliers expect to win the same parts of a forthcoming tender.
(114) With respect to the first point (the use of quotation-based shares instead of production shares), Commission considers the following.
(115) First, the market shares presented in the Article 6(1)(c) Decision, Statement of Objections and in this Decision are based on the methodology put forward by the Notifying Party itself in the Form CO.91 Only after the Commission expressed concerns about the competitive implications of the Transaction the Parties started considering a different way of calculating market shares.
change in the relative position of individual suppliers. In line with the Notifying Party’s own approach in the Form CO, market shares based on actual production are a more reliable source for assessing the competitive position of the Parties in a given year, as they are less affected by the cyclical swings of lumpy tender awards from year to year. Under this approach, the combined share of the Parties is very high and is expected to remain very high in the future (especially in the 6xxx alloys). Moreover, value-based market shares, which are particularly relevant in the current case due to the comparative advantage that the Parties appear to enjoy in high value applications, are even higher than volume shares.
(121) With respect to the Notifying Party’s second point (advantages of using nominated future sales compared expected future sales), as indicated in Section 2.2.2, the Commission considers that neither the Notifying Party’s approach (using nominated sales) nor the alternative approach (using total forecasted sales) is without limitations. For this reason, the Commission presents both sets of market shares in this Annex.
(122) More specifically with respect to the critique raised by the Notifying Party at points (a) and (b) of paragraph (113) above, the Commission considers that neither of these points implies that the shares based on total forecasted volumes or value introduce a systematic bias towards finding larger market shares for the Parties. For example, as regards (a), in case rivals’ expected future sales are based on assumptions about future vehicle production volumes that are more optimistic than the Notifying Party’s assumptions, market shares based on total forecasted volumes or value would in fact underestimate the position of the Parties.
(123) Since both methodologies may suffer from some imprecisions in predicting future outcomes, the Commission considers that it is sensible to consider the results of both methodologies to obtain as complete a picture as possible. The only fully complete market share figures are the shares of actual sales up to 2018, however, because they do not have to rely on extrapolations and estimations.
(124) In the Reply to Annex I of the SO95, the Notifying Party claims that Novelis’ sales […] distort the competitive assessment because taking these sales into account to derive Novelis’ share does not allow to reflect the actual competitive landscape since Constellium, Hydro, AMAG, Ma’aden96 and Profilglass compete on an equal footing with the Parties for each bidding opportunity. The Commission believes […] should be accounted for in the market share analysis. […].97 […].
(125) Finally, the Notifying Party claims that the high market shares reflect the Parties’ first mover advantage in Aluminium ABS supply. Focusing on market shares is therefore a poor proxy of the competitive interaction between the Parties going forward. In this respect the Commission notes that while the current combined market share of the Parties is very high, also for the future all market share measures (based on volumes or value, nominated or total forecasted, Form CO data or market reconstruction data) suggest that the Parties’ combined share will stabilise at a high level.
2.3. Market reconstruction: the choice of the most relevant capacity measure
(126) Before discussing the capacity shares (Section 2.4) and the supply-demand balance and pivotality analysis (Section 2.5), the Commission presents in this section the rationale for the main assumptions and scenarios presented in Section 2.4 and 2.5.
(127) The Commission considers that the most relevant measure of the amount of relevant capacity for Aluminium ABS customers in the EEA are the capacities from the Commission’s market reconstruction, adjusted to exclude CASH capacity currently used for exports outside the EEA98 and excluding CASH capacity currently used for non-ABS customers.
(128) The capacities from the market reconstruction are preferable to the capacity estimates provided by the Parties in the Form CO because the Commission collected data directly from each Aluminium ABS supplier.
(129) The Commission’s adjustments for CASH capacity used for exports outside the EEA and non- ABS customers are warranted because they reflect the fact that the Aluminium ABS suppliers use their CASH capacity installed in the EEA to serve also non-ABS customers and customers outside the EEA. Therefore, it cannot be assumed that the entirety of the plants currently serving the EEA are used for serving ABS customers in the EEA.
(130) The Commission considers that it has employed a very conservative approach that, if anything, overestimates the amount of capacity available for Aluminium ABS customers in the EEA for the following reasons:
(a) The Commission has used in its calculations the final ABS production capacity of Novelis and Aleris instead of the effective CASH capacity, as was argued by the Notifying Party.99 The Commission notes that this approach is favourable to the Notifying Party because it is likely that at least for some competitors the capacity figures provided to the Commission are based on effective capacity, which is generally higher than final ABS products capacity100.
(b) The data provided by each supplier in response to the market reconstruction indicates separately (i) the proportion of production sold outside the EEA and (ii) the proportion of production sold to non-ABS customers. While there is expected to be some overlap between non-EEA sales and non-ABS sales, it is not expected that suppliers only export non-ABS sales. In order to avoid any double-counting while being conservative, the Commission has therefore taken the most favourable view for the Parties and subtracted only the larger of the two figures from the overall CASH capacity available in the plants serving the EEA (even though the sum of non-EEA sales and non-ABS sales which should have been excluded is expected to be larger).101
(c) The Commission has conservatively allocated the entirety of the spare capacity of each supplier to be available for EEA automotive supplies, irrespective of whether the CASH capacity at the plant is currently used also for non-automotive or non-EEA sales. This is conservative, because in reality at least part of this spare capacity would likely also be used for non-automotive and non-EEA sales.
(d) At a given point in time, not all suppliers are qualified with all OEMs (or for all applications for a given OEM). Moreover, when new capacity has been installed, a new plant still has to be qualified by OEMs for each specific application. Nevertheless, the Commission has included all installed capacity in the analysis. This may results in an over-estimation of the options actually available individual OEMs.102
(e) The Commission assumes that the entirety of the CASH capacity available can be used to serve customers, while in reality it is common for industrial suppliers to leave a buffer of spare capacity to be able to address unexpected swings in volumes requests from their customers or faults in the production of certain batches.
(f) […].103
(131) […].104
(132) […].105 […].
(133) […].106
(134) In the Reply to the SO and the Reply to the Letter of Facts, the Notifying Party raised a number of objections to the approach adopted by the Commission. These objections are discussed and rebutted in the remainder of this section.
(135) In any event, since the main findings of the Commission are not materially affected by the numerous points raised by the Notifying Party, the Commission presents in this Annex107 results based on a number of alternative scenarios:
(a) ‘Form CO’: Capacity data based on the Parties’ estimates (as submitted with the Form CO);
(b) ‘Reconstruction EEA auto’: Capacity data from the Commission’s market reconstruction, with the exclusion of CASH capacity relating to sales to exports outside the EEA108 and sales to non-ABS customers. This is the scenario that the Commission considers as most appropriate to describe the competitive conditions in the European Aluminium ABS market;
(c) ‘Reconstruction’: Capacity data from the Commission’s market reconstruction, without any adjustments for exports outside the EEA or sales to non-ABS customers;
(d) ‘Reconstruction sensitivity’: Capacity data from the Commission’s market reconstruction, without any adjustment for exports or sales to non-ABS customers. Since one competitor sold over the period 2016–2018 a non-negligible amount of 5xxx alloys based on batch annealing capacity, in this sensitivity the Commission also added to each year between 2016 and 2023 an amount of batch annealing capacity equal to the maximum volumes of 5xxx alloys sold by this competitor over the period 2015-2018.
(136) The Commission stresses that this version ‘Reconstruction sensitivity’ is not a sensitivity to its analysis but merely a scenario combining the Notifying Party’s most extreme assumptions regarding capacity and demand. The Commission considers the gap between demand and capacity resulting from this version of the analysis to be wholly inconsistent with the evidence on the gap in supply and demand observed in the internal documents. 109
2.3.1. Sales to non-ABS customers
(137) In the SO Reply, the Notifying Party states that […].110
(138) The Commission maintains that CASH capacity dedicated non-ABS products cannot be considered to be readily available to meet demand from Aluminium ABS customers.
(139) […].
(140) The Commission further observes that a competitor to the Parties stated that each of its ‘CASH line[s] produces a mix of products of ABS, aerospace and other products’.111 This competitor also explains that ‘not fulfilling its commitments, particularly with automotive and aerospace customers would lead to seriously harming its credibility and losing business in the future years’. While the argument is particularly valid for automotive and aerospace customers, the Commission considers that the fact that rivals have been consistently using their CASH capacity also for non-ABS applications suggests that they would likely not seamlessly divert such capacity towards ABS in the event of a 10% price increase in ABS.
(141) More importantly, the Commission considers that although pre-merger margins have been higher for ABS than for non-ABS customers other than aerospace, Aluminium suppliers have, over the last 3 years, consistently used part of their CASH capacities to serve these lower margin customers rather than to bid more aggressively in ABS tenders (which are typically won at price levels implying a significant margin). It is unclear why such behaviour would change post-Transaction, in the event of a small but significant price increase by Novelis.
(142) The Commission finally notes that only the CASH capacity currently utilised for non-ABS applications is excluded, but that it assumes that all of the spare CASH capacity is available for ABS applications (as opposed to allocating part of it to more lucrative applications such as aerospace).
Figure 4: […]112
[…]
(143) The Notifying Party also takes issue with the way the Commission implemented the adjustment of CASH capacity to exclude capacity devoted to non-ABS products. In particular, the Notifying Party considers that the Commission methodology of assuming that a fixed share of capacity (equal to the average of the historical share during 2016-2018) is used for non-ABS carries an implicit assumption that sales of non-ABS products will grow by […]% between 2018 and 2023. In the Notifying Party’s view this is excessive, considering that the ABS market is expected to grow by […] during the same period.
(144) The Commission considers that its assumption that also the non-ABS market grows (albeit not as fast as the ABS market) is acceptable and in any event by no means likely to change the Commission's conclusions in case the ABS market grew slower than it is implicitly assumed by the Commission's calculations. As an illustration, the Commission notes that if one took the extreme assumption that the non-ABS market does not grow at all after 2018 (i.e. sales volumes to the non-ABS market remain at the 2018 levels), the Commission's non-ABS capacity calculations would lead to an overestimation of future sales volumes in the non-ABS market by the following: […].113 Given the limited size of these potentially imprecise predictions, even in the extreme scenario in which sales in the non-ABS market remained fixed at the 2018 level, the Commission's conclusions regarding the Parties’ combined capacity share and the pivotality of Novelis would not change.
(145) Last, the Notifying Party notes that for one competitor the ‘% of capacity used for non-ABS sales’ decreases after 2016 and reaches zero in 2018. The Notifying Party then argues that the Commission's methodology assuming that a constant % of production is devoted to non-ABS in future years (equal to the average % in 2016–2018) does not pick up this trend. As explained for the exports adjustment (see Section2.3.3) the Commission considers that three years cannot be taken as an indication of a particular trend in a lumpy market, where winning or losing a tender can materially affect these numbers. Taking the Notifying Party's suggestion would mean assuming that this competitor continues to sell no volumes in the non-ABS market across 2018-2023. There is no indication that the competitor is withdrawing from these markets. In any event, the Commission notes that the sales amounts to the non-ABS market involved in this claim by the Notifying Party are […]. Therefore, the Commission considers that no adjustments are needed in this respect.
(146) In light of the above, the Commission considers that the exclusion of sales relating to non-ABS customers is appropriate, that the methodology used by the Commission is robust and that the variations proposed by the Notifying Party are either unwarranted or immaterial.
(147) In any event, the Commission also reports a sensitivity scenario in which all CASH capacity is considered, with no adjustment for sales to non-ABS customers (scenario ‘Reconstruction’). This scenario provides an extreme sensitivity as explained in Section 3.
2.3.2. Batch annealing capacity by one competitor
(148) The Notifying Party claims that based on the material made available in the data room, one competitor appears to have sold in the past non-negligible volumes of 5xxx alloys by using batch annealing. Therefore, the Notifying Party submits that the maximum amount of 5xxx produced by batch annealing that this competitor has sold over the last 3 years should be added to the total industry capacity for Aluminium ABS.
(149) The Commission disagrees with this view for the following reasons:
(a) The evidence on the Commission’s file114 suggests that the quality of Aluminium ABS produced via batch annealing is low and not accepted by all OEMs. A majority of automotive customers responding to the market investigation disagree with the statement ‘that a CASH manufacturing line can be substituted by other manufacturing processes (for example batch annealing)’ while preserving quality, consistency, and cost- effectiveness.
(b) Novelis […].115 […].116 […].117
(150) Therefore, batch annealed 5xxx alloys cannot be considered as an adequate alternative to respective products produced on a CASH line. The Commission considers that its approach of excluding the capacity relating to batch annealing is appropriate.
(151) In any event, the Commission also reports a sensitivity scenario in which the maximum amount of 5xxx produced by batch annealing that has been successfully sold by this rival over the last 3 years is added to the total industry capacity for each year of the period 2019-2023 (scenario ‘Reconstruction sensitivity’). This scenario provides an extreme sensitivity as explained in Section 3.
2.3.3. Exports outside the EEA
(152) First, the Notifying Party claims that exports are opportunistic and given by the overcapacity present in Europe. Therefore, CASH capacity in the EEA should not be adjusted for exports.118
(153) The Commission considers that exports are not necessarily opportunistic but the result of a market strategy. Over the last three years (2016–2018), Aluminium ABS suppliers have been exporting to outside the EEA. […].119
(154) While one competitor indicated that once the ramp up of its US plant is fully operational it aims at using the EEA plants for the EEA only120, the other competitors have no plants outside the EEA and therefore any contracts won outside the EEA would need to be necessarily served from the EEA. Indeed, one of these competitors even indicated that it is aiming at getting bigger contracts outside of Europe in the future.121
(155) The Commission also notes that in key markets towards which the exports are directed (US and China) […]. Therefore, there is no reason why capacity typically used for such contracts should be redirected to the EEA.
Figure 5: […]122
[…]
(156) Second, the Notifying Party argues that the Commission has used incorrect figures for Aleris' exports. In Table 3 of the Reply to the quantitative part of the Letter of Facts (reported as Table 16 below), the Notifying Party compares what they label as ‘Aleris actual exports’ (provided in the Reply to RFI 40) with the Commission's estimation based on the data provided by Aleris for the Commission's market reconstruction. […].
(157) In this respect, the Commission notes that it is unclear how the Aleris exports provided in the Reply to RFI 40 have been calculated. The Notifying Party labels them as ‘actual’ but the Commission considers also its estimates ‘actuals’ as they have been calculated based on Aleris' own data, by multiplying Aleris' 2018 utilised CASH capacity times a percentage indicating the share of utilised CASH capacity that according to Aleris was used for non-EEA sales. Both columns in Table 17 below are based on Aleris' own data and any inconsistency cannot be attributed to the Commission.
(158) More importantly, the Commission's 2018 estimate for Aleris' exports ([…]) is also consistent with two other ways of calculating exports. First, if the 2018 exports are calculated as ‘production minus ABS sales’ (the approach that elsewhere the Notifying Party advocates that should be used123), the result is […]124 minus […]125 equal to […]). Second, if the 2018 exports are calculated as ‘utilised CASH capacity minus ABS sales’, the result is […]126 […]127 […]). The figure of 21.7kt calculated by the Commission is exactly in between these two alternative estimates, which are all way higher than the […] reported by Aleris in RFI 40.
Table 16. […]
[…]
(159) The Notifying Party also claims that the Commission's estimates of Aleris' exports must be wrong because for 2018 the Commission's estimated capacity for Aleris […].128 The Commission notes that the discrepancy is by a mere […], which is a perfectly acceptable margin of error considering that for that year it is well known that Aleris was operating […].
(160) Third, the Notifying Party takes issue with the way the Commission implemented the adjustment for exports both for the past years and for the future years.
(161) As regards the past years, the Notifying Party considers that exports should be calculated as ‘ABS production minus EEA ABS sales’ as opposed to, as the Commission does, multiplying the direct measure ‘% of utilised CASH capacity exported outside the EEA’ with the utilised CASH capacity, based on the information provided by each supplier.
(162) The Commission considers that both approaches are conceivable in principle. However, the approach advocated by the Notifying Party generates export estimates that are at odds (significantly lower) with other sources of information available. For instance, for 2018, in response to PN RFI 3 question 6b, the Notifying Party indicated that one competitor exported more than 20kt outside the EEA. Based on the market reconstruction, the Commission estimated exports to amount to 25kt. With its proposed alternative method (ABS production minus EEA ABS sales), the Notifying Party estimates no exports for 2018 (see Figure 2 of the Reply to the quantitative part of the Letter of Facts).
(163) The Commission cannot provide a detailed explanation for the discrepancy between the two methods because the Notifying Party's method relies on data (sheet ‘Production’ of the market reconstruction template) that was not used in the SO and has therefore not been subject to the same degree of scrutiny by the Commission as the data that was utilised (sheets ‘Sales’ and ‘Capacity’ of the market reconstruction template).
(164) In conclusion, the Commission considers its method to be more appropriate.
(165) As regards the future years, the Notifying Party takes issue with the fact that the Commission assumes that suppliers will export outside the EEA a percentage of CASH capacity equal to the average of the percentages exported over the previous three years (2016–2018). The Notifying Party argues that this method does not reflect the fact that for one competitor the proportion of CASH capacity exported has decreased between 2016 and 2018. In this respect, the Commission first notes that 3 years are not enough to suggest any trend, especially in an industry with lumpy large contracts where winning one tender outside the EEA (or a request by an EEA customer to ship some volumes to the US) can completely change the percentage of CASH capacity exported outside the EEA (especially considering that most of the largest competitors of the Notifying Party only have plants in the EEA and therefore can only serve non-EEA contracts via exports).129
(166) In light of the above, the Commission considers that it is appropriate to exclude exports outside the EEA from the CASH capacity available for customers in the EEA and that its methodology to perform such exclusion is sound.
(167) In any event, the Commission also reports a sensitivity scenario in which all CASH capacity is considered, with no adjustment for exports to non-EEA customers (scenario ‘Reconstruction sensitivity’). This scenario provides an extreme sensitivity as explained in Section 3.
2.3.4. Imports into the EEA
(168) The Notifying Party claims that if the Commission performs adjustments for exports it should also account for imports.130 The Commission notes that as suggested by the Parties in the Form CO, the capacity of Ma’aden, entirely located outside the EEA has been accounted for.
(169) As far as the Commission is aware, imports into the EEA have Ma’aden as the main source. Ma’aden is not currently selling material volumes in the EEA but both the Parties in their submissions and the Commission in its reconstruction account for Ma’aden capacity ([…] according to the Notifying Party once fully ramped up).
(170) In the Reply to Annex I of the SO, the Notifying Party also notes that imports of Nanshan should be added to the available EEA ABS capacity. The Commission notes that Nanshan’s sales into the EEA are negligible and forecasted to remain so ([…] in 2018, […] in 2023), that Nanshan is not mentioned in the capacity data provided with the Form CO and Nanshan does not appear in the supply-demand documents of Novelis.131 Therefore, the Commission does not make an adjustment for Nanshan in its market reconstruction.
(171) Finally, the Commission notes that in its sensitivity analysis in which exports outside the EEA are not excluded, the imports from Nanshan should not be included in that scenario.
2.3.5. Full ramp up occurring after 2023
(172) The Notifying Party argues that the Commission neglects already announced future capacity expansions by a competitor’s plant that according to the reconstruction will reach final capacity after 2023 (end of the market reconstruction).132 The Commission notes that it is inappropriate to account in its ‘steady state analysis’ for the fully ramped up capacity of this competitor. The Commission is using the available data for 2019–2023 as an indication of what would be the situation of the industry in the longer term and to do so it uses demand forecasts for 2019–2023 against effectively available CASH capacity over 2019–2023. While it is true that one competitor’s plant will have more capacity in the future after 2023, it is also true that in that period (after 2023) demand will be higher. So there is no reason why the Commission should consider additional capacity coming into the market after 2023 and comparing it to demand forecasts for the period up to 2023.
2.4. Capacity shares
2.4.1. Notifying Parties' data
(173) In addition to sales-based market shares, the Commission has also considered different suppliers’ production capacities in its competitive assessment. Capacity levels are important (a) because they permit calculating capacity shares, an alternative measure of the Parties’ competitive position in the market and (b) because they allow assessing the extent of overcapacity in the market. This section considers the derivation of capacity shares, whereas the next section considers an analysis of overcapacity in the market.
(174) The Notifying Party has provided estimates of the capacity for Aluminium ABS (measured as effective CASH line capacity133) by supplier, including expected future capacity expansions. Concretely, the Notifying Party has calculated CASH capacity by starting from the theoretical
(b) Second, the larger of the Parties (Novelis) uses all of its EEA CASH capacity for sales to […] while some rivals may use some of their EEA CASH capacity to serve […].
(c) Third, […].135 […]. (d)
(d) […].
2.4.2. Market reconstruction
Figure 6: […]136
[…]
Figure 7: […]
[…]
(178) As in the case of market shares, the Commission has also engaged in a market reconstruction for capacity volumes to verify the data provided by the Notifying Party. Specifically, the Commission has calculated the capacity for Aluminium ABS (measured as CASH line capacity) by supplier, including expected future capacity expansions, based on the data provided by market participants. These figures are reported in Table 19 and Table 20.
(179) As a preliminary remark, the Commission notes that a number of limitations in the data provided by the Parties’ competitors may lead to an under-estimation of the Parties’ capacity shares.
(a) Competitors’ reported capacity might be higher than the actual sales volumes resulting thereof, because of so-called process scrap.137 As a result, total capacity reported by certain of the Parties’ competitors (certainly so in the case of Hydro) would over-estimate the actual final kilotons of product that can in reality be produced.138
(b) Moreover, double counting of capacity can occur in case the same ton of FRP is treated first at one plant and then again at another plant of the same competitor. Similarly, there are cases in which the same ton of FRP is passed twice through a given machine. The Commission understands that the Notifying Party has avoided such double counting when providing their data.
(180) As a second remark, the Commission notes that not all of the CASH capacity at each supplier’s plants is used for the Aluminium ABS market defined by the Commission. Instead, some of the capacity is in reality used (i) for sales relating to automotive customers outside the EEA or (ii) for non-automotive (for example aerospace) customers.139 In the market reconstruction exercise, the Commission has therefore calculated the capacity of each supplier as the total effective CASH capacity minus the production that is unavailable for EEA automotive customers, as it is used for non-EEA or non-automotive customers.
(181) The result of the market reconstruction is displayed in Table 19 and Table 20. A number of conclusions can be drawn from these figures.
(a) First, the amount of capacity available for EEA customers of Aluminium ABS is significantly lower than the amount indicated by the Notifying Party (Table 20).
for the duration of a long term contract, then the Parties’ committed capacities are not relevant in assessing the likely price effects of a merger. Therefore, one should look at the shares of uncommitted capacities when assessing the likely competitive effects of this transaction.144
(183) The Commission disagrees with the Notifying Party’s view. Section 2.5.3 sets out in detail the Commission’s argument as to why looking at uncommitted capacity gives an incorrect view of the likely effects of the Transaction.
2.5. Analysis of capacity in the market for Aluminium ABS
2.5.1. Supply-demand balance
(184) In addition to considering the relative competitive positioning of the Parties, the Commission has also studied the balance between supply and demand in the market for Aluminium ABS in the EEA based on the data collected on available capacity and the projections of demand by the Aluminium ABS suppliers.
(185) The Commission notes that this section is purely a description of the supply-demand situation in the Aluminium ABS market. The analysis supporting the Commission’s concerns that rivals do not have sufficient spare capacity to defeat any unilateral effects from the Transaction is presented in Section 2.5.3, where the Commission shows that the rivals of Novelis do not have enough capacity to offset price increases from the transaction. Therefore, the Commission stresses that, as discussed in detail in Section 2.5.3, the relevant question is whether rivals’ capacity, rather than capacity in the industry overall (including the capacity of the Parties) is enough to cover the entire market demand.
(186) As regards demand, the Commission has used projections provided by Novelis during the investigation based on internal documents generated in the course of business. Novelis originally provided demand forecasts with the Form CO at the time of the Notification. The Commission considers that these are the most reliable estimates of the Aluminium ABS demand as they are (i) based on Novelis internal documents and (ii) based on documents produced before the merger proceedings.
(187) After the Article 6(1)(c) Decision, Novelis has revised its demand projections downwards compared to the projections reported in the Form CO. While the Commission considers these forecasts inferior compared to the Form CO ones (as they are not produced in tempore non suspecto), in what follows the Commission considers both Novelis’ old (Form CO) and updated (RFI 36) demand projections.
(188) After the Commission issued the SO, the Parties submitted a Ducker 2019 report145 containing even lower demand forecasts compared to the ones (already updated once) previously submitted by the Notifying Party. The Commission does not consider the Ducker 2019 forecasts in its main analysis for a number of reasons.
(189) First, the Commission considers Novelis’ own internal forecasts (especially those produced in the regular course of business) to be preferable. […].
(190) Second, even if the Notifying Party notes that the Decision relies on information from Ducker a number of times, the Commission considers that, for the purpose of investigating supply- demand balances in the industry, the information from Ducker likely underestimates demand. The Ducker report is produced for European Aluminium, the association of the European aluminium suppliers. From email exchanges between European Aluminium and its members it is clear that one of the objectives of the association is sending a reassuring message to OEMs that the capacity available in the Aluminium ABS market will be sufficient to cover their demand.146
(191) Third, consistently with the point above, the demand reported by the Ducker 2019 study for 2019 ([…] kt147) is already considerably lower than the available information on actual sales for 2018 based on the market reconstruction ([…] kt148) and […] than the Parties estimates of the 2019 demand ([…]kt149).
(192) Fourth, according to the previous Ducker 2016 study, the 2016 demand is […]kt.150 This is substantially lower than the historic figures for the realised 2016 ABS demand according to the reconstruction by the Commission ([…] kt151) and […] than the historic figures for the realised 2016 ABS demand provided by the Notifying Party itself ([…] kt152). This shows that either Ducker tends to underestimate demand systematically or the Notifying Party is suggesting to use figures for demand from Ducker that are not directly comparable with the demand forecasts they previously provided.153
(193) In any event, the Commission notes that its main conclusion that Novelis is pivotal pre-merger and would be even more pivotal post merger (that is, rivals do not have enough capacity to offset price increases from the transaction) is not affected by whether the Form CO demand forecasts, the updated forecasts from RFI 36 or the Ducker 2019 forecasts are used. To show this, Appendix I154 contains showing the pivotality of Novelis even when the Ducker 2019 demand forecasts are used. The Commission stresses that this is not intended to be a sensitivity to its analysis but merely a scenario combining the Notifying Party’s most extreme assumptions regarding capacity and demand. The Commission considers the gap between demand and capacity resulting from this version of the analysis to be inconsistent with the evidence on the gap in supply and demand observed in the internal documents.155
(194) As regards capacities, the Commission has used the figures provided by the Notifying Party in the Form CO156 as well as the results of the Commission’s own market reconstruction.
(195) As discussed in Section 2.3, the Commission considers that to understand the competitive dynamics in the EEA market for Aluminium ABS the most appropriate capacity figures are the ones based on the Commission’s market reconstruction, adjusted to exclude capacity used for non-ABS applications and non-EEA customers (scenario ‘Reconstruction EEA auto’).
(196) In terms of methodological approach, for a proper assessment of the supply-demand balance in the market for Aluminium ABS, it is important to note that demand is contracted long before actual production. In particular, a tender is typically nominated at least two years before the
Figure 9: […]161 162
[…]
Figure 10: […]163
[…]
(202) Finally, based on the data collected via the market reconstruction, the Commission has also calculated the level of current spare CASH capacity for each supplier of Aluminium ABS by subtracting, for each supplier’s production plant, the current total production from the total CASH capacity reconstructed by the Commission.
(203) Looking at spare CASH capacity is a very conservative proxy for the spare capacity available for Aluminium ABS in the EEA because the overall spare capacity available could be used also for non ABS applications (for example aerospace) or for non-EEA ABS sales.
(204) Total spare CASH capacity is less than […] ([…]% of total 2018 EEA CASH capacity). The Merged Entity accounts for […] of this spare capacity and the Parties’ competitors account for only […]% of the total 2018 CASH capacity (see Figure 11).
(205) The Commission therefore concludes that the actual availability of spare capacity in the market appears to be far more limited than portrayed by the Notifying Party Even absent the proposed Transaction, strong suppliers such as the Parties would be in the convenient situation of facing only limited competition from excess capacities in the market. In view of the highly concentrated nature of the market and the Parties’ own large market shares, this severe lack of excess capacities in a growing market reinforces the conclusion that the proposed Transaction is likely to harm competition materially.
Figure 11: […]164 165
[…]
2.5.2. Pivotality analysis
(206) In order to test more formally whether rivals’ spare capacity would be sufficient to offset any attempt to increase prices by Novelis and Aleris after the Transaction, the Commission has performed a pivotality analysis. Such an analysis assesses whether post-Transaction, the capacity of the Parties’ rivals would be sufficient to cover overall market demand. If not, then the Parties would be a ‘pivotal’ supplier post-Transaction (that is, their supply would be indispensable for market demand to be covered).
(207) It is well known from the economics literature, and consistent with the Commission’s case practice, that in markets with capacity constraints, pivotal firms enjoy an appreciable degree of market power.166 This is because even in a worst-case scenario, where rivals successfully win orders filling their entire capacity, the pivotal supplier would nonetheless be de facto the only supplier for the remaining part of demand that cannot be served by rivals. Pivotal suppliers are therefore in a position to exercise an appreciable degree of pricing power in the market, being aware that the market (that is, customers) are dependent on their supply.
(208) Small suppliers have a strong incentive to undercut competitors because if they fail to do so they risk ending up with no sales (as their competitors can fully cover the entire market demand). To the contrary, pivotal suppliers (those who face some degree of residual demand that cannot be covered by competitors) face a trade-off between pricing aggressively to capture some of the demand for which they face competition from competitors and keep prices high to exploit the portion of (residual) demand that cannot be covered by rivals. The larger the portion of residual demand faced by the incumbent supplier, the larger the amount of demand for which the incumbent knows is de facto the only supplier and therefore the larger the incentive to keep prices high and avoid undercutting competitors.
(209) The degree of market power exercised by a pivotal supplier depends on its degree of pivotality (that is, on the extent to which rivals are insufficient to cover total market demand). A merger may therefore cause anti-competitive effects by making a supplier pivotal that previously was not or by conferring to a supplier that was already pivotal even more control over indispensable production facilities.
(210) The Commission’s pivotality calculations, based on the figures for market shares, total demand and capacity suggest the following:
(a) First, as displayed in Table 22–Table 24, Novelis is already pivotal pre-Transaction. That is, it faces significant residual demand that cannot be covered by its rivals. This indicates that Novelis already has an appreciable degree of market power prior to the Transaction, which is consistent with […].167
(b) Second, Novelis would become even more pivotal post-Transaction. That is, it would face even more demand (an additional […]kt circa, corresponding to […]% of demand in 2019168) for which it would be de facto the only supplier.
(211) These conclusions are particularly strong when the CASH capacity from the Commission’s market reconstruction adjusted for sales to non-EEA and non-ABS customers are used (Table 24). The Commission considers this to be the most appropriate measure for capacity available to ABS customers in the EEA. However, the conclusions hold also when considering the figures provided by the Notifying Party or the figures from the Commission’s market reconstruction, not adjusted for sales to non-EEA and non-ABS customers (Table 22 and Table 23).
(212) The Commission notes that the results presented in paragraph (210) are conservative, as have been based on the new demand forecasts presented by the Notifying Party during the investigation, which are lower than the demand forecasts presented in the Form CO (see the figures below Table 22 and Table 23, which show both demand forecasts).
(214) Finally, the Commission also notes that the analysis presented above was not undertaken at the level of individual alloys or automotive components as capacity figures are available only at the overall automotive level. To the extent that the relationship between capacity and demand differs at the level of (say) 6xxx skin where the Parties are close competitors, or even at the narrower level of specific components or products, the conclusions based on the numbers presented above would underestimate the Commission’s concerns for such narrower segments.
2.5.3. The Commission's response to the Notifying Party’s arguments
(215) In response to the Article 6(1)(c) Decision175, the Notifying Party argued that:
(a) Capacity can change in response to demand developments and should in the future existing capacity be insufficient to meet expected demand, new capacity is likely to be added;
(b) The Commission’s pivotality analysis looks at demand and capacity in the market ‘in the aggregate’ and this fails to account for the fact that competition occurs on a tender by tender basis;
(c) An appropriate pivotality analysis should be based on non-nominated demand (that is demand for which a supplier has not yet been selected) and uncommitted capacity. Based on this alternative approach, rivals would have sufficient uncommitted capacity to cover non-nominated demand until 2023.
(216) In relation to the arguments presented by the Notifying Party in the Reply to the Article 6(1)(c) Decision, the Commission considers the following.
(217) As regards point (a) that capacity will be added whenever needed, the Commission notes that the analysis presented in Section 2.5 is to be considered as an analysis of the competitive effects of the Transaction taking as given the amount of capacity currently installed (or foreseen176) in the market up to 2023. On this basis, the Commission has concluded that the rivals of Novelis would not have enough capacity to offset any price effects post-Transaction. The impact of the Transaction on the Parties’ incentives to expand capacity are discussed in Section 2.6.
(218) As regards point (b) that competition occurs on a tender by tender basis, the Commission considers this view to be incorrect in the context of capacity constraints. The suppliers’ behaviour in each individual tender is significantly affected by their expectation about subsequent tenders. In a market in which capacity is not unlimited, each competitor knows that losing a tender today means that some rival is committing capacity and there will be a higher chance of winning (at a higher price) a subsequent tender due to the more limited amount of uncommitted rival capacity for the subsequent tender. Conversely, the degree to which competitors will compete prices down on an individual tender is limited by the fact that committing capacity at a low price today foregoes the option of winning a subsequent tender (that could have been supplied with the committed capacity) at a higher price. Therefore, under competition with certain capacity endowments it does not make sense for suppliers to compete each other’s price down very aggressively on individual tenders without considering competition in future tenders.
(219) As an illustration, consider a market with two subsequent tenders for 100kt each, with two suppliers having a capacity of 100kt each. In the Notifying Party’s characterization of competition in such a market, at least the first tender would be competitive, because rivals of each bidder have sufficient spare capacity to cover the first tender’s demand. In reality, however, both tenders will result in the monopoly price. This is because each supplier knows that losing the first tender implies (with certainty) that one will win the second tender at the monopoly price. Therefore, the outcome of both tenders would be at the monopoly price in this stylised example.
(220) […].177 […].
Figure 12 […]
[…]
(221) As regards point (c) that the pivotality analysis should focus on uncommitted capacity and non-nominated demand, the Commission considers that this approach provides a distorted snapshot of a more complex process of dynamic competition, which is captured by comparing total demand and supply. In particular, this approach is uninformative about the question of whether rivals have enough spare capacity to cover the entire demand of the upcoming tenders (say, the tenders that will be nominated in the next five years).
(222) First, the purpose of pivotality analysis is to consider whether (in the steady state equilibrium of an industry), a given supplier is necessary to ensure that all customers can be served. If this is the case, then the pivotal supplier enjoys market power, as described in the previous section. By definition, pivotality analysis therefore compares total demand with total supply in an industry, as only this aggregate analysis will be able to draw a conclusion on whether a supplier’s capacities are effectively indispensable for the demand side of the market to be served.
(223) Limiting a comparison of demand and supply to outstanding non-nominated production (as proposed by the Notifying Party) is likely to bias the outcome of a pivotality analysis severely in favour of finding no competition problem. Consider, for example, a stable market with 80kt demand and two identical suppliers with 50kt capacity each. Both suppliers are therefore pivotal (and hence possesses some market power). Assuming that suppliers hold the same market share, the long run equilibrium of this market is that each supplier sells 40kt per year, while maintaining a spare capacity of 10kt each. Assume that during each time period, 20kt of demand come up for new nomination again (of which both suppliers previously served 10kt). For these new outstanding tenders, each supplier has a total available capacity of 20kt (consisting of 10kt released capacity plus 10kt general spare capacity). If one were to focus only on outstanding tenders during a fixed time period rather than total demand (as proposed by the Notifying Party) one would therefore come to the erroneous conclusion that none of the two suppliers is pivotal, since outstanding capacities per supplier (20kt) are sufficient to cover outstanding demand (20kt). The analysis proposed by the Notifying Party would therefore grossly underestimate the market power of the two suppliers. While the Notifying Party’s proposed methodology would consider no supplier as pivotal, in reality both suppliers are pivotal for 37.5% of demand (30kt of 80kt), which could not be served by the respective rival. The Notifying Party’s methodology therefore systematically underestimates the actual degree of market power by incorrectly truncating the analysis and excluding actual current production from the picture.
(224) In the Reply to the SO, the Notifying Party claims that the Commission’s illustration is not applicable to the present case because the Notifying Party has showed that ‘rivals have enough spare capacity to supply all upcoming demand’. As explained in detail in paragraph (227)-
(232) below, the Commission considers that the Parties analysis does not show that rivals have enough capacity to supply all upcoming demand’. First, the Notifying Party’s analysis does not account for all upcoming demand because it considers only the upcoming demand corresponding to the tenders currently registered as ‘open’ in Novelis’ bidding data –this is only a subset of the upcoming tenders. Second, the time frame the Notifying Party considers is ‘truncated’ at 2023 and therefore focuses on a period for which a very large share of demand has already been determined by tenders that have taken place in the past. To the contrary, the bulk of the production corresponding to tenders occurring over the next few years will take place after 2023.
(225) […].178
(226) Third, if the approach proposed by the Notifying were correct, the Notifying Party’s conclusion that not even the largest firm in the market (Novelis) is pivotal would suggest that tender outcomes should be rather competitive (as each competitor knows that by pricing above costs it will likely be undercut by a rival willing to better utilise its spare capacity). To the contrary, the margins at which […].
(227) […]:
(a) […].
(b) […].
(228) In relation to point (a), the methodology proposed by the Notifying Party covers the open demand for production taking place between 2019 and 2023. This cannot be taken as the basis for assessing the relevant capacities for tenders that will be nominated in the near future (let alone those that will take place further in the future).
(229) Production occurring over the 2019–2023 period is mostly about competition that occurred in the past and therefore focussing on uncommitted capacity and non-nominated demand for production occurring over the 2019–2023 period is not very informative about competition (tender processes) that will take place over the same period (for tenders nominated during the next five years – from 2019 to 2023 – production will take place over 2021–2030.
(230) For example, tenders nominated in 2020 (that is, six months from now), for which SOP would be in 2022, since SOP is typically at least two years after nomination, would lead to a production until 2027–2029 under a normal 5–7 year production cycle. In other words, even for very current tenders, the Notifying Party’s proposed methodology (which covers open capacities until 2023) would address only a small part of the production cycle. Tenders that are two years out in the future are instead effectively not covered at all by Notifying Party’s methodology.
(231) In relation to point (b), the Commission considers that the approach is in principle more appropriate but the methodology suffers from the flaw that Novelis’ bidding data does not comprise the entirety of upcoming tenders, as it can only account for the volumes relating to those tenders that are open at present and, as such, recorded already in the bidding data of Novelis. The only upcoming tenders registered in Novelis’ bidding data are tenders for which production is expected to start in June 2021 at the latest. As producers issue more tenders over time, additional supply requirements will emerge which are currently not yet recorded in Novelis’ bidding data.
(232) Moreover, the time frame considered for production is truncated at up to 2023, while already a tender occurring in 2021, for which production would typically be 2023–2028, is mostly outside the period analysed.
(233) The Commission also notes that even if one (incorrectly) used the Notifying Party’s approach towards assessing pivotality (that is, using uncommitted capacity and non-nominated demand), based on the Commission’s market reconstruction figures for CASH capacity that is available for EEA automotive customers,179 the Notifying Party would be pivotal as of 2022.180
(234) In light of the above, the Commission concludes that a pivotality analysis comparing total capacity and total demand in the market is the accurate way of assessing whether a supplier is pivotal in the market. In markets with multi-year production cycles, it will always be the case that a certain amount of capacity is currently committed. What matters for the competitiveness of the market, however, is the balance of supply and demand over the production cycle.
(235) Based on the evidence presented in this section, the Commission therefore considers that Novelis has significant market power already pre-Transaction and that the Transaction would further strengthen this market power by materially increasing the market’s dependence on one single firm’s capacity. Contrary to the Parties’ claims, rivals’ spare capacity is insufficient to prevent the Transaction from increasing the Parties’ market power post-Transaction. In view of the substantial increase in capacity share, the Transaction is instead likely to cause significant damage for the competitive interaction in the market for Aluminium ABS.
(236) In the Reply to Annex I of the SO, the Notifying Party argues that the Commissions’ pivotality analysis ignores the fact that there can be no price effects on nominated volumes. It argues that a significant part of the Parties’ capacities is committed for a number of years and there can be no price effects on volumes that are already committed.
(237) The Commission agrees that there cannot be price effects on contracts already signed. 181 Indeed, the Commission did not claim in the SO (nor does it in this Decision) that the Transaction would lead to price effects for contracts already signed nor that the main harm from the Transaction would be price increases for production taking place in the next five years. This production is indeed mostly the result of competition that occurred in the past.182 The Commission’s main concern is the outcome of the tenders that will take place in the next five years (as contracts continuously come to expiry and have to be re-negotiated). If the Transaction occurred in a spot market and was therefore capable of affecting even the prices of production taking place immediately after the Transaction, the Commission’s concerns would have simply been even more pronounced.
(238) The Notifying Party also claims that the Commission’s pivotality analysis fails to take into account the dynamic nature of capacity expansion and the fact that such expansion responds to demand, particularly in a growing market such as Aluminium ABS. Security of supply is very important and sophisticated OEMs would be able to counter act a blunt attempt to withhold capacity from the market by resorting to imports, sponsoring entry or use steel.
(a) As regards imports, the Commission notes that it is already accounting for […] kt of capacity (mostly empty) by Ma'aden located in Saudi Arabia and that this has been assumed to be purely dedicated to the EEA. Moreover, the reasons why imports are not a sufficient competitive constraint are set out in Section 8.3.10.3 of the Decision.
(b) As regards the constraint from steel, the Commission points to the analysis presented in Section 2.1, where the conclusion is that competition from steel is not sufficient to constrain small but significant price increases in the Aluminium ABS market.
(c) As regards the ability of entry or expansion to prevent anticompetitive price increases resulting from the merger, the Commission notes that the Notifying Party’s […] own operation suggests otherwise. Indeed, while steadily increasing demand and high prices have attracted expansion by others in the past, […].
(239) The Notifying Party argues that the Commission’s theory of harm is not supported by past evidence.
(a) The Notifying Party argues that Novelis did not act in the past as a residual monopolist, because if Novelis was acting as a residual monopolist one would expect its capacity utilisation to be lower than that of its competitors. […]. Moreover, comparing spare capacity of Novelis and rivals in the recent past (say 2018) creates confusion because the spare capacity of rivals is inflated by the fact that many of them had just expanded their plants.
(b) […].
(c) The Notifying Party argues that if the Commission’s theory were correct, one would observe in the market periods (cycles) during which Novelis loses most tenders (waiting for the rivals to fill all their capacity), followed by periods during which Novelis charges very large prices (exploiting the fact that the rivals’ capacities are now full). In this respect, the Commission notes that the Notifying Party’s claim is based on an extreme characterization of Novelis’ behaviour that nowhere is made by the Commission in the SO or this Decision. The Commission’s claim that an increase in Novelis’ pivotality is likely to give Novelis additional pricing power (due to the fact that it would face after the Transaction less rival capacity) does not rely on Novelis completely withdrawing from the market for a certain bidding cycle (waiting for rivals to fill their capacity) and then charging the monopoly price on the remaining captive demand. This is further discussed at paragraph (240).
(240) The Notifying Party argues that the Commission’s theory of harm has unrealistic implications.
(a) According to the Notifying Party, for the Commission’s theory of harm to be valid, after the merger Novelis must refrain from winning any volumes until 2024, resulting in a drastic decrease of its market share. Then, Novelis would increase its price on the remainder of the demand for which it would be de facto the only supplier. Under this dynamic, the critical price increase that Novelis would have to apply on the volumes over which it would be pivotal is […]%, which is unrealistically high. In this respect, the Commission notes once again that the nowhere in the SO or this Decision does the Commission imply that the risk of price increases due to Novelis’ increased pivotality would come from Novelis completely withdrawing from the market for a certain bidding cycle (waiting for rivals to fill their capacity) and then charging the monopoly price on the remaining captive demand. The Commission considers that Novelis will continue to bid after the merger in the same way as it has been doing pre-merger, with the additional awareness that it would now face significantly less rival capacity compared to absent the Transaction. This in the Commission view would lead to less aggressively bidding behaviour by Novelis in each tender (as opposed to refraining from bidding in some tenders and then charging unrealistically high amounts in others, as spelled out in the extreme characterization made by the Notifying Party).
(b) According to the Notifying Party, using the Commission’s own numbers and methodology, every Aluminium ABS suppliers would be pivotal in 2023, suggesting that any Transaction in this industry would be problematic. First, as a preliminary remark, the Commission notes that by saying that all suppliers in the industry are pivotal is not a convincing argument in favour of clearing a Transaction: this amounts to saying that each supplier in the market is necessary to cover customers’ demand. Second, the Commission notes that while all suppliers of Aluminium ABS would be pivotal in 2023, the same does not hold for any of the other years. According to the Notifying party’s own calculations, […]. Last, and very importantly, pivotality is a matter of degree. What matters is not just the status of pivotal player but also the extent of the pivotality, that is, how much the capacity of your rivals falls short of being able to cover the entire market demand. Just like for market concentration, it matters whether and how much pivotal player is prior to the Transaction and the delta brought by the Transaction. In this case, the former element has an important dimension, because it is not the same thing to add Aleris’ capacity to AMAG or to Novelis.183
Figure 13: Years for which competitors are pivotal according184
[…]
(241) The Notifying Party complains that the Commission did not perform any analysis of the price effects of the Transaction.
(242) In this regard, the Commission notes that it is not necessary for the Commission to conduct concrete merger price simulations in order to find that a transaction restricts competition. This is particularly so in the case of homogeneous goods markets, where the economic literature shows that market structure (e.g., as measured by capacity shares or HHIs) is a particularly good indicator for firms’ ability to raise prices.185
(243) In the current case, the Commission has assessed the likelihood of significant anticompetitive effects, among other by: (i) assessing the high combined market and capacity share of the Merged Entity, (ii) assessing the relatively low elasticity of demand towards other goods (in particular, steel), (iii) assessing the lack of spare capacity of rivals (and the ensuing need of customers to rely on the Merged Entity), (iv) assessing the existence of appreciable barriers to entry. In homogeneous goods markets, these factors are the main drivers of anticompetitive effects.
(244) In addition, the Commission has developed and verified two self-standing theories of harm (based on a restriction of price competition and based on a restriction of future capacity competition). The Commission has also collected a substantial amount of internal documents supporting the anticompetitive effects arising from these two theories. Moreover, the Commission has assessed the degree of pre-merger market power exercised by the Parties. Finally, the Commission has assessed in significant detail (and dismissed) the alleged lack of closeness of the Parties on the basis of bidding data.
(245) The Commission therefore considers that a precise quantification of price effects in the short and long run would merely be a distraction from the conclusion that inevitably flows from the evidence collected in this case: that the Proposed Transaction would materially harm competition.186
2.6. Capacity competition
2.6.1.The impact of the Transaction on capacity competition
(246) In addition to the direct detrimental effect on price competition between suppliers of Aluminium ABS, based on existing (and planned) levels of capacity,187 the proposed Transaction is also likely to create more long-term competitive harm by stifling capacity competition. Indeed, the economic theory of capacity competition predicts188 that as the number of producers decreases, the total capacity that firms are willing to supply to the market will decrease, which causes upward pressure on prices. This softening effect on capacity competition is known to be particularly strong if, as in this case, the market shares of the merging parties are high. This is because the capacity choices of firms with a larger market have a comparatively larger impact on market prices.189 Moreover, when capacity extensions put pressure on market prices, larger producers will have to absorb a comparatively larger part of this pressure on profits. When deciding on when and to what extent to increase their capacity, firms with large market share will therefore be particularly careful in trading off the benefit in terms of organic growth potential with the downside relating to the effect that extra capacity has on the market price.
(247) In a context in which demand for Aluminium ABS is constantly growing, the idea that adding the capacity (and current sales) of Aleris to Novelis would create a supplier with lower incentives to engage in capacity expansions should not be interpreted as a concern that the Merged Entity would stop investing and no longer add capacity in the market. The market is growing significantly and the Commission considers that all aluminium ABS suppliers will continue investing after the transaction in order to keep up with this growing demand. However, the Merged Entity’s incentives to expand after the Transaction would be significantly lower than the incentives that Novelis and Aleris would have absent the Transaction, in particular in relation to those investments aimed at growing organically at the expense of rival suppliers. In this respect, the Commission presents in Section 8.3.9 of the Decision evidence that Novelis […].
(248) Therefore, in a growing market as the one for aluminium ABS, the reference in paragraph (246) to a reduction in supply and consequent upward pressure on prices should be interpreted as a lower and/or delayed amount of capacity being added to the market compared to the situation absent the Transaction, with consequent prices above the level at which they would be absent the transaction.
2.6.2. The Commission's response to the Notifying Party’s arguments
(249) In their Reply to the Article 6(1)(c) Decision, the Notifying Party disputes any prospect of softened capacity competition flowing from the Transaction. In particular, they criticise the analysis of spare capacity presented by the Commission on the grounds that, whenever demand approaches a level at which capacity would be tight:
(a) powerful OEMs would be able to induce Novelis and Aleris (or one of their competitors) to build extra capacity, and
(b) it would be in the interest of suppliers of Aluminium ABS to ensure that enough capacity is available in the market, as alternatively OEMs would be able to switch back to steel (or switch from steel to Aluminium ABS to a lower extent).
(250) The Commission disagrees with this assessment. As discussed in Section 2.1 above, OEMs with a preference for Aluminium ABS (especially those with preference driven by exogenous factors such as CO2 emissions regulations) cannot seamlessly switch between Aluminium ABS and steel in response to small but significant changes in relative prices. On the contrary, the functional differences between the two materials imply that a hypothetical monopolist over Aluminium ABS would possess appreciable market power with respect to its pricing. In the Commission’s view, the constraint exerted by the possibility of OEMs switching back to steel is therefore clearly not sufficient to prevent the Merged Entity from increasing its negotiation power vis-à-vis customers through the Transaction.
(251) The Commission acknowledges that, as in many other industries, new Aluminium ABS capacity is normally built only when some contractual pre-commitment is given by an OEM to fill part of the new capacity for a given time period. In view of the heavy capital investments required for capacity extensions, securing such initial customers is important to mitigate the commercial risk associated with a plant extension. However, the existence of such pre- commitments does not undermine the ability of suppliers to decide for themselves how much capacity they will build, at what time, and which terms to require from customers to agree to a capacity extension at all. On the contrary, suppliers will unilaterally consider a variety of factors that have a strategic impact on their long-term profitability when deciding whether to build additional capacity. For instance, profit maximising suppliers will take into account factors such as (i) expected aggregate capacity in the industry going forward, (ii) projections of demand growth or decline for the industry, (iii) possible strategic responses of competitors triggered by own capacity decisions, and (iv) the future willingness to pay and substitution opportunities of potential customers.
(252) Such long-term strategic considerations are far more pertinent for capacity choices than (say) for short-term price decisions. The Commission notes, in particular, that newly constructed factory buildings and production lines have a substantially longer lifespan than the duration of typical contractual pre-commitments of launch customers. When deciding whether to expand capacity today, suppliers are therefore faced with an appreciable risk of future imbalances between demand and supply, for example due to the fact that they have no certainty that the customer will buy again from them at the end of the production cycle and/or due to the highly cyclical nature of automotive demand (which implies that demand can severely and unexpectedly decline in a recession or accelerate in a boom).
(253) Moreover, although the evidence available to the Commission190 suggests that capacity expansions are typically realised with a partial volume pre-commitment by one or more lead customers, a significant remaining share of the capacity expansion is typically uncommitted.
(254) For the above reasons, suppliers will not simply let OEMs dictate them, based on OEMs’ current short-run interests, how much capacity they will build for the long run. Instead, suppliers negotiate hard with OEMs about undertaking possible extensions and the commercial terms of potential launch contracts. While doing so, they take careful account of their own long-term strategic interests, including how capacity extensions will affect likely future prices in the market.
(255) As noted at the beginning of this section, when merging parties have high market shares, economic theory predicts that concentration will bring about a material lessening of incentives to engage in capacity competition. Since capacity extensions today will weaken the industry’s ability to maintain high prices in the future, companies with larger market share will be more wary about the impact of their capacity choices on the industry equilibrium.191 Given the highly concentrated nature of the market for Aluminium ABS and the Parties’ own high market shares, the proposed transaction is therefore likely to restrict not only short-run price competition but also long-run capacity competition.
(256) In the Reply to Annex I of the SO, the Notifying Party argued that demand uncertainty undermines the theoretical foundation of the Commission’s dynamic theory of harm.192
(257) In relation to this claim, the Commission first points out that its theory of harm is not only grounded in theoretical economic models (e.g. Cournot model) but most importantly based on extensive evidence that the available capacity is closely related to the market price (the higher the available capacity in the market, the lower the equilibrium price) and that consequently the larger the current market share of a given player, the lower the incentive to expand capacity. And therefore, adding significant market share to a supplier who’s already very large can only further decrease the incentive to increase capacity.193
(258) Moreover, the Commission notes that uncertainty is quite common in a large variety of markets to which the merger literature applies. Importantly, the literature cited by the Notifying Party does not suggest that mergers are less harmful in the face of demand uncertainty.194 On the contrary, uncertainty can make mergers even less harmful, depending on the circumstances.195
(259) The Notifying Party argued that the Commission’s theory of harm relating to capacity expansions ignores the possibility of expansion by rivals or entry.
(260) The Commission disagrees with this view. On the contrary, the economic theory of capacity competition shows that mergers involving firms with high capacity shares lead to substantial price increases despite the reaction of rivals (e.g., through expansion of their own capacities).196 To the extent that the Notifying Party argues that substantial new entry would prevent anticompetitive price increases resulting from the merger, […].
(261) The Notifying Party argued that the Commission ignores competition from steel in its analysis of dynamic theory of harm. This point is discussed in Section 2.1.3.5.
(262) The Notifying Party argued that the Commission’s theory of harm relating to capacity expansions is not supported by the evidence. It argues that under the Commission’s assumption that Novelis is pivotal already pre-Transaction, Novelis should have low incentives to expand capacity. However, Novelis expanded its capacity by […]% between 2016 and 2019. In this respect, the Commission stresses again that its theory of harm does not go as far as saying that Novelis (or, more generally, the largest player by far in a market) has no incentives to expand capacity. This is an extreme claim that the Commission has not made anywhere in the SO or this Decision, especially considering that the market is growing strongly and it makes no sense for any player to stop investing.197 The Commission only claims that the largest player by far (Novelis) has less incentives to expand capacity compared to smaller rivals and these incentives are further reduced if its size were to increase due to the Transaction. […].198
2.7. Analysis of the bidding data submitted to the Commission
2.7.1. Description of the bidding data submitted to the Commission
(263) On 30 November 2018, the Notifying Party submitted to the Commission bidding data for Novelis and Aleris and an analysis of this data prepared by CRA.199
(264) On 18 February 2019, the Notifying Party submitted a final version of the analysis together with the Form CO, in the context of the Notification.
(265) In response to the Article 6(1)(c) Decision (in which the Commission presented its preliminary views on the bidding analysis performed by the Notifying Party), the Notifying Party started to work on an updated and revised version of both the Novelis bidding data and on a new bidding dataset for Aleris.
(266) On 19 May 2019, almost six months after the original submission of 30 November 2018, the Notifying Party submitted an updated bidding analysis by CRA (as well as updated bidding data), suggesting that the previously submitted data and analyses should be disregarded.200
(267) The Commission notes that the new bidding data provided by Novelis and Aleris differ in some very material respects from the data previously submitted.
(268) As regards Novelis, the main changes in the data were the following:
(a) Some tenders have been eliminated:
i. Some tenders were found to be duplicates and therefore erroneously included in the previous version of the data;
ii. Some tenders had been cancelled and therefore erroneously included in the previous version of the data;
iii. One tender was a fake simulation by Novelis and did not correspond to a real tender; This was erroneously included in the previous version of the data;
iv. Some tenders were present in the underlying raw datasets of Novelis but did not appear in the underlying raw datasets of Novelis updated during the Transaction procedure;
v. Etc.
(b) Some tenders have been added:
i. Novelis added some programs it bid for since the previous version of the data was submitted;
ii. Some tenders that were missing in the underlying raw data of Novelis were added as part of the matching exercise with the updated Aleris bidding data;
iii. Etc.
(c) Some tenders have been modified:
i. Tenders that have been closed in the meantime, so the status has been turned to ‘closed’ and an indication of the supplier has been added;
ii. Tenders for which margin information was missing and has been added;
iii. Etc.
(269) As regards Aleris, the Commission first notes that the old dataset was based on data gathered in the regular course of business while the new dataset has been built from start by the economic consultants of Aleris for the purpose of the investigation and following the Article 6(1)(c) Decision.
(270) As a result of the changes implemented by the Notifying Party, the Novelis bidding data changed materially compared to the previous version (see Figure 14):
(a) The total number of observations for the analysis of tenders won […].
(b) Drastic changes resulted at the level of individual OEMs data: i. […]201 […];
ii. […];
iii. […];
iv. […];
v. […];
vi. […].
Figure 14: […]202
[…]
(271) […]:
(a) […];
(b) […].
Figure 15: […]203
[…]
(272) For completeness, in the remainder of this section the Commission will assess both the evidence presented by the Notifying Party based on the old bidding data (which the Notifying Party claims should be disregarded), as well as the new bidding data.
2.7.2. Win-loss analysis
(273) As a preliminary remark, the Commission notes that the market for Aluminium ABS is a basic industry characterised by capacity constraints.
(274) In such a market, unless there is evidence that either the Parties products are intrinsically very differentiated or that the Parties current and future plans involve targeting different customer groups (neither of which is the case in the present case), the loss ratios are likely to simply reflect differences in capacities as opposed to intrinsic differences in the products or strategies of the various Aluminium ABS suppliers.
(275) In light of this, based on the evidence presented in the following sections, the Commission concludes that there is a significant level of competitive interaction between Novelis and Aleris, especially for the 6xxx segment and for a number of OEMs that are a common target of both merging parties.
2.7.2.1. The first CRA win-loss analysis
(276) The Notifying Party has submitted a report204 presenting a win-loss analysis based on the bidding data of Novelis and Aleris. The analysis looks at what proportion of tenders lost by Novelis (respectively, Aleris) is won by Aleris (respectively, Novelis).
(277) The Notifying Party concludes that Novelis and Aleris are not close competitors because they attract from each other a proportion of volumes that is lower than what would be implied by their market shares (especially Novelis).
(278) Specifically, Novelis wins […]% of the tenders lost by Aleris, while the comparable figure implied by Novelis’ market share would be […]%.205 Similarly, Aleris wins […]% of the tenders lost by Novelis, while the comparable figure implied by Aleris’ market share would be […]%.206
(279) The Commission has a number of remarks in relation to these results presented by the Parties.
(280) First, it should be stressed that in a context with market shares that are as high as in this case, a finding that the loss ratios between the Parties are lower than what would be implied by market shares does not rule out the possibility of strong competitive interaction between the Parties. Indeed, according to the data submitted, even based on the Notifying Party’s own figures, the measured loss ratios are far from trivial. […]207 […].208
(281) Second, the Commission expanded the analysis presented by the Notifying Party and focused on a segment in which the Parties appear to be particularly close competitors –the 6xxx alloy. Indeed, in the 5xxx alloy segment Aleris has a very small share and, based on the bidding data of Novelis, […]. If the analysis is limited to the 6xxx alloy only, the measured loss ratios are […] ([…]% from Aleris to Novelis and […]% from Novelis to Aleris). Otherwise stated, the combined share implied by the measured diversions is [40-50]%.209
(282) In the Reply to the Article 6(1)(c) Decision, the Notifying Party argues that:
(a) The evidence provided by the Commission on loss ratios confirms that the Parties are not close competitors, as the loss ratios imply a combined share that is significantly lower than the actual combined market shares;
(b) The loss ratios calculated by the Commission are based on all tenders with Start of Production (‘SOP’) […]. In a dynamic market such as the Aluminium ABS, it is more relevant to focus on more recently awarded tenders. […].
(283) In relation to the arguments presented by the Notifying Party in the Reply to the Article 6(1)(c) Decision, the Commission considers the following.
(a) There is a fundamental disagreement between the Commission’s and the Notifying Party’s assessment of the same evidence available. In the Commission’s view, loss ratios of […]%210 suggest that when one of the Parties bids and does not win, […]. The Commission considers that this evidence shows a degree of competitive interaction between the merging parties. The evidence on loss ratios should not be interpreted as a relative concept (for example relative to what would be implied by market shares or relative to the loss ratio to other rivals). A merger between two competitors with the above loss ratios, in a concentrated market characterised by capacity constraints is very well likely of generating significant price effects even in the event the Parties are not the closest competitors –especially considering that the evidence from the loss ratios reflects more the available capacities to each supplier rather than reflecting intrinsic differences in their product offerings.211
(b) As regards the critique that the Commission looks at all tenders for the period with date […], the Commission considers that calculating the loss ratios over a period of […] increases the precision of the estimate and avoids that in a bidding market with lumpy orders the results be skewed by one or two larger tenders. To the contrary, by calculating loss ratios by year, the Notifying Party bases their results on a limited number of tenders with consequent no guarantee that any given year is representative of the true position of any given supplier (for example a supplier may not participate significantly in tenders in a specific year if the OEMs it targets have not issued important tenders in that year212). Indeed, while the Notifying Party focuses on the results […] (loss ratio Novelis to Aleris equal to […]% and Aleris to Novelis […]%), the figures […] provide a completely different picture (loss ratio Novelis to Aleris is equal to […]% […] and […]% […], while Aleris to Novelis is […]% […] and […]% […]). Finally, the Commission also notes that focusing on the period […] is the approach put forward by the Notifying Party in the Form CO and the CRA submission of […].
(c) In a market characterised by capacity constraints, suppliers tend to be selective and avoid putting competitive bids for each tender by each OEM. Therefore, it is particularly informative to look at loss ratios by OEM to see if the Parties are closer for certain OEMs than for others. The evidence suggests that the Parties are particularly close competitors for 6xxx alloys for the following very important OEMs:
i. […];
ii. […];
iii. […];
iv. […];
v. […].
(284) Overall, based on the evidence above, the Commission concludes that the win-loss analysis supports the conclusion that there is significant competitive interaction between the Parties.
2.7.2.2. The second CRA win-loss analysis
(285) Based on the significantly revised bidding data (see Section 2.7), the Notifying Party claims that for the period with SOP dates […] the loss ratio213 from Novelis to Aleris is […]% and from Aleris to Novelis is […]%.
(286) These figures are lower than the loss ratios calculated based on the original bidding data214 (see Section 2.7.2.1) and therefore the Notifying Party claims that the new bidding data shows even more than the original data that the Parties are not close competitors.
(287) The Commission concludes that the evidence on loss ratios from the revised bidding data is not sufficient to characterise Novelis and Aleris as distant competitors and dispel competition concerns, especially in light of their very large share of capacity in the market, as well as limited spare capacity of rivals.
(288) First, as regards the loss ratio from Novelis to Aleris, the Commission considers that the change compared to the previous version of the bidding data ([…]) is minimal and does not allow concluding that Aleris is a distant competitor from Novelis. In light of the very large market shares and capacity shares of Novelis, as well as the limited capacity of Novelis’ rivals (see Sections 2.2, 2.3 and 2.5) the Commission does not consider that a loss ratio of […]% constitutes any evidence that Aleris does not pose a competitive constraint on Novelis.
(289) Second, the Commission stresses that Aleris has a capacity share equal to approximately [10- 20]% of the market. Therefore, it is natural to expect that Aleris must select a number of OEMs and tenders to target, […]. Therefore, in a market with strong asymmetries in the capacities of different suppliers, the loss ratios reflect differences in capacities as opposed to intrinsic differences in the products of Novelis and Aleris.
(290) Third, in the 6xxx segment, where Novelis and Aleris are closer because […], the loss ratio from Novelis to Aleris is […]%. This means that […].
(291) Fourth, to confirm the significant bidding overlaps between Novelis and Aleris at the level of individual OEMs, the Commission notes that the loss ratio from Novelis to Aleris in the 6xxx segment is […]. As regards bidding overlaps at the individual OEM level, the Commission notes that the new data submitted by Novelis after the Article 6(1)(c) Decision considerably differs from the data submitted with the Notification. Based on the previous data, the loss ratios from Novelis to Aleris were significant also for BMW.
(292) As regards the loss ratio from Aleris to Novelis, the loss ratios from Aleris to Novelis is […]% overall and […]% in 6xxx. The Notifying Party considers that an indication that Novelis and Aleris are not distant competitors but the Commission disagrees for the following reasons.
(293) First, the Commission notes that Novelis being by far the largest of the two merging parties, the relevant question is rather whether Aleris poses a competitive constraint on Novelis rather than whether Novelis poses a competitive constraint on Aleris. And in relation to this question the Commission considers that the [10-20]% capacity share of Aleris, as well as the evidence on the loss ratios discussed at paragraphs (287) to (291) support the conclusion that Aleris is a significant constraint on Novelis.
(294) Second the Commission notes that the new calculations are based on a completely new dataset produced during the investigation (after the Article 6(1)(c) Decision ). As stressed in Section 2.7.1, […], the difference being tenders of which it does not keep track in the regular course of business.
(295) Third, , the Commission considers that the loss ratio from Aleris to Novelis is affected by the […].215 […].
(296) Fourth, at the level of individual OEMs, the Commission considers that Novelis has captured significant share of tenders […]216 […].
(297) Fifth, the Commission notes that in the reply to request for information 45 the Parties confirmed that […]217 […], while in fact the Request for Quotation (‘RFQ’) to Aluminium ABS suppliers was cancelled as ultimately those components of the Volvo car were designed by using steel. In the Commission’s opinion tenders for which the RFQ to Aluminium ABS suppliers were cancelled should be dropped from the dataset for the purpose of calculating loss ratios, […].218 Correcting for this would materially increase the loss ratio from Aleris to Novelis and would […].
(298) Overall, the Commission concludes that the evidence on loss ratios from the revised bidding data (as well as the evidence from the original bidding data) is not sufficient to characterise Novelis and Aleris as distant competitors. The Commission concludes that the win-loss analysis supports the conclusion that there is significant competitive interaction between the Parties, especially in the 6xxx segments, and for a number of OEMs that appear to be the focus of both suppliers […].
2.7.2.3. The Commission's response to the Notifying Party’s arguments
(299) In the Reply to Annex I of the Statement of Objections (‘SO’), the Notifying Party argued that the loss ratios between the Parties show that the Parties are not close competitors and proposed a new calculation method for weighted diversion ratios. These objections are discussed and rebutted in the remainder of this section.
(300) First, the Notifying Party claims that the Commission misinterprets loss ratios as diversion ratios. The Commission disagrees with the Notifying Party’s claim, because it was never claimed that should be interpreted as diversion ratios. The Commission looked into the loss ratios only to rebut the argument of the Notifying Party that Novelis and Aleris, despite their very large combined market share, are distant competitors. Based on the evidence from loss ratios (considering both the old and the new set of bidding data provided), the Commission concludes that there is a significant extent of competitive interaction between Novelis to Aleris, especially in the 6xxx segment and for certain important OEMs that are targeted by both merging parties.
(301) Second, the Notifying Party argues that loss ratios are significantly lower than what one would expect if diversion within Aluminium ABS was proportional to market shares. The Commission’s calculations show that the loss ratio from Novelis to Aleris is in line with the diversion implied by capacity market shares. Based on the average capacity share219 of Novelis and Aleris during the same period, the loss ratio implied by the Parties’ […].220 Based on the updated bidding data (which is the dataset yielding the lowest loss ratios between the Parties), the loss ratio from Novelis to Aleris for the period with SOP dates […], as mentioned in paragraph (285). Contrary to what the Notifying Party submits, the loss ratio is thus consistent with and actually higher than one would expect if diversion was proportional to the Parties’ capacity shares. As stressed in paragraphs (289)–(291), this is consistent with economic theory and the fact that, in a basic industry characterised by strong asymmetries in the capacities of different suppliers like the present one, the loss ratios are likely to simply reflect differences in capacities221 as opposed to intrinsic differences in the products of Novelis and Aleris.
(302) Third, the Notifying Party considers that the Commission’s claim that the loss ratios are high is without any economic basis. The loss ratios used by the Commission are not themselves indicative of any price effect and are not sufficient to conclude that the Transaction will result in a SIEC. In this regard, the Commission does not pretend that a single piece of information like the loss ratios can be used in itself as an indication that the Transaction is likely to lead to price effects. The Commission simply assessed the loss ratios (in response to a submission by the Notifying Party) in relation the question of whether in addition to a very high combined share there is also significant competitive interaction between the Parties. As explained in paragraph (288), the fact that Aleris attracts […] of the volumes lost by Novelis is a piece of information that contributes to supporting the view that Aleris is a competitive constraint on Novelis that is worth preserving.
(303) Fourth, the Notifying Party argues that the Commission’s per customer approach is not relevant for establishing a SIEC. The relevant test is to assess whether the Transaction would have significant anticompetitive effects in the market overall and thus a weighted average loss ratio of the individual OEM-level loss ratios should be considered. In this respect, the Commission notes that finding significant anticompetitive effects in the market overall does not necessarily mean finding likelihood of a significant price increase on each and every customer (especially since the market is characterised by price discrimination). A finding that loss ratios are significant for certain important OEMs is a very pertinent finding in support of a SIEC.
(304) In addition the Commission considers that the methodology for weighting the OEM-level loss ratios proposed by the Notifying Party in the Reply to the SO is flawed. Taking as an example the loss ratio from Novelis to Aleris, the OEM-level loss ratios are weighted by the share represented by the OEM among all volumes won by Aleris.
(305) However, for correct weighting approach the OEM-level loss ratios must be weighted by the share represented by the OEM among all volumes lost by Aleris (as the loss ratio from Novelis to Aleris is calculated as a proportion of the volumes lost by Novelis). Indeed, this corrected methodology yields a weighted loss ratio that corresponds to the overall loss ratio obtained by collapsing together all OEMs in the bidding data. This is the approach initially proposed by the Notifying Party in its economic reports prior to the Reply to the SO and the approach followed by the Commission in the SO and this Decision.
(306) The Commission therefore concludes that the in-depth analysis of diversion ratios presented by the Notifying Party does not show that both Parties exert limited competitive pressure on each other.
(307) Fifth, the Notifying Party stresses that competition from steel is not taken into account by the Commission. In this respect, the Commission notes that loss ratios (which, as a metric, have been proposed by the Notifying Party in the first place) indicate the proportion of volumes Novelis bid for but did not win and whether a significant share of these losses is captured by Aleris. At the bidding stage, the OEM has already decided to procure Aluminium ABS and not steel and therefore the relevant set amongst which the loss ratios should be calculated is Aluminium ABS. This is indeed the approach proposed by the Notifying Party in the two CRA reports of 30 November 2018222 and 19 May 2019223. The Commission, cannot take alleged switches to steel into account, as these do not exist in the data the Notifying Party volunteered to submit to show lack of significant competitive interaction. The submitted bidding data only contains examples where the Parties won or lost competition against other Aluminium ABS producers.224 If anything, the Commission considers that the evidence from the bidding data reinforces the Commission’s claim that at the bidding stage there is no competitive interaction between Aluminium ABS and steel.
2.7.3. Econometric analysis
2.7.3.1. The first CRA econometric analysis
(308) The Notifying Party used the bidding data from Novelis to present an econometric analysis of the impact of Aleris’ participation in tenders on Novelis’ prices, margins and probability of winning.225,226
(309) Based on the results presented in Figure 16, they conclude that Aleris does not have an impact on Novelis’ prices, margins and probability of winning. Therefore, in the view of the Notifying Party, Novelis and Aleris are not close competitors.
Figure 16: Aleris’ impact on Novelis’ price and margins
[…]
Source: CRA report of 30 November 2018
(310) In relation to the Notifying Party’s findings, the Commission considers that there are very significant limitations in the availability of bidding data from Novelis (for example prices are available only for a small proportion of the tenders in which Novelis participated). This is problematic because the econometric techniques adopted by the Notifying Party require the availability of a sufficiently rich dataset (in terms of number of observations and variability across these observations). The absence of such a dataset (as appears to be the case here227) suggests that rather than showing that Aleris’ presence has no effect on Novelis’ prices and margins, the results presented by the Notifying Party suggest that the data available is not sufficiently rich to allow sufficiently robust statistical inference on this question. Therefore, the Commission considers that the analysis conducted by the Parties does not allow concluding that Aleris does not have an economic impact on Novelis.
(311) In the following, the Commission sets out a number of more detailed issues identified in relation to the econometric analysis of the Parties.
(312) First, the full dataset of Novelis contains data relating to […] tenders ([…]228). The econometric analysis performed by the Notifying Party is limited to the tenders for which a winner has already been announced and price-cost data is available, which reduces the sample to […] tenders only. This is corresponds to […] and […] tender parts only for the analysis of the tenders won and lost by Novelis, respectively. Therefore, the Notifying Party’s analysis is based on […]%229 of the tender parts for which Novelis has bid.
(313) Second, the analysis potentially suffers from an omitted variable bias, since it includes a variable capturing the participation by Aleris but not variables capturing the participation of other competitors (Hydro, Constellium, etc.). The Notifying Party claims that this makes their results ‘conservative’ because, they explain, the omission of the participation by other rivals possibly means that the variable indicating Aleris’ participation also picks up part of the effect of the presence of other competitors. As a result, the Notifying Party’s claim that their analysis possibly overstates the impact of Aleris. The Commission considers that indeed, […]. To the extent that there is a strong correlation in the participations of Aleris and the participation of Constellium and Hydro, the variable indicating Aleris’ participation would also capture a large part of the effect of Hydro and Constellium’s participation, resulting in a potential overestimation of the impact of Aleris on Novelis. However, under this premise, […].
(314) Third, the analysis of the Notifying Party focuses on all tenders in which Novelis has participated, including the tenders for 5xxx. In this segment, Aleris has a very negligible share and therefore it is expected that Aleris exerts significantly more limited impact on Novelis compared to the impact that it exerts on Novelis in the 6xxx segment. Indeed, […]. However, there is no sufficient variability in the data to study the impact of Aleris on Novelis for the 6xxx segment only (see Figure 17).
Figure 17: […]230
[…]
(315) Fourth, the models estimated by the Notifying Party include several variables (volumes, an indication of whether the customer is a Tier 1, OEM231 fixed effects) that are strongly correlated or even perfectly correlated with Aleris’ participation. This is expected to reduce the significance of the coefficient on Aleris’ participation as well as possibly modify its magnitude.
(316) Indeed, in the presence of strong correlation between two variables it becomes very difficult to statistically separate the effect of one variable (say, Aleris' participation) from the effect of the other variable (say, the size of the tender volume). In these cases, the econometric analysis likely suggests that both variables do not have a statistically significant effect on Novelis prices, but the correct interpretation is that there is not enough variability in the data for the analysis to be able to statistically separate the two effects.
(317) When the two variables are perfectly correlated (as is the case for the Aleris' participation and certain OEM fixed effects232) certain tenders are even completely disregarded by the econometric analysis and the information they contain is not used to estimate the effect of Aleris' presence on Novelis' prices. This may lead to a bias in the estimated impact of Aleris' presence.
(318) Specifically:
(a) Tier 1: […]. The inclusion of variable Tier1 in the model specification is also debatable, because the Notifying Party […].
(b) The Commission also notes that when restricting the analysis to the tenders involving Tier1 customers only, Aleris’ participation has a negative impact on the prices and margins bid by Novelis when focusing on the tenders won by Novelis.
(c) OEM fixed effects (FEs): The Notifying Party includes in some specifications a FE for each OEM. These FEs are problematic because they are very correlated, and in some cases perfectly correlated, with Aleris’ participation (see Figure 18). Technically, using OEM fixed effects in the regression implies that the observations relating to OEMs […].
Figure 18: […]233
[…]
(319) Fifth, an econometric analysis focusing on Novelis’ pricing behaviour on all OEMs risks ‘polluting’ the potentially strong constraining effect of Aleris on Novelis for certain common customers, because it also includes tenders relating to customers for which Novelis participates but without a strong incentive/willingness to win.
(320) This idea is also confirmed by the high variability in the loss ratios across OEMs shown in Section 2.7.2.
(321) While the richness in the dataset is too limited to study the impact of Aleris’ participation at the level of these individual OEMs, […].234 […].
Figure 19: […]235
[…]
(322) Sixth, the Commission notes that as explained by the Notifying Party in the reply to request for information 45, the participation of Aleris in the tenders in which Novelis has participated has been defined at the overall tender level.236 That is, considering a tender comprising 20 components, if Aleris participated, say, in only 5, Aleris has been indicated as having participated for all components of that tender. This is possibly a severe measurement error in the participation of Aleris, which can generate considerable bias in the estimate of the impact of Aleris’ presence.
(323) The Commission therefore considers that the first bidding analysis presented by the Notifying Party does not allow concluding that Novelis and Aleris are distant competitors and Aleris does not have a material competitive impact on Novelis’ conduct.
2.7.3.2. The second CRA econometric analysis
(324) On 19 May 2019, based on new bidding data of Novelis, the Notifying Party provided an update of their previous econometric analysis of the impact of Aleris’ participation on Novelis’ prices, margins and probability of winning.237
(325) The Notifying Party claims that their new analysis confirms the previous finding that the presence of Aleris has no significant effect on the prices, margins and winning probability of Novelis.
(326) The Commission disagrees with the Notifying Party’s conclusion and considers that the new econometric analysis does not allow concluding that Novelis and Aleris are distant competitors and Aleris does not have a material competitive impact on Novelis’ conduct.
(327) First, contrary to the approach taken in the previous report (where separate analyses had been presented for the tenders won and lost by Novelis), in the new submission the Notifying Party presents an analysis that pools together both the tenders won and the tenders lost by Novelis.238 This is a less preferable approach compared to the previous approach of focusing on the tenders won by Novelis for at least two reasons:
(a) The tenders won by Novelis are those in which the Novelis price was the one ultimately paid by the consumer;
(b) Including in the analysis the tenders lost by Novelis risks including many tenders in which Novelis did participate but with limited willingness to win (for example due to capacity constraints) and for which Novelis therefore put a high price irrespectively of Aleris’ presence.
(328) If the approach adopted in the previous report is applied to the new data (that is, analysing separately tenders won and tenders lost by Novelis), the main model used by the Notifying Party in the previous report shows that amongst the tenders won by Novelis the presence of Aleris has a statistically and economically significant effect on the prices and margins of Novelis. While such analysis was presented in the first table of the relevant section of the previous report of 30 November 2018 (table 7) the results of this same analysis are presented in an Appendix as the 8th table of the relevant section (table 14, see Figure 20).
Figure 20: […]239
[…]
(329) For the avoidance of doubt, the Commission, does not present the results above as definitive evidence that it is possible, based on the available data, to estimate with sufficient statistical certainty that Aleris has an economically and statistically significant effect on Novelis’ prices and margins. The Commission presents these results as an illustration that, as explained in the case of the first CRA econometric analysis, the data is not rich enough to be able to estimate with sufficient confidence the effect of Aleris on Novelis. The Notifying Party presented a number of analyses based on which the effect of Aleris on Novelis is not significant and the Commission presents here a number of alternative plausible assumptions/analysis under which Aleris turns out to have an economically and statistically significant effect on Novelis.
(330) The Notifying Party claims that the results should be disregarded because if one tender […] is dropped from the analysis, the results become again statistically non-significant.240 The Commission disagrees with the view that this tender should be removed.
(331) The […] tender that the Notifying Party now claims should be eliminated, was present also in the old dataset but in the report of 30 November 2018 there is no reference to the fact that it would be appropriate to drop the tender.
(332) The Notifying Party appears to have devoted such in-depth analysis ([…]) only to this single tender even though the same argument may apply to the other tenders, triggering further removals or even additions. In this respect the Commission concludes that:
(a) A systematic analysis of for all tenders is impossible because […];
(b) The […] tender that the Notifying Party wants to exclude is the tender […]. This makes its exclusion the most effective way of making the results again not statistically significant;
(c) The […] tender that the Notifying Party wants to exclude is for a customer ([…]) that responded to the Commission’s Questionnaire to Automotive customers that Novelis and Aleris are the closest competitors for each alloy in aluminium ABS. 241
(333) Second, the Commission notes that the econometrics results presented in tables 7, 8, 11, 12, 13 and 14 ([…]) of the report of 19 May 2019 change drastically if tenders corresponding to the 5xxx segment are removed from the bidding data.
(334) As explained in Section 2.2, Aleris has a very negligible share for 5xxx products and as a result the competitive pressure that it exerts on Novelis may be significantly lower for these products compared to 6xxx products. Indeed, in the updated bidding data of Novelis submitted by the Notifying Party, […]. The Notifying Party’s econometric analyses therefore combine, as in the former analysis of bidding data, two segments in which Aleris’ presence and expected impact on Novelis’ prices or margins may differ to a significant extent.
(335) In order to assess separately the impact of Aleris participation on Novelis’ prices and margins between 6xxx tenders and 5xxx tenders, the Commission has run the exact same specifications as presented by the Notifying Party in tables 7, 8, 11, 12, 13 and 14 on a dataset limited to 6xxx tenders. The removal of 5xxx tenders brings the total number of observations from […] to […] in the econometric analyses.
(336) All specifications, either for the base regression assessing the impact of Aleris participation on Novelis’ prices (table 7) or sensitivities of this regression (tables 11, 13 and 14), show a significant impact of Aleris participation. For the base regression, coefficients of Aleris participation are significant at the […]% level and range from […] to […]. This means that Aleris’ participation has generated on average a decrease in Novelis’ prices by […]%.
(337) Regarding the regressions assessing the impact of Aleris participation on Novelis’ margins, all specifications except Model 4 of regressions on gross and net margins for lost tenders show a significant impact of Aleris participation. For the base regression (table 8), the coefficients of Aleris participation are significant at the […] level and range from […] to […]. This means that Aleris’ participation […].
(338) Following the Notifying Party’s reasoning and preferred specifications, the Commission would interpret these results as showing that Aleris participation […].
(339) In the Reply to Annex I of the SO, the Notifying Party claims that the results become again statistically not significant if a set of fixed effects for each Tier 1 is added to the regression. The Commission considers that it is not surprising, in light of the limited variation in the data, that when adding more regressors to the estimating equation results become at some point non significant. The Commission’s point that the Notifying Party own analysis shows statistically significant results when focussing on the 6xxx segment was intended to illustrate how sensitive the Notifying Party’s results are to modelling assumptions. The Commission’s point was not intended to show that the econometric analysis unambiguously shows that Aleris has an impact on Novelis. To the contrary, the Commission main point is indeed that there is not enough variation in the data to answer this question.
(340) Third, the Commission notes that the econometrics results presented in Table 16 (impact of Aleris participation on Novelis’ prices or margins when dropping the […] tender) of the report of 19 May 2019 change drastically if the […] is included. […].242
(341) […].
Table 25: […]
[…]
Table 26: […]
[…]
(342) In the Reply to the Letter of Facts, the Notifying Party argues that […]. However, the Commission replicated the results presented in Table 25 and Table 26 by using logarithmic volumes in place of absolute volumes and the result that […] Novelis’ prices and margins remains after the transformation.
(343) Last, the Commission notes that as explained by the Notifying Party in the reply to RFI 45, the participation of Aleris in the tenders in which Novelis has participated has been defined at the overall tender level.243 That is, considering a tender comprising 20 components, if Aleris participated, say, in only 5, Aleris has been indicated as having participated for all components of that tender. This is possibly a severe measurement error in the participation of Aleris, which can generate considerable bias in the estimate of the impact of Aleris’ presence.
(344) Based on the points noted in paragraphs (327) to (343) the Commission considers that the new bidding analysis presented by the Notifying Party does not allow concluding that Novelis and Aleris are distant competitors and Aleris does not have a material competitive impact on Novelis’ conduct. If anything, several specifications of the models presented by the Notifying Party (or minor modifications of these models) show that Aleris has had an impact on the prices charged by Novelis.
(345) As a final consideration, the Commission notes that in the Reply to Annex I of the SO, the Notifying Party claims that the theoretical critiques presented by the Commission in the SO are not sufficient to support a finding of any harm based on the bidding analysis. In this respect, the Commission reiterates its view that the critiques presented in the SO and in this Decision are intended to substantiate that, contrary to the Notifying Party’s view, the Notifying Party’s econometric analysis of the bidding data does not allow the conclusion that Aleris does not have an impact on Novelis. The Commission emphasises that the available data are not rich enough to obtain statistically significant results in light of the complex amount of variables that need to be controlled for in the regression analysis. The references made by the Commission to results becoming statistically significant are not intended as a definitive proof of the impact of Aleris on Novelis’ prices or margins, but rather an indication that small reasonable changes in the Notifying Party’s analysis generate results that do not support the Notifying Party’s view that Aleris’ participation has an impact on Novelis. In conclusion, the Commission considers that the econometric analysis of the bidding data neither allows concluding with reliability that Aleris has a significant effect on Novelis’ prices and margin, but neither can it be used to claim the opposite (as the Notifying Party does).
4. APPENDIX II – THE IMPACT OF STRICTER REGULATORY REQUIREMENTS ON THE ELASTICITY OF ALUMINIUM ABS
(350) The Commission considers that in the coming years the elasticity of Aluminium ABS is likely to materially further decrease. This is because progressively more stringent regulatory requirements are expected to substantially increase OEMs’ demand for Aluminium ABS in the future, as acknowledged by the Notifying Party. For instance, Novelis’ own demand projection for Europe […].249
(351) From an economic perspective, the projected increase in demand for Aluminium ABS is highly likely to decrease its demand elasticity. Intuitively, the more strongly consumers desire (or even require) a given product, the more their demand will be inelastic to a small increase in the product’s price.
(352) To illustrate this, consider a linear demand function p(q) = a − bq , where p denotes the price, q is the quantity demanded and a and b are positive constants representing the intercept and slope of the demand function, respectively.
(353) Demand can grow for two reasons in this example: (a) either there is an increase in a, so the product becomes more attractive relative to other products over time; or (b) there is an increase in b, so the product’s addressable market grows over time.250 Here, we are interested in an increase in a, as regulation will increase the attractiveness of aluminium relative to steel for OEMs.251 It is therefore necessary to assess how a potential increase in a affects the product’s elasticity of demand. Formally, the elasticity of demand is given by
(354) Accordingly, an increase in leads to a decrease in the elasticity of demand, as 𝑑e
/𝑑a = −p/(a −p)2 < 0. 252 As regulation foresees progressively more stringent emission targets for OEMs over time, the elasticity of Aluminium ABS demand can therefore be expected to decline going forward. The competitive constraint exerted by steel is thus likely to decline even further in the upcoming years.
(355) In the Reply to the quantitative part of the Letter of Facts, the Notifying Party contested the Commission’s claim that elasticity is likely to be even lower going forward. In the Notifying Party’s view, the Commission’s claim is based on specific assumptions on the form of demand and supply. Moreover, the Notifying Party notes that in their view demand will be more elastic in the future because (1) […].
(356) In relation to the claim that the Commission’s claim rests on specific demand and supply assumptions, the Commission considers that the demand and supply conditions under which the demand elasticity increases as demand increases are unrealistic and regularly dismissed in the literature.253 Intuitively, the more strongly consumers desire (or even require) a given product, the more their demand will be inelastic to a small increase in the product’s price.
(357) As regards the claim that […], the Commission considers that it is far from clear that this will be the case. Since regulation is at the fleet level as opposed applying to each individual car model, further demand for Aluminium ABS could well mainly come from increase Aluminium share of material in large car models that already today employ Aluminium ABS to an extent. Moreover, even if demand were to come […], this would by no means imply that the elasticity of demand would increase. On the contrary also additional demand deriving […] is likely to decrease the elasticity of demand, as OEMs are less able to avoid purchasing Aluminium ABS for such vehicles.
(358) Finally, the Commission disagrees with the claim that the introduction of electric vehicles will reduce the need to lightweight, as extensively discussed in Section 6.2.2.3 of the Decision.254
Case M. 9076 – NOVELIS / ALERIS COMMITMENTS TO THE EUROPEAN COMMISSION
Pursuant to Article 8(2) of Council Regulation (EC) No 139/2004 (the Merger Regulation), Novelis Inc. (Novelis) and Aleris Corporation (Aleris; Novelis and Aleris are jointly referred to as the Parties) hereby enter into the following Commitments (the Commitments) vis-à-vis the European Commission (the Commission) with a view to render the acquisition of sole control over Aleris by Novelis (the Concentration) compatible with the internal market and the functioning of the EEA Agreement.
This text shall be interpreted in light of the Commission’s decision pursuant to Article 8(2) of the Merger Regulation to declare the Concentration compatible with the internal market and the functioning of the EEA Agreement (the Decision), in the general framework of European Union law, in particular in light of the Merger Regulation, and by reference to the Commission Notice on remedies acceptable under Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004 (the Remedies Notice).
The Commitments shall take effect upon the date of the Decision, provided that if the completion of the Transaction does not subsequently take place for whatever reason and is thereby abandoned, the Parties shall not be bound by these Commitments.
A. DEFINITIONS
1. For the purpose of the Commitments, the following terms shall have the following meaning:
ABS: aluminum automotive body sheets.
ABS Business: Aleris’ operations that concern the production of ABS in the EEA and sale to ABS customers.
Affiliated Undertakings: undertakings controlled by the Parties and/or by the ultimate parents of the Parties, whereby the notion of control shall be interpreted pursuant to Article 3 of the Merger Regulation and in light of the Commission Consolidated Jurisdictional Notice under Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings (the Consolidated Jurisdictional Notice).
Aleris: Aleris Corporation, with its registered office at 25825, Science Park Drive, suite 400, Cleveland, Ohio 44122-7392, USA.
Antimonopoly Law: the antimonopoly law of China.
Assets: the assets that contribute to the current operation or are necessary to ensure the viability and competitiveness of the Divestment Business as indicated in Section B, paragraph 7 (a), (b) and (c) and described more in detail in the Schedule.
Best Efforts: best effort obligations shall be interpreted in light of the Commission’s decision pursuant to Article 8(2) of the Merger Regulation to declare the Transaction compatible with the internal market and the functioning of the EEA Agreement, the Merger Regulation and the general principles of EU law. Any interpretation that may be given to this term under the law of other jurisdictions is not relevant for the purpose of interpreting and/or implementing the Commitments only.
Casthouse: all assets utilised for the casting of slabs.
Closing: the transfer of the legal title of the Divestment Business to the Purchaser.
Closing Period: the period of […] months from (i) the approval of the Purchaser and the terms of sale by the Commission or (ii) the Foreign Closing Date, whichever is the later, but ending in all instances the latest […] months from the approval of the Purchaser and the terms of sale by the Commission.
Combined Entity: entity resulting from the acquisition of sole control over Aleris by Novelis.
Confidential Information: any business secrets, know-how, commercial information, or any other information of a proprietary nature that is not in the public domain.
Conflict of Interest: any conflict of interest that impairs the Trustee’s objectivity and independence in discharging its duties under the Commitments.
Divestment Business: the business as defined in Section B, paragraphs 6 to 9, and in the Schedule, which Novelis commits to divest.
Divestiture Trustee: one or more natural or legal person(s) who is/are approved by the Commission and appointed by Novelis and who has/have received from the Parties the exclusive Trustee Mandate to sell the Divestment Business to a Purchaser at no minimum price.
Duffel Plant: plant located in A Stocletlaan 87, Duffel, 2570, Belgium, owned by Aleris Aluminium Duffel BVBA.
Effective Date: the date of adoption of the Decision.
First Divestiture Period: the period of […] months from the Effective Date.
Foreign Closing Date: the date that all of the applicable approvals, clearances or waiting periods under the HSR Act and the Antimonopoly Law, or agreements with the applicable governmental entity to refrain from closing relating to the Transaction shall have been obtained, expired or been terminated, as applicable, and no governmental order preventing closing shall be in effect.
FRPs: flat-rolled products.
Hold Separate Manager: the person appointed by the Parties for the Divestment Business to manage the day-to-day business under the supervision of the Monitoring Trustee.
HSR Act: the Hart-Scott-Rodino Antitrust Improvements Act of 1976, as amended, of the United States.
Key Personnel: all personnel necessary to maintain the viability and competitiveness of the Divestment Business, as listed in the Schedule, including the Hold Separate Manager.
Koblenz Plant: plant located in Carl-Spaeter-Strasse 10, Koblenz, 56070, Germany, owned by Aleris Rolled Products Germany GmbH.
Monitoring Trustee: one or more natural or legal person(s) who is/are approved by the Commission and appointed by Novelis, and who has/have the duty to monitor the Parties’ compliance with the conditions and obligations attached to the Decision.
Novelis: Novelis Inc., incorporated under the laws of Canada, with its registered office at 231 Church Street, Mississauga, Ontario L5M 1N1, Canada, and registered under number 960553-3.
Other FRPs: FRPs manufactured specifically at the Duffel Plant, which are not ABS.
Other FRPs Business: Aleris’ operations that concern the production and sale in the merchant market by the Duffel Plant of FRPs that are not ABS. The Other FRPs Business does not include the production and sale of FRPs carried out at any other Aleris’ or Novelis’ plants. In particular, the Other FRPs Business does not include the aerospace and the heat exchanger FRPs that […].
Parties: Novelis, Aleris and their respective Affiliated Undertakings also referred to as the Combined Entity.
Personnel: all staff currently employed by the Divestment Business or otherwise necessary to ensure the viability of the Divestment Business, including staff seconded to the Divestment Business, shared personnel as well as the additional personnel listed in the Schedule.
Purchaser: the entity approved by the Commission as acquirer of the Divestment Business in accordance with the criteria set out in Section D.
Purchaser Criteria: the criteria laid down in paragraph 22 of these Commitments that the Purchaser must fulfill in order to be approved by the Commission.
Schedule: the schedule to these Commitments describing more in detail the Divestment Business.
Transaction: the acquisition of sole control over Aleris by Novelis.
Transaction Closing: transfer of the legal title of Aleris to Novelis.
Trustee(s): the Monitoring Trustee and/or the Divestiture Trustee as the case may be.
Trustee Divestiture Period: the period of […] months from the end of the First Divestiture Period.
Voerde Plant: plant located in Schleusenstraße 11, Voerde (Niederrhein), 56562, Germany, owned by Aleris Casthouse Germany GmbH.
B. THE COMMITMENT TO DIVEST AND THE DIVESTMENT BUSINESS
2. In order to maintain effective competition, Novelis commits to divest, or procure the divestiture of the Divestment Business by the end of the Trustee Divestiture Period as a going concern to a Purchaser and on terms of sale approved by the Commission in accordance with the procedure described in paragraphs 23 and 24 of these Commitments. To carry out the divestiture, Novelis commits to find a Purchaser and to enter into a final binding sale and purchase agreement for the sale of the Divestment Business within the First Divestiture Period. If Novelis has not entered into such an agreement at the end of the First Divestiture Period, Novelis shall grant the Divestiture Trustee an exclusive mandate to sell the Divestment Business in accordance with the procedure described in paragraph 36 in the Trustee Divestiture Period.
3. The Transaction shall not be implemented before Novelis or the Divestiture Trustee has entered into a final binding sale and purchase agreement for the sale of the Divestment Business and the Commission has approved the purchaser (i.e. as an upfront buyer) and the terms of sale in accordance with paragraphs 23 and 24.
4. Novelis shall be deemed to have complied with this commitment if:
(a) by the end of the Trustee Divestiture Period, Novelis or the Divestiture Trustee has entered into a final binding sale and purchase agreement and the Commission approves the proposed purchaser and the terms of sale as being consistent with the Commitments in accordance with the procedure described in paragraphs 23 and 24; and
(b) the Closing of the sale of the Divestment Business to the Purchaser takes place within the Closing Period.
5. In order to maintain the structural effect of the Commitments, Novelis shall, for a period of 10 years after Closing, not acquire, whether directly or indirectly, the possibility of exercising influence (as defined in paragraph 43 of the Remedies Notice, footnote 3) over the whole or part of the Divestment Business, unless, following the submission of a reasoned request from Novelis showing good cause and accompanied by a report from the Monitoring Trustee (as provided in paragraph 50 of these Commitments), the Commission finds that the structure of the market has changed to such an extent that the absence of influence over the Divestment Business is no longer necessary to render the proposed concentration compatible with the internal market.
Structure and definition of the Divestment Business
6. The Divestment Business consists of (1) Aleris’ ABS Business, and (2) Other FRPs Business located at Aleris’ Duffel Plant.
7. The legal and functional structure of the Divestment Business as operated to date is described in the Schedule. The Divestment Business, described in more detail in the Schedule, includes all assets and staff that contribute to the current operation or are necessary to ensure the viability and competitiveness of the Divestment Business, in particular:
(a) all tangible assets (including the Duffel Plant and the land where the Duffel Plant is located) and intangible assets (including intellectual property rights, leasehold interests, and all easements and rights of way or access appurtenant to the Duffel Plant and land);
(b) all licences, permits and authorisations issued by any governmental organisation for the benefit of the Divestment Business;
(c) all contracts, leases, commitments and customer orders of the Divestment Business; all customer, credit and other records of the Divestment Business; and
(d) the Personnel.
8. The Divestment Business also includes the benefit, for a transitional period of up to […] after Closing and on terms and conditions equivalent to those at present afforded to the Divestment Business, of all current arrangements under which Aleris or its Affiliated Undertakings supply products or services to the Divestment Business, as detailed in the Schedule, unless otherwise agreed with the Purchaser. The period for such transitional supply arrangements may be further extended up to […], on a […] basis, if reasonably justified by the Purchaser under the oversight of Monitoring Trustee and Commission to preserve the viability and competitiveness of the Divestment Business. For the avoidance of doubt, in the performance of the contractual arrangements entered into pursuant to this clause, the Combined Entity shall prioritise the production and services to be supplied to the Purchaser over those to its own business and shall ensure a timely and effective supply. Strict firewall procedures will be adopted so as to ensure that any competitively sensitive information related to, or arising from such supply arrangements will not be shared with, or passed on to, anyone outside the relevant unit providing the product and service operations.
9. In addition, the Divestment Business includes certain […] arrangements with the Combined Entity on terms and conditions […], as detailed in the Schedule, to ensure the continuity of […] the Koblenz Plant retained by the Combined Entity, as well as of other […] arrangements.
C. RELATED COMMITMENTS
Preservation of viability, marketability and competitiveness
10. Until Closing, the Parties shall preserve or procure the preservation of the economic viability, marketability and competitiveness of the Divestment Business, in accordance with good business practice, and shall minimise as far as possible any risk of loss of competitive potential of the Divestment Business. In particular:
(a) the Parties undertake not to carry out any action that might have a significant adverse impact on the value, management or competitiveness of the Divestment Business or that might alter the nature and scope of activity, or the industrial or commercial strategy or the investment policy of the Divestment Business;
(b) the Parties undertake to make available, or procure to make available, sufficient resources for the development of the Divestment Business, on the basis and continuation of the existing business plans, including maintaining the Divestment Business’s current links with Aleris’s intercompany cash pool, and financing; and
(c) the Parties undertake to take all reasonable steps, or procure that all reasonable steps are being taken, including appropriate incentive schemes (based on industry practice), to encourage all Key Personnel to remain with the Divestment Business, and not to solicit or move any Personnel to the Parties’ remaining business. Where, nevertheless, individual members of the Key Personnel exceptionally leave the Divestment Business, the Parties shall provide a reasoned proposal to replace the person or persons concerned to the Commission and the Monitoring Trustee. The Parties must be able to demonstrate to the Commission that the replacement is well suited to carry out the functions exercised by those individual members of the Key Personnel. The replacement shall take place under the supervision of the Monitoring Trustee, who shall report to the Commission.
(d) Novelis undertakes to make available to the Purchaser CAPEX funding in the amounts specified below to fund the following three investment projects at the Duffel Plant:
i. CAPEX of EUR […] to install at the Duffel Plant a scalper […];
ii. CAPEX of EUR […] to modify the hot rolling mill currently installed at the Duffel Plant (the Duffel Hot Roll Mill) to enable it to hot roll automotive slabs […].
iii. CAPEX of EUR […] to modify the Casthouse currently installed (including if necessary the addition of new equipment) at the Duffel Plant (the Duffel Casthouse) to enable it to (1) cast aluminum slabs […] and (2) cast, in addition to the existing capacity, an additional volume of aluminum slabs […] in the last 12 months prior to 30 June 2019 and supplied to the Duffel Plant. […].
(e) The CAPEX funds will be at the disposal of the Purchaser, provided that: (i) the Purchaser demonstrates that the funds will be used to fund the three investment projects at the Duffel Plant mentioned above; and (ii) the Purchaser is in the final stages of implementing the investment projects requiring the funds within […] of the date of Closing. The relevant CAPEX funds for the three above described projects will be put in escrow.
Hold-separate obligations
11. The Parties commit, from the Effective Date until Closing, to keep the Divestment Business separate from the business(es) that the Parties are retaining, and to ensure that unless explicitly permitted under these Commitments: (i) management and staff of the business(es) retained by the Parties have no involvement in the Divestment Business; and (ii) the Key Personnel and Personnel of the Divestment Business have no involvement in any businesses retained by the Parties and do not report to any individual outside the Divestment Business.
12. Until Closing, the Parties shall assist the Monitoring Trustee in ensuring that the Divestment Business is managed as a distinct and saleable entity separate from the business(es) which the Parties are retaining.
13. Immediately after the adoption of the Decision, the Parties shall appoint a Hold Separate Manager. The Hold Separate Manager, who shall be part of the Key Personnel, shall manage the Divestment Business independently and in the best interest of the business with a view to ensuring its continued economic viability, marketability and competitiveness and its independence from the businesses retained by the Parties. The Hold Separate Manager shall closely cooperate with and report to the Monitoring Trustee and, if applicable, the Divestiture Trustee. Any replacement of the Hold Separate Manager shall be subject to the procedure laid down in paragraph 10(c) of these Commitments. The Commission may, after having heard the Parties, require the Parties to replace the Hold Separate Manager.
14. To ensure that the Divestment Business is held and managed as a separate entity, the Monitoring Trustee shall exercise Aleris’ (or Novelis’ if applicable) rights as shareholder in the legal entity or entities that constitute the Divestment Business (except for its rights in respect of dividends that are due before Closing), with the aim of acting in the best interest of the business, which shall be determined on a stand-alone basis, as an independent financial investor, and with a view to fulfilling the Parties’ obligations under the Commitments. Furthermore, the Monitoring Trustee shall have the power to replace members of the supervisory board or non-executive directors of the board of directors, who have been appointed on behalf of Aleris (or Novelis if applicable). Upon request of the Monitoring Trustee, Aleris (or Novelis if applicable) shall resign as a member of the boards or shall cause such members of the boards to resign.
Ring-fencing
15. From the Effective Date, the Parties shall implement, or procure to implement, all necessary measures to ensure that they do not, after the Effective Date, obtain any Confidential Information relating to the Divestment Business and that any such Confidential Information obtained by the Parties before the Effective Date will be eliminated and not be used by the Parties. This includes measures vis-à-vis Aleris’ (of Novelis’ if applicable) appointees on the supervisory board and/or board of directors of the Divestment Business. In particular, the participation of the Divestment Business in any central information technology network shall be severed to the extent possible, without compromising the viability of the Divestment Business. The Parties may obtain or keep information relating to the Divestment Business which is reasonably necessary for the divestiture of the Divestment Business or the disclosure of which to the Parties is required by law.
Non-solicitation clause
16. The Parties undertake, subject to customary limitations, not to solicit, and to procure that Affiliated Undertakings do not solicit, the Key Personnel transferred with the Divestment Business for a period of […] after Closing.
17. The Parties undertake, for a period of […] from the date of Closing, not to solicit, and to procure that Affiliated Undertakings do not solicit, any customers for the existing purchase orders and volumes they have with the Divestment Business on the Effective Date.
Clarification
18. Nothing in the above paragraphs 11–17 is intended to affect (1) any third party debt instrument, or (2) any intercompany debt, cash pool or other similar facilities now in effect, to which Aleris Aluminium Duffel BVBA is a party, guarantor, or obligor. Those instruments shall remain in full force and effect through the closing of the Transaction. For the avoidance of doubt, Aleris Aluminium Duffel BVBA shall remain a subsidiary of Aleris Corporation at all times until Closing and shall continue to be integrated in the external financial reporting of Aleris Corporation and nothing herein is intended to, or shall be construed as, limiting, changing, or affecting that relationship and such financial reporting until Closing.
Due diligence
19. In order to enable potential purchasers to carry out a reasonable due diligence of the Divestment Business, the Parties shall, subject to customary confidentiality assurances and dependent on the stage of the divestiture process:
(a) provide to potential purchasers sufficient information as regards the Divestment Business; and
(b) provide to potential purchasers sufficient information relating to the Personnel and allow them reasonable access to the Personnel.
Reporting
20. Novelis shall submit written reports in English on potential purchasers of the Divestment Business and developments in the negotiations with such potential purchasers to the Commission and the Monitoring Trustee no later than 10 days after the end of every month following the Effective Date (or otherwise at the Commission’s request). Novelis shall submit a list of all potential purchasers having expressed interest in acquiring the Divestment Business to the Commission at each and every stage of the divestiture process, as well as a copy of all the offers made by potential purchasers within five days of their receipt. Novelis shall inform the Commission within two working days of the applicable approvals, clearances or waiting periods under the HSR Act and the Antimonopoly Law, or agreements with the applicable governmental entity to refrain from closing relating to the Transaction have been obtained, expired or been terminated, as applicable, and no governmental order preventing closing is in effect, or other final decisions by the applicable governmental entity concerning the approval of the Transaction.
21. Novelis shall inform the Commission and the Monitoring Trustee on the preparation of the data room documentation and the due diligence procedure and shall submit a copy of any information memorandum to the Commission and the Monitoring Trustee before sending the memorandum out to potential purchasers.
D. THE PURCHASER
22. In order to be approved by the Commission, the Purchaser must fulfill the following criteria:
(a) the Purchaser shall be independent of and unconnected to the Parties and their Affiliated Undertakings (this being assessed having regard to the situation following the divestiture);
(b) the Purchaser shall have the financial resources, proven expertise (in particular in the aluminum and/or in the flat rolled products sector) and incentive to maintain and develop the Divestment Business as a viable and active competitive force in competition with the Combined Entity and other competitors; having adequately proven to the Commission that the Purchaser will effectively and cost efficiently replace the current supply relationships between the Duffel Plant and other Aleris plants in the medium term, whether through investment in the Duffel Plant (with or without the CAPEX funding specified above in paragraph 10(d) and (e)), existing or planned capabilities in the Purchaser’s other plants, or third-party supply relationships; and
(c) the acquisition of the Divestment Business by the Purchaser must neither be likely to create, in light of the information available to the Commission, prima facie competition concerns nor give rise to a risk that the implementation of the Commitments will be delayed. In particular, the Purchaser must reasonably be expected to obtain all necessary approvals from the relevant regulatory authorities for the acquisition of the Divestment Business.
23. The final binding sale and purchase agreement (as well as ancillary agreements) relating to the divestment of the Divestment Business shall be conditional on the Commission’s approval, and may be made conditional on the Commission granting a waiver under the second sentence of paragraph 24 below. When Novelis has reached an agreement with a purchaser, it shall submit a fully documented and reasoned proposal, including a copy of the final agreement(s), within one week to the Commission and the Monitoring Trustee. Novelis must be able to demonstrate to the Commission that the purchaser fulfils the Purchaser Criteria and that the Divestment Business is being sold in a manner consistent with the Commission's Decision and the Commitments. For the approval, the Commission shall verify that the purchaser fulfils the Purchaser Criteria and that the Divestment Business is being sold in a manner consistent with the Commitments including their objective to bring about a lasting structural change in the market.
24. The Commission may approve the sale of the Divestment Business without one or more Assets or parts of the Personnel, or by substituting one or more Assets or parts of the Personnel with one or more different assets or different personnel if this does not affect the viability and competitiveness of the Divestment Business after the sale, taking account of the proposed purchaser. In particular, the Commission shall waive, in whole or in part, the Purchaser CAPEX funding commitment set out in paragraph 10(d) and (e) above, if the Purchaser demonstrates that it already has assets or access to inputs, that make the CAPEX funding unnecessary and provided that the waiver of such CAPEX funding does not affect the viability and competitiveness of the Divestment Business after the sale.
E. TRUSTEE
I. Appointment procedure
25. Novelis shall appoint a Monitoring Trustee to carry out the functions specified in these Commitments for a Monitoring Trustee. Novelis commits not to close the Transaction before the appointment of a Monitoring Trustee.
26. If Novelis has not entered into a binding sale and purchase agreement regarding the Divestment Business […] before the end of the First Divestiture Period or if the Commission has rejected a purchaser proposed by Novelis at that time or thereafter, Novelis shall appoint a Divestiture Trustee. The appointment of the Divestiture Trustee shall take effect upon the commencement of the Trustee Divestiture Period.
27. The Trustee shall:
i. at the time of appointment, be independent of the Parties and their Affiliated Undertakings;
ii. possess the necessary qualifications to carry out its mandate, for example have sufficient relevant experience as an investment banker or consultant or auditor; and
iii. neither have nor become exposed to a Conflict of Interest.
28. The Trustee shall be remunerated by Novelis in a way that does not impede the independent and effective fulfillment of its mandate. In particular, where the remuneration package of a Divestiture Trustee includes a success premium linked to the final sale value of the Divestment Business, such success premium may only be earned if the divestiture takes place within the Trustee Divestiture Period.
Proposal by Novelis
29. No later than […] after the Effective Date, Novelis shall submit the name or names of one or more natural or legal persons whom Novelis proposes to appoint as the Monitoring Trustee to the Commission for approval. No later than […] before the end of the First Divestiture Period or on request by the Commission, Novelis shall submit a list of one or more persons whom Novelis proposes to appoint as Divestiture Trustee to the Commission for approval. The proposal shall contain sufficient information for the Commission to verify that the person or persons proposed as Trustee fulfill the requirements set out in paragraph 27 and shall include:
(a) the full terms of the proposed mandate, which shall include all provisions necessary to enable the Trustee to fulfil its duties under these Commitments;
(b) the outline of a work plan which describes how the Trustee intends to carry out its assigned tasks; and
(c) an indication whether the proposed Trustee is to act as both Monitoring Trustee and Divestiture Trustee or whether different trustees are proposed for the two functions.
Approval or rejection by the Commission
30. The Commission shall have the discretion to approve or reject the proposed Trustee(s) and to approve the proposed mandate subject to any modifications it deems necessary for the Trustee to fulfil its obligations. If only one name is approved, Novelis shall appoint or cause to be appointed the person or persons concerned as Trustee, in accordance with the mandate approved by the Commission. If more than one name is approved, Novelis shall be free to choose the Trustee to be appointed from among the names approved. The Trustee shall be appointed within one week of the Commission’s approval, in accordance with the mandate approved by the Commission.
New proposal by Novelis
31. If all the proposed Trustees are rejected, Novelis shall submit the names of at least two more natural or legal persons within one week of being informed of the rejection, in accordance with paragraphs 25 and 30 of these Commitments.
Trustee nominated by the Commission
32. If all further proposed Trustees are rejected by the Commission, the Commission shall nominate a Trustee, whom Novelis shall appoint, or cause to be appointed, in accordance with a trustee mandate approved by the Commission.
II. Functions of the Trustee
33. The Trustee shall assume its specified duties and obligations in order to ensure compliance with the Commitments. The Commission may, on its own initiative or at the request of the Trustee or Novelis, give any orders or instructions to the Trustee in order to ensure compliance with the conditions and obligations attached to the Decision.
Duties and obligations of the Monitoring Trustee
34. The Monitoring Trustee shall:
(i) propose in its first report to the Commission a detailed work plan describing how it intends to monitor compliance with the obligations and conditions attached to the Decision;
(ii) oversee, in close co-operation with the Hold Separate Manager, the on-going management of the Divestment Business with a view to ensuring its continued economic viability, marketability and competitiveness and monitor compliance by the Parties with the conditions and obligations attached to the Decision. To that end the Monitoring Trustee shall:
(a) monitor the preservation of the economic viability, marketability and competitiveness of the Divestment Business, and the keeping separate of the Divestment Business from the business retained by the Parties, in accordance with paragraphs 10 and 11 of these Commitments, including the power to verify that the CAPEX funding is made available by Novelis and is used by the Purchaser in line with the principles described in paragraph 10(d) and (e);
(b) supervise the management of the Divestment Business as a distinct and saleable entity, in accordance with paragraph 12 of these Commitments;
(c) with respect to Confidential Information:
- determine all necessary measures to ensure that the Parties do not after the Effective Date obtain any Confidential Information relating to the Divestment Business,
- in particular strive for the severing of the Divestment Business’ participation in a central information technology network to the extent possible, without compromising the viability of the Divestment Business,
- make sure that any Confidential Information relating to the Divestment Business obtained by the Parties before the Effective Date is eliminated and will not be used by the Parties and decide whether such information may be disclosed to or kept by the Parties as the disclosure is reasonably necessary to allow Novelis to carry out the divestiture or as the disclosure is required by law; and
(d) monitor the splitting of assets and the allocation of Personnel between the Divestment Business and the Combined Entity or Affiliated Undertakings;
(iii) propose to the Parties such measures as the Monitoring Trustee considers necessary to ensure the Parties’ compliance with the conditions and obligations attached to the Decision, in particular the maintenance of the full economic viability, marketability or competitiveness of the Divestment Business, the holding separate of the Divestment Business and the non-disclosure of competitively sensitive information;
(iv) review and assess potential purchasers as well as the progress of the divestiture process and verify that, dependent on the stage of the divestiture process:
(a) potential purchasers receive sufficient and correct information relating to the Divestment Business and the Personnel in particular by reviewing, if available, the data room documentation, the information memorandum and the due diligence process; and
(b) potential purchasers are granted reasonable access to the Personnel.
(v) act as a contact point for any requests by third parties, in particular potential purchasers, in relation to the Commitments;
(vi) provide to the Commission, sending Novelis a non-confidential copy at the same time, a written report within 15 days after the end of every month that shall cover the operation and management of the Divestment Business as well as the splitting of assets and the allocation of Personnel so that the Commission can assess whether the business is held in a manner consistent with the Commitments and the progress of the divestiture process as well as potential purchasers;
(vii) promptly report in writing to the Commission, sending the Parties a non-confidential copy at the same time, if it concludes on reasonable grounds that the Parties is failing to comply with these Commitments;
(viii) within one week after receipt of the documented proposal referred to in paragraphs 23 and 24 of these Commitments, submit to the Commission, sending Novelis a non-confidential copy at the same time, a reasoned opinion as to the suitability and independence of the proposed purchaser and the viability of the Divestment Business after the sale and as to whether the Divestment Business is sold in a manner consistent with the conditions and obligations attached to the Decision, in particular, if relevant, whether the sale of the Divestment Business without one or more Assets or not all of the Personnel affects the viability of the Divestment Business after the sale, taking account of the proposed purchaser;
(ix) assume the other functions assigned to the Monitoring Trustee under the conditions and obligations attached to the Decision.
35. If the Monitoring and Divestiture Trustee are not the same persons, the Monitoring Trustee and the Divestiture Trustee shall cooperate closely with each other during and for the purpose of the preparation of the Trustee Divestiture Period in order to facilitate each other's tasks.
Duties and obligations of the Divestiture Trustee
36. Within the Trustee Divestiture Period, the Divestiture Trustee shall sell at no minimum price the Divestment Business to a purchaser, provided that the Commission has approved both the purchaser and the final binding sale and purchase agreement (and ancillary agreements) as in line with the Commission's Decision and the Commitments in accordance with paragraphs 22 to 24 of these Commitments. The Divestiture Trustee shall include in the sale and purchase agreement (as well as in any ancillary agreements) such terms and conditions as it considers appropriate for an expedient sale in the Trustee Divestiture Period. In particular, the Divestiture Trustee may include in the sale and purchase agreement such customary representations and warranties and indemnities as are reasonably required to effect the sale. The Divestiture Trustee shall protect the legitimate financial interests of Novelis, subject to Novelis’ unconditional obligation to divest at no minimum price in the Trustee Divestiture Period.
37. In the Trustee Divestiture Period (or otherwise at the Commission’s request), the Divestiture Trustee shall provide the Commission with a comprehensive monthly report written in English on the progress of the divestiture process. Such reports shall be submitted within 15 days after the end of every month with a simultaneous copy to the Monitoring Trustee and a non-confidential copy to Novelis.
III. Duties and obligations of the Parties
38. The Parties shall provide and shall cause its advisors to provide the Trustee with all such co- operation, assistance and information as the Trustee may reasonably require to perform its tasks. The Trustee shall have full and complete access to the Parties or the Divestment Business’ books, records, documents, management or other personnel, facilities, sites and technical information necessary for fulfilling its duties under the Commitments and the Parties and the Divestment Business shall provide the Trustee upon request with copies of any document. Novelis and the Divestment Business shall make available to the Trustee one or more offices on their premises and shall be available for meetings in order to provide the Trustee with all information necessary for the performance of its tasks.
39. The Parties shall provide the Monitoring Trustee with all managerial and administrative support that it may reasonably request on behalf of the management of the Divestment Business. This shall include all administrative support functions relating to the Divestment Business which are currently carried out at headquarters level. Novelis shall provide and shall cause its advisors to provide the Monitoring Trustee, on request, with the information submitted to potential purchasers, in particular give the Monitoring Trustee access to the data room documentation and all other information granted to potential purchasers in the due diligence procedure. Novelis shall inform the Monitoring Trustee on possible purchasers, submit lists of potential purchasers at each stage of the selection process, including the offers made by potential purchasers at those stages, and keep the Monitoring Trustee informed of all developments in the divestiture process.
40. The Parties shall grant or procure Affiliated Undertakings to grant comprehensive powers of attorney, duly executed, to the Divestiture Trustee to effect the sale (including ancillary agreements), the Closing and all actions and declarations which the Divestiture Trustee considers necessary or appropriate to achieve the sale and the Closing, including the appointment of advisors to assist with the sale process. Upon request of the Divestiture Trustee, the Parties shall cause the documents required for effecting the sale and the Closing to be duly executed.
41. The Parties shall indemnify the Trustee and its employees and agents (each an Indemnified Party) and hold each Indemnified Party harmless against, and hereby agrees that an Indemnified Party shall have no liability to the Parties for, any liabilities arising out of the performance of the Trustee’s duties under the Commitments, except to the extent that such liabilities result from the willful default, recklessness, gross negligence or bad faith of the Trustee, its employees, agents or advisors.
42. At the expense of Novelis, the Trustee may appoint advisors (in particular for technical expertise, corporate finance or legal advice), subject to Novelis’ approval (this approval not to be unreasonably withheld or delayed) if the Trustee considers the appointment of such advisors necessary or appropriate for the performance of its duties and obligations under the Mandate, provided that any fees and other expenses incurred by the Trustee are reasonable. Should Novelis refuse to approve the advisors proposed by the Trustee the Commission may approve the appointment of such advisors instead, after having heard Novelis. Only the Trustee shall be entitled to issue instructions to the advisors. Paragraph 41 of these Commitments shall apply mutatis mutandis. In the Trustee Divestiture Period, the Divestiture Trustee may use advisors who served Novelis during the Divestiture Period if the Divestiture Trustee considers this in the best interest of an expedient sale.
43. The Parties agree that the Commission may share Confidential Information proprietary to the Parties with the Trustee. The Trustee shall not disclose such information and the principles contained in Article 17 (1) and (2) of the Merger Regulation apply mutatis mutandis.
44. The Parties agree that the contact details of the Monitoring Trustee are published on the website of the Commission's Directorate-General for Competition and they shall inform interested third parties, in particular any potential purchasers, of the identity and the tasks of the Monitoring Trustee.
45. For a period of 10 years from the Effective Date, the Commission may request all information from the Parties that is reasonably necessary to monitor the effective implementation of these Commitments.
IV. Replacement, discharge and reappointment of the Trustee
46. If the Trustee ceases to perform its functions under the Commitments or for any other good cause, including the exposure of the Trustee to a Conflict of Interest:
(a) the Commission may, after hearing the Trustee and Novelis, require Novelis to replace the Trustee; or
(b) Novelis may, with the prior approval of the Commission, replace the Trustee.
47. If the Trustee is removed according to paragraph 46 of these Commitments, the Trustee may be required to continue in its function until a new Trustee is in place to whom the Trustee has effected a full hand over of all relevant information. The new Trustee shall be appointed in accordance with the procedure referred to in paragraphs 25-32 of these Commitments.
48. Unless removed according to paragraph 46 of these Commitments, the Trustee shall cease to act as Trustee only after the Commission has discharged it from its duties after all the Commitments with which the Trustee has been entrusted have been implemented. However, the Commission may at any time require the reappointment of the Monitoring Trustee if it subsequently appears that the relevant remedies might not have been fully and properly implemented.
F. THE REVIEW CLAUSE
49. The Commission may extend the time periods foreseen in the Commitments in response to a request from Novelis or, in appropriate cases, on its own initiative. Where Novelis requests an extension of a time period, it shall submit a reasoned request to the Commission no later than 1 month before the expiry of that period, showing good cause. This request shall be accompanied by a report from the Monitoring Trustee, who shall, at the same time send a non-confidential copy of the report to Novelis. Only in exceptional circumstances shall Novelis be entitled to request an extension within the last month of any period.
50. The Commission may further, in response to a reasoned request from Novelis showing good cause waive, modify or substitute, in exceptional circumstances, one or more of the undertakings in these Commitments. This request shall be accompanied by a report from the Monitoring Trustee, who shall, at the same time send a non-confidential copy of the report to Novelis. The request shall not have the effect of suspending the application of the undertaking and, in particular, of suspending the expiry of any time period in which the undertaking has to be complied with.
G. ENTRY INTO FORCE
51. The Commitments shall take effect upon the date of adoption of the Decision.
[Signed]
duly authorized for an on behalf of Novelis [Date]
[Signed]
duly authorized for an on behalf of Aleris
[Date]
Case COMP/M. 9076 – NOVELIS / ALERIS
SCHEDULE
The Divestment Business
1. In accordance with paragraph 6 of these Commitments, the Divestment Business consists of (1) Aleris’ ABS Business, and (2) Other FRPs Business located at Aleris’ Duffel Plant.
2. The Divestment Business is incorporated as Aleris Aluminium Duffel BVBA, which owns and operates the ABS Business and Other FRPs Business. The Divestment Business includes, but is not limited to:
i. all tangible assets that are necessary to operate the Divestment Business, including the Duffel Plant and the land where the Duffel Plant is located, as detailed in Annex 1;
ii. all intangible assets, including intellectual property, leasehold interests, and all easements and rights of way or access appurtenant to the Duffel Plant and land, necessary to operate the Divestment Business, as detailed in Annex 2;
iii. all licenses, permits and authorizations issued by any governmental organization for the benefit of the Divestment Business;
iv. all contracts, leases and commitments of the Divestment Business, including supply agreements of metal products (primary ingot and slabs), as detailed in Annex 3;
v. all customer lists (including the customers detailed in Annex 4), customer contracts and/or purchase orders related to the Divestment Business. Novelis and Aleris undertake to use Best Efforts to transfer to the Purchaser all Aleris’ relationships with current Divestment Business customers. Customers shall have their existing agreements novated to the Purchaser and shall be released from any further purchasing obligations from the Combined Entity;
vi. in line with applicable employment laws and other relevant legislation, the Personnel and Key Personnel necessary to operate and ensure the viability of the Divestment Business, as specified in Annex 5;
vii. all EEA ABS-related R&D activities as detailed in Annex 6, as well as all Other FRPs-related R&D activities conducted at the Duffel Research Center. For avoidance of doubt, this also includes (i) all ABS projects that the R&D center in Aachen is currently working on with the Duffel Research Center; and (ii) the ABS-related assets located at Aachen as well as the associated employees, unless Novelis demonstrates that it is not feasible to include it in the Divestment Business.
3. In accordance with paragraph 8 of these Commitments, at the option of the Purchaser, for a transitional period of up to […] from Closing, or such further extended period, the Combined Entity will enter into the following contractual arrangements, […], with the Purchaser and supply under the arrangements:
i. agreement for the supply of […], covering the volumes and widths required by the Duffel Plant from these […] according to Aleris’ existing business plans or volume projections, and in any event and at the option of the Purchaser not less than the amounts supplied in the last 12 months prior to 30 June 2019.
ii. agreement for the […], covering the volumes required by the Duffel Plant according to Aleris’ existing business plans or volume projections, and in any event and at the option of the Purchaser not less than the amounts supplied in the last 12 months prior to 30 June 2019.
iii. agreement for the supply of […], covering the volumes required by the Duffel Plant according to Aleris’ existing business plans or volume projections, and in any event and at the option of the Purchaser not less than the amounts supplied in the last 12 months prior to 30 June 2019.
iv. agreements for the provision of services that the Purchaser may require during a startup phase, which may comprise: (a) assistance with typical human resources tasks, such as payroll and other HR related services; (b) order management, production planning, order handling and finance services (such as billing, accounts receivable etc.); (c) information services and software licenses necessary for sales handling and running the Divestment Business; (d) IT and telephony services; and (e) assistance with organizing logistics for supplies of FRPs to the Divestment Business’ customers.
In the performance of the contractual arrangements entered into pursuant to this clause, the Combined Entity shall prioritise the production and services to be supplied to the Purchaser over those to its own business and shall ensure a timely and effective supply.
Assets excluded from the Divestment Business
4. For the avoidance of doubt, the Divestment Business shall not include the following, as detailed in
Annex 7:
i. the […] at the Koblenz Plant;
ii. the […] at the Voerde Plant;
iii. the […] at the Koblenz Plant;
iv. the […] at the Koblenz Plant;
For the avoidance of doubt, the Divestment Business does not include any assets, contracts or staff related to the production and sale of non-ABS FRPs by any of Aleris’ and Novelis’ plants other than the Duffel Plant and not related to the production and sale of non-ABS FRPs by the Duffel plant. In particular, the Other FRPs Business does not include the aerospace and the heat exchanger FRPs that the […].
5. In accordance with paragraph 9 of these Commitments, the Combined Entity and the Purchaser will also enter into the following contractual arrangements:
i. […] agreement for the processing of […] products with the […] located at the Duffel Plant and specified in Annex 8 on a […] basis for a minimum duration of […] after Closing. […].
ii. for a transitional period of up to […] from Closing, an agreement for the shipment of […], and in any event and at the option of the Purchaser not less than the amounts supplied in the last 12 months prior to 30 June 2019. […].
iii. for a transitional period of up to […] from Closing, an agreement for the […] by the Duffel Plant on a […] basis according to Aleris’ existing business plans or volume projections. […].
6. If there is any asset or personnel which is not be covered by paragraph 2 and 3 of this Schedule but which is both used (exclusively or not) in the Divestment Business and necessary for the continued viability and competitiveness of the Divestment Business, that asset or adequate substitute will be offered to potential purchasers.
ANNEX 1 – TANGIBLE ASSETS RELATED TO THE DIVESTMENT BUSINESS: THE DUFFEL PLANT INSTALLED EQUIPMENT
[…]
ANNEX 2 – PATENTS RELATED TO THE DIVESTMENT BUSINESS
[…]
ANNEX 3 – KEY SUPPLY CONTRACTS RELATED TO THE DIVESTMENT BUSINESS
[…]
ANNEX 4 – CUSTOMER LIST RELATED TO THE DIVESTMENT BUSINESS
[…]
ANNEX 5 – PERSONNEL AND KEY PERSONNEL RELATED TO THE DIVESTMENT BUSINESS
[…]
ANNEX 6 – R&D ACTIVITIES RELATED TO EEA ABS AND OTHER FRPS R&D ACTIVITIES CONDUCTED AT THE DUFFEL RESEARCH CENTER
[…]
ANNEX 7 – ASSETS EXCLUDED FROM THE DIVESTMENT BUSINESS
[…]
ANNEX 8 – EQUIPMENT LOCATED AT THE DUFFEL PLANT USED FOR PROCESSING INTERMEDIATE AEROSPACE PRODUCTS
[…]
1 OJ L 24, 29.1.2004, p. 1 (‘the Merger Regulation’). With effect from 1 December 2009, the Treaty on the Functioning of the European Union (‘TFEU’) has introduced certain changes, such as the replacement of ‘Community’ by ‘Union’ and ‘common market’ by ‘internal market’. The terminology of the TFEU will be used throughout this decision.
2 Publication in the Official Journal of the European Union No C 70, 25.02.2019, p. 12.
3 References to Novelis throughout this Decision, in particular in the context of references to internal documents, will also include Hindalco.
4 Also known as ’aluminum’ in certain versions of English. The Commission uses the European terminology throughout this Decision. In direct quotes, the terminology used by the market participant has been maintained.
5 […].
6 […].
7 […].
8 Phase I questionnaire to competitors, DocID2073; Phase I questionnaire to customers of other FRPs, DocID2092; Phase I questionnaire to automotive customers, DocID2094.
9 Phase II – Questionnaire to OEMs, DocID2071; Phase II questionnaire to customers of other FRPs, DocID2083; Phase II questionnaire to competitors of other FRPs, DocID2107.
10 OJ L 133, 30.4.2004, p.1.
11 CRA data room report, DocID2454.
12 CRA data room report – Letter of Facts, DocID2880.
13 Aluminium minerals also have non-metallic applications. However, the present case only concerns metallic aluminium and the other applications are not discussed. ‘Aluminium’ in this Decision invariably refers to metallic aluminium unless explicitly indicated to the contrary.
14 This smelting process is carried out in electrolytic cells, in which the carbon cathode placed in the bottom of the cells forms the negative electrode. Anodes, which are made of carbon, are consumed during the electrolytic process when the anode reacts with the oxygen in the alumina to form CO2.
15 7xxx series (Al-Zn) are heat-treatable alloys in which zinc is the principal alloying element (although other elements, such as copper and magnesium may be specified). The 7xxx series are the strongest aluminium alloys and widely used in the aerospace industry, but currently not used to any meaningful extent for automotive applications.
16 Form CO, paragraph 388.
17 Replies to question 43 of Questionnaire to Automotive Customers, DocID2094; Reply to request for information 1, Annex 24, pages 6–11 and 48, DocID74-665.
18 There might be cases where ‘Tier 2’ suppliers purchase Aluminium ABS and sell the resulting products to a ‘Tier 1’ supplier, which, in turn, sells the resulting products to OEMs. However, for the purpose of the present Decision, no distinction is made between Tier 1 and Tier 2 suppliers, and they are both referred to as Tier suppliers.
19 Form CO, paragraph 104 et seq.
20 Reply to the request for information 12, DocID699.
21 Reply to request for information 12, DocID699.
22 Reply to request for information 42, DocID2137; minutes of calls with customers on 28.11.2018, DocID140, 5.12.2018, DocID792, 15.5.2019, DocID1990 and 2.4.2019, DocID1414; and minutes of a call with a competitor on 19.12.2018, DocID733.
23 Minutes of a call with a competitor on 19.12.2018, DocID733 and minutes of call with customers on 17.4.2019, DocID1420 and on 2.4.2019, DocID1978.
24 Minutes of a call with a competitor on 19.12.2018, DocID733.
25 Minutes of a call with a customer on 17.4.2019, DocID1420.
26 Minutes of a call with a customer on 17.4.2019, DocID1420.
27 Minutes of a call with a competitor on 19.122018, DocID733.
28 Minutes of a call with a customer on 17.4.2019, DocID1420.
29 See Section 8.3.7.2 for discussion of CASH line capacity as the relevant capacity metric (also in contrast to batch annealing capacity).
30 Reply to request for information 22, DocID1018.
31 […].
32 Reply to request for information 32, DocID1663-7.
33 […].
34 […].
35 Aluminium FRP manufacturers neither fix nor influence the quotation of LME prices.
36 Conversion revenues or price, also referred to as ‘fabrication’ or ‘processing’ price, is a term commonly used in the Aluminium ABS industry, and, more generally, the Aluminium FRP industry. It is the price charged to customers for the fabrication of a certain Aluminium FRP and includes the cost of casting, rolling and CASH processing. This price excludes metal price (LME).
37 For example, according to […] provided in Annex Q4-N1 to the Reply to request for information 32, DocID1663-78).
38 […].
39 […].
40 Minutes of a call with a competitor on 15.3.2018, DocID900.
41 […].
42 Minutes of a call with a competitor on 15.3.2018, DocID900.
43 Minutes of calls with customers on 12.4.2019, DocID1031; on 11.4.2019, DocID1444; and on 29.5.2019, DocID2224.
44 Minutes of a call with a competitor on 19.12.2018, DocID733.
45 Minutes of a call with a competitor on 15.3.2018, DocID900.
46 Form CO, paragraph 76.
47 Form CO, paragraph 77.
48 Reply to request for information 32, DocID1663-7.
49 Minutes of a call with a customer on 2.4.2019, DocID1978.
50 Minutes of a call with a customer on 15.5.2019, DocID1990.
51 Form CO, paragraph 7.
52 OJ L 282, 19.10.2016, p. 4. The Paris Agreement sets out, inter alia, a long-term goal in line with the objective to keep the global average temperature increase well below 2 °C above pre-industrial levels and to pursue efforts to keep it to 1.5 °C above pre-industrial levels.
53 Regulation (EC) 443/2009 of the European Parliament and of the Council of 23 April 2009 setting emission performance standards for new passenger cars as part of the Community’s integrated approach to reduce CO2 emissions from light-duty vehicles (the ‘2009 Regulation on CO2 emissions standards’) (OJ L 140, 5.6.2009, page 1).
54 See https://ec.europa.eu/clima/policies/transport/vehicles/cars en.
55 Regulation (EU) 2019/631 of the European Parliament and of the Council of 17 April 2019 setting CO2 emission performance standards for new passenger cars and for new light commercial vehicles, and repealing Regulations (EC) No 443/2009 and (EU) No 510/2011 (OJ L 111, 25.4.2019, page 13).
56 Reply to the Letter of Facts, paragraphs 33–34.
57 Minutes of calls with customers on 28.11.2018, DocID140; 24.1.2019, DocID822; 15.5.2019, DocID1990; and 17.4.2019, DocID1420; and with competitors on 14.5.2019, DocID1960.
58 Minutes of a call with a customer on 28.11.2018, DocID140.
59 Minutes of a call with a customer on 15.5.2019, DocID1990.
60 Minutes of a call with a competitor on 12.12.2018, DocID737.
61 Form CO, ‘Goldman Sachs Equity Research 2018’, DocID49, slide 9.
62 In previous decisions, the term ‘Aluminium ABS’ has not been used. Nonetheless, in Alcan / Pechiney (II), ‘automotive sheet’ refers to a product used in the production of automotive body parts. M.3226 – Alcan / Pechiney (II), paragraph 60.
63 See, for example, M.4605 – Hindalco / Novelis, paragraph 13.; M.3226 – Alcan / Pechiney (II), paragraphs 58–65; M.2702 – Norsk Hydro / VAW, paragraph 12; and M.1663 – Alcan / Alusuisse, recital 54.
64 See, for example, M.2111 – Alcoa / British Aluminium, paragraph 9.
65 See, for example, for example, M.4605 – Hindalco / Novelis, paragraph 13; M.3226 – Alcan / Pechiney (II), paragraph 61; and M.2702 – Norsk Hydro / VAW, paragraph 11.
66 Form CO, paragraph 65.
67 Certain other types of Aluminium FRPs also require for instance continuous annealing (for example certain products supplied to the aerospace industry).
68 See the names of the respondents to Phase II - Questionnaire to competitors in other (standard) FRP, DocID2083.
69 As explained in Section 8.3.4.1, the JV between Alcoa and Ma’aden supplies some types of Aluminium ABS to the EEA market.
70 See for example, Reply to request for information 32, Annex 11, ‘VORI0017 20171024’, DocID1660, pages 2–3; and Reply to request for information 19 ‘ALE00721969.pdf’, slide 2, DocID976-47416.
71 Replies to question 33 of Questionnaire to Automotive Customers, DocID2094; Minutes of calls with customers on 28.11.2018, DocID140; on 24.1.2019, DocID822; on 17.4.2019, DocID1420; on 15.5.2019, DocID1990; and on 20.5.2019, DocID2123.
72 […].
73 Reply to the Letter of Facts, paragraphs 5-104; Reply to the SO, paragraphs 38-133; Reply to the Article 6(1)(c) Decision, paragraphs 122-150; Form CO, paragraphs 79-116.
74 Reply to the SO, paragraphs 48-60.
75 Reply to the SO, paragraphs 70-78.
76 Reply to the SO, paragraphs 61-69; Reply to the Article 6(1)(c) Decision, paragraphs 129-135.
77 Reply to the SO, paragraphs 64-65.
78 Reply to the SO, paragraph 63.
79 Reply to the SO, paragraphs 121-127 and Section 2.1 of its Annex I.
80 Reply to the Letter of Facts, paragraphs 5-13; Reply to the SO, paragraphs 103-120; Reply to the Article 6(1)(c) Decision, paragraphs 136-144.
81 Reply to the SO, paragraphs 82-102; Reply to the Article 6(1)(c) Decision, paragraphs 153-155.
82 M.3225 – Alcan / Pechiney (II), paragraph 60. See also M.4605 – Hindalco / Novelis, paragraph 13.
83 M.3225 – Alcan / Pechiney (II), paragraphs 60 and 164–5.
84 M.8713 – Tata Steel/ThyssenKrupp/JV, Commission decision of 11.6.2019. See press release of 11 June 2019, http://europa.eu/rapid/press-release IP-19-2948 en htm.
85 M.3225 – Alcan / Pechiney (II), paragraph 60.
86 Reply to the SO, paragraph 39. See also Reply to the Letter of Facts, paragraph 17.
87 M.8713 – Tata Steel/ThyssenKrupp/JV, Commission decision of 11.6.2019. See press release of 11 June 2019 , http://europa.eu/rapid/press-release IP-19-2948 en.htm.
88 Reply to question 1 of Questionnaire to Competitors, DocID2073, pages 6, 33.
89 Minutes of a call with a competitor 12.12.2018, DocID737, paragraph 1.
90 Reply to question 1 of Questionnaire to Competitors, DocID2073, pages 85.
91 Minutes of a call with a competitor 14.5.2019, DocID1960, paragraph 1.
92 Some of the steel suppliers, such as Voestalpine use Aluminium ABS in their downstream component manufacturing or distribute such products. However, they are not active in the production of Aluminium ABS itself. See, for instance, minutes of a call with a competitor on 11 April 2019, DocID1444.
93 […].
94 […].
95 […].
96 […].
97 […].
98 In line with public international law and trade agreements, including in particular the GATT/WTO agreements, TDIs can take the form of anti-dumping, anti-subsidy or safeguard measures. Anti- dumping measures are always adopted in relation to imports from specific countries, safeguard measures in principle on imports from all countries.
99 Safeguard measures can be applied under the following conditions: if, as a result of unforeseen developments, a product is being imported into the EU in such increased quantities and/or on such terms and conditions as to cause, or threaten to cause, serious injury to EU producers of like or directly competitive products. Safeguard measures may only be imposed to the extent and for such time as may be necessary to prevent or remedy the injury.
100 Commission implementing regulation (EU) 2019/159 of 31 January 2019 imposing definitive safeguard measures against imports of certain steel products (OJ L 31, 1.2.2019, p. 27).
101 Commission implementing regulation (EU) 2019/159 of 31 January 2019 imposing definitive safeguard measures against imports of certain steel products (OJ L 31, 1.2.2019, p. 27), recital 49.
102 Anti-dumping measures are imposed on imports that are found to be dumped and cause injury to a Union industry. Dumping is defined as selling a good for export at less than its normal value. The normal value is either the product’s price as sold on the home market of the non-EU company, or a price based on the cost of production and profit.
103 Commission Implementing Regulation (EU) 2018/186 of 7 February 2018 imposing a definitive anti- dumping duty and collecting definitively the provisional duty imposed on imports of certain corrosion resistant steels originating in the People's Republic of China (OJ L 34, 8.2.2018, p. 16).
104 Among others, Commission Implementing Regulation (EU) 2017/2213 of 30 November 2017 amending Commission Implementing Regulation (EU) 2017/271 extending the definitive anti-dumping duty imposed by Council Regulation (EC) No 925/2009 on imports of certain aluminium foil originating in the People’s Republic of China to imports of slightly modified certain aluminium foil.
105 Commission Implementing Regulation (EU) 2017/109 of 23 January 2017 imposing a definitive anti- dumping duty on imports of certain aluminium road wheels originating in the People’s Republic of China following an expiry review pursuant to Article 11(2) of Regulation (EU) 2016/1036 of the European Parliament and of the Council (OJ L 18, 24.1.2017, p. 1).
106 Reply to the SO, paragraph 43.
107 Form CO, paragraph 113.
108 Reply to question 10 of Questionnaire to Automotive Customers, DocID2094.
109 Minutes of a call with a customer on 28.11.2018, DocID140.
110 Reply to question 10.1 of Questionnaire to Automotive Customers, DocID2094.
111 Minutes of a call with a customer on 2.4.2019, DocID1414.
112 Minutes of a call with a customer, 2.4.2019, DocID1414.
113 Reply to the SO, paragraphs 45-46.
114 See, for example, minutes of calls with customers on 2.4.2019, DocID1978; and on 15.5.2019, DocID1990.
115 […].
116 Reply to question 1 of Phase II Questionnaire to OEM, DocID2071.
117 Minutes of a call with a customer on 15.5.2019, DocID1990.
118 Reply to question 1 of Phase II Questionnaire to OEM, DocID2071.
119 Reply to question 11.1 of Questionnaire to Automotive Customers, DocID2094.
120 Reply to question 1 of Phase II Questionnaire to OEM, DocID2071.
121 Reply to question 11.1 of Questionnaire to Automotive Customers, DocID2094, page 59.
122 Minutes of a call with a customer on 12.12.2018, DocID1837.
123 Minutes of a call with a customer on 17.4.2019, DocID1420.
124 Minutes of a call with a customer on 2.4.2019, DocID1414.
125 Minutes of a call with a competitor, 23.05.2019, DocID2172.
126 Seeking Alpha, Constellium Q4 2018 Earnings Call Transcript, DocID1998.
127 Minutes of a call with a competitor, 23.05.2019, DocID2172.
128 […].
129 DocID1041-3134 (The Parties’ reply to the Commission’s request for information RFI 18, ‘NOV- EU00386035.pdf’).
130 […].
131 […].
132 […].
133 Also known as ‘fenders’ in certain versions of English.
134 Also known as a ‘hood’ in certain versions of English.
135 Also known as a ‘trunk’ in certain versions of English.
136 The Commission uses the European terminology throughout this Decision. In direct quotes, the terminology used by the market participant has been maintained.
137 ‘Clarification questions, meeting with the European Commission on May 21, 2019’, submitted by the Notifying Party on 21 May 2019, DocID1862, slide 16.
138 DocID18 ‘Goldman Sachs Equity Research 2018 Cars 2025 Vol.5 ‘Lighter, faster, cheaper’, slide 9, Exhibit 6.
139 Reply to the SO, paragraph 29.
140 Reply to the SO, Figure 1.
141 SO, paragraphs 138-139.
142 Minutes of a call with a customer on 2.4.2019, DocID1978, paragraph 3.
143 Reply to the Letter of Facts, paragraphs 18-21.
144 Minutes of a call with a customer on 15.5.2019, DocID1990, paragraph 2.
145 Minutes of a call with a customer on 17.4.2019, DocID1420, paragraph 2.
146 Minutes of a call with a customer on 2.4.2019, DocID1978, paragraph 4.
147 Although the OEM refers to aluminium in general, and not specifically to Aluminium ABS, the Novelis’ internal document referred to in recital (158) and Figure 13 shows that the reference is to be understood as to Aluminium ABS because the OEMs plans to issue a tender for Aluminium ABS for its next Kadjar model.
148 Reply to the Letter of Facts, paragraphs 33-43.
149 […].
150 […].
151 […].
152 […].
153 […].
154 […].
155 […].
156 Reply to the SO, paragraphs 55-60.
157 […].
158 Form CO, Annex 13b, ‘20180407 McKinsey Light Weight Materials in Automotive vf.pdf’, DocID145- 424.
159 DocID2501 (Ducker, ‘Aluminium Content in European Cars_EA_140619.pdf’).
160 […].
161 […].
162 Minutes of a call with a customer, 20.5.2019, DocID2123.
163 Minutes of a call with a customer, 24.1.2019, DocID822.
164 Minutes of a call with a customer, 17.4.2019, DocID1420.
165 Minutes of a call with a customer, 28.11.2018, DocID140.
166 Minutes of a call with a competitor, 14.5.2019, DocID1960.
167 Reply to request for information 25, Annex 1, slide 25, DocID1111.
168 Reply to the Article 6(1)(c) Decision, page 39.
169 Replies to question 2 of Phase II – Questionnaire to OEMs, DocID2071 and DocID2109.
170 Minutes of a call with an OEM on 15.5.2019, DocID1990
171 […].
172 See for example, Reply to the Letter of Facts, paragraphs 25-28, 31-47, and 69-85.
173 […].
174 Reply to the SO, paragraphs 64-65.
175 […].
176 […].
177 Reply to the Letter of Facts, paragraphs 29-30.
178 See Annex I, Section 4.
179 Reply to request for information 32, ‘Annex Q9-1.pptx’, DocID1754-9.
180 Reply to the Letter of Facts, paragraphs 37-42; Reply to the Article 6(1)(c) Decision, paragraphs 129– 135.
181 Replies to question 2 of Phase II – Questionnaire to OEMs, DocID2071.
182 […].
183 […].
184 […].
185 Reply to the SO, paragraphs 70-78.
186 ‘Clarification questions. Meeting with the European Commission, May 21, 2019’, submitted by the Notifying Party on 25 May 2019, DocID1862, slide 6.
187 See for example: Reply to the SO, paragraphs 2, 48, 70-72.
188 Minutes of a call with a customer, 15.5.2019, DocID1990.
189 Minutes of a call with a customer, 2.4.2019, DocID1414.
190 Minutes of a call with a customer, 20.5.2019, DocID2123, […].
191 Minutes of a call with a customer, 17.4.2019, DocID1420.
192 Ducker Worldwide is a consulting and research company. It offers among other services market intelligence. In Europe, it is tasked by the European Aluminium Association to produce every three years a report on aluminium content in vehicles.
193 Reply to request for information 40, Annex Q10, slide 4.
194 Reply to request for information 40, Annex Q10, slide 4.
195 Reply to question 9.1 of Questionnaire to Automotive Customers, DocID2094.
196 Minutes of a call with a customer, 2.4.2019, DocID1978.
197 Minutes of a call with a customer, 2.4.2019, DocID1414.
198 Reply to the SO, paragraphs 80-81.
199 […].
200 Reply to the Article 6(1)(c) Decision, paragraphs 153–155; Reply to the SO, paragraphs 82-102.
201 […].
202 […].
203 […].
204 […].
205 […].
206 […].
207 […].
208 […].
209 […].
210 […].
211 […].
212 […].
213 The claim of the Notifying Party in its Reply to the Letter of Facts, paragraphs 98-99, that a ‘slower growth for Aluminium ABS’ is projected is not in contradiction to this finding.
214 Reply to request for information 36, ‘CRA – Automotive market shares – formatted.xlsx’, DocID1782- 12.
215 […].
216 […].
217 See Commission Notice on the definition of relevant market for the purposes of Community competition law (OJ C 372, 9.12.1997, p.5), paragraphs 13, 15–9.
218 See for example: Form CO, paragraphs 94–5; Reply to the Article 6(1)(c) Decision, paragraphs 136–44; Reply to the SO, paragraphs 106–109.
219 The growth of Aluminium ABS appears to be underestimated because sales in 2012 have been linearly interpolated from sales in 2015–2017 (due to the lack of other available data), despite the fact that in the period 2012–2013 a flat demand, if not a decrease in demand might have occurred, due to the reduced number of vehicles produced in the same year.
220 […].
221 […].
222 Minutes of a call with a customer on 20.5.2019, DocID2123.
223 Minutes of a call with a customer on 2.4.2019, DocID1978.
224 […].
225 […].
226 […].
227 […].
228 […].
229 Reply to the Letter of Facts, paragraph 102.
230 See for example, Reply to the Letter of Facts, paragraphs 28, 31.
231 Regulation (EU) 2019/631 of the European Parliament and of the Council of 17 April 2019 setting CO2 emission performance standards for new passenger cars and for new light commercial vehicles, and repealing Regulations (EC) No 443/2009 and (EU) No 510/2011 (OJ L 111, 25.4.2019, page 13).
232 Reply to the SO, paragraphs 117-120.
233 See for example: Reply to the Letter of Facts, paragraphs 49, 53, 62, 68, 77, 81–82; Reply to the SO, paragraphs 28–30.
234 […].
235 Metal stamping is a manufacturing process used to convert flat metal sheets into specific shapes through a forming technique.
236 Minutes of a call with a competitor on 15.3.2018, DocID900.
237 Minutes of calls with customers on 12.4.2019, DocID1031; on 11.4.2019, DocID1444; and on 29.5.2019, DocID2224.
238 Replies to questions 9 and 10 of Questionnaire to Automotive Customers, DocID2094.
239 Reply to the SO, paragraphs 128-133.
240 M.3225 – Alcan / Pechiney (II); M.4605 – Hindalco / Novelis.
241 Reply to the Article 6(1)(c) Decision, paragraphs 106–18.
242 DocID 1677-3 (Reply to RFI 32, Annex 13-2 […] 12-5-2017 FINAL), slides 8-9.
243 See, for example, Figure 6, Figure 38, Figure 42, and Figure 46.
244 Replies to question 15 of the Questionnaire to Automotive Customers, DocID2094.
245 Replies to question 15 of Questionnaire to Automotive Customers, DocID2094.
246 Replies to question 15.1 of Questionnaire to Automotive Customers, DocID2094.
247 Replies to question 15 of Questionnaire to Automotive Customers, DocID2094.
248 Replies to question 15 of Questionnaire to Automotive Customers, DocID2094.
249 Minutes of a call with a competitor on 15.3.2018, DocID900; on 30.4.2019, DocID1806; and Minutes of a call with a customer on 5.12.2018, DocID792.
250 Form CO, paragraph 74.
251 Minutes of call with a competitor on 14.5.2019. DocID1960, paragraph 4.
252 Commission Notice on the definition of relevant market for the purposes of Community competition law (OJ C 372, 9.12.1997, p.5), paragraph 20.
253 Minutes of a call with a competitor on 15.3.2019, DocID900, paragraph 1.
254 Minutes of a call with a competitor on 15.3.2019, DocID900, paragraph 3.
255 See for examples: Minutes of a call with a competitor on 30.4.2019, DocID1806, paragraph 17; Minutes of a call with a competitor on 23.5.2019, DocID2172, paragraphs 13–15.
256 […].
257 […].
258 […].
259 […].
260 […].
261 […].
262 […].
263 Commission Notice on the definition of relevant market for the purpose of Community competition law, OJ C 372, 9 December 1997, p.5, paragraph 20.
264 The Notifying Party submits that Standard FRPs do not include Aluminium ABS, Aluminium FRPs for aerospace, Aluminium FRPs for automotive heat exchangers, lithographic sheets, Aluminium FRPs for beverage and food cans and aluminium foil (Form CO, paragraph 49).
265 Form CO, paragraphs 135–8 and Reply to Article 6(1)(c) Decision, paragraph 165.
266 Form CO, paragraph 135.
267 Form CO, paragraph 145.
268 Reply to Article 6(1)(c) Decision, paragraph 175.
269 Aluminium anodising sheet is, to a limited extent, also used in industrial applications (for example housing for electric installations), electronics (for example smart phones) and niche applications such as suitcases (Form CO, paragraph 144).
270 Reply to request for information 35.
271 Form CO, paragraph 142.
272 Reply to request for information 35.
273 Reply to request for information 8.
274 Reply to Article 6(1)(c) Decision, paragraph 168.
275 See, for example, M.3225 – Alcan/Pechiney (II), paragraph 61.
276 Replies to question 5.1 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
277 Replies to question 8.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
278 Minutes of a call with a competitor on 15.3.2019, DocID900.
279 Minutes of a call with a competitor on 14.3.2019, DocID889.
280 Minutes of a call with a competitor on 30.4.2019, DocID806.
281 Minutes of a call with a competitor on 14.5.2019, DocID2126.
282 Replies to question 12.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
283 Minutes of a call with a competitor on 14.3.2019, DocID889.
284 Replies to question 17 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
285 Replies to questions 12 and 13 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
286 Replies to question 41.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
287 Replies to question 22 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
288 Replies to question 9.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
289 Replies to question 9.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
290 Replies to question 9.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
291 Minutes of a call with a competitor on 14.3.2019, DocID889.
292 Replies to question 20.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
293 Replies to question 23.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
294 Replies to question 34 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
295 Replies to question 41.1 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107.
296 Minutes of a call with a competitor on 15.3.2019, DocID900.
297 Replies to question 27 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
298 Commission Notice on the definition of the relevant market for the purposes of Community competition law (OJ C372, 9.12.1997, p.5), paragraphs 8 and seq. and 28 and seq.
299 Commission Notice on the definition of the relevant market for the purposes of Community competition law (OJ C 372, 9.12.1997, p.5), paragraphs 44-52.
300 Form CO, paragraphs 119 and seq.
301 Form CO, paragraph 120.
302 Form CO, paragraph 122.
303 For the purpose of the present Decision, the notation ‘kt’ indicates thousand of tonnes.
304 Form CO, paragraph 127.
305 Form CO, paragraph 428.
306 Reply to the Letter of Facts, paragraph 108;
307 M.3225 – Alcan/Pechiney (II), paragraph 66.
308 See for example, case M.4605 – Hindalco/Novelis, paragraph 13.
309 For the purpose of the present case, defining the relevant geographic market as restricted to the EEA or as also including Switzerland does not make any difference as there is no consumption of Aluminium ABS in Switzerland.
310 Minutes of a call with a customer on 28.11.2018, DocID140.
311 Minutes of a call with a competitor on 12.12.2018, DocID737.
312 Minutes of a call with a customer on 12.12.2018, DocID1837.
313 Minutes of a call with a customer on 20.5.2019, DocID2123.
314 […].
315 […].
316 Replies to question 20 of Questionnaire to Automotive Customers, DocID2094.
317 Minutes of a call with a customer, 17.4.2019, DocID1420.
318 Replies to question 18.1 of Questionnaire to Automotive Customers, DocID2094.
319 Minutes of a call with a customer, 2.4.2019, DocID1978.
320 Minutes of a call with a customer, 15.5.2019, DocID1990.
321 Replies to question 23 of Questionnaire to Automotive Customers, DocID2094.
322 Aging is a process by which heat treatable alloys such as 6xxx hardens changing its mechanical properties.
323 Form CO, paragraph 302.
324 Replies to question 23.1 of Questionnaire to Automotive Customers, DocID2094.
325 Replies to question 23.1 of Questionnaire to Automotive Customers, DocID2094.
326 Replies to question 23.1 of Questionnaire to Automotive Customers, DocID2094.
327 Replies to question 23.1 of Questionnaire to Automotive Customers, DocID2094.
328 Minutes of a call with a customer on 17.4.2018, DocID1420.
329 Minutes of a call with a customer on 12.12.2018, DocID1837.
330 Replies to question 23.1 of Questionnaire to Automotive Customers, DocID2094.
331 Minutes of a call with a competitor on 19.12.2018, DocID733.
332 Minutes of a call with a competitor on 12.12.2018, DocID737.
333 Minutes of a call with a customer on 28.11.2018, DocID140.
334 […].
335 Council Regulation (EEC) No 2658/87 on the tariff and statistical nomenclature and on the Common Customs Tariff (OJ L 292, 30.10.1998, page 1).
336 Regulation (EEC) No 2840/72 of the Council of 19 December 1972 concluding an Agreement between the European Economic Community and the Swiss Confederation and adopting provisions for its implementation and concluding an additional Agreement concerning the validity, for the Principality of Liechtenstein, of the Agreement between the European Economic Community and the Swiss Confederation of 22 July 1972 (OJ L 300, 31.12.1972, p. 188).
337 Form CO, paragraph 153.
338 See cases M.3225 Alcan / Pechiney (II), paragraph 68; and M.2702 Norsk Hydro / VAW, paragraph 14.
339 Replies to questions 65 and 66 of the Questionnaire to Competitors, DocID2073 and replies to questions 9 and 10 of the Questionnaire to Customers of other FRPs, DocID2092.
340 See, for example, replies to questions 28 and 29 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083; and replies to questions 31 and 32 of Phase II Questionnaire to competitors in other (standard) FRPs, DocID2107.
341 For the sake of having a more robust assessment of the Parties’ position in the relevant market, the capacity of Alcoa/Ma’aden, considering its relative geographic proximity and regular supplies into the EEA, is also included within capacity shares.
342 Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings (OJ C 31, 05.02.2004, pages 5-18), in particular paragraphs 4 and 22.
343 Horizontal Merger Guidelines, paragraph 2.
344 Horizontal Merger Guidelines, paragraph 17.
345 Horizontal Merger Guidelines, paragraph 24.
346 Horizontal Merger Guidelines, paragraph 25.
347 M.7278 – GE/Alstom (2015), Annex I - The Commission's Economic Analysis of Bidding Data, recitals 7 et seq.
348 See discussion of the role of capacity constraints in Section 2.4 of Annex I.
349 Form CO, paragraph 156.
350 Reply to the SO, paragraphs 26-32.
351 […].
352 Form CO, paragraph 156.
353 Reply to the SO, paragraph 147.
354 Form CO, paragraph 156.
355 Reply to the Article 6(1)(c) Decision, paragraph 84.
356 Reply to the SO, paragraph 436.
357 Form CO, paragraph 7.
358 Reply to the SO, paragraph 261.
359 […].
360 Form CO, paragraph 156.
361 Reply to the SO, paragraph 158.
362 Form CO, paragraph 156.
363 Reply to the SO, paragraph 444.
364 Reply to the SO, paragraph 450.
365 Reply to the SO, paragraph 483.
366 5 in France (Issoire, Montreuil Juigné, Neuf-Brisach, Nuits-Saint-Georges and Ussel); 6 in Germany (Burg, Crailsheim, Neckarsulm Dahenfeld, Gottmadingen, Landau, Singen); 1 in Switzerland (Valais), 2 in Slovakia (Levice and Zilina) and 1 in Czech Republic (Decin).
367 Excerpt from Constellium’s website (18.6.2019), DocID2205.
368 Hydro’s Annual report 2018, page 55, DocID2201.
369 Excerpt from Hydro’s website (18.6.2019), DocID2205.
370 Excerpt from AMAG’s website (18.6.2019), DocID2200.
371 Excerpt from AMAG’s website (18.6.2019), DocID2200.
372 Minutes of a call with a competitor on 14.5.2019, DocID1960.
373 In 2016, Alcoa Inc. completed the separation of its business into two independent, public listed companies – Alcoa Corporation and Arconic Inc. (the new name for Alcoa Inc.). Following the separation, Alcoa Corporation holds the stake in Alcoa/Ma’aden previously held by Alcoa Inc. and Arconic retained the global rolled products (other than the 25.1% ownership stake in Alcoa/Ma’aden and a rolling mill in Indiana, US).
374 Ma’aden’s press release of 23 May 2010, DocID2204.
375 Ma’aden’s press release of 19 February 2019, DocID2203.
376 Horizontal Merger Guidelines, parargraph 14.
377 In M.3225 – Alcan / Pechiney (II), paragraph 75, the Commission considered capacity shares and, in that case, production shares as indicative of market power.
378 ‘Presentation for the State-of-Play Meeting, 9 April 2019’, submitted by the Notifying Party on 9 April 2019. Reply to the Article 6(1)(c) Decision, paragraphs 35–50; Reply to the SO, paragraphs 189-204.
379 Horizontal Merger Guidelines, paragraph 27.
380 Horizontal Merger Guidelines, paragraph 26.
381 See Form CO, Annexes 10a and 10b. For […].
382 In the Decision, the expressions ‘production capacity’, ‘supply capacity’, ‘manufacturing capacity’, and ‘capacity’ are used interchangeably.
383 M.3225 – Alcan/Pechiney (II), paragraphs 72 et seq.
384 […].
385 Reply to the Article 6(1)(c) Decision, paragraphs 28-29.
386 M.7564 – Mahle Behr / Delphi Thermal Systems Business, paragraph 33.
387 As noted by the Notifying Party in the Reply to the Letter of Facts, paragraph 115.
388 […].
389 With respect to the Notifying Party’s claim in the Reply to the Letter of Facts, that there is significant buy side concentration, the Commission refers to Section 8.3.11.1, which explains that supply side concentration in the present case is in fact higher than buy side concentration.
390 Form CO, paragraph 167.
391 […].
392 Minutes of a call with a customer 2.4.2019, DOC ID 1978.
393 Response to the Commission’s request for information RFI 47, question 5.
394 Reply to the Letter of Facts, paragraph 117.
395 The capacity data provided by the Notifying Party as well as the capacity data collected by the Commission also include the planned/announced capacity expansions by each rival, up to 2023.
396 Horizontal Merger Guidelines, paragraph 17.
397 Commission Notice on the definition of relevant market for the purposes of Community competition law (OJ C 372, 9.12.1997, p.5), paragraph 55.
398 Form CO, Annex 43 'Share calculation methodology', paragraph 5, DocID145-36.
399 For the purpose of this Decision, the expressions 'conversion revenue', margin on metal' and *conversion premium' are used by the Commission interchangeably.
400 For Novelis see for example Form CO, Annex 28 [...].
401 Horizontal Merger Guidelines, paragraphs 19-20.
402 Reply to request for information 19, 'Global Market strategy update March-18', slide 6, DocID 97645135.
403 Minutes of a call with a competitor on 23.5.2019, DocID2172, paragraph 3.
404 For the purpose of the present assessment, the Commission uses what in Annex I is referred to as "reconstruction EEA Auto" scenario. This scenario considers capacities adjusted to account for a share of capacity that is used for exports outside the EEA and for non-ABS applications, as provided by the various Aluminium ABS suppliers during the market reconstruction. The Commission considers this scenario to be the most appropriate to describe the capacity available to ABS customers in the EEA. Annex I contains a sensitivity which does not adjust the capacities of the EEA ABS suppliers for exports or non-ABS applications.
405 Form CO, paragraphs 166-185; Reply to the Article 6(1)(c) Decision, paragraphs 38-47.
406 Form CO, footnote 95.
407 Form CO, paragraph 167.
408 […].
409 […].
410 […].
411 […].
412 […].
413 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
414 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
415 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
416 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
417 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
418 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
419 Replies to question 25 of Questionnaire to Automotive Customers, DocID2094.
420 Replies to question 56.1 of Questionnaire to Automotive Customers, DocID2094.
421 More details are available in Annex I, Section 2.5.2.
422 For example, see Daisuke Hirata (2009), ‘Asymmetric Bertrand-Edgeworth Oligopoly and Mergers’, B.E. Journal of Theoretical Economics, Vol. 9, No. 1, pp. 1935-1704. See also Case M.6471 Outukumpu/Inoxum (Commission decision of 7 November 2012).
423 […].
424 […].
425 Form CO, paragraph 246.
426 […].
427 The Commission conducted its reconstruction of spare capacity on 2018 data as 2018 is the last complete year with actual data available and therefore allows for a consideration actual of capacity data and utilisation rates of Aluminium ABS suppliers.
428 […].
429 […].
430 Form CO, paragraph 257.
431 […].
432 […].
433 […].
434 […].
435 Form CO, paragraphs 257-303.
436 […].
437 Reply to request for information 19, ‘ALE01068474.pptx’, DocID977-7821.
438 Form CO, paragraph 250.
439 […].
440 […].
441 […].
442 […].
443 […].
444 […].
445 […].
446 […].
447 Reply to the Article 6(1)(c) Decision, paragraph 53.
448 Reply to request for information 22, Annex Q4-2, […] .
449 Form CO, paragraph 249.
450 […].
451 […].
452 […].
453 Replies to question 43 of Questionnaire to Automotive Customers, DocID2094.
454 […].
455 Passivation refers to surface treatment in order to make the material less reactive to the environment.
456 Reply to question 24 of Questionnaire to Competitors, DocID2073.
457 Reply to question 44 of Questionnaire to Automotive Customers, DocID2094.
458 Reply to question 44 of Questionnaire to Automotive Customers, DocID2094.
459 Customer mentioned in DocID145-116 (Form CO, Annex 10a (Novelis), […] .
460 Reply to request for information 32, Annex Q1e, […] , DocID1663-26.
461 […].
462 […].
463 […].
464 Reply to the Letter of Facts, paragraph 137.
465 Minutes of a call with a competitor on 23.5.2019, DocID2172.
466 […].
467 […].
468 Minutes of a call with a competitor on 23.5.2019, DocID2172, paragraph 3.
469 Minutes of a call with a competitor on 23.5.2019, DocID2172, paragraph 5.
470 Minutes of a call with a competitor on 23.5.2019, DocID2089.
471 […].
472 […].
473 […].
474 Reply to the Article 6(1)(c) Decision, paragraphs 4-5.
475 Minutes of a call with a competitor on 23.5.2019, DocID2089.
476 Minutes of a call with a customer on 12.12.2019, DocID1837.
477 Minutes of a call with a customer on 20.5.2019, DocID2123.
478 Minutes of a call with a customer on 29.5.2019, DocID2224.
479 Minutes of a call with a customer on 28.11.2018, DocID140.
480 Minutes of a call with a customer on 2.4.2019, DocID1978.
481 Minutes of a call with a customer on 24.1.2019, DocID822.
482 Minutes of a call with a customer on 15.5.2019, DocID1990.
483 […].
484 […].
485 […].
486 […].
487 […].
488 […].
489 […].
490 […].
491 […].
492 […].
493 […].
494 […].
495 […].
496 According to the Commission’s analysis of the Notifying Party’s data.
497 Horizontal Merger Guidelines, paragraph 17.
498 In coming years until 2023, the combined capacity share of the Merged Entity will also remain very high and substantially larger than the capacity of its next competitor as evidenced in Table 11.
499 […].
500 […].
501 Reply to the Letter of Facts, paragraphs 145-146, OEMs buyer power is further discussed in section 8.3.11.
502 Pivotality is further developped in section 8.3.6.
503 Replies to question 7 of the Phase II Questionnaire to OEM, question 7. DocID2071.
504 Reply to the Article 6(1)(c) Decision, paragraphs 23–2.
505 […].
506 […].
507 […].
508 […].
509 […].
510 […].
511 […].
512 […].
513 […].
514 […].
515 […].
516 […].
517 […].
518 […].
519 […].
520 […].
521 […].
522 […].
523 […].
524 […].
525 […].
526 […].
527 […].
528 Uncommitted capacity of a supplier in future years is a concept different from its (current or expected) spare capacity. Uncommitted capacity is capacity which has not yet been contracted or committed to a customer – in many cases because certain tenders have not been concluded yet by OEMs, whereas spare capacity is to be understood as the capacity which exceeds the volume the supplier in question supplies overall in a given year.
529 […].
530 […].
531 […].
532 Reply to request for information 36, ‘CRA – Automotive market shares – formatted.xlsx’, DocID1782- 12.
533 […].
534 […].
535 […].
536 Reply to the Article 6(1)(c) Decision, paragraph 84.
537 Form CO, paragraph 70.
538 ‘Evidence paper’, submitted by the Notifying Party on 6 June 2019, paragraph 5.
539 The parties data is consistent with the commission market reconstruction figures by volume based forecasted sales (Table 10 of the annex). The market reconstruction by volume based on nominated sales (Table 11 of the annex) is not relevant because it does not include the volumes linked to the latest long term contract between Novelis and […].
540 Reply to the Letter of Facts, paragraph 157.
541 […].
542 […].
543 […].
544 […].
545 […].
546 […].
547 Minutes of a call with a customer on 15.5.2019, DocID1990.
548 Form CO, paragraph 156.
549 Replies to questions 58 and 58.1 of Questionnaire to Automotive Customers, DocID2073.
550 Replies to questions 59 of Questionnaire to Automotive Customers, DocID2073.
551 Minutes of a call with a customer on 24.1.2019, DocID822.
552 Reply to question 64 of Questionnaire to Automotive Customers, DocID2529.
553 Minutes of a call with a customer on 12.12.2018, DocID1837.
554 Minutes of a call with a customer on 29.5.2019, DocID2224.
555 Minutes of a call with a customer on 5.12.2018, DocID792.
556 Reply to question 59 of Questionnaire to Automotive Customers, DocID2433.
557 Reply to question 61 of Questionnaire to Automotive Customers, DocID2433.
558 Reply to question 39 of Phase II – Questionnaire to OEMs, DocID2071.
559 Minutes of a call with a customer on 15.5.2019, DocID1990.
560 Minutes of a call with a customer on 15.5.2019, DocID1990, paragraph 17.
561 For instance, Form CO, paragraph 165 and ‘Evidence paper’, submitted by the Notifying Party on 6 June 2019, paragraphs 54 et seq.
562 Reply to request for information 32, DocID1663-7.
563 Minutes of a call with a customer, 28.11.2018, DocID140.
564 Minutes of a call with a competitor on 12.12.2018, DocID737, paragraph 17.
565 Minutes of a call with a competitor on 12.12.2018, DocID1837, paragraph 11.
566 Minutes of a call with a competitor on 28.12.2018, DocID140, paragraph 27.
567 […].
568 Minutes of a call with a customer on 15.5. 2019, DocID02224.
569 Minutes of a call with a customer on 1.5.2019, DocID1990.
570 Replies to question 18 of Questionnaire to Competitors, DocID2073.
571 […].
572 […].
573 […].
574 ‘Unilateral effects and price discrimination’, submitted by the Notifying Party on 7 June 2019, page 3.
575 Reply to Request for Information 12, DocID699.
576 Minutes of a call with a customer on 29.5.2019, DocID2224.
577 […].
578 […].
579 Discussed in Section 2.5 of Annex I.
580 For example, see David M. Kreps and Jose A. Scheinkman (1983), ‘Quantity Precommitment and Betrand Competition Yield Cournot Outcomes’, Bell Journal of Economics, Vol. 14, No. 2, pp. 326- 337.
581 A 10% expansion by a larger supplier is different from a 10% expansion by a smaller supplier.
582 See Section 2.1.2 of Annex I.
583 […].
584 […].
585 […].
586 […].
587 […].
588 […].
589 […].
590 […].
591 […].
592 […].
593 […].
594 […].
595 […].
596 […].
597 […].
598 […].
599 […].
600 […].
601 […].
602 […].
603 […].
604 […].
605 […].
606 […].
607 […].
608 […].
609 […].
610 […].
611 […].
612 […].
613 […].
614 […].
615 […].
616 […].
617 […].
618 […].
619 […].
620 […].
621 Highlighting added by the Commission.
622 […].
623 […].
624 […].
625 […].
626 […].
627 […].
628 […].
629 […].
630 […].
631 […].
632 […].
633 […].
634 […].
635 […].
636 […].
637 […].
638 […].
639 […].
640 […].
641 […].
642 Reply to the Letter of Facts, paragraph 178.
643 The Notifying Party further submits in the Reply to the Letter of Facts, paragraph 178, that […].
644 […].
645 […].
646 […].
647 […].
648 […].
649 […].
650 […].
651 […].
652 […].
653 […].
654 […].
655 […].
656 […].
657 […].
658 Horizontal Merger Guidelines, paragraph 32.
659 Horizontal Merger Guidelines, paragraph 34.
660 DocID2177 (‘Constellium Holdco's (CSTM) CEO Jean-Marc Germain on Q4 2018 Results - Earnings Call Transcript’), page 1.
661 Minutes of a call with a competitor, 23.5.2019, DocID2089, paragraphs 8-9.
662 Minutes of a call with a competitor, 23.5.2019, DocID2172, paragraphs 41 and 44.
663 Minutes of a call with a competitor, 24.5.2019, DocID2115, paragraphs 1-8.
664 […].
665 Reply to the SO, paragraph 443.
666 Minutes of a call with a customer, 15.5.2019, DocID1990.
667 Minutes of a call with a customer, 2.4.2019, DocID1414.
668 Minutes of a call with an association, 20.5.2019, DocID2058.
669 Minutes of a call with an association, 20.5.2019, DocID2058.
670 Reply to request for information 18, ‘NOV-EU00131682.msg’, DocID1038-31682.
671 […].
672 […].
673 […].
674 […].
675 […].
676 […].
677 […].
678 […].
679 […].
680 […].
681 […].
682 […].
683 […].
684 […].
685 […].
686 […].
687 DocID145-609 (Form CO, Annex 54(i)_EEA ABS Production Asset Footprint).
688 DocID145-609 (Form CO, Annex 54(i)_EEA ABS Production Asset Footprint).
689 […].
690 […].
691 Reply to the SO, paragraphs 449-485; Form CO, paragraph 128.
692 […].
693 […].
694 […].
695 […].
696 Manufacturers serving aerospace customers appear to be an exception because manufacturing Aluminium FRP for aerospace also requires a CASH line. However, other barriers to entry explained in the present section, rather than manufacturing equipment, equally apply to manufacturers serving aerospace business.
697 Minutes of a call with a competitor 12.12.2018, DocID737, paragraph 11.
698 Form CO, paragraphs 389-391.
699 […].
700 […].
701 Minutes of a call with a competitor on 19.12.2018, DocID733, paragraph 6.
702 […].
703 […].
704 Minutes of a call with a competitor on 14.5.2019, DocID1960, paragraph 12.
705 Minutes of a call with a competitor on 14.5.2019, DocID1960, paragraph 5.
706 Minutes of a call with a competitor 23.5.2019, DocID2089, paragraph 8.
707 Minutes of a call with a competitor 23.5.2019, DocID2172, paragraph 27.
708 Minutes of a call with a competitor 15.3.2019, DocID900, paragraph 1.
709 Minutes of a call with a competitor 15.3.2019, DocID900, paragraph 3.
710 Minutes of a call with a competitor 15.3.2019, DocID900, paragraph 5.
711 Minutes of a call with a competitor on 14 May 2019, DocID1960, paragraph 10.
712 Minutes of a call with a competitor 30.4.2019, DocID1806, paragraphs 16-17.
713 Minutes of a call with a competitor 30.4.2019, DocID1806, paragraph 20.
714 Minutes of a call with a competitor on 23.5.2019, DocID2491, paragraph 14.
715 Minutes of a call with a competitor 23.5.2019, DocID2172, paragraphs 13-15.
716 […].
717 Minutes of a call with a competitor 30.4.2019, DocID1806, paragraph 19.
718 […].
719 […].
720 […].
721 […].
722 […].
723 […].
724 […].
725 Reply to the SO, paragraph 444.
726 […].
727 […].
728 Horizontal Merger Guidelines, paragraph 64.
729 […].
730 […].
731 Reply to the Article 6(1)(c) Decision, paragraph 17.
732 Reply to the SO, paragraph 483.
733 Form CO, paragraph 319.
734 Reply to the SO, paragraphs 485-486.
735 […].
736 […].
737 […].
738 […].
739 […].
740 […].
741 Reply to the SO, paragraph 487.
742 […].
743 […].
744 […].
745 […].
746 […].
747 Reply to question 58.1 of Questionnaire to Automotive Customers, DocID2094.
748 Reply to question 61 of Questionnaire to Automotive Customers, DocID2094.
749 […].
750 […].
751 […].
752 […].
753 […].
754 […].
755 […].
756 […].
757 […].
758 […].
759 […].
760 […].
761 […].
762 […].
763 Reply to the SO, paragraph 3.
764 Reply to request for information 49, question 3, DocID2701.
765 Minutes of a call with a customer, 29.5.2019, DocID2224.
766 […].
767 […].
768 […].
769 […].
770 […].
771 […].
772 […].
773 […].
774 […].
775 […].
776 Minutes of a call with a competitor 23.5.2019, DocID2089.
777 Horizontal Merger Guidelines, paragraph 65.
778 Reply to the SO, paragraph 507.
779 […].
780 […].
781 Minutes of a call with a customer, 29.5.2019, DocID2224.
782 Reply to question 26 of Phase II – Questionnaire to OEMs, DocID2071.
783 Minutes of a call with a customer 20.5.2019, DocID2123.
784 Minutes of a call with a customer 24.1.2019 DocID822.
785 Minutes of a call with a customer 28.11.2018 DocID140.
786 Minutes of a call with a customer 15.5.2019, DocID1990.
787 Minutes of a call with a customer 24.1.2019, DocID822.
788 Replies to question 27 of Phase II – Questionnaire to OEMs, DocID2071.
789 Reply to the SO, paragraphs 508-513.
790 Reply to the SO, paragraph 513.
791 […].
792 […].
793 […].
794 […].
795 Replies to question 20 of the Phase II – Questionnaire to OEMs, DocID2071.
796 Minutes of a call with a customer, 20.5.2019, DocID2123.
797 Reply to the SO, paragraph 505.
798 Reply to question 20.1 of Questionnaire to OEMs, DocID2469.
799 Replies to question 16.1 of the Phase II – Questionnaire to OEMs, DocID2071.
800 Replies to question 18.1 of the Phase II – Questionnaire to OEMs, DocID2071.
801 Minutes of a call with a customer 20.5.2019 DocID2123.
802 Replies to question 17 of the Phase II – Questionnaire to OEMs, DocID2071.
803 Horizontal Merger Guidelines, paragraph 67.
804 […].
805 Form CO, paragraph 136.
806 Form CO, paragraph 137.
807 […].
808 Reply to request for information 1.
809 Form CO, paras 345-348.
810 Reply to Article 6(1)(c) Decision, paragraph 188.
811 […].
812 Replies to question 16 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
813 Replies to question 16 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
814 See, for example, minutes of a call with a competitor on 30.4.2019, DocID806.
815 Replies to question 38 of Phase II Questionnaire to competitors in other (standard) FRP, DocID2107. See also replies to questions 21 and 22 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
816 Minutes of a call with a competitor on 24.5.2019, DocID2115.
817 […].
818 Form CO, paras 349-352.
819 Reply to Article 6(1)(c) Decision, paragraph 182.
820 […].
821 Replies to question 38 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
822 Replies to question 38 of Phase II Questionnaire to customers of other (standard) FRP, DocID2083.
823 Minutes of a call with a competitor on 14.3.2019, DocID889.
824 Minutes of a call with a competitor on 3.6.2019, DocID2163.
825 Minutes of a call with a customer on 11.6.2019, DocID2181.
826 Horizontal Merger Guidelines, paragraph 77.
827 Horizontal Merger Guidelines, paragraph 87.
828 Horizontal Merger Guidelines, paragraph 79.
829 Horizontal Merger Guidelines, paragraph 85.
830 Horizontal Merger Guidelines, paragraph 86.
831 […].
832 […].
833 OJ C 267, 22.10.2008, p.1.
834 Remedies Notice, paragraph 5.
835 Remedies Notice, paragraph 9.
836 Remedies Notice, paragraph 9.
837 Remedies Notice, paragraph 12.
838 Remedies Notice, paragraph 9.
839 Remedies Notice, paragraph 11.
840 Remedies Notice, paragraph 22.
841 Remedies Notice, paragraphs 23–25.
842 Remedies Notice, paragraph 26.
843 Remedies Notice, paragraph 32.
844 Remedies Notice, paragraph 30.
845 Remedies Notice, paragraph 47.
846 As described in the Schedule of the Committments of 9 August 2019.
847 Form RM of 9 August 2019, paragraphs 3-4.
848 Form RM of 9 August 2019, paragraph 20.
849 Form RM of 9 August 2019, paragraph 21.
850 Form RM of 9 August 2019, paragraph 22.
851 Form RM of 9 August 2019, paragraph 23.
852 Form RM of 9 August 2019, paragraphs 74-75.
853 Form RM of 13 August 2019, paragraph 3-4.
854 Form RM of 13 August 2019, paragraph 27.
855 Form RM of 13 August 2019, paragraph 34.
856 Reply to the Market Test to Competitors, question 1.1 DocID3287.
857 Reply to the Market Test to Customers, question 1.1, DocID3288.
858 Replies to the Market Test to Customers, questions 2-3, DocID3288 and Replies to the Market Test to Competitors, questions 2-3, DocID3287.
859 Replies to the Market Test to Customers, question 5, DocID3288.
860 See for example Replies to Market Test to Customers, question 14.1, DocID3288.
861 See for example Replies to Market Test to Competitors, question 3.1, DocID3287 and Replies to Market Test to Customers, question 9.1, DocID3288.
862 Replies to questions 8, 11 and 14 of MT2 – Market Test Customers, DocID3288; and replies to questions 4, 8 and 11 of MT1 – Market Test Competitors, DocID3287.
863 […].
864 Replies to questions 5, 9 and 12 of MT1 – Market Test to Competitors, DocID3287; and Replies to questions 7, 10 and 13 of MT2 – Market Test to Customers, DocID3288.
865 Replies to questions 6, 10 and 13 of MT1 – Market Test to Competitors, DocID3287.
866 Replies to questions 7, 10 and 13 of MT1 – Market Test to Competitors, questions 5, 9 and 12, DocID3287; and Replies to M2 – Market Test to Customers, DocID3288.
ANNEX
1 Section 8.2.2 of the Decision.
2 See paragraph 27 of the Horizontal Merger Guidelines.
3 See paragraph 32-35 of the Horizontal Merger Guidelines.
4 There are a number of limitations to these general observations. For instance, persistent excess capacities can only prevent competition concerns if underutilised capacities have comparable efficiency levels as utilised plants, since facilities with significantly higher unit costs would not be able to exert a comparable competitive constraint.
5 Reply to request for information 36.
6 Requests for information were sent to all the major suppliers of Aluminium ABS.
7 Based on the data package received on 30 November 2018 (zip file ‘M9076 - stata codes and data - bidding analysis.zip’).
8 Including a CRA report titled ‘Analysis of Novelis and Aleris bidding data’, dated 30 November 2018. [Doc Id:145-55].
9 Including a CRA report titled ‘Annex 4 – Observations on Annex 1 of the Decision’, dated 4 April 2019. [Doc Id:923].
10 Including a CRA report titled ‘Response to Annex I of the SO’, dated 17 July 2019 [Doc Id: 2559].
11 Including a CRA report titled ‘Response to Letter of Facts’, dated 19 August 2019 [Doc Id: 3048].
12 […].
13 See Figure 1.
14 See slide 17 of the slides deck ‘Apple - Oral Hearing Deck as presented on 23.07.19 - Confidential.pdf’, presented by the Notifying Party at the oral hearing.
15 See also Section 6.2.2.3.5 of the Decision.
16 The Notifying Party explained that in order to make the price of steel and aluminium comparable, two factors have to be taken into account: (i) aluminium has a lower density than steel (about a third), so that comparatively less aluminium is needed for a given application compared to steel and (ii) for certain applications different amounts of material may be needed in order to achieve certain desired material properties (comparatively more aluminium may be needed in order to achieve the same strength as steel). [...].
17 [...].
18 [...]
19 Form CO paragraph 113.
20 See Form CO Section 6.2.1.b.
21 The elasticity of demand of Aluminium ABS must fall below a certain threshold in order for Aluminium ABS to constitute a separate product market. Indeed, if steel exerted a tight competitive constraint on Aluminium ABS producers, one would expect the elasticity of demand for Aluminium ABS to be relatively high, as OEMs can easily switch to steel.
22 To see the difference, note that a hypothetical monopolist may be able to raise price by (say) 8% without losing profits compared to the current profit level, but might find it even more profitable to raise prices by only 4%.
23 That is, the ratio of the margin (price minus the relevant costs) and price.
24 The price increase typically considered is 5 or 10%.
25 For example, see Gregory J. Werden (1998), 'Demand Elasticities in Antitrust Analysis', Antitrust Law Journal, Vol. 66, No. 2, pp. 363-414.
26 Id.
27 [...]
28 See paragraph 18 of the EC Notice on the definition of the relevant market.
29 [...]
30 For example, a survey of the empirical literature on price elasticities, which covered a significant variety of products, found that product group elasticities are often scattered around one half. Kenneth W. Clements (2008), 'Price elasticities of demand are minus one-half, Economics Letters, Vol. 99, No. 3, pp. 490-493. The actual elasticity of Aluminium ABS demand would therefore have to be materially larger than common product group elasticities (1.4-2.9 instead of 0.5) for Aluminium ABS to be too narrow a product market definition.
31 […].
32 […].
33 […].
34 For example, see Serge Moresi, Steven Salop and John Woodbury (2017), ‘Market Defintion in Merger Analysis’, available at https://papers.ssrn.com/sol3/papers.cfm?abstract id=2906111, and the references cited therein.
35 And therefore the candidate market is too narrow and must be expanded.
36 U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, at §4.1.3.
37 Where failure of the SSNIP test means that steel would belong to the same product market.
38 This follows because under the same assumptions used for the critical loss formulas, the actual recapture ratio of a candidate market is equal to SM = 1- me. See Serge Moresi and Hans Zenger (2018), 'Recapture Ratios in Merger Analysis', Economics Letters, Vol. 170, pp. 136-138.
39 See Table 4 and Table 5.
40 See for instance Table 2 and Table 3.
41 […].
42 See discussion in Section 2.1.3.2. The Notifying Party’s estimate of […] is presented in Table 3 of the Reply to Annex I of the SO.
43 This is because the elasticity estimated by the Notifying Party, […], would be lower than the threshold elasticity, […].
44 The Commission has used Novelis' margins in the SSNIP test because (i) the average margins across all Aluminium ABS suppliers are not available and (ii) the fact that Novelis' margins are [...] makes the analysis favourable to the Notifying Party as explained in this section.
45 Reply to Annex I of the SO, p. 11-12.
46 Form CO, paragraph 113.
47 [...]
48 Form Co, paragraph 167.
49 [...].
50 [...]
51 [...]
52 The calculations were made available in the data room that opened after the Commission issued the Letter of Facts.
53 [...]
54 See Table 2 and Table 3. This comparison becomes even starker once one acknowledges [...] that the EBITDA margin is a more argin measure for medium term comparisons than the more short-term oriented gross margin (see Section 2.1.3.1). Indeed, using the EBITDA margin, the threshold elasticities range from [...] to [...], with a mean estimate of [...].
55 [...].
56 [...]
57 See Table 1 of the CRA report titled 'Response to Letter of Facts', dated 19 August 2019 [Doc Id: 3048].
58 […].
59 See paragraph (41),
60 See paragraph (41), […].
61 […].
62 […].
63 […].
64 […].
65 […].
66 Incidentally, this also explains why the Commission views the large price difference between Aluminium ABS and steel as corroborating evidence that steel and Aluminium ABS are not in the same market. Producing the same component with Aluminium ABS rather than steel is around […]% more expensive (Response, p. 11). There would be no reason to spend so much more on a component if it were not for the pressing need of OEMs to reduce weight to satisfy regulatory requirements. This high willingness to pay for Aluminium ABS components therefore suggests that, […], OEMs do not find it easy to seamlessly switch volumes to steel.
67 […].
68 […].
69 […].
70 […].
71 See SO paragraph 477.
72 Form CO, Figure 27.
73 See Figure 96 of the Decision, as well as more generally Section 8.3.9.5 of the Decision.
74 […].
75 […].
76 […].
77 […].
78 […].
79 […].
80 As explicitly mentioned in Figure 1 of the reply to RFI 32.
81 See Section 2.6 of Annex I of the SO.
82 […].
83 See Figure 96 of the Decision.
84 The Commission considers that the bidding nature of a market does not per se invalidate the informative value of market shares. In particular, where market shares do not fluctuate but remain very high for a significant number of years (as is the case here) market shares can and should be taken as a prima facie indication of market power. In this case, also capacity shares are informative about the capability of the different suppliers to successfully place volumes and place a competitive constraint on rivals.
85 The Commission notes however that even for basic industries products different suppliers may differ in terms of efficiency of their o serve specific customer needs. For instance, the Parties appear to particularly successful and skilled at serving customers' needs of 6xxx skin products.
86 For the Parties' rivals the figures are based on the Notifying Party's best estimates about rivals' nominated sales.
87 It appears that Aleris currently has […] only a minimal presence in 5xxx.
88 It appears that Aleris currently has […] only a minimal presence in 5xxx.
89 It appears that Aleris currently has […] only a minimal presence in 5xxx.
90 See discussion in Section 2.2.2.
91 The Notifying Party presented in the Form CO market shares based on actual sales in any given year of production (that is, if Novelis won a tender in 2016, for the production of 300kt split in equal amounts during 2018, 2019 and 2020, the methodology would attribute 100kt to each of 2018, 2019, and 2020).
92 The average lifetime of a project is circa 5-7 years. This means that OEMs re-tender out for the supply of Aluminium ABS for the production of a given car model [...] every 5-7 years. See Form CO paragraph 113.
93 The figures for capacity utilisation on average across all Aluminium ABS supplier slightly differ from the figures provided in the Statement of Objections because of a small calculation correction implemented by the Commission after the Statement of Objections.
94 [...]
95 Doc Id:2564
96 In this Annex, the Commission’s references to Ma’aden denote references to Alcoa/Ma’aden.
97 […].
98 Except for one competitor who indicated that in the long run they aim at using the EEA plant(s) for serving EEA customers. For this competitor, the share of exports is assumed to be zero as of 2019. This is conservative because it cannot be assumed with certainty that the competitor will achieve its forecast of having no exports outside the EEA as of 2019.
99 The latter is larger than final ABS production capacity.
100 […].
101 That is, the Commission has calculated the proportion of sales to ‘non-EEA or non-automotive customers’ (to be subtracted from the total CASH capacity to obtain the CASH capacity available for Aluminium ABS in the EEA) as the maximum between the ‘proportion of sales to non-EEA customers’ and the ‘proportion of sales to non-automotive customers’.
102 This effect is exacerbated by OEM’s desire to multi-source. As a result of such multi-sourcing, there is effectively less competitive capacity available to choose from than total spare capacity may seem to suggest, since OEMs have to award more than one supplier to ensure security of supply.
103 […].
104 […].
105 […].
106 […].
107 Section 2.4 and 2.5, as well as Appendix I.
108 Except for one competitor who indicated that in the long run they aim at using the EEA plant(s) for serving EEA customers. For this competitor, the share of exports is assumed to be zero as of 2019. This is conservative because it cannot be assumed with certainty that the competitor will achieve its forecast of having no exports outside the EEA as of 2019.
109 See for instance Figure 9 and Figure 10.
110 […].
111 Minutes of a call with a competitor on 23 May 2019, DocID2172.
112 Reply to request for information 18, ‘NOV-EU00440598.pptx’, DocID1298-57, reported in the Letter of Facts.
113 […].
114 See also Section 8.3.7.2 of the Decision.
115 […].
116 […].
117 […].
118 See Reply to Annex I of the SO, Section 3.1.2.
119 […].
120 In the minutes of 23 May 2019 the competitor indicated that: ‘The Company’s own EEA capacity is also used for North America until full ramp-up of its NA line […].The current forecast is that European plants will in the future supply European customers and North American plants North American customers’. Therefore, in its main scenario (‘Reconstruction EEA auto’), for this competitor the Commission has assumed that the share of exports is zero as of 2019. This is conservative because it cannot be assumed with certainty that the competitor will achieve its forecast of having no exports outside the EEA as of 2019.
121 These confidential minutes were made available in the data room that opened after the Commission issued the Letter of Facts.
122 Form CO annex ‘M.9076 - Annex 63 - Executive PMT - 20-09-2018.pdf’, slide 24.
123 See paragraph (161).
124 […].
125 […].
126 […].
127 […].
128 […].
129 […].
130 See Reply to Annex I of the SO, page 22.
131 See Reply to RFI 32, Q1e.
132 See Reply to Annex I of the SO, page 24.
133 […].
134 See Reply to request for information 40.
135 […].
136 […].
137 Process scrap refers to production sub-processes such as cutting and edge trimming. Process scrap also relates to production failures that may affect a supplier’s plant.
138 […].
139 […]
140 See Table 18.
141 Or even 0-5%, based on CASH capacity available for EEA customers of Aluminium ABS (Table 20).
142 See Table 18.
143 The Commission notes that the capacity shares for 2019 and 2020 slightly differ from the figures presented in the Letter of Facts due to a minor clerical error indicated by the Notifying Party during the data room that opened after the Commission issued the Letter of Facts. Specifically, the Parties combined share for 2019 has been changed from [50-60] to (50-60]% and the capacity share for 2020 has been changed from [50-60]% to [50-60]%.
144 See Section 3.3. of the Reply to Annex I of the SO.
145 See Reply to RFI 46.
146 See for instance the two slides attached to this email exchange between with European Aluminium and its members (Doc Id:1039- 12385 )
147 […].
148 […].
149 […].
150 […].
151 […]
152 […].
153 There are indeed many demand forecasts from Ducker, depending on (i) whether net or gross shipments are used (gross being the amount sent to the OEMs and net being the amount that ends up in a car) and (ii) whether sheet for vehicle content or ABS for vehicle content is used. The Notifying Party suggests the use of gross ABS shipments but as discussed in this paragraph the demand volumes resulting from this assumption are low even compared to the historically realised sales of Aluminium ABS.
154 See Section 3.2.
155 See for instance Figure 9 and Figure 10.
156 Or updated as per the additional material provided by the Parties in reply to subsequent RFIs (e.g. RFI 36 for nominated demand and RFI 31 for the Parties’ effective capacity).
157 Form CO paragraph 108.
158 Form CO paragraph 113.
159 Excel file 'Pivotality v3 - Data room xlsx', made available in the data room
160 Excel file 'Pivotality v3 - Data room xlsx', made available in the data room
161 […].
162 […].
163 […].
164 Excel file ‘Pivotality v1 - Data room xlsx’ made available in the data room.
165 […].
166 For example, see Daisuke Hirata (2009), ‘Asymmetric Bertrand-Edgeworth Oligopoly and Mergers’, B.E. Journal of Theoretical Economics, Vol. 9, No. 1, pp. 1935-1704. See also Case M.6471 Outukumpu/Inoxum (Commission decision of 7 November 2012).
167 […].
168 […].
169 […].
170 […].
171 […].
172 […].
173 […].
174 […].
175 Annex 4 of the Reply to the Article 6(1)(c) Decision.
176 That is, capacity expansions already planned. Indeed in the response to the Commission’s RFI for the market reconstruction competitors did not only indicate for future years the capacity currently installed but also the amounts of capacity that they are planning to add over the time period up to 2023 (included).
177 […].
178 […].
179 That is, excluding sales to non-ABS and non-EEA customers.
180 These figures were made available in the data room that opened after the Commission issued its Letter of Facts.
181 […].
182 With the exception of cases where tenders for start of production in 2020 are still open, as detailed in Sections 8.3.8.2 and 8.3.8.2 of the Decision. For these tenders, the Transaction could have an immediate effect due to the removal of Aleris as a competitor for upcoming tenders with start of production in the near future.
183 […].
184 […].
185 E.g., see Jean Tirole (1989), ‘The Theory of Industrial Organization’, MIT Press.
186 As an aside, the Commission also notes that the Notifying Party did not attempt to quantify the precise impact of the Transaction on prices either. To the extent that the Notifying Party argues that there would be no material price effects despite the large body of evidence collected by the Commission, it therefore fails to support this contention.
187 Discussed in Section 2.5.
188 For example, see David M. Kreps and Jose A. Scheinkman (1983), ‘Quantity Precommitment and Betrand Competition Yield Cournot Outcomes’, Bell Journal of Economics, Vol. 14, No. 2, pp. 326-337.
189 A 10% expansion by a larger supplier is different from a 10% expansion by a smaller supplier.
190 […].
191 In contrast, smaller players in less concentrated markets are too small to be able to stabilise market prices through their capacity choices.
192 In section 4.3 of the Reply to the quantitative part of the Letter of Facts, the Notifying Party introduced a new claim that demand uncertainty would not only undermine the Commission’s concerns relating to capacity expansions but also the Commission’s analysis
of Novelis’ pivotality. However, the Commission notes that the analysis presented in that section is based on an analysis of uncommitted capacity and non-nominated demand, which the Commission does not consider to be the correct approach for assessing the effect of the Transaction (see discussion in Section2.5.3).
193 See Section 2.6.1.
194 It simply suggests that some of the standard results obtained assuming no demand uncertainty may not necessarily apply when one assumed uncertainty.
195 In case capacity costs are high (as is arguably the case in the aluminium market), the paper by Young cited by the Notifying Party in fact suggests that uncertainty leads to the same outcome as in the absence of uncertainty (namely, Cournot competition, in which case mergers of parties with high market shares generate substantial anticompetitive effects).
196 E.g., see D.M. Kreps and J. Scheinkman (1983), ‘Quantity Precommitment and Bertrand Competition Yields Cournot Outcomes’, Bell Journal of Economics 14, pp. 326-337.
197 If anything, the claim at odds with reality is the Notifying Party claim that in a growing market such as Aluminium ABS Aleris (or its alternative purchaser) would not expand capacity.
198 […].
199 CRA report titled ‘Analysis of Novelis and Aleris bidding data’, dated 30 November 2018.
200 CRA report titled ‘Analysis of Novelis and Aleris updated bidding data’, dated 19 May 2019.
201 […].
202 Source: Novelis old and new bidding data.
203 Source: Aleris bidding data and reply to request for information 38.
204 CRA report titled ‘Analysis of Novelis and Aleris bidding data’, dated 30 November 2018.
205 […].
206 […].
207 […].
208 […].
209 […].
210 […].
211 For instance, in a 3 player market with firms A and B merging, a SIEC would very likely arise even if the diversion ratio between A and B is 25% and the diversion from A and B to C is 75%. What matters is the 25% diversion, irrespectively of the fact that it is C the closest competitor of the Parties and irrespectively of how high or low the diversion ratio of 25% may be compared to the market shares of the Parties.
212 Which is very likely considering that the manufacturing cycle is 5-7 years.
213 That is, the proportion of tenders won by one merging party, out of all tenders lost by the other merging party.
214 […].
215 […].
216 […].
217 […].
218 […].
219 The Commission considers the capacity figures from the market reconstruction to be preferable to the capacity estimates provided by the Parties in the Form CO because in its reconstruction the Commission collected data directly from each Aluminium ABS supplier.
220 […].
221 Or, as in the case of Novelis, […].
222 CRA report titled ‘Analysis of Novelis and Aleris bidding data’, dated 30 November 2018.
223 CRA report titled ‘Analysis of Novelis and Aleris updated bidding data’, dated 19 May 2019.
224 The only exception is […].
225 An econometric analysis is a statistical technique that allows to study the correlation between two variables (in this case the presence of Aleris and the prices, margins, and probability of winning of Novelis), controlling for a number of other factors potentially affecting the prices, margins, and probability of winning of Novelis (such as Novelis' costs, the type of product and the volumes being tendered out, the customer type, etc.).
226 CRA report titled ‘Analysis of Novelis and Aleris bidding data’, dated 30 November 2018.
227 […].
228 […].
229 […].
230 Source: Novelis old bidding data.
231 Original Equipment Manufacturer.
232 […].
233 Source: Novelis old bidding data.
234 […].
235 Source: Novelis old bidding data.
236 Especially in the email sent on 28 June 2019 at 11:48, the parties explained that […].
237 CRA report titled ‘Analysis of Novelis and Aleris updated bidding data’, dated 19 May 2019.
238 Using a fixed effect variable to identify whether the tender was won or lost by Novelis.
239 Source: Novelis old bidding data.
240 […].
241 Reply to request for information 47, question 9 and Annex Q-9, DocID2652.
242 […].
243 Especially in the email sent on 28 June 2019 at 11:48, the Notifying Party explained that the participation of Aleris was not defined at the level of individual components because it ‘would involve matching all parts that appear in both Parties’ datasets, which would be impracticable, given the different spellings involved which ruled out a mechanical matching’.
244 As explained in Section 2.3.
245 See for instance Figure 9 and Figure 10.
246 […].
247 Excel files ‘Pivotality v3 – Data room xlsx’, made available in the data room.
248 In the scenario after the Transaction ‘Novelis’ refers to the Merged Entity comprising Novelis and Aleris.
249 […].
250 Of course, any combination of (a) and (b) is possible, too.
251 By contrast, a change in would reflect a change in market size, e.g., because total car consumption grows or declines over time.
252 By comparison, a change in caused by changing car demand does not affect the elasticity of demand, as is independent of .
253 More specifically, this would require demand to be more convex at the current equilibrium point than an isoelastic demand curve. In mathematical terms, demand would therefore have to be ‘superconvex’. Such pathological demand forms (while theoretically possible) are routinely dismissed in the economic literature as practically implausible, among other because they imply that alternatives to some product A must become less attractive for customers of A if the price of A increases. E.g., see Jerry Hausman (2010), ‘2010 Merger Guidelines: Empirical Analysis’, Antitrust Source, October 2010, pp. 1-5.
254 Sub-section titled ‘Demand-side: Potential substitutability during design phase does not warrant finding of a combined market for Aluminium ABS and flat steel products used in automotive bodies’.